Prepare mqprio_dump() for RTNL avoidance.
Use RCU instead of RTNL, and no longer acquire each children spinlock.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Toke Høiland-Jørgensen <toke@toke.dk>
Link: https://patch.msgid.link/20260510091455.4039245-8-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Prepare mq_dump_common() for RTNL avoidance.
Use RCU instead of RTNL, and no longer acquire each children spinlock.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Toke Høiland-Jørgensen <toke@toke.dk>
Link: https://patch.msgid.link/20260510091455.4039245-7-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add qstats_backlog_sub() and qstats_backlog_add() helpers
and use them instead of open-coding them.
These helpers use WRITE_ONCE() to prevent store-tearing.
Also use WRITE_ONCE() in fq_reset() and qdisc_reset()
when sch->qstats.backlog is cleared.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Toke Høiland-Jørgensen <toke@toke.dk>
Link: https://patch.msgid.link/20260510091455.4039245-4-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Helpers to increment or decrement sch->q.qlen, with appropriate
WRITE_ONCE() to prevent store tearing.
Add other WRITE_ONCE() when sch->q.qlen is changed.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Toke Høiland-Jørgensen <toke@toke.dk>
Link: https://patch.msgid.link/20260510091455.4039245-3-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The protodown functionality allows user space to turn off the carrier of
a net device:
# ip link add name dummy1 up type dummy
# ip link add name macvlan1 up link dummy1 type macvlan mode bridge
# ip link set dev macvlan1 protodown on
$ ip -br link show dev macvlan1
macvlan1@dummy1 DOWN 0a:5c:a3:05:c7:86 <NO-CARRIER,BROADCAST,MULTICAST,UP>
Different applications can set different protodown reasons, which
prevents an application from turning on the carrier of a net device as
long as others want it down:
# ip link set dev macvlan1 protodown_reason 1 on
# ip link set dev macvlan1 protodown_reason 2 on
# ip link set dev macvlan1 protodown off
Error: Cannot clear protodown, active reasons.
# ip link set dev macvlan1 protodown_reason 2 off
# ip link set dev macvlan1 protodown off
Error: Cannot clear protodown, active reasons.
# ip link set dev macvlan1 protodown_reason 1 off
# ip link set dev macvlan1 protodown off
$ ip -br link show dev macvlan1
macvlan1@dummy1 UP 0a:5c:a3:05:c7:86 <BROADCAST,MULTICAST,UP,LOWER_UP>
Unfortunately, this mechanism is not very useful when the carrier of a
net device can be toggled by toggling the carrier of its lower device:
# ip link set dev macvlan1 protodown on
$ ip -br link show dev macvlan1
macvlan1@dummy1 DOWN 0a:5c:a3:05:c7:86 <NO-CARRIER,BROADCAST,MULTICAST,UP>
# ip link set dev dummy1 carrier off
# ip link set dev dummy1 carrier on
$ ip -br link show dev macvlan1
macvlan1@dummy1 UP 0a:5c:a3:05:c7:86 <BROADCAST,MULTICAST,UP,LOWER_UP>
Obviously, this is not the intended behavior and it is unlikely to be
relied on by anyone. In fact, it is a problem for applications like FRR
that use protodown with macvlan on top of a bridge as part of Virtual
Router Redundancy Protocol (VRRP).
Solve this by preventing a net device configured with protodown on from
gaining carrier by making netif_carrier_on() a NOP when protodown is
turned on.
Output with the patch:
# ip link add name dummy1 up type dummy
# ip link add name macvlan1 up link dummy1 type macvlan mode bridge
# ip link set dev macvlan1 protodown on
$ ip -br link show dev macvlan1
macvlan1@dummy1 DOWN 0a:5c:a3:05:c7:86 <NO-CARRIER,BROADCAST,MULTICAST,UP>
# ip link set dev dummy1 carrier off
# ip link set dev dummy1 carrier on
$ ip -br link show dev macvlan1
macvlan1@dummy1 DOWN 0a:5c:a3:05:c7:86 <NO-CARRIER,BROADCAST,MULTICAST,UP>
# ip link set dev macvlan1 protodown off
$ ip -br link show dev macvlan1
macvlan1@dummy1 UP 0a:5c:a3:05:c7:86 <BROADCAST,MULTICAST,UP,LOWER_UP>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Link: https://patch.msgid.link/20260507105906.891817-3-idosch@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
fq_codel_dump_class_stats() acquires qdisc spinlock only when requested
to follow flow->head chain.
As we did in sch_cake recently, add the missing READ_ONCE()/WRITE_ONCE()
annotations.
Fixes: edb09eb17e ("net: sched: do not acquire qdisc spinlock in qdisc/class stats dump")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20260504163842.1162001-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
"tc qdisc show ... handle xxx" filtering can be done by the kernel.
A followup patch can do the same for tcm_parent.
iproute2/tc needs a small companion patch.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20260503114515.2460477-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
We soon will no longer hold RTNL in qdisc dumps.
Add READ_ONCE()/WRITE_ONCE() annotations.
Note: taprio already uses RCU to protect most of its fields.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260501064247.2027688-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
cake_dump_class_stats() runs without qdisc spinlock being held.
In this first patch, I add READ_ONCE()/WRITE_ONCE() annotations for:
- flow->head
- flow->dropped
- b->backlogs[]
Fixes: 046f6fd5da ("sched: Add Common Applications Kept Enhanced (cake) qdisc")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Toke Høiland-Jørgensen <toke@toke.dk>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20260430061610.3503483-2-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When sfb has children (eg qfq qdisc) whose peek() callback is
qdisc_peek_dequeued(), we could get a kernel panic. When the parent of such
qdiscs (eg illustrated in patch #3 as tbf) wants to retrieve an skb from
its child (sfb in this case), it will do the following:
1a. do a peek() - and when sensing there's an skb the child can offer, then
- the child in this case(sfb) calls its child's (qfq) peek.
qfq does the right thing and will return the gso_skb queue packet.
Note: if there wasnt a gso_skb entry then qfq will store it there.
1b. invoke a dequeue() on the child (sfb). And herein lies the problem.
- sfb will call the child's dequeue() which will essentially just
try to grab something of qfq's queue.
[ 127.594489][ T453] KASAN: null-ptr-deref in range [0x0000000000000048-0x000000000000004f]
[ 127.594741][ T453] CPU: 2 UID: 0 PID: 453 Comm: ping Not tainted 7.1.0-rc1-00035-gac961974495b-dirty #793 PREEMPT(full)
[ 127.595059][ T453] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
[ 127.595254][ T453] RIP: 0010:qfq_dequeue+0x35c/0x1650 [sch_qfq]
[ 127.595461][ T453] Code: 00 fc ff df 80 3c 02 00 0f 85 17 0e 00 00 4c 8d 73 48 48 89 9d b8 02 00 00 48 b8 00 00 00 00 00 fc ff df 4c 89 f2 48 c1 ea 03 <80> 3c 02 00 0f 85 76 0c 00 00 48 b8 00 00 00 00 00 fc ff df 4c 8b
[ 127.596081][ T453] RSP: 0018:ffff88810e5af440 EFLAGS: 00010216
[ 127.596337][ T453] RAX: dffffc0000000000 RBX: 0000000000000000 RCX: dffffc0000000000
[ 127.596623][ T453] RDX: 0000000000000009 RSI: 0000001880000000 RDI: ffff888104fd82b0
[ 127.596917][ T453] RBP: ffff888104fd8000 R08: ffff888104fd8280 R09: 1ffff110211893a3
[ 127.597165][ T453] R10: 1ffff110211893a6 R11: 1ffff110211893a7 R12: 0000001880000000
[ 127.597404][ T453] R13: ffff888104fd82b8 R14: 0000000000000048 R15: 0000000040000000
[ 127.597644][ T453] FS: 00007fc380cbfc40(0000) GS:ffff88816f2a8000(0000) knlGS:0000000000000000
[ 127.597956][ T453] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 127.598160][ T453] CR2: 00005610aa9890a8 CR3: 000000010369e000 CR4: 0000000000750ef0
[ 127.598390][ T453] PKRU: 55555554
[ 127.598509][ T453] Call Trace:
[ 127.598629][ T453] <TASK>
[ 127.598718][ T453] ? mark_held_locks+0x40/0x70
[ 127.598890][ T453] ? srso_alias_return_thunk+0x5/0xfbef5
[ 127.599053][ T453] sfb_dequeue+0x88/0x4d0
[ 127.599174][ T453] ? ktime_get+0x137/0x230
[ 127.599328][ T453] ? srso_alias_return_thunk+0x5/0xfbef5
[ 127.599480][ T453] ? qdisc_peek_dequeued+0x7b/0x350 [sch_qfq]
[ 127.599670][ T453] ? srso_alias_return_thunk+0x5/0xfbef5
[ 127.599831][ T453] tbf_dequeue+0x6b1/0x1098 [sch_tbf]
[ 127.599988][ T453] __qdisc_run+0x169/0x1900
The right thing to do in #1b is to grab the skb off gso_skb queue.
This patchset fixes that issue by changing #1b to use qdisc_dequeue_peeked()
method instead.
Fixes: e13e02a3c6 ("net_sched: SFB flow scheduler")
Signed-off-by: Victor Nogueria <victor@mojatatu.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260430152957.194015-3-jhs@mojatatu.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When red qdisc has children (eg qfq qdisc) whose peek() callback is
qdisc_peek_dequeued(), we could get a kernel panic. When the parent of such
qdiscs (eg illustrated in patch #3 as tbf) wants to retrieve an skb from
its child (red in this case), it will do the following:
1a. do a peek() - and when sensing there's an skb the child can offer, then
- the child in this case(red) calls its child's (qfq) peek.
qfq does the right thing and will return the gso_skb queue packet.
Note: if there wasnt a gso_skb entry then qfq will store it there.
1b. invoke a dequeue() on the child (red). And herein lies the problem.
- red will call the child's dequeue() which will essentially just
try to grab something of qfq's queue.
[ 78.667668][ T363] KASAN: null-ptr-deref in range [0x0000000000000048-0x000000000000004f]
[ 78.667927][ T363] CPU: 1 UID: 0 PID: 363 Comm: ping Not tainted 7.1.0-rc1-00033-g46f74a3f7d57-dirty #790 PREEMPT(full)
[ 78.668263][ T363] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
[ 78.668486][ T363] RIP: 0010:qfq_dequeue+0x446/0xc90 [sch_qfq]
[ 78.668718][ T363] Code: 54 c0 e8 dd 90 00 f1 48 c7 c7 e0 03 54 c0 48 89 de e8 ce 90 00 f1 48 8d 7b 48 b8 ff ff 37 00 48 89 fa 48 c1 e0 2a 48 c1 ea 03 <80> 3c 02 00 74 05 e8 ef a1 e1 f1 48 8b 7b 48 48 8d 54 24 58 48 8d
[ 78.669312][ T363] RSP: 0018:ffff88810de573e0 EFLAGS: 00010216
[ 78.669533][ T363] RAX: dffffc0000000000 RBX: 0000000000000000 RCX: 0000000000000000
[ 78.669790][ T363] RDX: 0000000000000009 RSI: 0000000000000004 RDI: 0000000000000048
[ 78.670044][ T363] RBP: ffff888110dc4000 R08: ffffffffb1b0885a R09: fffffbfff6ba9078
[ 78.670297][ T363] R10: 0000000000000003 R11: ffff888110e31c80 R12: 0000001880000000
[ 78.670560][ T363] R13: ffff888110dc4150 R14: ffff888110dc42b8 R15: 0000000000000200
[ 78.670814][ T363] FS: 00007f66a8f09c40(0000) GS:ffff888163428000(0000) knlGS:0000000000000000
[ 78.671110][ T363] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 78.671324][ T363] CR2: 000055db4c6a30a8 CR3: 000000010da67000 CR4: 0000000000750ef0
[ 78.671585][ T363] PKRU: 55555554
[ 78.671713][ T363] Call Trace:
[ 78.671843][ T363] <TASK>
[ 78.671936][ T363] ? __pfx_qfq_dequeue+0x10/0x10 [sch_qfq]
[ 78.672148][ T363] ? __pfx__printk+0x10/0x10
[ 78.672322][ T363] ? srso_alias_return_thunk+0x5/0xfbef5
[ 78.672496][ T363] ? lockdep_hardirqs_on_prepare+0xa8/0x1a0
[ 78.672706][ T363] ? srso_alias_return_thunk+0x5/0xfbef5
[ 78.672875][ T363] ? trace_hardirqs_on+0x19/0x1a0
[ 78.673047][ T363] red_dequeue+0x65/0x270 [sch_red]
[ 78.673217][ T363] ? srso_alias_return_thunk+0x5/0xfbef5
[ 78.673385][ T363] tbf_dequeue.cold+0xb0/0x70c [sch_tbf]
[ 78.673566][ T363] __qdisc_run+0x169/0x1900
The right thing to do in #1b is to grab the skb off gso_skb queue.
This patchset fixes that issue by changing #1b to use qdisc_dequeue_peeked()
method instead.
Fixes: 77be155cba ("pkt_sched: Add peek emulation for non-work-conserving qdiscs.")
Reported-by: Manas <ghandatmanas@gmail.com>
Reported-by: Rakshit Awasthi <rakshitawasthi17@gmail.com>
Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260430152957.194015-2-jhs@mojatatu.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
My prior patch missed few READ_ONCE()/WRITE_ONCE() annotations.
Fixes: 5154561d9b ("net/sched: sch_pie: annotate data-races in pie_dump_stats()")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260430080056.35104-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
There is no point dumping qdiscs for all devices when user space
wants them for a single device:
tc -s -d qdisc show dev eth1
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260430023628.3216283-5-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Use for_each_netdev_dump() instead of for_each_netdev().
This is more scalable, and will ease RCU conversion.
This also offer better behavior when other threads
are adding or deleting netevices concurrently.
This enables dumping qdiscs for a single device
at user space request in the following patch.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260430023628.3216283-4-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Change tc_fill_qdisc() to return -EMSGSIZE when skb is too small.
Change tc_dump_qdisc_root() to propagate tc_fill_qdisc() error to its callers.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260430023628.3216283-3-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Change tc_fill_tclass() to return -EMSGSIZE when skb is too small.
Change its caller to propagate this error (instead of -EINVAL)
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260430023628.3216283-2-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
qstats_overlimit_inc() is only used to increment per cpu overlimits.
It can use this_cpu_inc() to avoid this_cpu_ptr() extra cost
and avoid potential store tearing.
Change qstats_overlimit_inc() name and its argument type.
Also add a WRITE_ONCE() in qdisc_qstats_overlimit() to prevent
store tearing.
$ scripts/bloat-o-meter -t vmlinux.0 vmlinux.1
add/remove: 0/0 grow/shrink: 0/7 up/down: 0/-91 (-91)
Function old new delta
tcf_skbmod_act 772 764 -8
tcf_police_act 733 725 -8
tcf_gate_act 318 310 -8
tcf_pedit_act 1295 1284 -11
tcf_mirred_to_dev 1126 1114 -12
tcf_ife_act 1077 1061 -16
tcf_mirred_act 1324 1296 -28
Total: Before=24274627, After=24274536, chg -0.00%
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20260428070919.3109557-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
cake_dump_stats() runs without qdisc spinlock being held.
In this final patch, I add READ_ONCE()/WRITE_ONCE() annotations
for cparams.target and cparams.interval.
Fixes: 046f6fd5da ("sched: Add Common Applications Kept Enhanced (cake) qdisc")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: "Toke Høiland-Jørgensen" <toke@toke.dk>
Link: https://patch.msgid.link/20260427083606.459355-6-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
cake_dump_stats() runs without qdisc spinlock being held.
In this fourth patch, I add READ_ONCE()/WRITE_ONCE() annotations
for the following fields:
- avg_peak_bandwidth
- buffer_limit
- buffer_max_used
- avg_netoff
- max_netlen
- max_adjlen
- min_netlen
- min_adjlen
- active_queues
- tin_rate_bps
- bytes
- tin_backlog
Other annotations are added in following patch, to ease code review.
Fixes: 046f6fd5da ("sched: Add Common Applications Kept Enhanced (cake) qdisc")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Toke Høiland-Jørgensen <toke@toke.dk>
Link: https://patch.msgid.link/20260427083606.459355-5-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
cake_dump_stats() runs without qdisc spinlock being held.
In this third patch, I add READ_ONCE()/WRITE_ONCE() annotations
for the following fields:
- packets
- tin_dropped
- tin_ecn_mark
- ack_drops
- peak_delay
- avge_delay
- base_delay
Other annotations are added in following patches, to ease code review.
Fixes: 046f6fd5da ("sched: Add Common Applications Kept Enhanced (cake) qdisc")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: "Toke Høiland-Jørgensen" <toke@toke.dk>
Link: https://patch.msgid.link/20260427083606.459355-4-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
cake_dump_stats() runs without qdisc spinlock being held.
In this second patch, I add READ_ONCE()/WRITE_ONCE() annotations
for the following fields:
- bulk_flow_count
- unresponsive_flow_count
- max_skblen
- flow_quantum
Other annotations are added in following patches, to ease code review.
Fixes: 046f6fd5da ("sched: Add Common Applications Kept Enhanced (cake) qdisc")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: "Toke Høiland-Jørgensen" <toke@toke.dk>
Link: https://patch.msgid.link/20260427083606.459355-3-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
cake_dump_stats() runs without qdisc spinlock being held.
In this first patch, I add READ_ONCE()/WRITE_ONCE() annotations
for the following fields:
- way_hits
- way_misses
- way_collisions
- sparse_flow_count
- decaying_flow_count
Other annotations are added in following patches, to ease code review.
Fixes: 046f6fd5da ("sched: Add Common Applications Kept Enhanced (cake) qdisc")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: "Toke Høiland-Jørgensen" <toke@toke.dk>
Link: https://patch.msgid.link/20260427083606.459355-2-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Currently, mctp_i2c_get_tx_flow_state() is called before the packet length
sanity check. This function marks a new flow as active in the MCTP core.
If the sanity check fails, mctp_i2c_xmit() returns early without calling
mctp_i2c_lock_nest(). This results in a mismatched locking state: the
flow is active, but the I2C bus lock was never acquired for it.
When the flow is later released, mctp_i2c_release_flow() will see the
active state and queue an unlock marker. The TX thread will then
decrement midev->i2c_lock_count from 0, causing it to underflow to -1.
This underflow permanently breaks the driver's locking logic, allowing
future transmissions to occur without holding the I2C bus lock, leading
to bus collisions and potential hardware hangs.
Move the mctp_i2c_get_tx_flow_state() call to after the length sanity
check to ensure we only transition the flow state if we are actually
going to proceed with the transmission and locking.
Fixes: f5b8abf9fc ("mctp i2c: MCTP I2C binding driver")
Signed-off-by: William A. Kennington III <william@wkennington.com>
Acked-by: Jeremy Kerr <jk@codeconstruct.com.au>
Link: https://patch.msgid.link/20260423074741.201460-1-william@wkennington.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
When a TAPRIO child qdisc is deleted via RTM_DELQDISC, taprio_graft()
is called with new == NULL and stores NULL into q->qdiscs[cl - 1].
Subsequent RTM_GETTCLASS dump operations walk all classes via
taprio_walk() and call taprio_dump_class(), which calls taprio_leaf()
returning the NULL pointer, then dereferences it to read child->handle,
causing a kernel NULL pointer dereference.
The bug is reachable with namespace-scoped CAP_NET_ADMIN on any kernel
with CONFIG_NET_SCH_TAPRIO enabled. On systems with unprivileged user
namespaces enabled, an unprivileged local user can trigger a kernel
panic by creating a taprio qdisc inside a new network namespace,
grafting an explicit child qdisc, deleting it, and requesting a class
dump. The RTM_GETTCLASS dump itself requires no capability.
Oops: general protection fault, probably for non-canonical address 0xdffffc0000000007: 0000 [#1] SMP KASAN NOPTI
KASAN: null-ptr-deref in range [0x0000000000000038-0x000000000000003f]
RIP: 0010:taprio_dump_class (net/sched/sch_taprio.c:2478)
Call Trace:
<TASK>
tc_fill_tclass (net/sched/sch_api.c:1966)
qdisc_class_dump (net/sched/sch_api.c:2326)
taprio_walk (net/sched/sch_taprio.c:2514)
tc_dump_tclass_qdisc (net/sched/sch_api.c:2352)
tc_dump_tclass_root (net/sched/sch_api.c:2370)
tc_dump_tclass (net/sched/sch_api.c:2431)
rtnl_dumpit (net/core/rtnetlink.c:6864)
netlink_dump (net/netlink/af_netlink.c:2325)
rtnetlink_rcv_msg (net/core/rtnetlink.c:6959)
netlink_rcv_skb (net/netlink/af_netlink.c:2550)
</TASK>
Fix this by substituting &noop_qdisc when new is NULL in
taprio_graft(), a common pattern used by other qdiscs (e.g.,
multiq_graft()) to ensure the q->qdiscs[] slots are never NULL.
This makes control-plane dump paths safe without requiring individual
NULL checks.
Since the data-plane paths (taprio_enqueue and taprio_dequeue_from_txq)
previously had explicit NULL guards that would drop/skip the packet
cleanly, update those checks to test for &noop_qdisc instead. Without
this, packets would reach taprio_enqueue_one() which increments the root
qdisc's qlen and backlog before calling the child's enqueue; noop_qdisc
drops the packet but those counters are never rolled back, permanently
inflating the root qdisc's statistics.
After this change *old can be a valid qdisc, NULL, or &noop_qdisc.
Only call qdisc_put(*old) in the first case to avoid decreasing
noop_qdisc's refcount, which was never increased.
