[v6.1] possible deadlock in trie_update_elem (2)

[syzbot]

Hello,

syzbot found the following issue on:

HEAD commit: 50cbba13faa2 Linux 6.1.159
git tree: linux-6.1.y
console output: https://syzkaller.appspot.com/x/log.txt?x=1584e074580000
kernel config: https://syzkaller.appspot.com/x/.config?x=6a5cbd268406bf43
dashboard link: https://syzkaller.appspot.com/bug?extid=575065ff114761dfb81b
compiler: Debian clang version 20.1.8 (++20250708063551+0c9f909b7976-1~exp1~20250708183702.136), Debian LLD 20.1.8

Unfortunately, I don't have any reproducer for this issue yet.

Downloadable assets:
disk image: https://storage.googleapis.com/syzbot-assets/b43fb8fc9816/disk-50cbba13.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/ba3d5dc8928b/vmlinux-50cbba13.xz
kernel image: https://storage.googleapis.com/syzbot-assets/2b4e001b3742/bzImage-50cbba13.xz

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+575065...@syzkaller.appspotmail.com

======================================================
WARNING: possible circular locking dependency detected
syzkaller #0 Not tainted
------------------------------------------------------
syz.6.2607/15125 is trying to acquire lock:
ffff88813fffeba0 (&pgdat->kswapd_wait){....}-{2:2}, at: __wake_up_common_lock kernel/sched/wait.c:137 [inline]
ffff88813fffeba0 (&pgdat->kswapd_wait){....}-{2:2}, at: __wake_up+0xf4/0x190 kernel/sched/wait.c:160

but task is already holding lock:
ffff888079b08238 (&trie->lock){....}-{2:2}, at: trie_update_elem+0xc4/0xe90 kernel/bpf/lpm_trie.c:335

which lock already depends on the new lock.


the existing dependency chain (in reverse order) is:

-> #3 (&trie->lock){....}-{2:2}:
__raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:110 [inline]
_raw_spin_lock_irqsave+0xa4/0xf0 kernel/locking/spinlock.c:162
trie_delete_elem+0x90/0x690 kernel/bpf/lpm_trie.c:467
0xffffffffa00006a6
bpf_dispatcher_nop_func include/linux/bpf.h:1012 [inline]
__bpf_prog_run include/linux/filter.h:603 [inline]
bpf_prog_run include/linux/filter.h:610 [inline]
__bpf_trace_run kernel/trace/bpf_trace.c:2285 [inline]
bpf_trace_run2+0x1cd/0x3b0 kernel/trace/bpf_trace.c:2324
trace_tlb_flush+0x143/0x190 include/trace/events/tlb.h:38
switch_mm_irqs_off+0x758/0xb10 arch/x86/mm/tlb.c:-1
context_switch kernel/sched/core.c:5229 [inline]
__schedule+0xe99/0x40b0 kernel/sched/core.c:6561
schedule+0xb9/0x180 kernel/sched/core.c:6637
worker_thread+0xef6/0x1250 kernel/workqueue.c:2460
kthread+0x29d/0x330 kernel/kthread.c:376
ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295

-> #2 (&rq->__lock){-.-.}-{2:2}:
_raw_spin_lock_nested+0x2e/0x40 kernel/locking/spinlock.c:378
raw_spin_rq_lock_nested+0x26/0x140 kernel/sched/core.c:537
raw_spin_rq_lock kernel/sched/sched.h:1355 [inline]
rq_lock kernel/sched/sched.h:1645 [inline]
task_fork_fair+0x5c/0x350 kernel/sched/fair.c:11901
sched_cgroup_fork+0x32c/0x3c0 kernel/sched/core.c:4685
copy_process+0x24b8/0x4020 kernel/fork.c:2459
kernel_clone+0x225/0x8b0 kernel/fork.c:2757
user_mode_thread+0xda/0x130 kernel/fork.c:2833
rest_init+0x23/0x300 init/main.c:710
start_kernel+0x0/0x539 init/main.c:905
start_kernel+0x490/0x539 init/main.c:1152
secondary_startup_64_no_verify+0xcf/0xdb

