linux-stable.git/net/core/dev.c, branch v4.19.232

net: annotate data-races on txq->xmit_lock_owner

2021-12-08T07:50:14+00:00

commit 7a10d8c810cfad3e79372d7d1c77899d86cd6662 upstream.

syzbot found that __dev_queue_xmit() is reading txq->xmit_lock_owner
without annotations.

No serious issue there, let's document what is happening there.

BUG: KCSAN: data-race in __dev_queue_xmit / __dev_queue_xmit

write to 0xffff888139d09484 of 4 bytes by interrupt on cpu 0:
 __netif_tx_unlock include/linux/netdevice.h:4437 [inline]
 __dev_queue_xmit+0x948/0xf70 net/core/dev.c:4229
 dev_queue_xmit_accel+0x19/0x20 net/core/dev.c:4265
 macvlan_queue_xmit drivers/net/macvlan.c:543 [inline]
 macvlan_start_xmit+0x2b3/0x3d0 drivers/net/macvlan.c:567
 __netdev_start_xmit include/linux/netdevice.h:4987 [inline]
 netdev_start_xmit include/linux/netdevice.h:5001 [inline]
 xmit_one+0x105/0x2f0 net/core/dev.c:3590
 dev_hard_start_xmit+0x72/0x120 net/core/dev.c:3606
 sch_direct_xmit+0x1b2/0x7c0 net/sched/sch_generic.c:342
 __dev_xmit_skb+0x83d/0x1370 net/core/dev.c:3817
 __dev_queue_xmit+0x590/0xf70 net/core/dev.c:4194
 dev_queue_xmit+0x13/0x20 net/core/dev.c:4259
 neigh_hh_output include/net/neighbour.h:511 [inline]
 neigh_output include/net/neighbour.h:525 [inline]
 ip6_finish_output2+0x995/0xbb0 net/ipv6/ip6_output.c:126
 __ip6_finish_output net/ipv6/ip6_output.c:191 [inline]
 ip6_finish_output+0x444/0x4c0 net/ipv6/ip6_output.c:201
 NF_HOOK_COND include/linux/netfilter.h:296 [inline]
 ip6_output+0x10e/0x210 net/ipv6/ip6_output.c:224
 dst_output include/net/dst.h:450 [inline]
 NF_HOOK include/linux/netfilter.h:307 [inline]
 ndisc_send_skb+0x486/0x610 net/ipv6/ndisc.c:508
 ndisc_send_rs+0x3b0/0x3e0 net/ipv6/ndisc.c:702
 addrconf_rs_timer+0x370/0x540 net/ipv6/addrconf.c:3898
 call_timer_fn+0x2e/0x240 kernel/time/timer.c:1421
 expire_timers+0x116/0x240 kernel/time/timer.c:1466
 __run_timers+0x368/0x410 kernel/time/timer.c:1734
 run_timer_softirq+0x2e/0x60 kernel/time/timer.c:1747
 __do_softirq+0x158/0x2de kernel/softirq.c:558
 __irq_exit_rcu kernel/softirq.c:636 [inline]
 irq_exit_rcu+0x37/0x70 kernel/softirq.c:648
 sysvec_apic_timer_interrupt+0x3e/0xb0 arch/x86/kernel/apic/apic.c:1097
 asm_sysvec_apic_timer_interrupt+0x12/0x20