Fixes: 665338b2a7 ("net/sched: taprio: dump class stats for the actual q->qdiscs[]")
Reported-by: Xiang Mei <xmei5@asu.edu>
Signed-off-by: Weiming Shi <bestswngs@gmail.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Tested-by: Weiming Shi <bestswngs@gmail.com>
Link: https://patch.msgid.link/20260422161958.2517539-3-bestswngs@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
fq_codel_dump_stats() acquires the qdisc spinlock a bit too late.
Move this acquisition before we fill tc_fq_pie_xstats with live data.
Alternative would be to add READ_ONCE() and WRITE_ONCE() annotations,
but the spinlock is needed anyway to scan q->new_flows and q->old_flows.
Fixes: ec97ecf1eb ("net: sched: add Flow Queue PIE packet scheduler")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20260423063527.2568262-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
choke_dump_stats() only runs with RTNL held.
It reads fields that can be changed in qdisc fast path.
Add READ_ONCE()/WRITE_ONCE() annotations.
Fixes: edb09eb17e ("net: sched: do not acquire qdisc spinlock in qdisc/class stats dump")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20260423062839.2524324-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reject requests with negative latency or jitter.
A negative value added to current timestamp (u64) wraps
to an enormous time_to_send, disabling dequeue.
The original UAPI used u32 for these values; the conversion to 64-bit
time values via TCA_NETEM_LATENCY64 and TCA_NETEM_JITTER64
allowed signed values to reach the kernel without validation.
Jitter is already silently clamped by an abs() in netem_change();
that abs() can be removed in a follow-up once this rejection is in
place.
Fixes: 99803171ef ("netem: add uapi to express delay and jitter in nanoseconds")
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20260418032027.900913-7-stephen@networkplumber.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
get_slot_next() computes a random delay between min_delay and
max_delay using:
get_random_u32() * (max_delay - min_delay) >> 32
This overflows signed 64-bit arithmetic when the delay range exceeds
approximately 2.1 seconds (2^31 nanoseconds), producing a negative
result that effectively disables slot-based pacing. This is a
realistic configuration for WAN emulation (e.g., slot 1s 5s).
Use mul_u64_u32_shr() which handles the widening multiply without
overflow.
Fixes: 0a9fe5c375 ("netem: slotting with non-uniform distribution")
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20260418032027.900913-6-stephen@networkplumber.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reject slot configurations that have no defensible meaning:
- negative min_delay or max_delay
- min_delay greater than max_delay
- negative dist_delay or dist_jitter
- negative max_packets or max_bytes
Negative or out-of-order delays underflow in get_slot_next(),
producing garbage intervals. Negative limits trip the per-slot
accounting (packets_left/bytes_left <= 0) on the first packet of
every slot, defeating the rate-limiting half of the slot feature.
Note that dist_jitter has been silently coerced to its absolute
value by get_slot() since the feature was introduced; rejecting
negatives here converts that silent coercion into -EINVAL. The
abs() can be removed in a follow-up.
Fixes: 836af83b54 ("netem: support delivering packets in delayed time slots")
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20260418032027.900913-5-stephen@networkplumber.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
netem_change() unconditionally reseeds the PRNG on every tc change
command. If TCA_NETEM_PRNG_SEED is not specified, a new random seed
is generated, destroying reproducibility for users who set a
deterministic seed on a previous change.
Move the initial random seed generation to netem_init() and only
reseed in netem_change() when TCA_NETEM_PRNG_SEED is explicitly
provided by the user.
Fixes: 4072d97ddc ("netem: add prng attribute to netem_sched_data")
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20260418032027.900913-4-stephen@networkplumber.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The queue limit check in netem_enqueue() uses q->t_len which only
counts packets in the internal tfifo. Packets placed in sch->q by
the reorder path (__qdisc_enqueue_head) are not counted, allowing
the total queue occupancy to exceed sch->limit under reordering.
Include sch->q.qlen in the limit check.
Fixes: f8d4bc4550 ("net/sched: netem: account for backlog updates from child qdisc")
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20260418032027.900913-3-stephen@networkplumber.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The 4-state Markov chain in loss_4state() has gaps at the boundaries
between transition probability ranges. The comparisons use:
if (rnd < a4)
else if (a4 < rnd && rnd < a1 + a4)
When rnd equals a boundary value exactly, neither branch matches and
no state transition occurs. The redundant lower-bound check (a4 < rnd)
is already implied by being in the else branch.
Remove the unnecessary lower-bound comparisons so the ranges are
contiguous and every random value produces a transition, matching
the GI (General and Intuitive) loss model specification.
This bug goes back to original implementation of this model.
Fixes: 661b79725f ("netem: revised correlated loss generator")
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20260418032027.900913-2-stephen@networkplumber.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
sfb_dump_stats() only runs with RTNL held,
reading fields that can be changed in qdisc fast path.
Add READ_ONCE()/WRITE_ONCE() annotations.
Alternative would be to acquire the qdisc spinlock, but our long-term
goal is to make qdisc dump operations lockless as much as we can.
tc_sfb_xstats fields don't need to be latched atomically,
otherwise this bug would have been caught earlier.
Fixes: edb09eb17e ("net: sched: do not acquire qdisc spinlock in qdisc/class stats dump")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260421141655.3953721-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
red_dump_stats() only runs with RTNL held,
reading fields that can be changed in qdisc fast path.
Add READ_ONCE()/WRITE_ONCE() annotations.
Alternative would be to acquire the qdisc spinlock, but our long-term
goal is to make qdisc dump operations lockless as much as we can.
tc_red_xstats fields don't need to be latched atomically,
otherwise this bug would have been caught earlier.
Fixes: edb09eb17e ("net: sched: do not acquire qdisc spinlock in qdisc/class stats dump")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20260421142309.3964322-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
fq_codel_dump_stats() acquires the qdisc spinlock a bit too late.
Move this acquisition before we fill st.qdisc_stats with live data.
Fixes: edb09eb17e ("net: sched: do not acquire qdisc spinlock in qdisc/class stats dump")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20260421142509.3967231-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
pie_dump_stats() only runs with RTNL held,
reading fields that can be changed in qdisc fast path.
Add READ_ONCE()/WRITE_ONCE() annotations.
Alternative would be to acquire the qdisc spinlock, but our long-term
goal is to make qdisc dump operations lockless as much as we can.
tc_pie_xstats fields don't need to be latched atomically,
otherwise this bug would have been caught earlier.
Fixes: edb09eb17e ("net: sched: do not acquire qdisc spinlock in qdisc/class stats dump")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20260421142944.4009941-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
hhf_dump_stats() only runs with RTNL held,
reading fields that can be changed in qdisc fast path.
Add READ_ONCE()/WRITE_ONCE() annotations.
Fixes: edb09eb17e ("net: sched: do not acquire qdisc spinlock in qdisc/class stats dump")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20260421143349.4052215-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Fix dualpi2_change() to correctly enforce updated limit and memlimit
values after a configuration change of the dualpi2 qdisc.
Before this patch, dualpi2_change() always attempted to dequeue packets
via the root qdisc (C-queue) when reducing backlog or memory usage, and
unconditionally assumed that a valid skb will be returned. When traffic
classification results in packets being queued in the L-queue while the
C-queue is empty, this leads to a NULL skb dereference during limit or
memlimit enforcement.
This is fixed by first dequeuing from the C-queue path if it is
non-empty. Once the C-queue is empty, packets are dequeued directly from
the L-queue. Return values from qdisc_dequeue_internal() are checked for
both queues. When dequeuing from the L-queue, the parent qdisc qlen and
backlog counters are updated explicitly to keep overall qdisc statistics
consistent.
Fixes: 320d031ad6 ("sched: Struct definition and parsing of dualpi2 qdisc")
Reported-by: "Kito Xu (veritas501)" <hxzene@gmail.com>
Closes: https://lore.kernel.org/netdev/20260413075740.2234828-1-hxzene@gmail.com/
Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
Link: https://patch.msgid.link/20260417152551.71648-1-chia-yu.chang@nokia-bell-labs.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
In advance_sched(), when should_change_schedules() returns true,
switch_schedules() is called to promote the admin schedule to oper.
switch_schedules() queues the old oper schedule for RCU freeing via
call_rcu(), but 'next' still points into an entry of the old oper
schedule. The subsequent 'next->end_time = end_time' and
rcu_assign_pointer(q->current_entry, next) are use-after-free.
Fix this by selecting 'next' from the new oper schedule immediately
after switch_schedules(), and using its pre-calculated end_time.
setup_first_end_time() sets the first entry's end_time to
base_time + interval when the schedule is installed, so the value
is already correct.
The deleted 'end_time = sched_base_time(admin)' assignment was also
harmful independently: it would overwrite the new first entry's
pre-calculated end_time with just base_time.
Fixes: a3d43c0d56 ("taprio: Add support adding an admin schedule")
Reported-by: Junxi Qian <qjx1298677004@gmail.com>
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
cake_update_flowkeys() is supposed to update the flow dissector keys
with the NAT-translated addresses and ports from conntrack, so that
CAKE's per-flow fairness correctly identifies post-NAT flows as
belonging to the same connection.
For the source port, this works correctly:
keys->ports.src = port;
But for the destination port, the assignment is reversed:
port = keys->ports.dst;
This means the NAT destination port is never updated in the flow keys.
As a result, when multiple connections are NATed to the same destination,
CAKE treats them as separate flows because the original (pre-NAT)
destination ports differ. This breaks CAKE's NAT-aware flow isolation
when using the "nat" mode.
The bug was introduced in commit b0c19ed608 ("sch_cake: Take advantage
of skb->hash where appropriate") which refactored the original direct
assignment into a compare-and-conditionally-update pattern, but wrote
the destination port update backwards.
Fix by reversing the assignment direction to match the source port
pattern.
Fixes: b0c19ed608 ("sch_cake: Take advantage of skb->hash where appropriate")
Signed-off-by: Dudu Lu <phx0fer@gmail.com>
Acked-by: Toke Høiland-Jørgensen <toke@toke.dk>
Link: https://patch.msgid.link/20260413110041.44704-1-phx0fer@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
In tcf_blockcast_redir(), when iterating block ports to redirect
packets to multiple devices, the mac_header_xmit flag is queried
from the wrong device. The loop sends to dev_prev but queries
dev_is_mac_header_xmit(dev) — which is the NEXT device in the
iteration, not the one being sent to.
This causes tcf_mirred_to_dev() to make incorrect decisions about
whether to push or pull the MAC header. When the block contains
mixed device types (e.g., an ethernet veth and a tunnel device),
intermediate devices get the wrong mac_header_xmit flag, leading to
skb header corruption. In the worst case, skb_push_rcsum with an
incorrect mac_len can exhaust headroom and panic.
The last device in the loop is handled correctly (line 365-366 uses
dev_is_mac_header_xmit(dev_prev)), confirming this is a copy-paste
oversight for the intermediate devices.
Fix by using dev_prev instead of dev for the mac_header_xmit query,
consistent with the device actually being sent to.
Fixes: 42f39036cd ("net/sched: act_mirred: Allow mirred to block")
Signed-off-by: Dudu Lu <phx0fer@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20260413084927.71353-1-phx0fer@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Merge in late fixes in preparation for the net-next PR.
Conflicts:
include/net/sch_generic.h
a6bd339dbb ("net_sched: fix skb memory leak in deferred qdisc drops")
ff2998f29f ("net: sched: introduce qdisc-specific drop reason tracing")
https://lore.kernel.org/adz0iX85FHMz0HdO@sirena.org.uk
drivers/net/ethernet/airoha/airoha_eth.c
1acdfbdb51 ("net: airoha: Fix VIP configuration for AN7583 SoC")
bf3471e6e6 ("net: airoha: Make flow control source port mapping dependent on nbq parameter")
Adjacent changes:
drivers/net/ethernet/airoha/airoha_ppe.c
f44218cd5e ("net: airoha: Reset PPE cpu port configuration in airoha_ppe_hw_init()")
7da62262ec ("inet: add ip_local_port_step_width sysctl to improve port usage distribution")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When looking up a flow table in act_ct in tcf_ct_flow_table_get(),
rhashtable_lookup_fast() internally opens and closes an RCU read critical
section before returning ct_ft.
The tcf_ct_flow_table_cleanup_work() can complete before refcount_inc_not_zero()
is invoked on the returned ct_ft resulting in a UAF on the already freed ct_ft
object. This vulnerability can lead to privilege escalation.
Analysis from zdi-disclosures@trendmicro.com:
When initializing act_ct, tcf_ct_init() is called, which internally triggers
tcf_ct_flow_table_get().
static int tcf_ct_flow_table_get(struct net *net, struct tcf_ct_params *params)
{
struct zones_ht_key key = { .net = net, .zone = params->zone };
struct tcf_ct_flow_table *ct_ft;
int err = -ENOMEM;
mutex_lock(&zones_mutex);
ct_ft = rhashtable_lookup_fast(&zones_ht, &key, zones_params); // [1]
if (ct_ft && refcount_inc_not_zero(&ct_ft->ref)) // [2]
goto out_unlock;
...
}
static __always_inline void *rhashtable_lookup_fast(
struct rhashtable *ht, const void *key,
const struct rhashtable_params params)
{
void *obj;
rcu_read_lock();
obj = rhashtable_lookup(ht, key, params);
rcu_read_unlock();
return obj;
}
At [1], rhashtable_lookup_fast() looks up and returns the corresponding ct_ft
from zones_ht . The lookup is performed within an RCU read critical section
through rcu_read_lock() / rcu_read_unlock(), which prevents the object from
being freed. However, at the point of function return, rcu_read_unlock() has
already been called, and there is nothing preventing ct_ft from being freed
before reaching refcount_inc_not_zero(&ct_ft->ref) at [2]. This interval becomes
the race window, during which ct_ft can be freed.
Free Process:
tcf_ct_flow_table_put() is executed through the path tcf_ct_cleanup() call_rcu()
tcf_ct_params_free_rcu() tcf_ct_params_free() tcf_ct_flow_table_put().
static void tcf_ct_flow_table_put(struct tcf_ct_flow_table *ct_ft)
{
if (refcount_dec_and_test(&ct_ft->ref)) {
rhashtable_remove_fast(&zones_ht, &ct_ft->node, zones_params);
INIT_RCU_WORK(&ct_ft->rwork, tcf_ct_flow_table_cleanup_work); // [3]
queue_rcu_work(act_ct_wq, &ct_ft->rwork);
}
}
At [3], tcf_ct_flow_table_cleanup_work() is scheduled as RCU work
static void tcf_ct_flow_table_cleanup_work(struct work_struct *work)
{
struct tcf_ct_flow_table *ct_ft;
struct flow_block *block;
ct_ft = container_of(to_rcu_work(work), struct tcf_ct_flow_table,
rwork);
nf_flow_table_free(&ct_ft->nf_ft);
block = &ct_ft->nf_ft.flow_block;
down_write(&ct_ft->nf_ft.flow_block_lock);
WARN_ON(!list_empty(&block->cb_list));
up_write(&ct_ft->nf_ft.flow_block_lock);
kfree(ct_ft); // [4]
module_put(THIS_MODULE);
}
tcf_ct_flow_table_cleanup_work() frees ct_ft at [4]. When this function executes
between [1] and [2], UAF occurs.
This race condition has a very short race window, making it generally
difficult to trigger. Therefore, to trigger the vulnerability an msleep(100) was
inserted after[1]
Fixes: 138470a9b2 ("net/sched: act_ct: fix lockdep splat in tcf_ct_flow_table_get")
Reported-by: zdi-disclosures@trendmicro.com
Tested-by: Victor Nogueira <victor@mojatatu.com>
Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20260410111627.46611-1-jhs@mojatatu.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Like pointed out by Sashiko [1], since commit ed76f5edcc ("net: sched:
protect filter_chain list with filter_chain_lock mutex") TC filters are
added to a shared block and published to datapath before their ->change()
function is called. This is a problem for cls_fw: an invalid filter
created with the "old" method can still classify some packets before it
is destroyed by the validation logic added by Xiang.
Therefore, insisting with repeated runs of the following script:
# ip link add dev crash0 type dummy
# ip link set dev crash0 up
# mausezahn crash0 -c 100000 -P 10 \
> -A 4.3.2.1 -B 1.2.3.4 -t udp "dp=1234" -q &
# sleep 1
# tc qdisc add dev crash0 egress_block 1 clsact
# tc filter add block 1 protocol ip prio 1 matchall \
> action skbedit mark 65536 continue
# tc filter add block 1 protocol ip prio 2 fw
# ip link del dev crash0
can still make fw_classify() hit the WARN_ON() in [2]:
WARNING: ./include/net/pkt_cls.h:88 at fw_classify+0x244/0x250 [cls_fw], CPU#18: mausezahn/1399
Modules linked in: cls_fw(E) act_skbedit(E)
CPU: 18 UID: 0 PID: 1399 Comm: mausezahn Tainted: G E 7.0.0-rc6-virtme #17 PREEMPT(full)
Tainted: [E]=UNSIGNED_MODULE
Hardware name: Red Hat KVM, BIOS 1.16.3-2.el9 04/01/2014
RIP: 0010:fw_classify+0x244/0x250 [cls_fw]
Code: 5c 49 c7 45 00 00 00 00 00 41 5d 41 5e 41 5f 5d c3 cc cc cc cc 5b b8 ff ff ff ff 41 5c 41 5d 41 5e 41 5f 5d c3 cc cc cc cc 90 <0f> 0b 90 eb a0 0f 1f 80 00 00 00 00 90 90 90 90 90 90 90 90 90 90
RSP: 0018:ffffd1b7026bf8a8 EFLAGS: 00010202
RAX: ffff8c5ac9c60800 RBX: ffff8c5ac99322c0 RCX: 0000000000000004
RDX: 0000000000000001 RSI: ffff8c5b74d7a000 RDI: ffff8c5ac8284f40
RBP: ffffd1b7026bf8d0 R08: 0000000000000000 R09: ffffd1b7026bf9b0
R10: 00000000ffffffff R11: 0000000000000000 R12: 0000000000010000
R13: ffffd1b7026bf930 R14: ffff8c5ac8284f40 R15: 0000000000000000
FS: 00007fca40c37740(0000) GS:ffff8c5b74d7a000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007fca40e822a0 CR3: 0000000005ca0001 CR4: 0000000000172ef0
Call Trace:
<TASK>
tcf_classify+0x17d/0x5c0
tc_run+0x9d/0x150
__dev_queue_xmit+0x2ab/0x14d0
ip_finish_output2+0x340/0x8f0
ip_output+0xa4/0x250
raw_sendmsg+0x147d/0x14b0
__sys_sendto+0x1cc/0x1f0
__x64_sys_sendto+0x24/0x30
do_syscall_64+0x126/0xf80
entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7fca40e822ba
Code: d8 64 89 02 48 c7 c0 ff ff ff ff eb b8 0f 1f 00 f3 0f 1e fa 41 89 ca 64 8b 04 25 18 00 00 00 85 c0 75 15 b8 2c 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 7e c3 0f 1f 44 00 00 41 54 48 83 ec 30 44 89
RSP: 002b:00007ffc248a42c8 EFLAGS: 00000246 ORIG_RAX: 000000000000002c
RAX: ffffffffffffffda RBX: 000055ef233289d0 RCX: 00007fca40e822ba
RDX: 000000000000001e RSI: 000055ef23328c30 RDI: 0000000000000003
RBP: 000055ef233289d0 R08: 00007ffc248a42d0 R09: 0000000000000010
R10: 0000000000000000 R11: 0000000000000246 R12: 000000000000001e
R13: 00000000000186a0 R14: 0000000000000000 R15: 00007fca41043000
</TASK>
irq event stamp: 1045778
hardirqs last enabled at (1045784): [<ffffffff864ec042>] __up_console_sem+0x52/0x60
hardirqs last disabled at (1045789): [<ffffffff864ec027>] __up_console_sem+0x37/0x60
softirqs last enabled at (1045426): [<ffffffff874d48c7>] __alloc_skb+0x207/0x260
softirqs last disabled at (1045434): [<ffffffff874fe8f8>] __dev_queue_xmit+0x78/0x14d0
Then, because of the value in the packet's mark, dereference on 'q->handle'
with NULL 'q' occurs:
BUG: kernel NULL pointer dereference, address: 0000000000000038
[...]
RIP: 0010:fw_classify+0x1fe/0x250 [cls_fw]
[...]
Skip "old-style" classification on shared blocks, so that the NULL
dereference is fixed and WARN_ON() is not hit anymore in the short
lifetime of invalid cls_fw "old-style" filters.
[1] https://sashiko.dev/#/patchset/20260331050217.504278-1-xmei5%40asu.edu
[2] https://elixir.bootlin.com/linux/v7.0-rc6/source/include/net/pkt_cls.h#L86
Fixes: faeea8bbf6 ("net/sched: cls_fw: fix NULL pointer dereference on shared blocks")
Fixes: ed76f5edcc ("net: sched: protect filter_chain list with filter_chain_lock mutex")
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/e39cbd3103a337f1e515d186fe697b4459d24757.1775661704.git.dcaratti@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
codel_dump_stats() only runs with RTNL held,
reading fields that can be changed in qdisc fast path.
Add READ_ONCE()/WRITE_ONCE() annotations.
Alternative would be to acquire the qdisc spinlock, but our long-term
goal is to make qdisc dump operations lockless as much as we can.
tc_codel_xstats fields don't need to be latched atomically,
otherwise this bug would have been caught earlier.
No change in kernel size:
$ scripts/bloat-o-meter -t vmlinux.0 vmlinux
add/remove: 0/0 grow/shrink: 1/1 up/down: 3/-1 (2)
Function old new delta
codel_qdisc_dequeue 2462 2465 +3
codel_dump_stats 250 249 -1
Total: Before=29739919, After=29739921, chg +0.00%
Fixes: 76e3cc126b ("codel: Controlled Delay AQM")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260407143053.1570620-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
tcf_csum_act() walks nested VLAN headers directly from skb->data when an
skb still carries in-payload VLAN tags. The current code reads
vlan->h_vlan_encapsulated_proto and then pulls VLAN_HLEN bytes without
first ensuring that the full VLAN header is present in the linear area.