-> #1 (&p->pi_lock){-.-.}-{2:2}:
__raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:110 [inline]
_raw_spin_lock_irqsave+0xa4/0xf0 kernel/locking/spinlock.c:162
try_to_wake_up+0x5c/0x1080 kernel/sched/core.c:4111
autoremove_wake_function+0x13/0xf0 kernel/sched/wait.c:419
__wake_up_common+0x2a4/0x4e0 kernel/sched/wait.c:107
__wake_up_common_lock kernel/sched/wait.c:138 [inline]
__wake_up+0x10e/0x190 kernel/sched/wait.c:160
rmqueue mm/page_alloc.c:3884 [inline]
get_page_from_freelist+0x1759/0x1ac0 mm/page_alloc.c:4325
__alloc_pages+0x1df/0x4e0 mm/page_alloc.c:5614
alloc_slab_page+0x5d/0x160 mm/slub.c:1799
allocate_slab mm/slub.c:1944 [inline]
new_slab+0x87/0x2c0 mm/slub.c:1997
___slab_alloc+0xbc6/0x1230 mm/slub.c:3154
__slab_alloc mm/slub.c:3240 [inline]
slab_alloc_node mm/slub.c:3325 [inline]
__kmem_cache_alloc_node+0x1a0/0x260 mm/slub.c:3398
__do_kmalloc_node mm/slab_common.c:935 [inline]
__kmalloc_node_track_caller+0x9e/0x230 mm/slab_common.c:956
kmalloc_reserve net/core/skbuff.c:446 [inline]
__alloc_skb+0x22a/0x7e0 net/core/skbuff.c:515
alloc_skb include/linux/skbuff.h:1271 [inline]
nsim_dev_trap_skb_build drivers/net/netdevsim/dev.c:748 [inline]
nsim_dev_trap_report drivers/net/netdevsim/dev.c:805 [inline]
nsim_dev_trap_report_work+0x28f/0xaf0 drivers/net/netdevsim/dev.c:851
process_one_work+0x898/0x1160 kernel/workqueue.c:2292
worker_thread+0xaa2/0x1250 kernel/workqueue.c:2439
kthread+0x29d/0x330 kernel/kthread.c:376
ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295

-> #0 (&pgdat->kswapd_wait){....}-{2:2}:
check_prev_add kernel/locking/lockdep.c:3090 [inline]
check_prevs_add kernel/locking/lockdep.c:3209 [inline]
validate_chain kernel/locking/lockdep.c:3825 [inline]
__lock_acquire+0x2cf8/0x7c50 kernel/locking/lockdep.c:5049
lock_acquire+0x1b4/0x490 kernel/locking/lockdep.c:5662
__raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:110 [inline]
_raw_spin_lock_irqsave+0xa4/0xf0 kernel/locking/spinlock.c:162
__wake_up_common_lock kernel/sched/wait.c:137 [inline]
__wake_up+0xf4/0x190 kernel/sched/wait.c:160
rmqueue mm/page_alloc.c:3884 [inline]
get_page_from_freelist+0x1759/0x1ac0 mm/page_alloc.c:4325
__alloc_pages+0x1df/0x4e0 mm/page_alloc.c:5614
alloc_slab_page+0x5d/0x160 mm/slub.c:1799
allocate_slab mm/slub.c:1944 [inline]
new_slab+0x87/0x2c0 mm/slub.c:1997
___slab_alloc+0xbc6/0x1230 mm/slub.c:3154
__slab_alloc mm/slub.c:3240 [inline]
slab_alloc_node mm/slub.c:3325 [inline]
__kmem_cache_alloc_node+0x1a0/0x260 mm/slub.c:3398
__do_kmalloc_node mm/slab_common.c:935 [inline]
__kmalloc_node+0xa0/0x240 mm/slab_common.c:943
kmalloc_node include/linux/slab.h:589 [inline]
bpf_map_kmalloc_node+0xb8/0x1a0 kernel/bpf/syscall.c:454
lpm_trie_node_alloc kernel/bpf/lpm_trie.c:291 [inline]
trie_update_elem+0x160/0xe90 kernel/bpf/lpm_trie.c:338
bpf_map_update_value+0x59e/0x670 kernel/bpf/syscall.c:228
generic_map_update_batch+0x569/0x850 kernel/bpf/syscall.c:1709
bpf_map_do_batch+0x466/0x600 kernel/bpf/syscall.c:-1
__sys_bpf+0x65f/0x6d0 kernel/bpf/syscall.c:-1
__do_sys_bpf kernel/bpf/syscall.c:5131 [inline]
__se_sys_bpf kernel/bpf/syscall.c:5129 [inline]
__x64_sys_bpf+0x78/0x90 kernel/bpf/syscall.c:5129
do_syscall_x64 arch/x86/entry/common.c:51 [inline]
do_syscall_64+0x4c/0xa0 arch/x86/entry/common.c:81
entry_SYSCALL_64_after_hwframe+0x68/0xd2

other info that might help us debug this:

Chain exists of:
&pgdat->kswapd_wait --> &rq->__lock --> &trie->lock

Possible unsafe locking scenario:

CPU0 CPU1
---- ----
lock(&trie->lock);
lock(&rq->__lock);
lock(&trie->lock);
lock(&pgdat->kswapd_wait);