read to 0xffff888139d09484 of 4 bytes by interrupt on cpu 1:
 __dev_queue_xmit+0x5e3/0xf70 net/core/dev.c:4213
 dev_queue_xmit_accel+0x19/0x20 net/core/dev.c:4265
 macvlan_queue_xmit drivers/net/macvlan.c:543 [inline]
 macvlan_start_xmit+0x2b3/0x3d0 drivers/net/macvlan.c:567
 __netdev_start_xmit include/linux/netdevice.h:4987 [inline]
 netdev_start_xmit include/linux/netdevice.h:5001 [inline]
 xmit_one+0x105/0x2f0 net/core/dev.c:3590
 dev_hard_start_xmit+0x72/0x120 net/core/dev.c:3606
 sch_direct_xmit+0x1b2/0x7c0 net/sched/sch_generic.c:342
 __dev_xmit_skb+0x83d/0x1370 net/core/dev.c:3817
 __dev_queue_xmit+0x590/0xf70 net/core/dev.c:4194
 dev_queue_xmit+0x13/0x20 net/core/dev.c:4259
 neigh_resolve_output+0x3db/0x410 net/core/neighbour.c:1523
 neigh_output include/net/neighbour.h:527 [inline]
 ip6_finish_output2+0x9be/0xbb0 net/ipv6/ip6_output.c:126
 __ip6_finish_output net/ipv6/ip6_output.c:191 [inline]
 ip6_finish_output+0x444/0x4c0 net/ipv6/ip6_output.c:201
 NF_HOOK_COND include/linux/netfilter.h:296 [inline]
 ip6_output+0x10e/0x210 net/ipv6/ip6_output.c:224
 dst_output include/net/dst.h:450 [inline]
 NF_HOOK include/linux/netfilter.h:307 [inline]
 ndisc_send_skb+0x486/0x610 net/ipv6/ndisc.c:508
 ndisc_send_rs+0x3b0/0x3e0 net/ipv6/ndisc.c:702
 addrconf_rs_timer+0x370/0x540 net/ipv6/addrconf.c:3898
 call_timer_fn+0x2e/0x240 kernel/time/timer.c:1421
 expire_timers+0x116/0x240 kernel/time/timer.c:1466
 __run_timers+0x368/0x410 kernel/time/timer.c:1734
 run_timer_softirq+0x2e/0x60 kernel/time/timer.c:1747
 __do_softirq+0x158/0x2de kernel/softirq.c:558
 __irq_exit_rcu kernel/softirq.c:636 [inline]
 irq_exit_rcu+0x37/0x70 kernel/softirq.c:648
 sysvec_apic_timer_interrupt+0x8d/0xb0 arch/x86/kernel/apic/apic.c:1097
 asm_sysvec_apic_timer_interrupt+0x12/0x20
 kcsan_setup_watchpoint+0x94/0x420 kernel/kcsan/core.c:443
 folio_test_anon include/linux/page-flags.h:581 [inline]
 PageAnon include/linux/page-flags.h:586 [inline]
 zap_pte_range+0x5ac/0x10e0 mm/memory.c:1347
 zap_pmd_range mm/memory.c:1467 [inline]
 zap_pud_range mm/memory.c:1496 [inline]
 zap_p4d_range mm/memory.c:1517 [inline]
 unmap_page_range+0x2dc/0x3d0 mm/memory.c:1538
 unmap_single_vma+0x157/0x210 mm/memory.c:1583
 unmap_vmas+0xd0/0x180 mm/memory.c:1615
 exit_mmap+0x23d/0x470 mm/mmap.c:3170
 __mmput+0x27/0x1b0 kernel/fork.c:1113
 mmput+0x3d/0x50 kernel/fork.c:1134
 exit_mm+0xdb/0x170 kernel/exit.c:507
 do_exit+0x608/0x17a0 kernel/exit.c:819
 do_group_exit+0xce/0x180 kernel/exit.c:929
 get_signal+0xfc3/0x1550 kernel/signal.c:2852
 arch_do_signal_or_restart+0x8c/0x2e0 arch/x86/kernel/signal.c:868
 handle_signal_work kernel/entry/common.c:148 [inline]
 exit_to_user_mode_loop kernel/entry/common.c:172 [inline]
 exit_to_user_mode_prepare+0x113/0x190 kernel/entry/common.c:207
 __syscall_exit_to_user_mode_work kernel/entry/common.c:289 [inline]
 syscall_exit_to_user_mode+0x20/0x40 kernel/entry/common.c:300
 do_syscall_64+0x50/0xd0 arch/x86/entry/common.c:86
 entry_SYSCALL_64_after_hwframe+0x44/0xae