If only part of an inner VLAN header is linearized, accessing
h_vlan_encapsulated_proto reads past the linear area, and the following
skb_pull(VLAN_HLEN) may violate skb invariants.
Fix this by requiring pskb_may_pull(skb, VLAN_HLEN) before accessing and
pulling each nested VLAN header. If the header still is not fully
available, drop the packet through the existing error path.
Fixes: 2ecba2d1e4 ("net: sched: act_csum: Fix csum calc for tagged packets")
Reported-by: Yifan Wu <yifanwucs@gmail.com>
Reported-by: Juefei Pu <tomapufckgml@gmail.com>
Co-developed-by: Yuan Tan <yuantan098@gmail.com>
Signed-off-by: Yuan Tan <yuantan098@gmail.com>
Suggested-by: Xin Liu <bird@lzu.edu.cn>
Tested-by: Ren Wei <enjou1224z@gmail.com>
Signed-off-by: Ruide Cao <caoruide123@gmail.com>
Signed-off-by: Ren Wei <n05ec@lzu.edu.cn>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/22df2fcb49f410203eafa5d97963dd36089f4ecf.1774892775.git.caoruide123@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
flow_change() calls tcf_block_q() and dereferences q->handle to derive
a default baseclass. Shared blocks leave block->q NULL, causing a NULL
deref when a flow filter without a fully qualified baseclass is created
on a shared block.
Check tcf_block_shared() before accessing block->q and return -EINVAL
for shared blocks. This avoids the null-deref shown below:
=======================================================================
KASAN: null-ptr-deref in range [0x0000000000000038-0x000000000000003f]
RIP: 0010:flow_change (net/sched/cls_flow.c:508)
Call Trace:
tc_new_tfilter (net/sched/cls_api.c:2432)
rtnetlink_rcv_msg (net/core/rtnetlink.c:6980)
[...]
=======================================================================
Fixes: 1abf272022 ("net: sched: tcindex, fw, flow: use tcf_block_q helper to get struct Qdisc")
Reported-by: Weiming Shi <bestswngs@gmail.com>
Signed-off-by: Xiang Mei <xmei5@asu.edu>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20260331050217.504278-2-xmei5@asu.edu
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
The old-method path in fw_classify() calls tcf_block_q() and
dereferences q->handle. Shared blocks leave block->q NULL, causing a
NULL deref when an empty cls_fw filter is attached to a shared block
and a packet with a nonzero major skb mark is classified.
Reject the configuration in fw_change() when the old method (no
TCA_OPTIONS) is used on a shared block, since fw_classify()'s
old-method path needs block->q which is NULL for shared blocks.
The fixed null-ptr-deref calling stack:
KASAN: null-ptr-deref in range [0x0000000000000038-0x000000000000003f]
RIP: 0010:fw_classify (net/sched/cls_fw.c:81)
Call Trace:
tcf_classify (./include/net/tc_wrapper.h:197 net/sched/cls_api.c:1764 net/sched/cls_api.c:1860)
tc_run (net/core/dev.c:4401)
__dev_queue_xmit (net/core/dev.c:4535 net/core/dev.c:4790)
Fixes: 1abf272022 ("net: sched: tcindex, fw, flow: use tcf_block_q helper to get struct Qdisc")
Reported-by: Weiming Shi <bestswngs@gmail.com>
Signed-off-by: Xiang Mei <xmei5@asu.edu>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20260331050217.504278-1-xmei5@asu.edu
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
In netem_enqueue(), the packet corruption logic uses
get_random_u32_below(skb_headlen(skb)) to select an index for
modifying skb->data. When an AF_PACKET TX_RING sends fully non-linear
packets over an IPIP tunnel, skb_headlen(skb) evaluates to 0.
Passing 0 to get_random_u32_below() takes the variable-ceil slow path
which returns an unconstrained 32-bit random integer. Using this
unconstrained value as an offset into skb->data results in an
out-of-bounds memory access.
Fix this by verifying skb_headlen(skb) is non-zero before attempting
to corrupt the linear data area. Fully non-linear packets will silently
bypass the corruption logic.
Fixes: c865e5d99e ("[PKT_SCHED] netem: packet corruption option")
Reported-by: Yifan Wu <yifanwucs@gmail.com>
Reported-by: Juefei Pu <tomapufckgml@gmail.com>
Signed-off-by: Yuan Tan <tanyuan98@outlook.com>
Signed-off-by: Xin Liu <bird@lzu.edu.cn>
Signed-off-by: Yuhang Zheng <z1652074432@gmail.com>
Signed-off-by: Yucheng Lu <kanolyc@gmail.com>
Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>
Link: https://patch.msgid.link/45435c0935df877853a81e6d06205ac738ec65fa.1774941614.git.kanolyc@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When building netlink messages, tc_chain_fill_node() never initializes
the tcm_info field of struct tcmsg. Since the allocation is not zeroed,
kernel heap memory is leaked to userspace through this 4-byte field.
The fix simply zeroes tcm_info alongside the other fields that are
already initialized.
Fixes: 32a4f5ecd7 ("net: sched: introduce chain object to uapi")
Signed-off-by: Yochai Eisenrich <echelonh@gmail.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20260328211436.1010152-1-echelonh@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
As IPv6 is built-in only, the ipv6_stub infrastructure is no longer
necessary.
Convert remaining ipv6_stub users to make direct function calls. The
fallback functions introduced previously will prevent linkage errors
when CONFIG_IPV6 is disabled.
Signed-off-by: Fernando Fernandez Mancera <fmancera@suse.de>
Tested-by: Ricardo B. Marlière <rbm@suse.com>
Link: https://patch.msgid.link/20260325120928.15848-9-fmancera@suse.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
m2sm() converts a u32 slope to a u64 scaled value. For large inputs
(e.g. m1=4000000000), the result can reach 2^32. rtsc_min() stores
the difference of two such u64 values in a u32 variable `dsm` and
uses it as a divisor. When the difference is exactly 2^32 the
truncation yields zero, causing a divide-by-zero oops in the
concave-curve intersection path:
Oops: divide error: 0000
RIP: 0010:rtsc_min (net/sched/sch_hfsc.c:601)
Call Trace:
init_ed (net/sched/sch_hfsc.c:629)
hfsc_enqueue (net/sched/sch_hfsc.c:1569)
[...]
Widen `dsm` to u64 and replace do_div() with div64_u64() so the full
difference is preserved.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Reported-by: Weiming Shi <bestswngs@gmail.com>
Signed-off-by: Xiang Mei <xmei5@asu.edu>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20260326204310.1549327-1-xmei5@asu.edu
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Fix a use-after-free in the clsact qdisc upon init/destroy rollback asymmetry.
The latter is achieved by first fully initializing a clsact instance, and
then in a second step having a replacement failure for the new clsact qdisc
instance. clsact_init() initializes ingress first and then takes care of the
egress part. This can fail midway, for example, via tcf_block_get_ext(). Upon
failure, the kernel will trigger the clsact_destroy() callback.
Commit 1cb6f0bae5 ("bpf: Fix too early release of tcx_entry") details the
way how the transition is happening. If tcf_block_get_ext on the q->ingress_block
ends up failing, we took the tcx_miniq_inc reference count on the ingress
side, but not yet on the egress side. clsact_destroy() tests whether the
{ingress,egress}_entry was non-NULL. However, even in midway failure on the
replacement, both are in fact non-NULL with a valid egress_entry from the
previous clsact instance.
What we really need to test for is whether the qdisc instance-specific ingress
or egress side previously got initialized. This adds a small helper for checking
the miniq initialization called mini_qdisc_pair_inited, and utilizes that upon
clsact_destroy() in order to fix the use-after-free scenario. Convert the
ingress_destroy() side as well so both are consistent to each other.
Fixes: 1cb6f0bae5 ("bpf: Fix too early release of tcx_entry")
Reported-by: Keenan Dong <keenanat2000@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Cc: Martin KaFai Lau <martin.lau@kernel.org>
Acked-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://patch.msgid.link/20260313065531.98639-1-daniel@iogearbox.net
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
When creating a flower classifier with an ipv4 address the
flow_dissector has both FLOW_DISSECTOR_KEY_IPV4_ADDRS and
FLOW_DISSECTOR_KEY_IPV6_ADDRS bits set in used_keys.
This happens because ipv4/ipv6 fields are a union and
FL_KEY_SET_IF_MASKED() will interpret either being set as both.
Removing the unions fixes this behavior without needing special handling
for union fields.
Example of a command that caused FLOW_DISSECTOR_KEY_IPV4_ADDRS and
FLOW_DISSECTOR_KEY_IPV6_ADDRS to be set:
tc filter add dev p1 ingress protocol ip flower skip_sw \
dst_ip 224.0.1.129 action trap
Signed-off-by: Cedric Jehasse <cedric.jehasse@luminex.be>
Link: https://patch.msgid.link/20260311-net-next-mv88e6xxx-tcam-v8-1-32dd5ba30002@luminex.be
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Syzbot reported a warning in u32_init_knode() [1].
Similar to commit 7cba18332e ("net: sched: cls_u32: Avoid memcpy()
false-positive warning") which addressed the same issue in u32_change(),
use unsafe_memcpy() in u32_init_knode() to work around the compiler's
inability to see into composite flexible array structs.
This silences the false-positive reported by syzbot:
memcpy: detected field-spanning write (size 32) of single field
"&new->sel" at net/sched/cls_u32.c:855 (size 16)
Since the memory is correctly allocated with kzalloc_flex() using
s->nkeys, this is purely a false positive and does not need a Fixes tag.
[1] https://syzkaller.appspot.com/bug?extid=d5ace703ed883df56e42
Reported-by: syzbot+d5ace703ed883df56e42@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/all/69a811b9.a70a0220.b118c.0019.GAE@google.com/T/
Reviewed-by: Simon Horman <horms@kernel.org>
Acked-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Signed-off-by: Jiayuan Chen <jiayuan.chen@shopee.com>
Link: https://patch.msgid.link/20260309123917.402183-1-jiayuan.chen@linux.dev
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Some modern cpus disable X86_FEATURE_RETPOLINE feature,
even if a direct call can still be beneficial.
Even when IBRS is present, an indirect call is more expensive
than a direct one:
Direct Calls:
Compilers can perform powerful optimizations like inlining,
where the function body is directly inserted at the call site,
eliminating call overhead entirely.
Indirect Calls:
Inlining is much harder, if not impossible, because the compiler
doesn't know the target function at compile time.
Techniques like Indirect Call Promotion can help by using
profile-guided optimization to turn frequently taken indirect calls
into conditional direct calls, but they still add complexity
and potential overhead compared to a truly direct call.
In this patch, I split tc_skip_wrapper in two different
static keys, one for tc_act() (tc_skip_wrapper_act)
and one for tc_classify() (tc_skip_wrapper_cls).
Then I enable the tc_skip_wrapper_cls only if the count
of builtin classifiers is above one.
I enable tc_skip_wrapper_act only it the count of builtin
actions is above one.
In our production kernels, we only have CONFIG_NET_CLS_BPF=y
and CONFIG_NET_ACT_BPF=y. Other are modules or are not compiled.
Tested on AMD Turin cpus, cls_bpf_classify() cost went
from 1% down to 0.18 %, and FDO will be able to inline
it in tcf_classify() for further gains.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Reviewed-by: Pedro Tammela <pctammela@mojatatu.com>
Reviewed-by: Victor Nogueira <victor@mojatatu.com>
Link: https://patch.msgid.link/20260307133601.3863071-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
rtnl_kfree_skbs() reduces RTNL and qdisc spinlock hold time.
skbs are freed later after RTNL has been released.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260306133154.678730-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Following typical script is extremely disruptive,
because each graft operation calls dev_deactivate()
which resets all the queues of the device.
QPARAM="limit 100000 flow_limit 1000 buckets 4096"
TXQS=64
for ETH in eth1
do
tc qd del dev $ETH root 2>/dev/null
tc qd add dev $ETH root handle 1: mq
for i in `seq 1 $TXQS`
do
slot=$( printf %x $(( i )) )
tc qd add dev $ETH parent 1:$slot fq $QPARAM
done
done
One can add "ip link set dev $ETH down/up" to reduce the disruption time:
QPARAM="limit 100000 flow_limit 1000 buckets 4096"
TXQS=64
for ETH in eth1
do
ip link set dev $ETH down
tc qd del dev $ETH root 2>/dev/null
tc qd add dev $ETH root handle 1: mq
for i in `seq 1 $TXQS`
do
slot=$( printf %x $(( i )) )
tc qd add dev $ETH parent 1:$slot fq $QPARAM
done
ip link set dev $ETH up
done
Or we can add a @reset_needed flag to dev_deactivate() and
dev_deactivate_many().
This flag is set to true at device dismantle or linkwatch_do_dev(),
and to false for graft operations.
In the future, we might only stop one queue instead of the whole
device, ie call dev_deactivate_queue() instead of dev_deactivate().
I think the problem (quadratic behavior) was added in commit
2fb541c862 ("net: sch_generic: aviod concurrent reset and enqueue op
for lockless qdisc") but this does not look serious enough to deserve
risky backports.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Yunsheng Lin <linyunsheng@huawei.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Reviewed-by: Toke Høiland-Jørgensen <toke@redhat.com>
Reviewed-by: Victor Nogueira <victor@mojatatu.com>
Link: https://patch.msgid.link/20260307163430.470644-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
teql_master_xmit() calls netdev_start_xmit(skb, slave) to transmit
through slave devices, but does not update skb->dev to the slave device
beforehand.
When a gretap tunnel is a TEQL slave, the transmit path reaches
iptunnel_xmit() which saves dev = skb->dev (still pointing to teql0
master) and later calls iptunnel_xmit_stats(dev, pkt_len). This
function does:
get_cpu_ptr(dev->tstats)
Since teql_master_setup() does not set dev->pcpu_stat_type to
NETDEV_PCPU_STAT_TSTATS, the core network stack never allocates tstats
for teql0, so dev->tstats is NULL. get_cpu_ptr(NULL) computes
NULL + __per_cpu_offset[cpu], resulting in a page fault.
BUG: unable to handle page fault for address: ffff8880e6659018
#PF: supervisor write access in kernel mode
#PF: error_code(0x0002) - not-present page
PGD 68bc067 P4D 68bc067 PUD 0
Oops: Oops: 0002 [#1] SMP KASAN PTI
RIP: 0010:iptunnel_xmit (./include/net/ip_tunnels.h:664 net/ipv4/ip_tunnel_core.c:89)
Call Trace:
<TASK>
ip_tunnel_xmit (net/ipv4/ip_tunnel.c:847)
__gre_xmit (net/ipv4/ip_gre.c:478)
gre_tap_xmit (net/ipv4/ip_gre.c:779)
teql_master_xmit (net/sched/sch_teql.c:319)
dev_hard_start_xmit (net/core/dev.c:3887)
sch_direct_xmit (net/sched/sch_generic.c:347)
__dev_queue_xmit (net/core/dev.c:4802)
neigh_direct_output (net/core/neighbour.c:1660)
ip_finish_output2 (net/ipv4/ip_output.c:237)
__ip_finish_output.part.0 (net/ipv4/ip_output.c:315)
ip_mc_output (net/ipv4/ip_output.c:369)
ip_send_skb (net/ipv4/ip_output.c:1508)
udp_send_skb (net/ipv4/udp.c:1195)
udp_sendmsg (net/ipv4/udp.c:1485)
inet_sendmsg (net/ipv4/af_inet.c:859)
__sys_sendto (net/socket.c:2206)
Fix this by setting skb->dev = slave before calling
netdev_start_xmit(), so that tunnel xmit functions see the correct
slave device with properly allocated tstats.
Fixes: 039f50629b ("ip_tunnel: Move stats update to iptunnel_xmit()")
Reported-by: Xiang Mei <xmei5@asu.edu>
Signed-off-by: Weiming Shi <bestswngs@gmail.com>
Link: https://patch.msgid.link/20260304044216.3517851-3-bestswngs@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Cross-merge networking fixes after downstream PR (net-7.0-rc3).
No conflicts.
Adjacent changes:
net/netfilter/nft_set_rbtree.c
fb7fb40163 ("netfilter: nf_tables: clone set on flush only")
3aea466a43 ("netfilter: nft_set_rbtree: don't disable bh when acquiring tree lock")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When/if a NIC resets, queues are deactivated by dev_deactivate_many(),
then reactivated when the reset operation completes.
fq_reset() removes all the skbs from various queues.
If we do not clear q->band_pkt_count[], these counters keep growing
and can eventually reach sch->limit, preventing new packets to be queued.
Many thanks to Praveen for discovering the root cause.
Fixes: 29f834aa32 ("net_sched: sch_fq: add 3 bands and WRR scheduling")
Diagnosed-by: Praveen Kaligineedi <pkaligineedi@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Neal Cardwell <ncardwell@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20260304015640.961780-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
DualPI2 drops packets during dequeue but was using kfree_skb_reason()
directly, bypassing trace_qdisc_drop. Convert to qdisc_dequeue_drop()
and add QDISC_DROP_L4S_STEP_NON_ECN to the qdisc drop reason enum.
- Set TCQ_F_DEQUEUE_DROPS flag in dualpi2_init()
- Use enum qdisc_drop_reason in drop_and_retry()
- Replace kfree_skb_reason() with qdisc_dequeue_drop()
Signed-off-by: Jesper Dangaard Brouer <hawk@kernel.org>
Reviewed-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://patch.msgid.link/177211351978.3011628.11267023360997620069.stgit@firesoul
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Rename QDISC_DROP_CAKE_FLOOD to QDISC_DROP_FLOOD_PROTECTION to use a
generic name without embedding the qdisc name. This follows the
principle that drop reasons should describe the drop mechanism rather
than being tied to a specific qdisc implementation.
The flood protection drop reason is used by qdiscs implementing
probabilistic drop algorithms (like BLUE) that detect unresponsive
flows indicating potential DoS or flood attacks. CAKE uses this via
its Cobalt AQM component.
Signed-off-by: Jesper Dangaard Brouer <hawk@kernel.org>
Reviewed-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://patch.msgid.link/177211347537.3011628.13759059534638729639.stgit@firesoul
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Rename FQ-specific drop reasons to generic names:
- QDISC_DROP_FQ_BAND_LIMIT -> QDISC_DROP_BAND_LIMIT
- QDISC_DROP_FQ_HORIZON_LIMIT -> QDISC_DROP_HORIZON_LIMIT
This follows the principle that drop reasons should describe the drop
mechanism rather than being tied to a specific qdisc implementation.
These concepts (priority band limits, timestamp horizon) could apply
to other qdiscs as well.
Remove the local macro define FQDR() and instead use the
full QDISC_DROP_* name to make it easier to navigate code.
Signed-off-by: Jesper Dangaard Brouer <hawk@kernel.org>
Reviewed-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://patch.msgid.link/177211346902.3011628.12523261489552097455.stgit@firesoul
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Convert SFQ to use the new qdisc-specific drop reason infrastructure.
This patch demonstrates how to convert a flow-based qdisc to use the
new enum qdisc_drop_reason. As part of this conversion:
- Add QDISC_DROP_MAXFLOWS for flow table exhaustion
- Rename FQ_FLOW_LIMIT to generic FLOW_LIMIT, now shared by FQ and SFQ
- Use QDISC_DROP_OVERLIMIT for sfq_drop() when overall limit exceeded
- Use QDISC_DROP_FLOW_LIMIT for per-flow depth limit exceeded
The FLOW_LIMIT reason is now a common drop reason for per-flow limits,
applicable to both FQ and SFQ qdiscs.
Signed-off-by: Jesper Dangaard Brouer <hawk@kernel.org>
Reviewed-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://patch.msgid.link/177211345946.3011628.12770616071857185664.stgit@firesoul
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Create new enum qdisc_drop_reason and trace_qdisc_drop tracepoint
for qdisc layer drop diagnostics with direct qdisc context visibility.
The new tracepoint includes qdisc handle, parent, kind (name), and
device information. Existing SKB_DROP_REASON_QDISC_DROP is retained
for backwards compatibility via kfree_skb_reason().
Convert qdiscs with drop reasons to use the new infrastructure.
Change CAKE's cobalt_should_drop() return type from enum skb_drop_reason
to enum qdisc_drop_reason to fix implicit enum conversion warnings.
Use QDISC_DROP_UNSPEC as the 'not dropped' sentinel instead of
SKB_NOT_DROPPED_YET. Both have the same compiled value (0), so the
comparison logic remains semantically equivalent.
Signed-off-by: Jesper Dangaard Brouer <hawk@kernel.org>
Reviewed-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://patch.msgid.link/177211345275.3011628.1974310302645218067.stgit@firesoul
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
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).
[1] https://lore.kernel.org/netdev/674b8cbfc385c6f37fb29a1de08d8fe5c2b0fbee.1771321118.git.pabeni@redhat.com/
[2] https://lore.kernel.org/netdev/cc6bfb4a-4a2b-42d8-b9ce-7ef6644fb22b@ovn.org/
Reported-by: GangMin Kim <km.kim1503@gmail.com>
Fixes: 3f14b377d0 ("net/sched: act_ct: fix skb leak and crash on ooo frags")
CC: stable@vger.kernel.org
Signed-off-by: Victor Nogueira <victor@mojatatu.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20260225134349.1287037-1-victor@mojatatu.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
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.