*** DEADLOCK ***

2 locks held by syz.6.2607/15125:
#0: ffffffff8c92b120 (rcu_read_lock){....}-{1:2}, at: rcu_lock_acquire include/linux/rcupdate.h:350 [inline]
#0: ffffffff8c92b120 (rcu_read_lock){....}-{1:2}, at: rcu_read_lock include/linux/rcupdate.h:791 [inline]
#0: ffffffff8c92b120 (rcu_read_lock){....}-{1:2}, at: bpf_map_update_value+0x375/0x670 kernel/bpf/syscall.c:227
#1: ffff888079b08238 (&trie->lock){....}-{2:2}, at: trie_update_elem+0xc4/0xe90 kernel/bpf/lpm_trie.c:335

stack backtrace:
CPU: 0 PID: 15125 Comm: syz.6.2607 Not tainted syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/25/2025
Call Trace:
<TASK>
dump_stack_lvl+0x168/0x22e lib/dump_stack.c:106
check_noncircular+0x274/0x310 kernel/locking/lockdep.c:2170
check_prev_add kernel/locking/lockdep.c:3090 [inline]
check_prevs_add kernel/locking/lockdep.c:3209 [inline]
validate_chain kernel/locking/lockdep.c:3825 [inline]
__lock_acquire+0x2cf8/0x7c50 kernel/locking/lockdep.c:5049
lock_acquire+0x1b4/0x490 kernel/locking/lockdep.c:5662
__raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:110 [inline]
_raw_spin_lock_irqsave+0xa4/0xf0 kernel/locking/spinlock.c:162
__wake_up_common_lock kernel/sched/wait.c:137 [inline]
__wake_up+0xf4/0x190 kernel/sched/wait.c:160
rmqueue mm/page_alloc.c:3884 [inline]
get_page_from_freelist+0x1759/0x1ac0 mm/page_alloc.c:4325
__alloc_pages+0x1df/0x4e0 mm/page_alloc.c:5614
alloc_slab_page+0x5d/0x160 mm/slub.c:1799
allocate_slab mm/slub.c:1944 [inline]
new_slab+0x87/0x2c0 mm/slub.c:1997
___slab_alloc+0xbc6/0x1230 mm/slub.c:3154
__slab_alloc mm/slub.c:3240 [inline]
slab_alloc_node mm/slub.c:3325 [inline]
__kmem_cache_alloc_node+0x1a0/0x260 mm/slub.c:3398
__do_kmalloc_node mm/slab_common.c:935 [inline]
__kmalloc_node+0xa0/0x240 mm/slab_common.c:943
kmalloc_node include/linux/slab.h:589 [inline]
bpf_map_kmalloc_node+0xb8/0x1a0 kernel/bpf/syscall.c:454
lpm_trie_node_alloc kernel/bpf/lpm_trie.c:291 [inline]
trie_update_elem+0x160/0xe90 kernel/bpf/lpm_trie.c:338
bpf_map_update_value+0x59e/0x670 kernel/bpf/syscall.c:228
generic_map_update_batch+0x569/0x850 kernel/bpf/syscall.c:1709
bpf_map_do_batch+0x466/0x600 kernel/bpf/syscall.c:-1
__sys_bpf+0x65f/0x6d0 kernel/bpf/syscall.c:-1
__do_sys_bpf kernel/bpf/syscall.c:5131 [inline]
__se_sys_bpf kernel/bpf/syscall.c:5129 [inline]
__x64_sys_bpf+0x78/0x90 kernel/bpf/syscall.c:5129
do_syscall_x64 arch/x86/entry/common.c:51 [inline]
do_syscall_64+0x4c/0xa0 arch/x86/entry/common.c:81
entry_SYSCALL_64_after_hwframe+0x68/0xd2
RIP: 0033:0x7f4aa278f749
Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 a8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f4aa353f038 EFLAGS: 00000246 ORIG_RAX: 0000000000000141
RAX: ffffffffffffffda RBX: 00007f4aa29e5fa0 RCX: 00007f4aa278f749
RDX: 0000000000000038 RSI: 0000200000000000 RDI: 000000000000001a
RBP: 00007f4aa2813f91 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 00007f4aa29e6038 R14: 00007f4aa29e5fa0 R15: 00007ffd5aed9b58
</TASK>


---
This report is generated by a bot. It may contain errors.
See https://goo.gl/tpsmEJ for more information about syzbot.
syzbot engineers can be reached at syzk...@googlegroups.com.

syzbot will keep track of this issue. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.

If the report is already addressed, let syzbot know by replying with:
#syz fix: exact-commit-title

If you want to overwrite report's subsystems, reply with:
#syz set subsystems: new-subsystem
(See the list of subsystem names on the web dashboard)

If the report is a duplicate of another one, reply with:
#syz dup: exact-subject-of-another-report

If you want to undo deduplication, reply with:
#syz undup