value changed: 0x00000000 -> 0xffffffff

Reported by Kernel Concurrency Sanitizer on:
CPU: 1 PID: 28712 Comm: syz-executor.0 Tainted: G        W         5.16.0-rc1-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Eric Dumazet 
Reported-by: syzbot 
Link: https://lore.kernel.org/r/20211130170155.2331929-1-eric.dumazet@gmail.com
Signed-off-by: Jakub Kicinski 
Signed-off-by: Greg Kroah-Hartman

net: sched: update default qdisc visibility after Tx queue cnt changes

2021-11-26T10:36:04+00:00

[ Upstream commit 1e080f17750d1083e8a32f7b350584ae1cd7ff20 ]

mq / mqprio make the default child qdiscs visible. They only do
so for the qdiscs which are within real_num_tx_queues when the
device is registered. Depending on order of calls in the driver,
or if user space changes config via ethtool -L the number of
qdiscs visible under tc qdisc show will differ from the number
of queues. This is confusing to users and potentially to system
configuration scripts which try to make sure qdiscs have the
right parameters.

Add a new Qdisc_ops callback and make relevant qdiscs TTRT.

Note that this uncovers the "shortcut" created by
commit 1f27cde313d7 ("net: sched: use pfifo_fast for non real queues")
The default child qdiscs beyond initial real_num_tx are always
pfifo_fast, no matter what the sysfs setting is. Fixing this
gets a little tricky because we'd need to keep a reference
on whatever the default qdisc was at the time of creation.
In practice this is likely an non-issue the qdiscs likely have
to be configured to non-default settings, so whatever user space
is doing such configuration can replace the pfifos... now that
it will see them.

Reported-by: Matthew Massey 
Reviewed-by: Dave Taht 
Signed-off-by: Jakub Kicinski 
Signed-off-by: David S. Miller 
Signed-off-by: Sasha Levin

net: Prevent infinite while loop in skb_tx_hash()

2021-11-02T17:26:45+00:00

commit 0c57eeecc559ca6bc18b8c4e2808bc78dbe769b0 upstream.

Drivers call netdev_set_num_tc() and then netdev_set_tc_queue()
to set the queue count and offset for each TC.  So the queue count
and offset for the TCs may be zero for a short period after dev->num_tc
has been set.  If a TX packet is being transmitted at this time in the
code path netdev_pick_tx() -> skb_tx_hash(), skb_tx_hash() may see
nonzero dev->num_tc but zero qcount for the TC.  The while loop that
keeps looping while hash >= qcount will not end.

Fix it by checking the TC's qcount to be nonzero before using it.

Fixes: eadec877ce9c ("net: Add support for subordinate traffic classes to netdev_pick_tx")
Reviewed-by: Andy Gospodarek 
Signed-off-by: Michael Chan 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

gro: ensure frag0 meets IP header alignment

2021-08-04T10:23:44+00:00

commit 38ec4944b593fd90c5ef42aaaa53e66ae5769d04 upstream.

After commit 0f6925b3e8da ("virtio_net: Do not pull payload in skb->head")
Guenter Roeck reported one failure in his tests using sh architecture.

After much debugging, we have been able to spot silent unaligned accesses
in inet_gro_receive()

The issue at hand is that upper networking stacks assume their header
is word-aligned. Low level drivers are supposed to reserve NET_IP_ALIGN
bytes before the Ethernet header to make that happen.

This patch hardens skb_gro_reset_offset() to not allow frag0 fast-path
if the fragment is not properly aligned.

Some arches like x86, arm64 and powerpc do not care and define NET_IP_ALIGN
as 0, this extra check will be a NOP for them.

Note that if frag0 is not used, GRO will call pskb_may_pull()
as many times as needed to pull network and transport headers.