Fixes: 1bddd758ba ("net/sched: sch_cake: share shaper state across sub-instances of cake_mq")
Signed-off-by: Jonas Köppeler <j.koeppeler@tu-berlin.de>
Acked-by: Toke Høiland-Jørgensen <toke@toke.dk>
Link: https://patch.msgid.link/20260226-cake-mq-skip-sync-bandwidth-unlimited-v1-2-01830bb4db87@tu-berlin.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Skip inter-instance sync when no rate limit is configured, as it serves
no purpose and only adds overhead.
Fixes: 1bddd758ba ("net/sched: sch_cake: share shaper state across sub-instances of cake_mq")
Signed-off-by: Jonas Köppeler <j.koeppeler@tu-berlin.de>
Acked-by: Toke Høiland-Jørgensen <toke@toke.dk>
Link: https://patch.msgid.link/20260226-cake-mq-skip-sync-bandwidth-unlimited-v1-1-01830bb4db87@tu-berlin.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
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: a51c328df3 ("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>
This converts some of the visually simpler cases that have been split
over multiple lines. I only did the ones that are easy to verify the
resulting diff by having just that final GFP_KERNEL argument on the next
line.
Somebody should probably do a proper coccinelle script for this, but for
me the trivial script actually resulted in an assertion failure in the
middle of the script. I probably had made it a bit _too_ trivial.
So after fighting that far a while I decided to just do some of the
syntactically simpler cases with variations of the previous 'sed'
scripts.
The more syntactically complex multi-line cases would mostly really want
whitespace cleanup anyway.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This is the exact same thing as the 'alloc_obj()' version, only much
smaller because there are a lot fewer users of the *alloc_flex()
interface.
As with alloc_obj() version, this was done entirely with mindless brute
force, using the same script, except using 'flex' in the pattern rather
than 'objs*'.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This was done entirely with mindless brute force, using
git grep -l '\<k[vmz]*alloc_objs*(.*, GFP_KERNEL)' |
xargs sed -i 's/\(alloc_objs*(.*\), GFP_KERNEL)/\1)/'
to convert the new alloc_obj() users that had a simple GFP_KERNEL
argument to just drop that argument.
Note that due to the extreme simplicity of the scripting, any slightly
more complex cases spread over multiple lines would not be triggered:
they definitely exist, but this covers the vast bulk of the cases, and
the resulting diff is also then easier to check automatically.
For the same reason the 'flex' versions will be done as a separate
conversion.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This is the result of running the Coccinelle script from
scripts/coccinelle/api/kmalloc_objs.cocci. The script is designed to
avoid scalar types (which need careful case-by-case checking), and
instead replace kmalloc-family calls that allocate struct or union
object instances:
Single allocations: kmalloc(sizeof(TYPE), ...)
are replaced with: kmalloc_obj(TYPE, ...)
Array allocations: kmalloc_array(COUNT, sizeof(TYPE), ...)
are replaced with: kmalloc_objs(TYPE, COUNT, ...)
Flex array allocations: kmalloc(struct_size(PTR, FAM, COUNT), ...)
are replaced with: kmalloc_flex(*PTR, FAM, COUNT, ...)
(where TYPE may also be *VAR)
The resulting allocations no longer return "void *", instead returning
"TYPE *".
Signed-off-by: Kees Cook <kees@kernel.org>
Commit 38a6f08657 ("net: sched: support hash selecting tx queue")
added SKBEDIT_F_TXQ_SKBHASH support. The inclusive range size is
computed as:
mapping_mod = queue_mapping_max - queue_mapping + 1;
The range size can be 65536 when the requested range covers all possible
u16 queue IDs (e.g. queue_mapping=0 and queue_mapping_max=U16_MAX).
That value cannot be represented in a u16 and previously wrapped to 0,
so tcf_skbedit_hash() could trigger a divide-by-zero:
queue_mapping += skb_get_hash(skb) % params->mapping_mod;
Compute mapping_mod in a wider type and reject ranges larger than U16_MAX
to prevent params->mapping_mod from becoming 0 and avoid the crash.
Fixes: 38a6f08657 ("net: sched: support hash selecting tx queue")
Cc: stable@vger.kernel.org # 6.12+
Signed-off-by: Ruitong Liu <cnitlrt@gmail.com>
Link: https://patch.msgid.link/20260213175948.1505257-1-cnitlrt@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Core & protocols
----------------
- A significant effort all around the stack to guide the compiler to
make the right choice when inlining code, to avoid unneeded calls for
small helper and stack canary overhead in the fast-path. This
generates better and faster code with very small or no text size
increases, as in many cases the call generated more code than the
actual inlined helper.
- Extend AccECN implementation so that is now functionally complete,
also allow the user-space enabling it on a per network namespace
basis.
- Add support for memory providers with large (above 4K) rx buffer.
Paired with hw-gro, larger rx buffer sizes reduce the number of
buffers traversing the stack, dincreasing single stream CPU usage by
up to ~30%.
- Do not add HBH header to Big TCP GSO packets. This simplifies the RX
path, the TX path and the NIC drivers, and is possible because
user-space taps can now interpret correctly such packets without the
HBH hint.
- Allow IPv6 routes to be configured with a gateway address that is
resolved out of a different interface than the one specified, aligning
IPv6 to IPv4 behavior.
- Multi-queue aware sch_cake. This makes it possible to scale the rate
shaper of sch_cake across multiple CPUs, while still enforcing a
single global rate on the interface.
- Add support for the nbcon (new buffer console) infrastructure to
netconsole, enabling lock-free, priority-based console operations that
are safer in crash scenarios.
- Improve the TCP ipv6 output path to cache the flow information, saving
cpu cycles, reducing cache line misses and stack use.
- Improve netfilter packet tracker to resolve clashes for most protocols,
avoiding unneeded drops on rare occasions.
- Add IP6IP6 tunneling acceleration to the flowtable infrastructure.
- Reduce tcp socket size by one cache line.
- Notify neighbour changes atomically, avoiding inconsistencies between
the notification sequence and the actual states sequence.
- Add vsock namespace support, allowing complete isolation of vsocks
across different network namespaces.
- Improve xsk generic performances with cache-alignment-oriented
optimizations.
- Support netconsole automatic target recovery, allowing netconsole
to reestablish targets when underlying low-level interface comes back
online.
Driver API
----------
- Support for switching the working mode (automatic vs manual) of a DPLL
device via netlink.
- Introduce PHY ports representation to expose multiple front-facing
media ports over a single MAC.
- Introduce "rx-polarity" and "tx-polarity" device tree properties, to
generalize polarity inversion requirements for differential signaling.
- Add helper to create, prepare and enable managed clocks.
Device drivers
--------------
- Add Huawei hinic3 PF etherner driver.
- Add DWMAC glue driver for Motorcomm YT6801 PCIe ethernet controller.
- Add ethernet driver for MaxLinear MxL862xx switches
- Remove parallel-port Ethernet driver.
- Convert existing driver timestamp configuration reporting to
hwtstamp_get and remove legacy ioctl().
- Convert existing drivers to .get_rx_ring_count(), simplifing the RX
ring count retrieval. Also remove the legacy fallback path.
- Ethernet high-speed NICs:
- Broadcom (bnxt, bng):
- bnxt: add FW interface update to support FEC stats histogram and
NVRAM defragmentation
- bng: add TSO and H/W GRO support
- nVidia/Mellanox (mlx5):
- improve latency of channel restart operations, reducing the used
H/W resources
- add TSO support for UDP over GRE over VLAN
- add flow counters support for hardware steering (HWS) rules
- use a static memory area to store headers for H/W GRO, leading to
12% RX tput improvement
- Intel (100G, ice, idpf):
- ice: reorganizes layout of Tx and Rx rings for cacheline
locality and utilizes __cacheline_group* macros on the new layouts
- ice: introduces Synchronous Ethernet (SyncE) support
- Meta (fbnic):
- adds debugfs for firmware mailbox and tx/rx rings vectors
- Ethernet virtual:
- geneve: introduce GRO/GSO support for double UDP encapsulation
- Ethernet NICs consumer, and embedded:
- Synopsys (stmmac):
- some code refactoring and cleanups
- RealTek (r8169):
- add support for RTL8127ATF (10G Fiber SFP)
- add dash and LTR support
- Airoha:
- AN8811HB 2.5 Gbps phy support
- Freescale (fec):
- add XDP zero-copy support
- Thunderbolt:
- add get link setting support to allow bonding
- Renesas:
- add support for RZ/G3L GBETH SoC
- Ethernet switches:
- Maxlinear:
- support R(G)MII slow rate configuration
- add support for Intel GSW150
- Motorcomm (yt921x):
- add DCB/QoS support
- TI:
- icssm-prueth: support bridging (STP/RSTP) via the switchdev
framework
- Ethernet PHYs:
- Realtek:
- enable SGMII and 2500Base-X in-band auto-negotiation
- simplify and reunify C22/C45 drivers
- Micrel: convert bindings to DT schema
- CAN:
- move skb headroom content into skb extensions, making CAN metadata
access more robust
- CAN drivers:
- rcar_canfd:
- add support for FD-only mode
- add support for the RZ/T2H SoC
- sja1000: cleanup the CAN state handling
- WiFi:
- implement EPPKE/802.1X over auth frames support
- split up drop reasons better, removing generic RX_DROP
- additional FTM capabilities: 6 GHz support, supported number of
spatial streams and supported number of LTF repetitions
- better mac80211 iterators to enumerate resources
- initial UHR (Wi-Fi 8) support for cfg80211/mac80211
- WiFi drivers:
- Qualcomm/Atheros:
- ath11k: support for Channel Frequency Response measurement
- ath12k: a significant driver refactor to support
multi-wiphy devices and and pave the way for future device support
in the same driver (rather than splitting to ath13k)
- ath12k: support for the QCC2072 chipset
- Intel:
- iwlwifi: partial Neighbor Awareness Networking (NAN) support
- iwlwifi: initial support for U-NII-9 and IEEE 802.11bn
- RealTek (rtw89):
- preparations for RTL8922DE support
- Bluetooth:
- implement setsockopt(BT_PHY) to set the connection packet type/PHY
- set link_policy on incoming ACL connections
- Bluetooth drivers:
- btusb: add support for MediaTek7920, Realtek RTL8761BU and 8851BE
- btqca: add WCN6855 firmware priority selection feature
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
-----BEGIN PGP SIGNATURE-----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=O31Y
-----END PGP SIGNATURE-----
Merge tag 'net-next-7.0' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next
Pull networking updates from Paolo Abeni:
"Core & protocols:
- A significant effort all around the stack to guide the compiler to
make the right choice when inlining code, to avoid unneeded calls
for small helper and stack canary overhead in the fast-path.
This generates better and faster code with very small or no text
size increases, as in many cases the call generated more code than
the actual inlined helper.
- Extend AccECN implementation so that is now functionally complete,
also allow the user-space enabling it on a per network namespace
basis.
- Add support for memory providers with large (above 4K) rx buffer.
Paired with hw-gro, larger rx buffer sizes reduce the number of
buffers traversing the stack, dincreasing single stream CPU usage
by up to ~30%.
- Do not add HBH header to Big TCP GSO packets. This simplifies the
RX path, the TX path and the NIC drivers, and is possible because
user-space taps can now interpret correctly such packets without
the HBH hint.
- Allow IPv6 routes to be configured with a gateway address that is
resolved out of a different interface than the one specified,
aligning IPv6 to IPv4 behavior.
- Multi-queue aware sch_cake. This makes it possible to scale the
rate shaper of sch_cake across multiple CPUs, while still enforcing
a single global rate on the interface.
- Add support for the nbcon (new buffer console) infrastructure to
netconsole, enabling lock-free, priority-based console operations
that are safer in crash scenarios.
- Improve the TCP ipv6 output path to cache the flow information,
saving cpu cycles, reducing cache line misses and stack use.
- Improve netfilter packet tracker to resolve clashes for most
protocols, avoiding unneeded drops on rare occasions.
- Add IP6IP6 tunneling acceleration to the flowtable infrastructure.
- Reduce tcp socket size by one cache line.
- Notify neighbour changes atomically, avoiding inconsistencies
between the notification sequence and the actual states sequence.
- Add vsock namespace support, allowing complete isolation of vsocks
across different network namespaces.
- Improve xsk generic performances with cache-alignment-oriented
optimizations.
- Support netconsole automatic target recovery, allowing netconsole
to reestablish targets when underlying low-level interface comes
back online.
Driver API:
- Support for switching the working mode (automatic vs manual) of a
DPLL device via netlink.
- Introduce PHY ports representation to expose multiple front-facing
media ports over a single MAC.
- Introduce "rx-polarity" and "tx-polarity" device tree properties,
to generalize polarity inversion requirements for differential
signaling.
- Add helper to create, prepare and enable managed clocks.
Device drivers:
- Add Huawei hinic3 PF etherner driver.
- Add DWMAC glue driver for Motorcomm YT6801 PCIe ethernet
controller.
- Add ethernet driver for MaxLinear MxL862xx switches
- Remove parallel-port Ethernet driver.
- Convert existing driver timestamp configuration reporting to
hwtstamp_get and remove legacy ioctl().
- Convert existing drivers to .get_rx_ring_count(), simplifing the RX
ring count retrieval. Also remove the legacy fallback path.
- Ethernet high-speed NICs:
- Broadcom (bnxt, bng):
- bnxt: add FW interface update to support FEC stats histogram
and NVRAM defragmentation
- bng: add TSO and H/W GRO support
- nVidia/Mellanox (mlx5):
- improve latency of channel restart operations, reducing the
used H/W resources
- add TSO support for UDP over GRE over VLAN
- add flow counters support for hardware steering (HWS) rules
- use a static memory area to store headers for H/W GRO,
leading to 12% RX tput improvement
- Intel (100G, ice, idpf):
- ice: reorganizes layout of Tx and Rx rings for cacheline
locality and utilizes __cacheline_group* macros on the new
layouts
- ice: introduces Synchronous Ethernet (SyncE) support
- Meta (fbnic):
- adds debugfs for firmware mailbox and tx/rx rings vectors
- Ethernet virtual:
- geneve: introduce GRO/GSO support for double UDP encapsulation
- Ethernet NICs consumer, and embedded:
- Synopsys (stmmac):
- some code refactoring and cleanups
- RealTek (r8169):
- add support for RTL8127ATF (10G Fiber SFP)
- add dash and LTR support
- Airoha:
- AN8811HB 2.5 Gbps phy support
- Freescale (fec):
- add XDP zero-copy support
- Thunderbolt:
- add get link setting support to allow bonding
- Renesas:
- add support for RZ/G3L GBETH SoC
- Ethernet switches:
- Maxlinear:
- support R(G)MII slow rate configuration
- add support for Intel GSW150
- Motorcomm (yt921x):
- add DCB/QoS support
- TI:
- icssm-prueth: support bridging (STP/RSTP) via the switchdev
framework
- Ethernet PHYs:
- Realtek:
- enable SGMII and 2500Base-X in-band auto-negotiation
- simplify and reunify C22/C45 drivers
- Micrel: convert bindings to DT schema
- CAN:
- move skb headroom content into skb extensions, making CAN
metadata access more robust
- CAN drivers:
- rcar_canfd:
- add support for FD-only mode
- add support for the RZ/T2H SoC
- sja1000: cleanup the CAN state handling
- WiFi:
- implement EPPKE/802.1X over auth frames support
- split up drop reasons better, removing generic RX_DROP
- additional FTM capabilities: 6 GHz support, supported number of
spatial streams and supported number of LTF repetitions
- better mac80211 iterators to enumerate resources
- initial UHR (Wi-Fi 8) support for cfg80211/mac80211
- WiFi drivers:
- Qualcomm/Atheros:
- ath11k: support for Channel Frequency Response measurement
- ath12k: a significant driver refactor to support multi-wiphy
devices and and pave the way for future device support in the
same driver (rather than splitting to ath13k)
- ath12k: support for the QCC2072 chipset
- Intel:
- iwlwifi: partial Neighbor Awareness Networking (NAN) support
- iwlwifi: initial support for U-NII-9 and IEEE 802.11bn
- RealTek (rtw89):
- preparations for RTL8922DE support
- Bluetooth:
- implement setsockopt(BT_PHY) to set the connection packet type/PHY
- set link_policy on incoming ACL connections
- Bluetooth drivers:
- btusb: add support for MediaTek7920, Realtek RTL8761BU and 8851BE
- btqca: add WCN6855 firmware priority selection feature"
* tag 'net-next-7.0' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1254 commits)
bnge/bng_re: Add a new HSI
net: macb: Fix tx/rx malfunction after phy link down and up
af_unix: Fix memleak of newsk in unix_stream_connect().
net: ti: icssg-prueth: Add optional dependency on HSR
net: dsa: add basic initial driver for MxL862xx switches
net: mdio: add unlocked mdiodev C45 bus accessors
net: dsa: add tag format for MxL862xx switches
dt-bindings: net: dsa: add MaxLinear MxL862xx
selftests: drivers: net: hw: Modify toeplitz.c to poll for packets
octeontx2-pf: Unregister devlink on probe failure
net: renesas: rswitch: fix forwarding offload statemachine
ionic: Rate limit unknown xcvr type messages
tcp: inet6_csk_xmit() optimization
tcp: populate inet->cork.fl.u.ip6 in tcp_v6_syn_recv_sock()
tcp: populate inet->cork.fl.u.ip6 in tcp_v6_connect()
ipv6: inet6_csk_xmit() and inet6_csk_update_pmtu() use inet->cork.fl.u.ip6
ipv6: use inet->cork.fl.u.ip6 and np->final in ip6_datagram_dst_update()
ipv6: use np->final in inet6_sk_rebuild_header()
ipv6: add daddr/final storage in struct ipv6_pinfo
net: stmmac: qcom-ethqos: fix qcom_ethqos_serdes_powerup()
...
-----BEGIN PGP SIGNATURE-----
iQIzBAABCAAdFiEE+soXsSLHKoYyzcli6rmadz2vbToFAmmGmrgACgkQ6rmadz2v
bTq6NxAAkCHosxzGn9GYYBV8xhrBJoJJDCyEbQ4nR0XNY+zaWnuykmiPP9w1aOAM
zm/po3mQB2pZjetvlrPrgG5RLgBCAUHzqVGy0r+phUvD3vbohKlmSlMm2kiXOb9N
T01BgLWsyqN2ZcNFvORdSsftqIJUHcXxU6RdupGD60sO5XM9ty5cwyewLX8GBOas
UN2bOhbK2DpqYWUvtv+3Q3ykxoStMSkXZvDRurwLKl4RHeLjXZXPo8NjnfBlk/F2
vdFo/F4NO4TmhOave6UPXvKb4yo9IlBRmiPAl0RmNKBxenY8j9XuV/xZxU6YgzDn
+SQfDK+CKQ4IYIygE+fqd4e5CaQrnjmPPcIw12AB2CF0LimY9Xxyyk6FSAhMN7wm
GTVh5K2C3Dk3OiRQk4G58EvQ5QcxzX98IeeCpcckMUkPsFWHRvF402WMUcv9SWpD
DsxxPkfENY/6N67EvH0qcSe/ikdUorQKFl4QjXKwsMCd5WhToeP4Z7Ck1gVSNkAh
9CX++mLzg333Lpsc4SSIuk9bEPpFa5cUIKUY7GCsCiuOXciPeMDP3cGSd5LioqxN
qWljs4Z88QDM2LJpAh8g4m3sA7bMhES3nPmdlI5CfgBcVyLW8D8CqQq4GEZ1McwL
Ky084+lEosugoVjRejrdMMEOsqAfcbkTr2b8jpuAZdwJKm6p/bw=
=cBdK
-----END PGP SIGNATURE-----
Merge tag 'bpf-next-7.0' of git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next
Pull bpf updates from Alexei Starovoitov:
- Support associating BPF program with struct_ops (Amery Hung)
- Switch BPF local storage to rqspinlock and remove recursion detection
counters which were causing false positives (Amery Hung)
- Fix live registers marking for indirect jumps (Anton Protopopov)
- Introduce execution context detection BPF helpers (Changwoo Min)
- Improve verifier precision for 32bit sign extension pattern
(Cupertino Miranda)
- Optimize BTF type lookup by sorting vmlinux BTF and doing binary
search (Donglin Peng)
- Allow states pruning for misc/invalid slots in iterator loops (Eduard
Zingerman)
- In preparation for ASAN support in BPF arenas teach libbpf to move
global BPF variables to the end of the region and enable arena kfuncs
while holding locks (Emil Tsalapatis)
- Introduce support for implicit arguments in kfuncs and migrate a
number of them to new API. This is a prerequisite for cgroup
sub-schedulers in sched-ext (Ihor Solodrai)
- Fix incorrect copied_seq calculation in sockmap (Jiayuan Chen)
- Fix ORC stack unwind from kprobe_multi (Jiri Olsa)
- Speed up fentry attach by using single ftrace direct ops in BPF
trampolines (Jiri Olsa)
- Require frozen map for calculating map hash (KP Singh)
- Fix lock entry creation in TAS fallback in rqspinlock (Kumar
Kartikeya Dwivedi)
- Allow user space to select cpu in lookup/update operations on per-cpu
array and hash maps (Leon Hwang)
- Make kfuncs return trusted pointers by default (Matt Bobrowski)
- Introduce "fsession" support where single BPF program is executed
upon entry and exit from traced kernel function (Menglong Dong)
- Allow bpf_timer and bpf_wq use in all programs types (Mykyta
Yatsenko, Andrii Nakryiko, Kumar Kartikeya Dwivedi, Alexei
Starovoitov)
- Make KF_TRUSTED_ARGS the default for all kfuncs and clean up their
definition across the tree (Puranjay Mohan)
- Allow BPF arena calls from non-sleepable context (Puranjay Mohan)
- Improve register id comparison logic in the verifier and extend
linked registers with negative offsets (Puranjay Mohan)
- In preparation for BPF-OOM introduce kfuncs to access memcg events
(Roman Gushchin)
- Use CFI compatible destructor kfunc type (Sami Tolvanen)
- Add bitwise tracking for BPF_END in the verifier (Tianci Cao)
- Add range tracking for BPF_DIV and BPF_MOD in the verifier (Yazhou
Tang)
- Make BPF selftests work with 64k page size (Yonghong Song)
* tag 'bpf-next-7.0' of git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next: (268 commits)
selftests/bpf: Fix outdated test on storage->smap
selftests/bpf: Choose another percpu variable in bpf for btf_dump test
selftests/bpf: Remove test_task_storage_map_stress_lookup
selftests/bpf: Update task_local_storage/task_storage_nodeadlock test
selftests/bpf: Update task_local_storage/recursion test
selftests/bpf: Update sk_storage_omem_uncharge test
bpf: Switch to bpf_selem_unlink_nofail in bpf_local_storage_{map_free, destroy}
bpf: Support lockless unlink when freeing map or local storage
bpf: Prepare for bpf_selem_unlink_nofail()
bpf: Remove unused percpu counter from bpf_local_storage_map_free
bpf: Remove cgroup local storage percpu counter
bpf: Remove task local storage percpu counter
bpf: Change local_storage->lock and b->lock to rqspinlock
bpf: Convert bpf_selem_unlink to failable
bpf: Convert bpf_selem_link_map to failable
bpf: Convert bpf_selem_unlink_map to failable
bpf: Select bpf_local_storage_map_bucket based on bpf_local_storage
selftests/xsk: fix number of Tx frags in invalid packet
selftests/xsk: properly handle batch ending in the middle of a packet
bpf: Prevent reentrance into call_rcu_tasks_trace()
...