Fixes: 0f6925b3e8da ("virtio_net: Do not pull payload in skb->head")
Fixes: 78a478d0efd9 ("gro: Inline skb_gro_header and cache frag0 virtual address")
Signed-off-by: Eric Dumazet 
Reported-by: Guenter Roeck 
Cc: Xuan Zhuo 
Cc: "Michael S. Tsirkin" 
Cc: Jason Wang 
Acked-by: Michael S. Tsirkin 
Tested-by: Guenter Roeck 
Signed-off-by: David S. Miller 
Signed-off-by: Matthieu Baerts 
Signed-off-by: Greg Kroah-Hartman

net: Treat __napi_schedule_irqoff() as __napi_schedule() on PREEMPT_RT

2021-07-20T14:15:59+00:00

[ Upstream commit 8380c81d5c4fced6f4397795a5ae65758272bbfd ]

__napi_schedule_irqoff() is an optimized version of __napi_schedule()
which can be used where it is known that interrupts are disabled,
e.g. in interrupt-handlers, spin_lock_irq() sections or hrtimer
callbacks.

On PREEMPT_RT enabled kernels this assumptions is not true. Force-
threaded interrupt handlers and spinlocks are not disabling interrupts
and the NAPI hrtimer callback is forced into softirq context which runs
with interrupts enabled as well.

Chasing all usage sites of __napi_schedule_irqoff() is a whack-a-mole
game so make __napi_schedule_irqoff() invoke __napi_schedule() for
PREEMPT_RT kernels.

The callers of ____napi_schedule() in the networking core have been
audited and are correct on PREEMPT_RT kernels as well.

Reported-by: Juri Lelli 
Signed-off-by: Sebastian Andrzej Siewior 
Reviewed-by: Thomas Gleixner 
Reviewed-by: Juri Lelli 
Signed-off-by: David S. Miller 
Signed-off-by: Sasha Levin

can: dev: Move device back to init netns on owning netns delete

2021-03-30T12:37:03+00:00

commit 3a5ca857079ea022e0b1b17fc154f7ad7dbc150f upstream.

When a non-initial netns is destroyed, the usual policy is to delete
all virtual network interfaces contained, but move physical interfaces
back to the initial netns. This keeps the physical interface visible
on the system.

CAN devices are somewhat special, as they define rtnl_link_ops even
if they are physical devices. If a CAN interface is moved into a
non-initial netns, destroying that netns lets the interface vanish
instead of moving it back to the initial netns. default_device_exit()
skips CAN interfaces due to having rtnl_link_ops set. Reproducer:

  ip netns add foo
  ip link set can0 netns foo
  ip netns delete foo

WARNING: CPU: 1 PID: 84 at net/core/dev.c:11030 ops_exit_list+0x38/0x60
CPU: 1 PID: 84 Comm: kworker/u4:2 Not tainted 5.10.19 #1
Workqueue: netns cleanup_net
[] (unwind_backtrace) from [] (show_stack+0x10/0x14)
[] (show_stack) from [] (dump_stack+0x94/0xa8)
[] (dump_stack) from [] (__warn+0xb8/0x114)
[] (__warn) from [] (warn_slowpath_fmt+0x7c/0xac)
[] (warn_slowpath_fmt) from [] (ops_exit_list+0x38/0x60)
[] (ops_exit_list) from [] (cleanup_net+0x230/0x380)
[] (cleanup_net) from [] (process_one_work+0x1d8/0x438)
[] (process_one_work) from [] (worker_thread+0x64/0x5a8)
[] (worker_thread) from [] (kthread+0x148/0x14c)
[] (kthread) from [] (ret_from_fork+0x14/0x2c)

To properly restore physical CAN devices to the initial netns on owning
netns exit, introduce a flag on rtnl_link_ops that can be set by drivers.
For CAN devices setting this flag, default_device_exit() considers them
non-virtual, applying the usual namespace move.

The issue was introduced in the commit mentioned below, as at that time
CAN devices did not have a dellink() operation.