The next commits will transition away from using the hop-by-hop
extension header to encode packet length for BIG TCP. Add wrappers
around ip6->payload_len that return the actual value if it's non-zero,
and calculate it from skb->len if payload_len is set to zero (and a
symmetrical setter).
The new helpers are used wherever the surrounding code supports the
hop-by-hop jumbo header for BIG TCP IPv6, or the corresponding IPv4 code
uses skb_ip_totlen (e.g., in include/net/netfilter/nf_tables_ipv6.h).
No behavioral change in this commit.
Signed-off-by: Alice Mikityanska <alice@isovalent.com>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260205133925.526371-2-alice.kernel@fastmail.im
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Using kmem_cache_free_bulk() in fq_gc() was not optimal.
1) It needs an array.
2) It is only saving cpu cycles for large batches.
The automatic array forces a stack canary, which is expensive.
In practice fq_gc was finding zero, one or two flows at most
per round.
Remove the array, use kmem_cache_free().
This makes fq_enqueue() smaller and faster.
$ scripts/bloat-o-meter -t vmlinux.old vmlinux.new
add/remove: 0/0 grow/shrink: 0/1 up/down: 0/-79 (-79)
Function old new delta
fq_enqueue 1629 1550 -79
Total: Before=24886583, After=24886504, chg -0.00%
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260204190034.76277-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Currently we are registering one dynamic lockdep key for each allocated
qdisc, to avoid false deadlock reports when mirred (or TC eBPF) redirects
packets to another device while the root lock is acquired [1].
Since dynamic keys are a limited resource, we can save them at least for
qdiscs that are not meant to acquire the root lock in the traffic path,
or to carry traffic at all, like:
- clsact
- ingress
- noqueue
Don't register dynamic keys for the above schedulers, so that we hit
MAX_LOCKDEP_KEYS later in our tests.
[1] https://github.com/multipath-tcp/mptcp_net-next/issues/451
Changes in v2:
- change ordering of spin_lock_init() vs. lockdep_register_key()
(Jakub Kicinski)
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/94448f7fa7c4f52d2ce416a4895ec87d456d7417.1770220576.git.dcaratti@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
After 076433bd78 ("net_sched: sch_fq: add fast path
for mostly idle qdisc") we need to remove one unlikely()
because q->internal holds all the fast path packets.
skb = fq_peek(&q->internal);
if (unlikely(skb)) {
q->internal.qlen--;
Calling INET_ECN_set_ce() is very unlikely.
These changes allow fq_dequeue_skb() to be (auto)inlined,
thus making fq_dequeue() faster.
$ scripts/bloat-o-meter -t vmlinux.0 vmlinux
add/remove: 2/2 grow/shrink: 0/1 up/down: 283/-269 (14)
Function old new delta
INET_ECN_set_ce - 267 +267
__pfx_INET_ECN_set_ce - 16 +16
__pfx_fq_dequeue_skb 16 - -16
fq_dequeue_skb 103 - -103
fq_dequeue 1685 1535 -150
Total: Before=24886569, After=24886583, chg +0.00%
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20260203214716.880853-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
several netfilter compilation units rely on implicit includes
coming from nf_conntrack_proto_gre.h.
Clean this up and add the required dependencies where needed.
nf_conntrack.h requires net_generic() helper.
Place various gre/ppp/vlan includes to where they are needed.
Signed-off-by: Florian Westphal <fw@strlen.de>
This is more of a preventive patch to make the code more consistent and
to prevent possible exploits that employ child qlen manipulations on qfq.
use cl_is_active instead of relying on the child qdisc's qlen to determine
class activation.
Fixes: 462dbc9101 ("pkt_sched: QFQ Plus: fair-queueing service at DRR cost")
Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20260114160243.913069-3-jhs@mojatatu.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Design intent of teql is that it is only supposed to be used as root qdisc.
We need to check for that constraint.
Although not important, I will describe the scenario that unearthed this
issue for the curious.
GangMin Kim <km.kim1503@gmail.com> managed to concot a scenario as follows:
ROOT qdisc 1:0 (QFQ)
├── class 1:1 (weight=15, lmax=16384) netem with delay 6.4s
└── class 1:2 (weight=1, lmax=1514) teql
GangMin sends a packet which is enqueued to 1:1 (netem).
Any invocation of dequeue by QFQ from this class will not return a packet
until after 6.4s. In the meantime, a second packet is sent and it lands on
1:2. teql's enqueue will return success and this will activate class 1:2.
Main issue is that teql only updates the parent visible qlen (sch->q.qlen)
at dequeue. Since QFQ will only call dequeue if peek succeeds (and teql's
peek always returns NULL), dequeue will never be called and thus the qlen
will remain as 0. With that in mind, when GangMin updates 1:2's lmax value,
the qfq_change_class calls qfq_deact_rm_from_agg. Since the child qdisc's
qlen was not incremented, qfq fails to deactivate the class, but still
frees its pointers from the aggregate. So when the first packet is
rescheduled after 6.4 seconds (netem's delay), a dangling pointer is
accessed causing GangMin's causing a UAF.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Reported-by: GangMin Kim <km.kim1503@gmail.com>
Tested-by: Victor Nogueira <victor@mojatatu.com>
Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20260114160243.913069-2-jhs@mojatatu.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Paolo pointed out that we can avoid separately allocating struct
cake_sched_config even in the non-mq case, by embedding it into struct
cake_sched_data. This reduces the complexity of the logic that swaps the
pointers and frees the old value, at the cost of adding 56 bytes to the
latter. Since cake_sched_data is already almost 17k bytes, this seems
like a reasonable tradeoff.
Suggested-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Fixes: bc0ce2bad3 ("net/sched: sch_cake: Factor out config variables into separate struct")
Link: https://patch.msgid.link/20260113143157.2581680-1-toke@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Cross-merge BPF and other fixes after downstream PR.
No conflicts.
Adjacent:
Auto-merging MAINTAINERS
Auto-merging Makefile
Auto-merging kernel/bpf/verifier.c
Auto-merging kernel/sched/ext.c
Auto-merging mm/memcontrol.c
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
This commit adds shared shaper state across the cake instances beneath a
cake_mq qdisc. It works by periodically tracking the number of active
instances, and scaling the configured rate by the number of active
queues.
The scan is lockless and simply reads the qlen and the last_active state
variable of each of the instances configured beneath the parent cake_mq
instance. Locking is not required since the values are only updated by
the owning instance, and eventual consistency is sufficient for the
purpose of estimating the number of active queues.
The interval for scanning the number of active queues is set to 200 us.
We found this to be a good tradeoff between overhead and response time.
For a detailed analysis of this aspect see the Netdevconf talk:
https://netdevconf.info/0x19/docs/netdev-0x19-paper16-talk-paper.pdf
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Jonas Köppeler <j.koeppeler@tu-berlin.de>
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://patch.msgid.link/20260109-mq-cake-sub-qdisc-v8-5-8d613fece5d8@redhat.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Add a cake_mq qdisc which installs cake instances on each hardware
queue on a multi-queue device.
This is just a copy of sch_mq that installs cake instead of the default
qdisc on each queue. Subsequent commits will add sharing of the config
between cake instances, as well as a multi-queue aware shaper algorithm.
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://patch.msgid.link/20260109-mq-cake-sub-qdisc-v8-3-8d613fece5d8@redhat.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Factor out all the user-configurable variables into a separate struct
and embed it into struct cake_sched_data. This is done in preparation
for sharing the configuration across multiple instances of cake in an mq
setup.
No functional change is intended with this patch.
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://patch.msgid.link/20260109-mq-cake-sub-qdisc-v8-2-8d613fece5d8@redhat.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
To enable the cake_mq qdisc to reuse code from the mq qdisc, export a
bunch of functions from sch_mq. Split common functionality out from some
functions so it can be composed with other code, and export other
functions wholesale. To discourage wanton reuse, put the symbols into a
new NET_SCHED_INTERNAL namespace, and a sch_priv.h header file.
No functional change intended.
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://patch.msgid.link/20260109-mq-cake-sub-qdisc-v8-1-8d613fece5d8@redhat.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
With CONFIG_CFI enabled, the kernel strictly enforces that indirect
function calls use a function pointer type that matches the
target function. As bpf_kfree_skb() signature differs from the
btf_dtor_kfunc_t pointer type used for the destructor calls in
bpf_obj_free_fields(), add a stub function with the correct type to
fix the type mismatch.
Signed-off-by: Sami Tolvanen <samitolvanen@google.com>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20260110082548.113748-8-samitolvanen@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Whenever a mirred redirect to self on egress happens, mirred allocates a
new skb (skb_to_send). The loop to self check was done after that
allocation, but was not freeing the newly allocated skb, causing a leak.
Fix this by moving the if-statement to before the allocation of the new
skb.
The issue was found by running the accompanying tdc test in 2/2
with config kmemleak enabled.
After a few minutes the kmemleak thread ran and reported the leak coming from
mirred.
Fixes: 1d856251a0 ("net/sched: act_mirred: fix loop detection")
Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20260101135608.253079-2-jhs@mojatatu.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Now that KF_TRUSTED_ARGS is the default for all kfuncs, remove the
explicit KF_TRUSTED_ARGS flag from all kfunc definitions and remove the
flag itself.
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Reviewed-by: Emil Tsalapatis <emil@etsalapatis.com>
Signed-off-by: Puranjay Mohan <puranjay@kernel.org>
Link: https://lore.kernel.org/r/20260102180038.2708325-3-puranjay@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
zdi-disclosures@trendmicro.com says:
The vulnerability is a race condition between `ets_qdisc_dequeue` and
`ets_qdisc_change`. It leads to UAF on `struct Qdisc` object.
Attacker requires the capability to create new user and network namespace
in order to trigger the bug.
See my additional commentary at the end of the analysis.
Analysis:
static int ets_qdisc_change(struct Qdisc *sch, struct nlattr *opt,
struct netlink_ext_ack *extack)
{
...
// (1) this lock is preventing .change handler (`ets_qdisc_change`)
//to race with .dequeue handler (`ets_qdisc_dequeue`)
sch_tree_lock(sch);
for (i = nbands; i < oldbands; i++) {
if (i >= q->nstrict && q->classes[i].qdisc->q.qlen)
list_del_init(&q->classes[i].alist);
qdisc_purge_queue(q->classes[i].qdisc);
}
WRITE_ONCE(q->nbands, nbands);
for (i = nstrict; i < q->nstrict; i++) {
if (q->classes[i].qdisc->q.qlen) {
// (2) the class is added to the q->active
list_add_tail(&q->classes[i].alist, &q->active);
q->classes[i].deficit = quanta[i];
}
}
WRITE_ONCE(q->nstrict, nstrict);
memcpy(q->prio2band, priomap, sizeof(priomap));
for (i = 0; i < q->nbands; i++)
WRITE_ONCE(q->classes[i].quantum, quanta[i]);
for (i = oldbands; i < q->nbands; i++) {
q->classes[i].qdisc = queues[i];
if (q->classes[i].qdisc != &noop_qdisc)
qdisc_hash_add(q->classes[i].qdisc, true);
}
// (3) the qdisc is unlocked, now dequeue can be called in parallel
// to the rest of .change handler
sch_tree_unlock(sch);
ets_offload_change(sch);
for (i = q->nbands; i < oldbands; i++) {
// (4) we're reducing the refcount for our class's qdisc and
// freeing it
qdisc_put(q->classes[i].qdisc);
// (5) If we call .dequeue between (4) and (5), we will have
// a strong UAF and we can control RIP
q->classes[i].qdisc = NULL;
WRITE_ONCE(q->classes[i].quantum, 0);
q->classes[i].deficit = 0;
gnet_stats_basic_sync_init(&q->classes[i].bstats);
memset(&q->classes[i].qstats, 0, sizeof(q->classes[i].qstats));
}
return 0;
}
Comment:
This happens because some of the classes have their qdiscs assigned to
NULL, but remain in the active list. This commit fixes this issue by always
removing the class from the active list before deleting and freeing its
associated qdisc
Reproducer Steps
(trimmed version of what was sent by zdi-disclosures@trendmicro.com)
```
DEV="${DEV:-lo}"
ROOT_HANDLE="${ROOT_HANDLE👎}"
BAND2_HANDLE="${BAND2_HANDLE:-20:}" # child under 1:2
PING_BYTES="${PING_BYTES:-48}"
PING_COUNT="${PING_COUNT:-200000}"
PING_DST="${PING_DST:-127.0.0.1}"
SLOW_TBF_RATE="${SLOW_TBF_RATE:-8bit}"
SLOW_TBF_BURST="${SLOW_TBF_BURST:-100b}"
SLOW_TBF_LAT="${SLOW_TBF_LAT:-1s}"
cleanup() {
tc qdisc del dev "$DEV" root 2>/dev/null
}
trap cleanup EXIT
ip link set "$DEV" up
tc qdisc del dev "$DEV" root 2>/dev/null || true
tc qdisc add dev "$DEV" root handle "$ROOT_HANDLE" ets bands 2 strict 2
tc qdisc add dev "$DEV" parent 1:2 handle "$BAND2_HANDLE" \
tbf rate "$SLOW_TBF_RATE" burst "$SLOW_TBF_BURST" latency "$SLOW_TBF_LAT"
tc filter add dev "$DEV" parent 1: protocol all prio 1 u32 match u32 0 0 flowid 1:2
tc -s qdisc ls dev $DEV
ping -I "$DEV" -f -c "$PING_COUNT" -s "$PING_BYTES" -W 0.001 "$PING_DST" \
>/dev/null 2>&1 &
tc qdisc change dev "$DEV" root handle "$ROOT_HANDLE" ets bands 2 strict 0
tc qdisc change dev "$DEV" root handle "$ROOT_HANDLE" ets bands 2 strict 2
tc -s qdisc ls dev $DEV
tc qdisc del dev "$DEV" parent 1:2 || true
tc -s qdisc ls dev $DEV
tc qdisc change dev "$DEV" root handle "$ROOT_HANDLE" ets bands 1 strict 1
```
KASAN report
```
==================================================================
BUG: KASAN: slab-use-after-free in ets_qdisc_dequeue+0x1071/0x11b0 kernel/net/sched/sch_ets.c:481
Read of size 8 at addr ffff8880502fc018 by task ping/12308
>
CPU: 0 UID: 0 PID: 12308 Comm: ping Not tainted 6.18.0-rc4-dirty #1 PREEMPT(full)
Hardware name: QEMU Ubuntu 25.04 PC (i440FX + PIIX, 1996), BIOS 1.16.3-debian-1.16.3-2 04/01/2014
Call Trace:
<IRQ>
__dump_stack kernel/lib/dump_stack.c:94
dump_stack_lvl+0x100/0x190 kernel/lib/dump_stack.c:120
print_address_description kernel/mm/kasan/report.c:378
print_report+0x156/0x4c9 kernel/mm/kasan/report.c:482
kasan_report+0xdf/0x110 kernel/mm/kasan/report.c:595
ets_qdisc_dequeue+0x1071/0x11b0 kernel/net/sched/sch_ets.c:481
dequeue_skb kernel/net/sched/sch_generic.c:294
qdisc_restart kernel/net/sched/sch_generic.c:399
__qdisc_run+0x1c9/0x1b00 kernel/net/sched/sch_generic.c:417
__dev_xmit_skb kernel/net/core/dev.c:4221
__dev_queue_xmit+0x2848/0x4410 kernel/net/core/dev.c:4729
dev_queue_xmit kernel/./include/linux/netdevice.h:3365
[...]
Allocated by task 17115:
kasan_save_stack+0x30/0x50 kernel/mm/kasan/common.c:56
kasan_save_track+0x14/0x30 kernel/mm/kasan/common.c:77
poison_kmalloc_redzone kernel/mm/kasan/common.c:400
__kasan_kmalloc+0xaa/0xb0 kernel/mm/kasan/common.c:417
kasan_kmalloc kernel/./include/linux/kasan.h:262
__do_kmalloc_node kernel/mm/slub.c:5642
__kmalloc_node_noprof+0x34e/0x990 kernel/mm/slub.c:5648
kmalloc_node_noprof kernel/./include/linux/slab.h:987
qdisc_alloc+0xb8/0xc30 kernel/net/sched/sch_generic.c:950
qdisc_create_dflt+0x93/0x490 kernel/net/sched/sch_generic.c:1012
ets_class_graft+0x4fd/0x800 kernel/net/sched/sch_ets.c:261
qdisc_graft+0x3e4/0x1780 kernel/net/sched/sch_api.c:1196
[...]
Freed by task 9905:
kasan_save_stack+0x30/0x50 kernel/mm/kasan/common.c:56
kasan_save_track+0x14/0x30 kernel/mm/kasan/common.c:77
__kasan_save_free_info+0x3b/0x70 kernel/mm/kasan/generic.c:587
kasan_save_free_info kernel/mm/kasan/kasan.h:406
poison_slab_object kernel/mm/kasan/common.c:252
__kasan_slab_free+0x5f/0x80 kernel/mm/kasan/common.c:284
kasan_slab_free kernel/./include/linux/kasan.h:234
slab_free_hook kernel/mm/slub.c:2539
slab_free kernel/mm/slub.c:6630
kfree+0x144/0x700 kernel/mm/slub.c:6837
rcu_do_batch kernel/kernel/rcu/tree.c:2605
rcu_core+0x7c0/0x1500 kernel/kernel/rcu/tree.c:2861
handle_softirqs+0x1ea/0x8a0 kernel/kernel/softirq.c:622
__do_softirq kernel/kernel/softirq.c:656
[...]
Commentary:
1. Maher Azzouzi working with Trend Micro Zero Day Initiative was reported as
the person who found the issue. I requested to get a proper email to add to the
reported-by tag but got no response. For this reason i will credit the person
i exchanged emails with i.e zdi-disclosures@trendmicro.com
2. Neither i nor Victor who did a much more thorough testing was able to
reproduce a UAF with the PoC or other approaches we tried. We were both able to
reproduce a null ptr deref. After exchange with zdi-disclosures@trendmicro.com
they sent a small change to be made to the code to add an extra delay which
was able to simulate the UAF. i.e, this:
qdisc_put(q->classes[i].qdisc);
mdelay(90);
q->classes[i].qdisc = NULL;
I was informed by Thomas Gleixner(tglx@linutronix.de) that adding delays was
acceptable approach for demonstrating the bug, quote:
"Adding such delays is common exploit validation practice"
The equivalent delay could happen "by virt scheduling the vCPU out, SMIs,
NMIs, PREEMPT_RT enabled kernel"
3. I asked the OP to test and report back but got no response and after a
few days gave up and proceeded to submit this fix.
Fixes: de6d25924c ("net/sched: sch_ets: don't peek at classes beyond 'nbands'")
Reported-by: zdi-disclosures@trendmicro.com
Tested-by: Victor Nogueira <victor@mojatatu.com>
Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/20251128151919.576920-1-jhs@mojatatu.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
In cake_drop(), qdisc_tree_reduce_backlog() is used to update the qlen
and backlog of the qdisc hierarchy. Its caller, cake_enqueue(), assumes
that the parent qdisc will enqueue the current packet. However, this
assumption breaks when cake_enqueue() returns NET_XMIT_CN: the parent
qdisc stops enqueuing current packet, leaving the tree qlen/backlog
accounting inconsistent. This mismatch can lead to a NULL dereference
(e.g., when the parent Qdisc is qfq_qdisc).
This patch computes the qlen/backlog delta in a more robust way by
observing the difference before and after the series of cake_drop()
calls, and then compensates the qdisc tree accounting if cake_enqueue()
returns NET_XMIT_CN.
To ensure correct compensation when ACK thinning is enabled, a new
variable is introduced to keep qlen unchanged.
Fixes: 15de71d06a ("net/sched: Make cake_enqueue return NET_XMIT_CN when past buffer_limit")
Signed-off-by: Xiang Mei <xmei5@asu.edu>
Reviewed-by: Toke Høiland-Jørgensen <toke@toke.dk>
Link: https://patch.msgid.link/20251128001415.377823-1-xmei5@asu.edu
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
-----BEGIN PGP SIGNATURE-----
iQFHBAABCgAxFiEEn/sM2K9nqF/8FWzzDHRl3/mQkZwFAmknHroTHG1rbEBwZW5n
dXRyb25peC5kZQAKCRAMdGXf+ZCRnENfCACmEL7NELoRi3RvXzlYOjDi0aEW2C79
k9hPBkzBor/W9JhBOf7+3nbDiL8nJdMQIs3FY6X/OSgyxEvGwgXb+PiFVBaHtDmx
PvJb7PjwNjpm4GnoPlQVUGBZxH4fDWz7JS34M1+BbY8k1ajN6ndHURHE6rtGMFIM
obClghS9i4K/Uww/bpSzj5cU4AmIuLIU8X6Oqq2X1PacgtGYP8WNh/eQ1xk03Nt8
RxrU0XTkPTsbS74Y4h+1bwNhxMwjLA97f2GgZ6GUa22lrzDnBpfqpBWSy0xIDJTe
GZgKpBOhbFSRjPH8f89cRXz47jvMfReC05ccuKC82+RK94Yge8F1YIxP
=HPfp
-----END PGP SIGNATURE-----
Merge tag 'linux-can-fixes-for-6.18-20251126' of git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can
Marc Kleine-Budde says:
====================
pull-request: can 2025-11-26
this is a pull request of 8 patches for net/main.
Seungjin Bae provides a patch for the kvaser_usb driver to fix a
potential infinite loop in the USB data stream command parser.
Thomas Mühlbacher's patch for the sja1000 driver IRQ handler's max
loop handling, that might lead to unhandled interrupts.
3 patches by me for the gs_usb driver fix handling of failed transmit
URBs and add checking of the actual length of received URBs before
accessing the data.
The next patch is by me and is a port of Thomas Mühlbacher's patch
(fix IRQ handler's max loop handling, that might lead to unhandled
interrupts.) to the sun4i_can driver.
Biju Das provides a patch for the rcar_canfd driver to fix the CAN-FD
mode setting.
The last patch is by Shaurya Rane for the em_canid filter to ensure
that the complete CAN frame is present in the linear data buffer
before accessing it.
* tag 'linux-can-fixes-for-6.18-20251126' of git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can:
net/sched: em_canid: fix uninit-value in em_canid_match
can: rcar_canfd: Fix CAN-FD mode as default
can: sun4i_can: sun4i_can_interrupt(): fix max irq loop handling
can: gs_usb: gs_usb_receive_bulk_callback(): check actual_length before accessing data
can: gs_usb: gs_usb_receive_bulk_callback(): check actual_length before accessing header
can: gs_usb: gs_usb_xmit_callback(): fix handling of failed transmitted URBs
can: sja1000: fix max irq loop handling
can: kvaser_usb: leaf: Fix potential infinite loop in command parsers
====================
Link: https://patch.msgid.link/20251126155713.217105-1-mkl@pengutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Use pskb_may_pull() to ensure a complete CAN frame is present in the
linear data buffer before reading the CAN ID. A simple skb->len check
is insufficient because it only verifies the total data length but does
not guarantee the data is present in skb->data (it could be in
fragments).
pskb_may_pull() both validates the length and pulls fragmented data
into the linear buffer if necessary, making it safe to directly
access skb->data.
Reported-by: syzbot+5d8269a1e099279152bc@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=5d8269a1e099279152bc
Fixes: f057bbb6f9 ("net: em_canid: Ematch rule to match CAN frames according to their identifiers")
Signed-off-by: Shaurya Rane <ssrane_b23@ee.vjti.ac.in>
Link: https://patch.msgid.link/20251126085718.50808-1-ssranevjti@gmail.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
cake, codel and fq_codel can drop many packets from dequeue().
Use qdisc_dequeue_drop() so that the freeing can happen
outside of the qdisc spinlock scope.
Add TCQ_F_DEQUEUE_DROPS to sch->flags.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20251121083256.674562-15-edumazet@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
prefetch the skb that we are likely to dequeue at the next dequeue().
Also call fq_dequeue_skb() a bit sooner in fq_dequeue().
This reduces the window between read of q.qlen and
changes of fields in the cache line that could be dirtied
by another cpu trying to queue a packet.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20251121083256.674562-10-edumazet@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Group together changes to qdisc fields to reduce chances of false sharing
if another cpu attempts to acquire the qdisc spinlock.
qdisc_qstats_backlog_dec(sch, skb);
sch->q.qlen--;
qdisc_bstats_update(sch, skb);
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20251121083256.674562-9-edumazet@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Use new qdisc_pkt_segs() to avoid a cache line miss in cake_enqueue()
for non GSO packets.
cake_overhead() does not have to recompute it.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20251121083256.674562-7-edumazet@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Avoid up to two cache line misses in qdisc dequeue() to fetch
skb_shinfo(skb)->gso_segs/gso_size while qdisc spinlock is held.
This gives a 5 % improvement in a TX intensive workload.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20251121083256.674562-6-edumazet@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
qdisc_pkt_len_init() is currently initalizing qdisc_skb_cb(skb)->pkt_len.
Add qdisc_skb_cb(skb)->pkt_segs initialization and rename this function
to qdisc_pkt_len_segs_init().
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20251121083256.674562-4-edumazet@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Add a new u16 field, next to pkt_len : pkt_segs
This will cache shinfo->gso_segs to speed up qdisc deqeue().
Move slave_dev_queue_mapping at the end of qdisc_skb_cb,
and move three bits from tc_skb_cb :
- post_ct
- post_ct_snat
- post_ct_dnat
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20251121083256.674562-2-edumazet@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Cross-merge networking fixes after downstream PR (net-6.18-rc7).
No conflicts, adjacent changes:
tools/testing/selftests/net/af_unix/Makefile
e1bb28bf13 ("selftest: af_unix: Add test for SO_PEEK_OFF.")
45a1cd8346 ("selftests: af_unix: Add tests for ECONNRESET and EOF semantics")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
-----BEGIN PGP SIGNATURE-----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=JJui
-----END PGP SIGNATURE-----
Merge tag 'bpf-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf
Pull bpf fixes from Alexei Starovoitov:
- Fix interaction between livepatch and BPF fexit programs (Song Liu)
With Steven and Masami acks.
- Fix stack ORC unwind from BPF kprobe_multi (Jiri Olsa)
With Steven and Masami acks.
- Fix out of bounds access in widen_imprecise_scalars() in the verifier
(Eduard Zingerman)
- Fix conflicts between MPTCP and BPF sockmap (Jiayuan Chen)
- Fix net_sched storage collision with BPF data_meta/data_end (Eric
Dumazet)
- Add _impl suffix to BPF kfuncs with implicit args to avoid breaking
them in bpf-next when KF_IMPLICIT_ARGS is added (Mykyta Yatsenko)
* tag 'bpf-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf:
selftests/bpf: Test widen_imprecise_scalars() with different stack depth
bpf: account for current allocated stack depth in widen_imprecise_scalars()
bpf: Add bpf_prog_run_data_pointers()
selftests/bpf: Add mptcp test with sockmap
mptcp: Fix proto fallback detection with BPF
mptcp: Disallow MPTCP subflows from sockmap
selftests/bpf: Add stacktrace ips test for raw_tp
selftests/bpf: Add stacktrace ips test for kprobe_multi/kretprobe_multi
x86/fgraph,bpf: Fix stack ORC unwind from kprobe_multi return probe
Revert "perf/x86: Always store regs->ip in perf_callchain_kernel()"
bpf: add _impl suffix for bpf_stream_vprintk() kfunc
bpf:add _impl suffix for bpf_task_work_schedule* kfuncs
selftests/bpf: Add tests for livepatch + bpf trampoline
ftrace: bpf: Fix IPMODIFY + DIRECT in modify_ftrace_direct()
ftrace: Fix BPF fexit with livepatch
syzbot found that cls_bpf_classify() is able to change
tc_skb_cb(skb)->drop_reason triggering a warning in sk_skb_reason_drop().
WARNING: CPU: 0 PID: 5965 at net/core/skbuff.c:1192 __sk_skb_reason_drop net/core/skbuff.c:1189 [inline]
WARNING: CPU: 0 PID: 5965 at net/core/skbuff.c:1192 sk_skb_reason_drop+0x76/0x170 net/core/skbuff.c:1214
struct tc_skb_cb has been added in commit ec624fe740 ("net/sched:
Extend qdisc control block with tc control block"), which added a wrong
interaction with db58ba4592 ("bpf: wire in data and data_end for
cls_act_bpf").
drop_reason was added later.
Add bpf_prog_run_data_pointers() helper to save/restore the net_sched
storage colliding with BPF data_meta/data_end.
Fixes: ec624fe740 ("net/sched: Extend qdisc control block with tc control block")
Reported-by: syzbot <syzkaller@googlegroups.com>
Closes: https://lore.kernel.org/netdev/6913437c.a70a0220.22f260.013b.GAE@google.com/
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Reviewed-by: Victor Nogueira <victor@mojatatu.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20251112125516.1563021-1-edumazet@google.com
Replace comma between expressions with semicolons.
Using a ',' in place of a ';' can have unintended side effects.
Although that is not the case here, it is seems best to use ';'
unless ',' is intended.
Found by inspection.
No functional change intended.
Compile tested only.
Signed-off-by: Chen Ni <nichen@iscas.ac.cn>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20251112072709.73755-1-nichen@iscas.ac.cn
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Cross-merge networking fixes after downstream PR (net-6.18-rc6).
No conflicts, adjacent changes in:
drivers/net/phy/micrel.c
96a9178a29 ("net: phy: micrel: lan8814 fix reset of the QSGMII interface")
61b7ade9ba ("net: phy: micrel: Add support for non PTP SKUs for lan8814")
and a trivial one in tools/testing/selftests/drivers/net/Makefile.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
After commit 100dfa74ca ("inet: dev_queue_xmit() llist adoption")
I started seeing many qdisc requeues on IDPF under high TX workload.
$ tc -s qd sh dev eth1 handle 1: ; sleep 1; tc -s qd sh dev eth1 handle 1:
qdisc mq 1: root
Sent 43534617319319 bytes 268186451819 pkt (dropped 0, overlimits 0 requeues 3532840114)
backlog 1056Kb 6675p requeues 3532840114
qdisc mq 1: root
Sent 43554665866695 bytes 268309964788 pkt (dropped 0, overlimits 0 requeues 3537737653)
backlog 781164b 4822p requeues 3537737653
This is caused by try_bulk_dequeue_skb() being only limited by BQL budget.
perf record -C120-239 -e qdisc:qdisc_dequeue sleep 1 ; perf script
...
netperf 75332 [146] 2711.138269: qdisc:qdisc_dequeue: dequeue ifindex=5 qdisc handle=0x80150000 parent=0x10013 txq_state=0x0 packets=1292 skbaddr=0xff378005a1e9f200
netperf 75332 [146] 2711.138953: qdisc:qdisc_dequeue: dequeue ifindex=5 qdisc handle=0x80150000 parent=0x10013 txq_state=0x0 packets=1213 skbaddr=0xff378004d607a500
netperf 75330 [144] 2711.139631: qdisc:qdisc_dequeue: dequeue ifindex=5 qdisc handle=0x80150000 parent=0x10013 txq_state=0x0 packets=1233 skbaddr=0xff3780046be20100
netperf 75333 [147] 2711.140356: qdisc:qdisc_dequeue: dequeue ifindex=5 qdisc handle=0x80150000 parent=0x10013 txq_state=0x0 packets=1093 skbaddr=0xff37800514845b00
netperf 75337 [151] 2711.141037: qdisc:qdisc_dequeue: dequeue ifindex=5 qdisc handle=0x80150000 parent=0x10013 txq_state=0x0 packets=1353 skbaddr=0xff37800460753300
netperf 75337 [151] 2711.141877: qdisc:qdisc_dequeue: dequeue ifindex=5 qdisc handle=0x80150000 parent=0x10013 txq_state=0x0 packets=1367 skbaddr=0xff378004e72c7b00
netperf 75330 [144] 2711.142643: qdisc:qdisc_dequeue: dequeue ifindex=5 qdisc handle=0x80150000 parent=0x10013 txq_state=0x0 packets=1202 skbaddr=0xff3780045bd60000
...
This is bad because :
1) Large batches hold one victim cpu for a very long time.
2) Driver often hit their own TX ring limit (all slots are used).
3) We call dev_requeue_skb()
4) Requeues are using a FIFO (q->gso_skb), breaking qdisc ability to
implement FQ or priority scheduling.
5) dequeue_skb() gets packets from q->gso_skb one skb at a time
with no xmit_more support. This is causing many spinlock games
between the qdisc and the device driver.
Requeues were supposed to be very rare, lets keep them this way.
Limit batch sizes to /proc/sys/net/core/dev_weight (default 64) as
__qdisc_run() was designed to use.
Fixes: 5772e9a346 ("qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Toke Høiland-Jørgensen <toke@redhat.com>
Acked-by: Jesper Dangaard Brouer <hawk@kernel.org>
Link: https://patch.msgid.link/20251109161215.2574081-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Fix a KMSAN kernel-infoleak detected by the syzbot .
[net?] KMSAN: kernel-infoleak in __skb_datagram_iter
In tcf_ife_dump(), the variable 'opt' was partially initialized using a
designatied initializer. While the padding bytes are reamined
uninitialized. nla_put() copies the entire structure into a
netlink message, these uninitialized bytes leaked to userspace.
Initialize the structure with memset before assigning its fields
to ensure all members and padding are cleared prior to beign copied.
This change silences the KMSAN report and prevents potential information
leaks from the kernel memory.
This fix has been tested and validated by syzbot. This patch closes the
bug reported at the following syzkaller link and ensures no infoleak.
Reported-by: syzbot+0c85cae3350b7d486aee@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=0c85cae3350b7d486aee
Tested-by: syzbot+0c85cae3350b7d486aee@syzkaller.appspotmail.com
Fixes: ef6980b6be ("introduce IFE action")
Signed-off-by: Ranganath V N <vnranganath.20@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20251109091336.9277-3-vnranganath.20@gmail.com
Acked-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Remove busylock spinlock and use a lockless list (llist)
to reduce spinlock contention to the minimum.
Idea is that only one cpu might spin on the qdisc spinlock,
while others simply add their skb in the llist.
After this patch, we get a 300 % improvement on heavy TX workloads.
- Sending twice the number of packets per second.
- While consuming 50 % less cycles.
Note that this also allows in the future to submit batches
to various qdisc->enqueue() methods.
Tested:
- Dual Intel(R) Xeon(R) 6985P-C (480 hyper threads).
- 100Gbit NIC, 30 TX queues with FQ packet scheduler.
- echo 64 >/sys/kernel/slab/skbuff_small_head/cpu_partial (avoid contention in mm)
- 240 concurrent "netperf -t UDP_STREAM -- -m 120 -n"
Before:
16 Mpps (41 Mpps if each thread is pinned to a different cpu)
vmstat 2 5
procs -----------memory---------- ---swap-- -----io---- -system-- ------cpu-----
r b swpd free buff cache si so bi bo in cs us sy id wa st
243 0 0 2368988672 51036 1100852 0 0 146 1 242 60 0 9 91 0 0
244 0 0 2368988672 51036 1100852 0 0 536 10 487745 14718 0 52 48 0 0
244 0 0 2368988672 51036 1100852 0 0 512 0 503067 46033 0 52 48 0 0
244 0 0 2368988672 51036 1100852 0 0 512 0 494807 12107 0 52 48 0 0
244 0 0 2368988672 51036 1100852 0 0 702 26 492845 10110 0 52 48 0 0
Lock contention (1 second sample taken on 8 cores)
perf lock record -C0-7 sleep 1; perf lock contention
contended total wait max wait avg wait type caller
442111 6.79 s 162.47 ms 15.35 us spinlock dev_hard_start_xmit+0xcd
5961 9.57 ms 8.12 us 1.60 us spinlock __dev_queue_xmit+0x3a0
244 560.63 us 7.63 us 2.30 us spinlock do_softirq+0x5b
13 25.09 us 3.21 us 1.93 us spinlock net_tx_action+0xf8
If netperf threads are pinned, spinlock stress is very high.
perf lock record -C0-7 sleep 1; perf lock contention
contended total wait max wait avg wait type caller
964508 7.10 s 147.25 ms 7.36 us spinlock dev_hard_start_xmit+0xcd
201 268.05 us 4.65 us 1.33 us spinlock __dev_queue_xmit+0x3a0
12 26.05 us 3.84 us 2.17 us spinlock do_softirq+0x5b
@__dev_queue_xmit_ns:
[256, 512) 21 | |
[512, 1K) 631 | |
[1K, 2K) 27328 |@ |
[2K, 4K) 265392 |@@@@@@@@@@@@@@@@ |
[4K, 8K) 417543 |@@@@@@@@@@@@@@@@@@@@@@@@@@ |
[8K, 16K) 826292 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@|
[16K, 32K) 733822 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ |
[32K, 64K) 19055 |@ |
[64K, 128K) 17240 |@ |
[128K, 256K) 25633 |@ |
[256K, 512K) 4 | |
After:
29 Mpps (57 Mpps if each thread is pinned to a different cpu)
vmstat 2 5
procs -----------memory---------- ---swap-- -----io---- -system-- ------cpu-----
r b swpd free buff cache si so bi bo in cs us sy id wa st
78 0 0 2369573632 32896 1350988 0 0 22 0 331 254 0 8 92 0 0
75 0 0 2369573632 32896 1350988 0 0 22 50 425713 280199 0 23 76 0 0
104 0 0 2369573632 32896 1350988 0 0 290 0 430238 298247 0 23 76 0 0
86 0 0 2369573632 32896 1350988 0 0 132 0 428019 291865 0 24 76 0 0
90 0 0 2369573632 32896 1350988 0 0 502 0 422498 278672 0 23 76 0 0
perf lock record -C0-7 sleep 1; perf lock contention
contended total wait max wait avg wait type caller
2524 116.15 ms 486.61 us 46.02 us spinlock __dev_queue_xmit+0x55b
5821 107.18 ms 371.67 us 18.41 us spinlock dev_hard_start_xmit+0xcd
2377 9.73 ms 35.86 us 4.09 us spinlock ___slab_alloc+0x4e0
923 5.74 ms 20.91 us 6.22 us spinlock ___slab_alloc+0x5c9
121 3.42 ms 193.05 us 28.24 us spinlock net_tx_action+0xf8
6 564.33 us 167.60 us 94.05 us spinlock do_softirq+0x5b
If netperf threads are pinned (~54 Mpps)
perf lock record -C0-7 sleep 1; perf lock contention
32907 316.98 ms 195.98 us 9.63 us spinlock dev_hard_start_xmit+0xcd
4507 61.83 ms 212.73 us 13.72 us spinlock __dev_queue_xmit+0x554
2781 23.53 ms 40.03 us 8.46 us spinlock ___slab_alloc+0x5c9
3554 18.94 ms 34.69 us 5.33 us spinlock ___slab_alloc+0x4e0
233 9.09 ms 215.70 us 38.99 us spinlock do_softirq+0x5b
153 930.66 us 48.67 us 6.08 us spinlock net_tx_action+0xfd
84 331.10 us 14.22 us 3.94 us spinlock ___slab_alloc+0x5c9
140 323.71 us 9.94 us 2.31 us spinlock ___slab_alloc+0x4e0
@__dev_queue_xmit_ns:
[128, 256) 1539830 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ |
[256, 512) 2299558 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@|
[512, 1K) 483936 |@@@@@@@@@@ |
[1K, 2K) 265345 |@@@@@@ |
[2K, 4K) 145463 |@@@ |
[4K, 8K) 54571 |@ |
[8K, 16K) 10270 | |
[16K, 32K) 9385 | |
[32K, 64K) 7749 | |
[64K, 128K) 26799 | |
[128K, 256K) 2665 | |
[256K, 512K) 665 | |
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Toke Høiland-Jørgensen <toke@redhat.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>
Tested-by: Jamal Hadi Salim <jhs@mojatatu.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20251014171907.3554413-7-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This reverts commits 0f022d32c3
and 44180feacc.
Prior patch in this series implemented loop detection
in act_mirred, we can remove q->owner to save some cycles
in the fast path.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Toke Høiland-Jørgensen <toke@redhat.com>
Reviewed-by: Victor Nogueira <victor@mojatatu.com>
Tested-by: Jamal Hadi Salim <jhs@mojatatu.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20251014171907.3554413-5-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Commit 0f022d32c3 ("net/sched: Fix mirred deadlock on device recursion")
added code in the fast path, even when act_mirred is not used.
Prepare its revert by implementing loop detection in act_mirred.
Adds an array of device pointers in struct netdev_xmit.
tcf_mirred_is_act_redirect() can detect if the array
already contains the target device.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Toke Høiland-Jørgensen <toke@redhat.com>
Tested-by: Jamal Hadi Salim <jhs@mojatatu.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20251014171907.3554413-4-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
tcfa_qstats is currently only used to hold drops and overlimits counters.
tcf_action_inc_drop_qstats() and tcf_action_inc_overlimit_qstats()
currently acquire a->tcfa_lock to increment these counters.
Switch to two atomic_t to get lock-free accounting.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20250901093141.2093176-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Also storing tcf_action into struct tcf_skbmod_params
makes sure there is no discrepancy in tcf_skbmod_act().
No longer block BH in tcf_skbmod_init() when acquiring tcf_lock.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250827125349.3505302-5-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Also storing tcf_action into struct tcf_tunnel_key_params
makes sure there is no discrepancy in tunnel_key_act().
No longer block BH in tunnel_key_init() when acquiring tcf_lock.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250827125349.3505302-4-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Also storing tcf_action into struct tcf_vlan_params
makes sure there is no discrepancy in tcf_vlan_act().