Fixes: e008b5fc8dc7 ("net: Simplfy default_device_exit and improve batching.")
Link: https://lore.kernel.org/r/20210302122423.872326-1-martin@strongswan.org
Signed-off-by: Martin Willi 
Signed-off-by: Marc Kleine-Budde 
Signed-off-by: Greg Kroah-Hartman

net: Disable NETIF_F_HW_TLS_RX when RXCSUM is disabled

2021-01-27T10:05:44+00:00

commit a3eb4e9d4c9218476d05c52dfd2be3d6fdce6b91 upstream.

With NETIF_F_HW_TLS_RX packets are decrypted in HW. This cannot be
logically done when RXCSUM offload is off.

Fixes: 14136564c8ee ("net: Add TLS RX offload feature")
Signed-off-by: Tariq Toukan 
Reviewed-by: Boris Pismenny 
Link: https://lore.kernel.org/r/20210117151538.9411-1-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski 
Signed-off-by: Greg Kroah-Hartman

net: disable netpoll on fresh napis

2020-09-12T11:40:23+00:00

[ Upstream commit 96e97bc07e90f175a8980a22827faf702ca4cb30 ]

napi_disable() makes sure to set the NAPI_STATE_NPSVC bit to prevent
netpoll from accessing rings before init is complete. However, the
same is not done for fresh napi instances in netif_napi_add(),
even though we expect NAPI instances to be added as disabled.

This causes crashes during driver reconfiguration (enabling XDP,
changing the channel count) - if there is any printk() after
netif_napi_add() but before napi_enable().

To ensure memory ordering is correct we need to use RCU accessors.

Reported-by: Rob Sherwood 
Fixes: 2d8bff12699a ("netpoll: Close race condition between poll_one_napi and napi_disable")
Signed-off-by: Jakub Kicinski 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

dev: Defer free of skbs in flush_backlog

2020-07-31T16:37:47+00:00

[ Upstream commit 7df5cb75cfb8acf96c7f2342530eb41e0c11f4c3 ]

IRQs are disabled when freeing skbs in input queue.
Use the IRQ safe variant to free skbs here.

Fixes: 145dd5f9c88f ("net: flush the softnet backlog in process context")
Signed-off-by: Subash Abhinov Kasiviswanathan 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

net: increment xmit_recursion level in dev_direct_xmit()

2020-07-01T03:17:04+00:00

[ Upstream commit 0ad6f6e767ec2f613418cbc7ebe5ec4c35af540c ]

Back in commit f60e5990d9c1 ("ipv6: protect skb->sk accesses
from recursive dereference inside the stack") Hannes added code
so that IPv6 stack would not trust skb->sk for typical cases
where packet goes through 'standard' xmit path (__dev_queue_xmit())

Alas af_packet had a dev_direct_xmit() path that was not
dealing yet with xmit_recursion level.

Also change sk_mc_loop() to dump a stack once only.

Without this patch, syzbot was able to trigger :