No longer block BH in tcf_vlan_init() when acquiring tcf_lock.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250827125349.3505302-3-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Followup of f45b45cbfa ("Merge branch
'net_sched-act-extend-rcu-use-in-dump-methods'")
We never grab tcf_lock from BH context in these modules:
act_connmark
act_csum
act_ct
act_ctinfo
act_mpls
act_nat
act_pedit
act_skbedit
No longer block BH when acquiring tcf_lock from init functions.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250827125349.3505302-2-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The WARN_ON trigger based on !cl->leaf.q->q.qlen is unnecessary in
htb_activate. htb_dequeue_tree already accounts for that scenario.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: William Liu <will@willsroot.io>
Reviewed-by: Savino Dicanosa <savy@syst3mfailure.io>
Link: https://patch.msgid.link/20250819033632.579854-1-will@willsroot.io
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The following setup can trigger a WARNING in htb_activate due to
the condition: !cl->leaf.q->q.qlen
tc qdisc del dev lo root
tc qdisc add dev lo root handle 1: htb default 1
tc class add dev lo parent 1: classid 1:1 \
htb rate 64bit
tc qdisc add dev lo parent 1:1 handle f: \
cake memlimit 1b
ping -I lo -f -c1 -s64 -W0.001 127.0.0.1
This is because the low memlimit leads to a low buffer_limit, which
causes packet dropping. However, cake_enqueue still returns
NET_XMIT_SUCCESS, causing htb_enqueue to call htb_activate with an
empty child qdisc. We should return NET_XMIT_CN when packets are
dropped from the same tin and flow.
I do not believe return value of NET_XMIT_CN is necessary for packet
drops in the case of ack filtering, as that is meant to optimize
performance, not to signal congestion.
Fixes: 046f6fd5da ("sched: Add Common Applications Kept Enhanced (cake) qdisc")
Signed-off-by: William Liu <will@willsroot.io>
Reviewed-by: Savino Dicanosa <savy@syst3mfailure.io>
Acked-by: Toke Høiland-Jørgensen <toke@toke.dk>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20250819033601.579821-1-will@willsroot.io
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When a user creates a dualpi2 qdisc it automatically sets a timer. This
timer will run constantly and update the qdisc's probability field.
The issue is that the timer acquires the qdisc root lock and runs in
hardirq. The qdisc root lock is also acquired in dev.c whenever a packet
arrives for this qdisc. Since the dualpi2 timer callback runs in hardirq,
it may interrupt the packet processing running in softirq. If that happens
and it runs on the same CPU, it will acquire the same lock and cause a
deadlock. The following splat shows up when running a kernel compiled with
lock debugging:
[ +0.000224] WARNING: inconsistent lock state
[ +0.000224] 6.16.0+ #10 Not tainted
[ +0.000169] --------------------------------
[ +0.000029] inconsistent {IN-HARDIRQ-W} -> {HARDIRQ-ON-W} usage.
[ +0.000000] ping/156 [HC0[0]:SC0[2]:HE1:SE0] takes:
[ +0.000000] ffff897841242110 (&sch->root_lock_key){?.-.}-{3:3}, at: __dev_queue_xmit+0x86d/0x1140
[ +0.000000] {IN-HARDIRQ-W} state was registered at:
[ +0.000000] lock_acquire.part.0+0xb6/0x220
[ +0.000000] _raw_spin_lock+0x31/0x80
[ +0.000000] dualpi2_timer+0x6f/0x270
[ +0.000000] __hrtimer_run_queues+0x1c5/0x360
[ +0.000000] hrtimer_interrupt+0x115/0x260
[ +0.000000] __sysvec_apic_timer_interrupt+0x6d/0x1a0
[ +0.000000] sysvec_apic_timer_interrupt+0x6e/0x80
[ +0.000000] asm_sysvec_apic_timer_interrupt+0x1a/0x20
[ +0.000000] pv_native_safe_halt+0xf/0x20
[ +0.000000] default_idle+0x9/0x10
[ +0.000000] default_idle_call+0x7e/0x1e0
[ +0.000000] do_idle+0x1e8/0x250
[ +0.000000] cpu_startup_entry+0x29/0x30
[ +0.000000] rest_init+0x151/0x160
[ +0.000000] start_kernel+0x6f3/0x700
[ +0.000000] x86_64_start_reservations+0x24/0x30
[ +0.000000] x86_64_start_kernel+0xc8/0xd0
[ +0.000000] common_startup_64+0x13e/0x148
[ +0.000000] irq event stamp: 6884
[ +0.000000] hardirqs last enabled at (6883): [<ffffffffa75700b3>] neigh_resolve_output+0x223/0x270
[ +0.000000] hardirqs last disabled at (6882): [<ffffffffa7570078>] neigh_resolve_output+0x1e8/0x270
[ +0.000000] softirqs last enabled at (6880): [<ffffffffa757006b>] neigh_resolve_output+0x1db/0x270
[ +0.000000] softirqs last disabled at (6884): [<ffffffffa755b533>] __dev_queue_xmit+0x73/0x1140
[ +0.000000]
other info that might help us debug this:
[ +0.000000] Possible unsafe locking scenario:
[ +0.000000] CPU0
[ +0.000000] ----
[ +0.000000] lock(&sch->root_lock_key);
[ +0.000000] <Interrupt>
[ +0.000000] lock(&sch->root_lock_key);
[ +0.000000]
*** DEADLOCK ***
[ +0.000000] 4 locks held by ping/156:
[ +0.000000] #0: ffff897842332e08 (sk_lock-AF_INET){+.+.}-{0:0}, at: raw_sendmsg+0x41e/0xf40
[ +0.000000] #1: ffffffffa816f880 (rcu_read_lock){....}-{1:3}, at: ip_output+0x2c/0x190
[ +0.000000] #2: ffffffffa816f880 (rcu_read_lock){....}-{1:3}, at: ip_finish_output2+0xad/0x950
[ +0.000000] #3: ffffffffa816f840 (rcu_read_lock_bh){....}-{1:3}, at: __dev_queue_xmit+0x73/0x1140
I am able to reproduce it consistently when running the following:
tc qdisc add dev lo handle 1: root dualpi2
ping -f 127.0.0.1
To fix it, make the timer run in softirq.
Fixes: 320d031ad6 ("sched: Struct definition and parsing of dualpi2 qdisc")
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Victor Nogueira <victor@mojatatu.com>
Link: https://patch.msgid.link/20250815135317.664993-1-victor@mojatatu.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This issue applies for the following qdiscs: hhf, fq, fq_codel, and
fq_pie, and occurs in their change handlers when adjusting to the new
limit. The problem is the following in the values passed to the
subsequent qdisc_tree_reduce_backlog call given a tbf parent:
When the tbf parent runs out of tokens, skbs of these qdiscs will
be placed in gso_skb. Their peek handlers are qdisc_peek_dequeued,
which accounts for both qlen and backlog. However, in the case of
qdisc_dequeue_internal, ONLY qlen is accounted for when pulling
from gso_skb. This means that these qdiscs are missing a
qdisc_qstats_backlog_dec when dropping packets to satisfy the
new limit in their change handlers.
One can observe this issue with the following (with tc patched to
support a limit of 0):
export TARGET=fq
tc qdisc del dev lo root
tc qdisc add dev lo root handle 1: tbf rate 8bit burst 100b latency 1ms
tc qdisc replace dev lo handle 3: parent 1:1 $TARGET limit 1000
echo ''; echo 'add child'; tc -s -d qdisc show dev lo
ping -I lo -f -c2 -s32 -W0.001 127.0.0.1 2>&1 >/dev/null
echo ''; echo 'after ping'; tc -s -d qdisc show dev lo
tc qdisc change dev lo handle 3: parent 1:1 $TARGET limit 0
echo ''; echo 'after limit drop'; tc -s -d qdisc show dev lo
tc qdisc replace dev lo handle 2: parent 1:1 sfq
echo ''; echo 'post graft'; tc -s -d qdisc show dev lo
The second to last show command shows 0 packets but a positive
number (74) of backlog bytes. The problem becomes clearer in the
last show command, where qdisc_purge_queue triggers
qdisc_tree_reduce_backlog with the positive backlog and causes an
underflow in the tbf parent's backlog (4096 Mb instead of 0).
To fix this issue, the codepath for all clients of qdisc_dequeue_internal
has been simplified: codel, pie, hhf, fq, fq_pie, and fq_codel.
qdisc_dequeue_internal handles the backlog adjustments for all cases that
do not directly use the dequeue handler.
The old fq_codel_change limit adjustment loop accumulated the arguments to
the subsequent qdisc_tree_reduce_backlog call through the cstats field.
However, this is confusing and error prone as fq_codel_dequeue could also
potentially mutate this field (which qdisc_dequeue_internal calls in the
non gso_skb case), so we have unified the code here with other qdiscs.
Fixes: 2d3cbfd6d5 ("net_sched: Flush gso_skb list too during ->change()")
Fixes: 4b549a2ef4 ("fq_codel: Fair Queue Codel AQM")
Fixes: 10239edf86 ("net-qdisc-hhf: Heavy-Hitter Filter (HHF) qdisc")
Signed-off-by: William Liu <will@willsroot.io>
Reviewed-by: Savino Dicanosa <savy@syst3mfailure.io>
Link: https://patch.msgid.link/20250812235725.45243-1-will@willsroot.io
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Cross-merge networking fixes after downstream PR (net-6.17-rc2).
No conflicts.
Adjacent changes:
drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c
d7a276a576 ("net: stmmac: rk: convert to suspend()/resume() methods")
de1e963ad0 ("net: stmmac: rk: put the PHY clock on remove")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The call to nla_strscpy() already zero-pads the tail of the destination
buffer which makes the additional memset(0) call redundant. Remove it.
Signed-off-by: Thorsten Blum <thorsten.blum@linux.dev>
Reviewed-by: Joe Damato <joe@dama.to>
Link: https://patch.msgid.link/20250811164039.43250-1-thorsten.blum@linux.dev
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
TCA_MQPRIO_TC_ENTRY_INDEX is validated using
NLA_POLICY_MAX(NLA_U32, TC_QOPT_MAX_QUEUE), which allows the value
TC_QOPT_MAX_QUEUE (16). This leads to a 4-byte out-of-bounds stack
write in the fp[] array, which only has room for 16 elements (0–15).
Fix this by changing the policy to allow only up to TC_QOPT_MAX_QUEUE - 1.
Fixes: f62af20bed ("net/sched: mqprio: allow per-TC user input of FP adminStatus")
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Maher Azzouzi <maherazz04@gmail.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Link: https://patch.msgid.link/20250802001857.2702497-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Syzbot reported a WARNING in taprio_get_start_time().
When link speed is 470,589 or greater, q->picos_per_byte becomes too
small, causing length_to_duration(q, ETH_ZLEN) to return zero.
This zero value leads to validation failures in fill_sched_entry() and
parse_taprio_schedule(), allowing arbitrary values to be assigned to
entry->interval and cycle_time. As a result, sched->cycle can become zero.
Since SPEED_800000 is the largest defined speed in
include/uapi/linux/ethtool.h, this issue can occur in realistic scenarios.
To ensure length_to_duration() returns a non-zero value for minimum-sized
Ethernet frames (ETH_ZLEN = 60), picos_per_byte must be at least 17
(60 * 17 > PSEC_PER_NSEC which is 1000).
This patch enforces a minimum value of 17 for picos_per_byte when the
calculated value would be lower, and adds a warning message to inform
users that scheduling accuracy may be affected at very high link speeds.
Fixes: fb66df20a7 ("net/sched: taprio: extend minimum interval restriction to entire cycle too")
Reported-by: syzbot+398e1ee4ca2cac05fddb@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=398e1ee4ca2cac05fddb
Signed-off-by: Takamitsu Iwai <takamitz@amazon.co.jp>
Link: https://patch.msgid.link/20250728173149.45585-1-takamitz@amazon.co.jp
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
-----BEGIN PGP SIGNATURE-----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=/O3j
-----END PGP SIGNATURE-----
Merge tag 'bpf-next-6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next
Pull bpf updates from Alexei Starovoitov:
- Remove usermode driver (UMD) framework (Thomas Weißschuh)
- Introduce Strongly Connected Component (SCC) in the verifier to
detect loops and refine register liveness (Eduard Zingerman)
- Allow 'void *' cast using bpf_rdonly_cast() and corresponding
'__arg_untrusted' for global function parameters (Eduard Zingerman)
- Improve precision for BPF_ADD and BPF_SUB operations in the verifier
(Harishankar Vishwanathan)
- Teach the verifier that constant pointer to a map cannot be NULL
(Ihor Solodrai)
- Introduce BPF streams for error reporting of various conditions
detected by BPF runtime (Kumar Kartikeya Dwivedi)
- Teach the verifier to insert runtime speculation barrier (lfence on
x86) to mitigate speculative execution instead of rejecting the
programs (Luis Gerhorst)
- Various improvements for 'veristat' (Mykyta Yatsenko)
- For CONFIG_DEBUG_KERNEL config warn on internal verifier errors to
improve bug detection by syzbot (Paul Chaignon)
- Support BPF private stack on arm64 (Puranjay Mohan)
- Introduce bpf_cgroup_read_xattr() kfunc to read xattr of cgroup's
node (Song Liu)
- Introduce kfuncs for read-only string opreations (Viktor Malik)
- Implement show_fdinfo() for bpf_links (Tao Chen)
- Reduce verifier's stack consumption (Yonghong Song)
- Implement mprog API for cgroup-bpf programs (Yonghong Song)
* tag 'bpf-next-6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next: (192 commits)
selftests/bpf: Migrate fexit_noreturns case into tracing_failure test suite
selftests/bpf: Add selftest for attaching tracing programs to functions in deny list
bpf: Add log for attaching tracing programs to functions in deny list
bpf: Show precise rejected function when attaching fexit/fmod_ret to __noreturn functions
bpf: Fix various typos in verifier.c comments
bpf: Add third round of bounds deduction
selftests/bpf: Test invariants on JSLT crossing sign
selftests/bpf: Test cross-sign 64bits range refinement
selftests/bpf: Update reg_bound range refinement logic
bpf: Improve bounds when s64 crosses sign boundary
bpf: Simplify bounds refinement from s32
selftests/bpf: Enable private stack tests for arm64
bpf, arm64: JIT support for private stack
bpf: Move bpf_jit_get_prog_name() to core.c
bpf, arm64: Fix fp initialization for exception boundary
umd: Remove usermode driver framework
bpf/preload: Don't select USERMODE_DRIVER
selftests/bpf: Fix test dynptr/test_dynptr_memset_xdp_chunks failure
selftests/bpf: Fix test dynptr/test_dynptr_copy_xdp failure
selftests/bpf: Increase xdp data size for arm64 64K page size
...
Both taprio and mqprio have code to validate respective entry index
attributes. The validation is indented to ensure that the attribute is
present, and that it's value is in range, and that each value is only
used once.
The purpose of this patch is to align the implementation of taprio with
that of mqprio as there seems to be no good reason for them to differ.
For one thing, this way, bugs will be present in both or neither.
As a follow-up some consideration could be given to a common function
used by both sch.
No functional change intended.
Except of tdc run: the results of the taprio tests
# ok 81 ba39 - Add taprio Qdisc to multi-queue device (8 queues)
# ok 82 9462 - Add taprio Qdisc with multiple sched-entry
# ok 83 8d92 - Add taprio Qdisc with txtime-delay
# ok 84 d092 - Delete taprio Qdisc with valid handle
# ok 85 8471 - Show taprio class
# ok 86 0a85 - Add taprio Qdisc to single-queue device
# ok 87 6f62 - Add taprio Qdisc with too short interval
# ok 88 831f - Add taprio Qdisc with too short cycle-time
# ok 89 3e1e - Add taprio Qdisc with an invalid cycle-time
# ok 90 39b4 - Reject grafting taprio as child qdisc of software taprio
# ok 91 e8a1 - Reject grafting taprio as child qdisc of offloaded taprio
# ok 92 a7bf - Graft cbs as child of software taprio
# ok 93 6a83 - Graft cbs as child of offloaded taprio
Cc: Vladimir Oltean <vladimir.oltean@nxp.com>
Cc: Maher Azzouzi <maherazz04@gmail.com>
Link: https://lore.kernel.org/netdev/20250723125521.GA2459@horms.kernel.org/
Signed-off-by: Simon Horman <horms@kernel.org>
Reviewed-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Link: https://patch.msgid.link/20250725-taprio-idx-parse-v1-1-b582fffcde37@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Currently, packets dropped by pfifo_fast due to queue overflow are
marked with a generic SKB_DROP_REASON_QDISC_DROP in __dev_xmit_skb().
This patch adds explicit drop reason SKB_DROP_REASON_QDISC_OVERLIMIT
for queue-full cases, providing better distinction from other qdisc drops.
Signed-off-by: Fan Yu <fan.yu9@zte.com.cn>
Reviewed-by: Cong Wang <xiyou.wangcong@gmail.com>
Link: https://patch.msgid.link/20250724212837119BP9HOs0ibXDRWgsXMMir7@zte.com.cn
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
DualPI2 provides L4S-type low latency & loss to traffic that uses a
scalable congestion controller (e.g. TCP-Prague, DCTCP) without
degrading the performance of 'classic' traffic (e.g. Reno,
Cubic etc.). It is to be the reference implementation of IETF RFC9332
DualQ Coupled AQM (https://datatracker.ietf.org/doc/html/rfc9332).
Note that creating two independent queues cannot meet the goal of
DualPI2 mentioned in RFC9332: "...to preserve fairness between
ECN-capable and non-ECN-capable traffic." Further, it could even
lead to starvation of Classic traffic, which is also inconsistent
with the requirements in RFC9332: "...although priority MUST be
bounded in order not to starve Classic traffic." DualPI2 is
designed to maintain approximate per-flow fairness on L-queue and
C-queue by forming a single qdisc using the coupling factor and
scheduler between two queues.
The qdisc provides two queues called low latency and classic. It
classifies packets based on the ECN field in the IP headers. By
default it directs non-ECN and ECT(0) into the classic queue and
ECT(1) and CE into the low latency queue, as per the IETF spec.
Each queue runs its own AQM:
* The classic AQM is called PI2, which is similar to the PIE AQM but
more responsive and simpler. Classic traffic requires a decent
target queue (default 15ms for Internet deployment) to fully
utilize the link and to avoid high drop rates.
* The low latency AQM is, by default, a very shallow ECN marking
threshold (1ms) similar to that used for DCTCP.
The DualQ isolates the low queuing delay of the Low Latency queue
from the larger delay of the 'Classic' queue. However, from a
bandwidth perspective, flows in either queue will share out the link
capacity as if there was just a single queue. This bandwidth pooling
effect is achieved by coupling together the drop and ECN-marking
probabilities of the two AQMs.
The PI2 AQM has two main parameters in addition to its target delay.
The integral gain factor alpha is used to slowly correct any persistent
standing queue error from the target delay, while the proportional gain
factor beta is used to quickly compensate for queue changes (growth or
shrinkage). Either alpha and beta are given as a parameter, or they can
be calculated by tc from alternative typical and maximum RTT parameters.
Internally, the output of a linear Proportional Integral (PI)
controller is used for both queues. This output is squared to
calculate the drop or ECN-marking probability of the classic queue.
This counterbalances the square-root rate equation of Reno/Cubic,
which is the trick that balances flow rates across the queues. For
the ECN-marking probability of the low latency queue, the output of
the base AQM is multiplied by a coupling factor. This determines the
balance between the flow rates in each queue. The default setting
makes the flow rates roughly equal, which should be generally
applicable.
If DUALPI2 AQM has detected overload (due to excessive non-responsive
traffic in either queue), it will switch to signaling congestion
solely using drop, irrespective of the ECN field. Alternatively, it
can be configured to limit the drop probability and let the queue
grow and eventually overflow (like tail-drop).
GSO splitting in DUALPI2 is configurable from userspace while the
default behavior is to split gso. When running DUALPI2 at unshaped
10gigE with 4 download streams test, splitting gso apart results in
halving the latency with no loss in throughput:
Summary of tcp_4down run 'no_split_gso':
avg median # data pts
Ping (ms) ICMP : 0.53 0.30 ms 350
TCP download avg : 2326.86 N/A Mbits/s 350
TCP download sum : 9307.42 N/A Mbits/s 350
TCP download::1 : 2672.99 2568.73 Mbits/s 350
TCP download::2 : 2586.96 2570.51 Mbits/s 350
TCP download::3 : 1786.26 1798.82 Mbits/s 350
TCP download::4 : 2261.21 2309.49 Mbits/s 350
Summart of tcp_4down run 'split_gso':
avg median # data pts
Ping (ms) ICMP : 0.22 0.23 ms 350
TCP download avg : 2335.02 N/A Mbits/s 350
TCP download sum : 9340.09 N/A Mbits/s 350
TCP download::1 : 2335.30 2334.22 Mbits/s 350
TCP download::2 : 2334.72 2334.20 Mbits/s 350
TCP download::3 : 2335.28 2334.58 Mbits/s 350
TCP download::4 : 2334.79 2334.39 Mbits/s 350
A similar result is observed when running DUALPI2 at unshaped 1gigE
with 1 download stream test:
Summary of tcp_1down run 'no_split_gso':
avg median # data pts
Ping (ms) ICMP : 1.13 1.25 ms 350
TCP download : 941.41 941.46 Mbits/s 350
Summart of tcp_1down run 'split_gso':
avg median # data pts
Ping (ms) ICMP : 0.51 0.55 ms 350
TCP download : 941.41 941.45 Mbits/s 350
Additional details can be found in the draft:
https://datatracker.ietf.org/doc/html/rfc9332
Signed-off-by: Koen De Schepper <koen.de_schepper@nokia-bell-labs.com>
Co-developed-by: Olga Albisser <olga@albisser.org>
Signed-off-by: Olga Albisser <olga@albisser.org>
Co-developed-by: Olivier Tilmans <olivier.tilmans@nokia.com>
Signed-off-by: Olivier Tilmans <olivier.tilmans@nokia.com>
Co-developed-by: Henrik Steen <henrist@henrist.net>
Signed-off-by: Henrik Steen <henrist@henrist.net>
Co-developed-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
Signed-off-by: Bob Briscoe <research@bobbriscoe.net>
Signed-off-by: Ilpo Järvinen <ij@kernel.org>
Acked-by: Dave Taht <dave.taht@gmail.com>
Link: https://patch.msgid.link/20250722095915.24485-4-chia-yu.chang@nokia-bell-labs.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The configuration and statistics dump of the DualPI2 Qdisc provides
information related to both queues, such as packet numbers and queuing
delays in the L-queue and C-queue, as well as general information such as
probability value, WRR credits, memory usage, packet marking counters, max
queue size, etc.
The following patch includes enqueue/dequeue for DualPI2.
Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
Link: https://patch.msgid.link/20250722095915.24485-3-chia-yu.chang@nokia-bell-labs.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
DualPI2 is the reference implementation of IETF RFC9332 DualQ Coupled
AQM (https://datatracker.ietf.org/doc/html/rfc9332) providing two
queues called low latency (L-queue) and classic (C-queue). By default,
it enqueues non-ECN and ECT(0) packets into the C-queue and ECT(1) and
CE packets into the low latency queue (L-queue), as per IETF RFC9332 spec.
This patch defines the dualpi2 Qdisc structure and parsing, and the
following two patches include dumping and enqueue/dequeue for the DualPI2.
Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
Link: https://patch.msgid.link/20250722095915.24485-2-chia-yu.chang@nokia-bell-labs.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
might_sleep could be trigger in the atomic context in qfq_delete_class.
qfq_destroy_class was moved into atomic context locked
by sch_tree_lock to avoid a race condition bug on
qfq_aggregate. However, might_sleep could be triggered by
qfq_destroy_class, which introduced sleeping in atomic context (path:
qfq_destroy_class->qdisc_put->__qdisc_destroy->lockdep_unregister_key
->might_sleep).
Considering the race is on the qfq_aggregate objects, keeping
qfq_rm_from_agg in the lock but moving the left part out can solve
this issue.
Fixes: 5e28d5a3f7 ("net/sched: sch_qfq: Fix race condition on qfq_aggregate")
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Xiang Mei <xmei5@asu.edu>
Link: https://patch.msgid.link/4a04e0cc-a64b-44e7-9213-2880ed641d77@sabinyo.mountain
Reviewed-by: Cong Wang <xiyou.wangcong@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://patch.msgid.link/20250717230128.159766-1-xmei5@asu.edu
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
htb_lookup_leaf has a BUG_ON that can trigger with the following:
tc qdisc del dev lo root
tc qdisc add dev lo root handle 1: htb default 1
tc class add dev lo parent 1: classid 1:1 htb rate 64bit
tc qdisc add dev lo parent 1:1 handle 2: netem
tc qdisc add dev lo parent 2:1 handle 3: blackhole
ping -I lo -c1 -W0.001 127.0.0.1
The root cause is the following:
1. htb_dequeue calls htb_dequeue_tree which calls the dequeue handler on
the selected leaf qdisc
2. netem_dequeue calls enqueue on the child qdisc
3. blackhole_enqueue drops the packet and returns a value that is not
just NET_XMIT_SUCCESS
4. Because of this, netem_dequeue calls qdisc_tree_reduce_backlog, and
since qlen is now 0, it calls htb_qlen_notify -> htb_deactivate ->
htb_deactiviate_prios -> htb_remove_class_from_row -> htb_safe_rb_erase
5. As this is the only class in the selected hprio rbtree,
__rb_change_child in __rb_erase_augmented sets the rb_root pointer to
NULL
6. Because blackhole_dequeue returns NULL, netem_dequeue returns NULL,
which causes htb_dequeue_tree to call htb_lookup_leaf with the same
hprio rbtree, and fail the BUG_ON
The function graph for this scenario is shown here:
0) | htb_enqueue() {
0) + 13.635 us | netem_enqueue();
0) 4.719 us | htb_activate_prios();
0) # 2249.199 us | }
0) | htb_dequeue() {
0) 2.355 us | htb_lookup_leaf();
0) | netem_dequeue() {
0) + 11.061 us | blackhole_enqueue();
0) | qdisc_tree_reduce_backlog() {
0) | qdisc_lookup_rcu() {
0) 1.873 us | qdisc_match_from_root();
0) 6.292 us | }
0) 1.894 us | htb_search();
0) | htb_qlen_notify() {
0) 2.655 us | htb_deactivate_prios();
0) 6.933 us | }
0) + 25.227 us | }
0) 1.983 us | blackhole_dequeue();
0) + 86.553 us | }
0) # 2932.761 us | qdisc_warn_nonwc();
0) | htb_lookup_leaf() {
0) | BUG_ON();
------------------------------------------
The full original bug report can be seen here [1].
We can fix this just by returning NULL instead of the BUG_ON,
as htb_dequeue_tree returns NULL when htb_lookup_leaf returns
NULL.
[1] https://lore.kernel.org/netdev/pF5XOOIim0IuEfhI-SOxTgRvNoDwuux7UHKnE_Y5-zVd4wmGvNk2ceHjKb8ORnzw0cGwfmVu42g9dL7XyJLf1NEzaztboTWcm0Ogxuojoeo=@willsroot.io/
Fixes: 512bb43eb5 ("pkt_sched: sch_htb: Optimize WARN_ONs in htb_dequeue_tree() etc.")
Signed-off-by: William Liu <will@willsroot.io>
Signed-off-by: Savino Dicanosa <savy@syst3mfailure.io>
Link: https://patch.msgid.link/20250717022816.221364-1-will@willsroot.io
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
A race condition can occur when 'agg' is modified in qfq_change_agg
(called during qfq_enqueue) while other threads access it
concurrently. For example, qfq_dump_class may trigger a NULL
dereference, and qfq_delete_class may cause a use-after-free.
This patch addresses the issue by:
1. Moved qfq_destroy_class into the critical section.
2. Added sch_tree_lock protection to qfq_dump_class and
qfq_dump_class_stats.
Fixes: 462dbc9101 ("pkt_sched: QFQ Plus: fair-queueing service at DRR cost")
Signed-off-by: Xiang Mei <xmei5@asu.edu>
Reviewed-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Also storing tcf_action into struct tcf_skbedit_params
makes sure there is no discrepancy in tcf_skbedit_act().
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20250709090204.797558-12-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Also storing tcf_action into struct tcf_police_params
makes sure there is no discrepancy in tcf_police_act().
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20250709090204.797558-11-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Also storing tcf_action into struct tcf_pedit_params
makes sure there is no discrepancy in tcf_pedit_act().
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20250709090204.797558-10-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Also storing tcf_action into struct tcf_nat_params
makes sure there is no discrepancy in tcf_nat_act().
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20250709090204.797558-9-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Also storing tcf_action into struct tcf_mpls_params
makes sure there is no discrepancy in tcf_mpls_act().
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20250709090204.797558-8-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Also storing tcf_action into struct tcf_ctinfo_params
makes sure there is no discrepancy in tcf_ctinfo_act().
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20250709090204.797558-7-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Commit 21c167aa0b ("net/sched: act_ctinfo: use percpu stats")
missed that stats_dscp_set, stats_dscp_error and stats_cpmark_set
might be written (and read) locklessly.
Use atomic64_t for these three fields, I doubt act_ctinfo is used
heavily on big SMP hosts anyway.
Fixes: 24ec483cec ("net: sched: Introduce act_ctinfo action")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Pedro Tammela <pctammela@mojatatu.com>
Link: https://patch.msgid.link/20250709090204.797558-6-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Also storing tcf_action into struct tcf_ct_params
makes sure there is no discrepancy in tcf_ct_act().
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20250709090204.797558-5-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Also storing tcf_action into struct tcf_csum_params
makes sure there is no discrepancy in tcf_csum_act().
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20250709090204.797558-4-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Also storing tcf_action into struct tcf_connmark_parms
makes sure there is no discrepancy in tcf_connmark_act().
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20250709090204.797558-3-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
netem_enqueue's duplication prevention logic breaks when a netem
resides in a qdisc tree with other netems - this can lead to a
soft lockup and OOM loop in netem_dequeue, as seen in [1].
Ensure that a duplicating netem cannot exist in a tree with other
netems.
Previous approaches suggested in discussions in chronological order:
1) Track duplication status or ttl in the sk_buff struct. Considered
too specific a use case to extend such a struct, though this would
be a resilient fix and address other previous and potential future
DOS bugs like the one described in loopy fun [2].
2) Restrict netem_enqueue recursion depth like in act_mirred with a
per cpu variable. However, netem_dequeue can call enqueue on its
child, and the depth restriction could be bypassed if the child is a
netem.
3) Use the same approach as in 2, but add metadata in netem_skb_cb
to handle the netem_dequeue case and track a packet's involvement
in duplication. This is an overly complex approach, and Jamal
notes that the skb cb can be overwritten to circumvent this
safeguard.
4) Prevent the addition of a netem to a qdisc tree if its ancestral
path contains a netem. However, filters and actions can cause a
packet to change paths when re-enqueued to the root from netem
duplication, leading us to the current solution: prevent a
duplicating netem from inhabiting the same tree as other netems.
[1] https://lore.kernel.org/netdev/8DuRWwfqjoRDLDmBMlIfbrsZg9Gx50DHJc1ilxsEBNe2D6NMoigR_eIRIG0LOjMc3r10nUUZtArXx4oZBIdUfZQrwjcQhdinnMis_0G7VEk=@willsroot.io/
[2] https://lwn.net/Articles/719297/
Fixes: 0afb51e728 ("[PKT_SCHED]: netem: reinsert for duplication")
Reported-by: William Liu <will@willsroot.io>
Reported-by: Savino Dicanosa <savy@syst3mfailure.io>
Signed-off-by: William Liu <will@willsroot.io>
Signed-off-by: Savino Dicanosa <savy@syst3mfailure.io>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20250708164141.875402-1-will@willsroot.io
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
To prevent a potential crash in agg_dequeue (net/sched/sch_qfq.c)
when cl->qdisc->ops->peek(cl->qdisc) returns NULL, we check the return
value before using it, similar to the existing approach in sch_hfsc.c.
To avoid code duplication, the following changes are made:
1. Changed qdisc_warn_nonwc(include/net/pkt_sched.h) into a static
inline function.
2. Moved qdisc_peek_len from net/sched/sch_hfsc.c to
include/net/pkt_sched.h so that sch_qfq can reuse it.
3. Applied qdisc_peek_len in agg_dequeue to avoid crashing.
Signed-off-by: Xiang Mei <xmei5@asu.edu>
Reviewed-by: Cong Wang <xiyou.wangcong@gmail.com>
Link: https://patch.msgid.link/20250705212143.3982664-1-xmei5@asu.edu
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Refine qdisc_pkt_len_init to include headers up through
the inner transport header when computing header size
for encapsulations. Also refine net/sched/sch_cake.c
borrowed from qdisc_pkt_len_init().
Signed-off-by: Fengyuan Gong <gfengyuan@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://patch.msgid.link/20250702160741.1204919-1-gfengyuan@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
tc_action_net_exit() got an rtnl exclusion in commit
a159d3c4b8 ("net_sched: acquire RTNL in tc_action_net_exit()")
Since then, commit 16af606739 ("net: sched: implement reference
counted action release") made this RTNL exclusion obsolete for
most cases.
Only tcf_action_offload_del() might still require it.
Move the rtnl locking into tcf_idrinfo_destroy() when
an offload action is found.
Most netns do not have actions, yet deleting them is adding a lot
of pressure on RTNL, which is for many the most contended mutex
in the kernel.
We are moving to a per-netns 'rtnl', so tc_action_net_exit()
will not be able to grab 'rtnl' a single time for a batch of netns.
Before the patch:
perf probe -a rtnl_lock
perf record -e probe:rtnl_lock -a /bin/bash -c 'unshare -n "/bin/true"; sleep 1'
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.305 MB perf.data (25 samples) ]
After the patch:
perf record -e probe:rtnl_lock -a /bin/bash -c 'unshare -n "/bin/true"; sleep 1'
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.304 MB perf.data (9 samples) ]
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Vlad Buslov <vladbu@nvidia.com>
Cc: Jiri Pirko <jiri@resnulli.us>
Cc: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Link: https://patch.msgid.link/20250702071230.1892674-1-edumazet@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Certain classful qdiscs may invoke their classes' dequeue handler on an
enqueue operation. This may unexpectedly empty the child qdisc and thus
make an in-flight class passive via qlen_notify(). Most qdiscs do not
expect such behaviour at this point in time and may re-activate the
class eventually anyways which will lead to a use-after-free.
The referenced fix commit attempted to fix this behavior for the HFSC
case by moving the backlog accounting around, though this turned out to
be incomplete since the parent's parent may run into the issue too.
The following reproducer demonstrates this use-after-free:
tc qdisc add dev lo root handle 1: drr
tc filter add dev lo parent 1: basic classid 1:1
tc class add dev lo parent 1: classid 1:1 drr
tc qdisc add dev lo parent 1:1 handle 2: hfsc def 1
tc class add dev lo parent 2: classid 2:1 hfsc rt m1 8 d 1 m2 0
tc qdisc add dev lo parent 2:1 handle 3: netem
tc qdisc add dev lo parent 3:1 handle 4: blackhole
echo 1 | socat -u STDIN UDP4-DATAGRAM:127.0.0.1:8888
tc class delete dev lo classid 1:1
echo 1 | socat -u STDIN UDP4-DATAGRAM:127.0.0.1:8888
Since backlog accounting issues leading to a use-after-frees on stale
class pointers is a recurring pattern at this point, this patch takes
a different approach. Instead of trying to fix the accounting, the patch
ensures that qdisc_tree_reduce_backlog always calls qlen_notify when
the child qdisc is empty. This solves the problem because deletion of
qdiscs always involves a call to qdisc_reset() and / or
qdisc_purge_queue() which ultimately resets its qlen to 0 thus causing
the following qdisc_tree_reduce_backlog() to report to the parent. Note
that this may call qlen_notify on passive classes multiple times. This
is not a problem after the recent patch series that made all the
classful qdiscs qlen_notify() handlers idempotent.
Fixes: 3f98113810 ("sch_hfsc: Fix qlen accounting bug when using peek in hfsc_enqueue()")
Signed-off-by: Lion Ackermann <nnamrec@gmail.com>
Reviewed-by: Jamal Hadi Salim <jhs@mojatatu.com>
Acked-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/d912cbd7-193b-4269-9857-525bee8bbb6a@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Replace the deprecated strncpy() with the two-argument version of
strscpy() as the destination is an array
and buffer should be NUL-terminated.
Signed-off-by: Pranav Tyagi <pranav.tyagi03@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250620103653.6957-1-pranav.tyagi03@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Since taprio’s taprio_dev_notifier() isn’t protected by an
RCU read-side critical section, a race with advance_sched()
can lead to a use-after-free.
Adding rcu_read_lock() inside taprio_dev_notifier() prevents this.
Fixes: fed87cc671 ("net/sched: taprio: automatically calculate queueMaxSDU based on TC gate durations")
Cc: stable@vger.kernel.org
Signed-off-by: Hyunwoo Kim <imv4bel@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/aEzIYYxt0is9upYG@v4bel-B760M-AORUS-ELITE-AX
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Current release - regressions:
- af_unix: allow passing cred for embryo without SO_PASSCRED/SO_PASSPIDFD
Current release - new code bugs:
- eth: airoha: correct enable mask for RX queues 16-31
- veth: prevent NULL pointer dereference in veth_xdp_rcv when peer
disappears under traffic
- ipv6: move fib6_config_validate() to ip6_route_add(), prevent invalid
routes
Previous releases - regressions:
- phy: phy_caps: don't skip better duplex match on non-exact match
- dsa: b53: fix untagged traffic sent via cpu tagged with VID 0
- Revert "wifi: mwifiex: Fix HT40 bandwidth issue.", it caused transient
packet loss, exact reason not fully understood, yet
Previous releases - always broken:
- net: clear the dst when BPF is changing skb protocol (IPv4 <> IPv6)
- sched: sfq: fix a potential crash on gso_skb handling
- Bluetooth: intel: improve rx buffer posting to avoid causing issues
in the firmware
- eth: intel: i40e: make reset handling robust against multiple requests
- eth: mlx5: ensure FW pages are always allocated on the local NUMA
node, even when device is configure to 'serve' another node
- wifi: ath12k: fix GCC_GCC_PCIE_HOT_RST definition for WCN7850,
prevent kernel crashes
- wifi: ath11k: avoid burning CPU in ath11k_debugfs_fw_stats_request()
for 3 sec if fw_stats_done is not set
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
-----BEGIN PGP SIGNATURE-----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=0mtm
-----END PGP SIGNATURE-----
Merge tag 'net-6.16-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Pull networking fixes from Jakub Kicinski:
"Including fixes from bluetooth and wireless.
Current release - regressions:
- af_unix: allow passing cred for embryo without SO_PASSCRED/SO_PASSPIDFD
Current release - new code bugs:
- eth: airoha: correct enable mask for RX queues 16-31
- veth: prevent NULL pointer dereference in veth_xdp_rcv when peer
disappears under traffic
- ipv6: move fib6_config_validate() to ip6_route_add(), prevent
invalid routes
Previous releases - regressions:
- phy: phy_caps: don't skip better duplex match on non-exact match
- dsa: b53: fix untagged traffic sent via cpu tagged with VID 0
- Revert "wifi: mwifiex: Fix HT40 bandwidth issue.", it caused
transient packet loss, exact reason not fully understood, yet
Previous releases - always broken:
- net: clear the dst when BPF is changing skb protocol (IPv4 <> IPv6)
- sched: sfq: fix a potential crash on gso_skb handling
- Bluetooth: intel: improve rx buffer posting to avoid causing issues
in the firmware
- eth: intel: i40e: make reset handling robust against multiple
requests
- eth: mlx5: ensure FW pages are always allocated on the local NUMA
node, even when device is configure to 'serve' another node
- wifi: ath12k: fix GCC_GCC_PCIE_HOT_RST definition for WCN7850,
prevent kernel crashes
- wifi: ath11k: avoid burning CPU in ath11k_debugfs_fw_stats_request()
for 3 sec if fw_stats_done is not set"
* tag 'net-6.16-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (70 commits)
selftests: drv-net: rss_ctx: Add test for ntuple rules targeting default RSS context
net: ethtool: Don't check if RSS context exists in case of context 0
af_unix: Allow passing cred for embryo without SO_PASSCRED/SO_PASSPIDFD.
ipv6: Move fib6_config_validate() to ip6_route_add().
net: drv: netdevsim: don't napi_complete() from netpoll
net/mlx5: HWS, Add error checking to hws_bwc_rule_complex_hash_node_get()
veth: prevent NULL pointer dereference in veth_xdp_rcv
net_sched: remove qdisc_tree_flush_backlog()
net_sched: ets: fix a race in ets_qdisc_change()
net_sched: tbf: fix a race in tbf_change()
net_sched: red: fix a race in __red_change()
net_sched: prio: fix a race in prio_tune()
net_sched: sch_sfq: reject invalid perturb period
net: phy: phy_caps: Don't skip better duplex macth on non-exact match
MAINTAINERS: Update Kuniyuki Iwashima's email address.
selftests: net: add test case for NAT46 looping back dst
net: clear the dst when changing skb protocol
net/mlx5e: Fix number of lanes to UNKNOWN when using data_rate_oper
net/mlx5e: Fix leak of Geneve TLV option object
net/mlx5: HWS, make sure the uplink is the last destination
...
Gerrard Tai reported a race condition in ETS, whenever SFQ perturb timer
fires at the wrong time.
The race is as follows:
CPU 0 CPU 1
[1]: lock root
[2]: qdisc_tree_flush_backlog()
[3]: unlock root
|
| [5]: lock root
| [6]: rehash
| [7]: qdisc_tree_reduce_backlog()
|
[4]: qdisc_put()
This can be abused to underflow a parent's qlen.
Calling qdisc_purge_queue() instead of qdisc_tree_flush_backlog()
should fix the race, because all packets will be purged from the qdisc
before releasing the lock.
Fixes: b05972f01e ("net: sched: tbf: don't call qdisc_put() while holding tree lock")
Reported-by: Gerrard Tai <gerrard.tai@starlabs.sg>
Suggested-by: Gerrard Tai <gerrard.tai@starlabs.sg>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20250611111515.1983366-5-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Gerrard Tai reported a race condition in TBF, whenever SFQ perturb timer
fires at the wrong time.
The race is as follows:
CPU 0 CPU 1
[1]: lock root
[2]: qdisc_tree_flush_backlog()
[3]: unlock root
|
| [5]: lock root
| [6]: rehash
| [7]: qdisc_tree_reduce_backlog()
|
[4]: qdisc_put()
This can be abused to underflow a parent's qlen.
Calling qdisc_purge_queue() instead of qdisc_tree_flush_backlog()
should fix the race, because all packets will be purged from the qdisc
before releasing the lock.
Fixes: b05972f01e ("net: sched: tbf: don't call qdisc_put() while holding tree lock")
Reported-by: Gerrard Tai <gerrard.tai@starlabs.sg>
Suggested-by: Gerrard Tai <gerrard.tai@starlabs.sg>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Zhengchao Shao <shaozhengchao@huawei.com>
Link: https://patch.msgid.link/20250611111515.1983366-4-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