[1]
[  153.567378] WARNING: CPU: 7 PID: 11273 at net/core/sock.c:721 sk_mc_loop+0x51/0x70
[  153.567378] Modules linked in: nfnetlink ip6table_raw ip6table_filter iptable_raw iptable_nat nf_nat nf_conntrack nf_defrag_ipv4 nf_defrag_ipv6 iptable_filter macsec macvtap tap macvlan 8021q hsr wireguard libblake2s blake2s_x86_64 libblake2s_generic udp_tunnel ip6_udp_tunnel libchacha20poly1305 poly1305_x86_64 chacha_x86_64 libchacha curve25519_x86_64 libcurve25519_generic netdevsim batman_adv dummy team bridge stp llc w1_therm wire i2c_mux_pca954x i2c_mux cdc_acm ehci_pci ehci_hcd mlx4_en mlx4_ib ib_uverbs ib_core mlx4_core
[  153.567386] CPU: 7 PID: 11273 Comm: b159172088 Not tainted 5.8.0-smp-DEV #273
[  153.567387] RIP: 0010:sk_mc_loop+0x51/0x70
[  153.567388] Code: 66 83 f8 0a 75 24 0f b6 4f 12 b8 01 00 00 00 31 d2 d3 e0 a9 bf ef ff ff 74 07 48 8b 97 f0 02 00 00 0f b6 42 3a 83 e0 01 5d c3 <0f> 0b b8 01 00 00 00 5d c3 0f b6 87 18 03 00 00 5d c0 e8 04 83 e0
[  153.567388] RSP: 0018:ffff95c69bb93990 EFLAGS: 00010212
[  153.567388] RAX: 0000000000000011 RBX: ffff95c6e0ee3e00 RCX: 0000000000000007
[  153.567389] RDX: ffff95c69ae50000 RSI: ffff95c6c30c3000 RDI: ffff95c6c30c3000
[  153.567389] RBP: ffff95c69bb93990 R08: ffff95c69a77f000 R09: 0000000000000008
[  153.567389] R10: 0000000000000040 R11: 00003e0e00026128 R12: ffff95c6c30c3000
[  153.567390] R13: ffff95c6cc4fd500 R14: ffff95c6f84500c0 R15: ffff95c69aa13c00
[  153.567390] FS:  00007fdc3a283700(0000) GS:ffff95c6ff9c0000(0000) knlGS:0000000000000000
[  153.567390] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  153.567391] CR2: 00007ffee758e890 CR3: 0000001f9ba20003 CR4: 00000000001606e0
[  153.567391] Call Trace:
[  153.567391]  ip6_finish_output2+0x34e/0x550
[  153.567391]  __ip6_finish_output+0xe7/0x110
[  153.567391]  ip6_finish_output+0x2d/0xb0
[  153.567392]  ip6_output+0x77/0x120
[  153.567392]  ? __ip6_finish_output+0x110/0x110
[  153.567392]  ip6_local_out+0x3d/0x50
[  153.567392]  ipvlan_queue_xmit+0x56c/0x5e0
[  153.567393]  ? ksize+0x19/0x30
[  153.567393]  ipvlan_start_xmit+0x18/0x50
[  153.567393]  dev_direct_xmit+0xf3/0x1c0
[  153.567393]  packet_direct_xmit+0x69/0xa0
[  153.567394]  packet_sendmsg+0xbf0/0x19b0
[  153.567394]  ? plist_del+0x62/0xb0
[  153.567394]  sock_sendmsg+0x65/0x70
[  153.567394]  sock_write_iter+0x93/0xf0
[  153.567394]  new_sync_write+0x18e/0x1a0
[  153.567395]  __vfs_write+0x29/0x40
[  153.567395]  vfs_write+0xb9/0x1b0
[  153.567395]  ksys_write+0xb1/0xe0
[  153.567395]  __x64_sys_write+0x1a/0x20
[  153.567395]  do_syscall_64+0x43/0x70
[  153.567396]  entry_SYSCALL_64_after_hwframe+0x44/0xa9
[  153.567396] RIP: 0033:0x453549
[  153.567396] Code: Bad RIP value.
[  153.567396] RSP: 002b:00007fdc3a282cc8 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
[  153.567397] RAX: ffffffffffffffda RBX: 00000000004d32d0 RCX: 0000000000453549
[  153.567397] RDX: 0000000000000020 RSI: 0000000020000300 RDI: 0000000000000003
[  153.567398] RBP: 00000000004d32d8 R08: 0000000000000000 R09: 0000000000000000
[  153.567398] R10: 0000000000000000 R11: 0000000000000246 R12: 00000000004d32dc
[  153.567398] R13: 00007ffee742260f R14: 00007fdc3a282dc0 R15: 00007fdc3a283700
[  153.567399] ---[ end trace c1d5ae2b1059ec62 ]---

f60e5990d9c1 ("ipv6: protect skb->sk accesses from recursive dereference inside the stack")
Signed-off-by: Eric Dumazet 
Reported-by: syzbot 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman