[v6.6] INFO: rcu detected stall in sendfile64

[syzbot]

Hello,

syzbot found the following issue on:

HEAD commit: 924b4a879cbb Linux 6.6.142
git tree: linux-6.6.y
console output: https://syzkaller.appspot.com/x/log.txt?x=149a90ae580000
kernel config: https://syzkaller.appspot.com/x/.config?x=90249d2d52c08134
dashboard link: https://syzkaller.appspot.com/bug?extid=419fa9f4a183b1e48463
compiler: Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8

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

Downloadable assets:
disk image: https://storage.googleapis.com/syzbot-assets/bd13425862f1/disk-924b4a87.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/eaa5de9e440b/vmlinux-924b4a87.xz
kernel image: https://storage.googleapis.com/syzbot-assets/5a9aa3293c15/bzImage-924b4a87.xz

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+419fa9...@syzkaller.appspotmail.com

rcu: INFO: rcu_preempt detected stalls on CPUs/tasks:
rcu: Tasks blocked on level-0 rcu_node (CPUs 0-1): P21771/1:b..l
rcu: (detected by 0, t=10502 jiffies, g=112149, q=126 ncpus=2)
task:syz.8.4586 state:R running task stack:26568 pid:21771 ppid:11064 flags:0x00004002
Call Trace:
<TASK>
context_switch kernel/sched/core.c:5381 [inline]
__schedule+0x1553/0x45a0 kernel/sched/core.c:6700
preempt_schedule_irq+0xbf/0x150 kernel/sched/core.c:7010
irqentry_exit+0x67/0x70 kernel/entry/common.c:438
asm_sysvec_apic_timer_interrupt+0x1a/0x20 arch/x86/include/asm/idtentry.h:687
RIP: 0010:__rcu_read_unlock+0xa/0xd0 kernel/rcu/tree_plugin.h:419
Code: 80 e1 07 80 c1 03 38 c1 7c e3 48 89 df e8 8e b4 6d 00 eb d9 66 2e 0f 1f 84 00 00 00 00 00 66 90 f3 0f 1e fa 41 57 41 56 41 55 <41> 54 53 49 bf 00 00 00 00 00 fc ff df 65 48 8b 3d 11 ff 91 7e 48
RSP: 0018:ffffc9000dca7780 EFLAGS: 00000246
RAX: ffffffff81b5be6b RBX: ffffffff81b5bc2e RCX: 0000000000080000
RDX: ffffc9001126b000 RSI: 000000000007ffff RDI: 0000000000080000
RBP: ffffc9000dca78a0 R08: ffffea00013313f7 R09: 1ffffd400026627e
R10: dffffc0000000000 R11: fffff9400026627f R12: 1ffff92001b94ef8
R13: dffffc0000000000 R14: ffffea00013313c0 R15: 0000000000000001
rcu_read_unlock include/linux/rcupdate.h:818 [inline]
filemap_get_entry+0x36a/0x3f0 mm/filemap.c:1870
shmem_get_folio_gfp+0x2a5/0x2aa0 mm/shmem.c:1975
shmem_get_folio mm/shmem.c:2164 [inline]
shmem_file_splice_read+0x250/0xc60 mm/shmem.c:2954
splice_direct_to_actor+0x26b/0x8c0 fs/splice.c:1069
do_splice_direct+0x1d5/0x2f0 fs/splice.c:1194
do_sendfile+0x5f2/0xef0 fs/read_write.c:1254
__do_sys_sendfile64 fs/read_write.c:1322 [inline]
__se_sys_sendfile64+0x145/0x1a0 fs/read_write.c:1308
do_syscall_x64 arch/x86/entry/common.c:46 [inline]
do_syscall_64+0x55/0xb0 arch/x86/entry/common.c:76
entry_SYSCALL_64_after_hwframe+0x68/0xd2
RIP: 0033:0x7fa5e899ce59
RSP: 002b:00007fa5e97d0028 EFLAGS: 00000246 ORIG_RAX: 0000000000000028
RAX: ffffffffffffffda RBX: 00007fa5e8c15fa0 RCX: 00007fa5e899ce59
RDX: 0000000000000000 RSI: 0000000000000005 RDI: 0000000000000006
RBP: 00007fa5e8a32d6f R08: 0000000000000000 R09: 0000000000000000
R10: 000000000003ffff R11: 0000000000000246 R12: 0000000000000000
R13: 00007fa5e8c16038 R14: 00007fa5e8c15fa0 R15: 00007ffd6581ebf8
</TASK>
rcu: rcu_preempt kthread starved for 10409 jiffies! g112149 f0x0 RCU_GP_WAIT_FQS(5) ->state=0x0 ->cpu=1
rcu: Unless rcu_preempt kthread gets sufficient CPU time, OOM is now expected behavior.
rcu: RCU grace-period kthread stack dump:
task:rcu_preempt state:R running task stack:27656 pid:17 ppid:2 flags:0x00004000
Call Trace:
<TASK>
context_switch kernel/sched/core.c:5381 [inline]
__schedule+0x1553/0x45a0 kernel/sched/core.c:6700
schedule+0xbd/0x170 kernel/sched/core.c:6774
schedule_timeout+0x188/0x2d0 kernel/time/timer.c:2168
rcu_gp_fqs_loop+0x313/0x1590 kernel/rcu/tree.c:1667
rcu_gp_kthread+0x9d/0x3b0 kernel/rcu/tree.c:1866
kthread+0x2fa/0x390 kernel/kthread.c:388
ret_from_fork+0x48/0x80 arch/x86/kernel/process.c:152
ret_from_fork_asm+0x11/0x20 arch/x86/entry/entry_64.S:293
</TASK>
rcu: Stack dump where RCU GP kthread last ran:
Sending NMI from CPU 0 to CPUs 1:
NMI backtrace for cpu 1
CPU: 1 PID: 0 Comm: swapper/1 Not tainted syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 04/18/2026
RIP: 0010:pv_native_safe_halt+0xf/0x10 arch/x86/kernel/paravirt.c:148
Code: 88 1f 02 c3 cc cc cc cc cc cc cc f3 0f 1e fa 0f 0b 66 2e 0f 1f 84 00 00 00 00 00 f3 0f 1e fa 66 90 0f 00 2d 43 e4 3f 00 fb f4 <c3> 66 0f 1f 00 55 41 57 41 56 41 54 53 50 8b 2f eb 2e 41 89 de 80
RSP: 0018:ffffc90000187de0 EFLAGS: 000002c2
RAX: e8ffe56010701500 RBX: ffffffff8162ae21 RCX: e8ffe56010701500
RDX: 0000000000000001 RSI: ffffffff8acacbe0 RDI: ffffffff8b1c9c20
RBP: ffffc90000187f20 R08: ffff8880b8f36bab R09: 1ffff110171e6d75
R10: dffffc0000000000 R11: ffffed10171e6d76 R12: 1ffff92000030fc8
R13: dffffc0000000000 R14: 1ffff110037cb780 R15: 0000000000000000
FS: 0000000000000000(0000) GS:ffff8880b8f00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f133381f040 CR3: 0000000021149000 CR4: 00000000003506e0
Call Trace:
<TASK>
arch_safe_halt arch/x86/include/asm/paravirt.h:108 [inline]
default_idle+0x13/0x20 arch/x86/kernel/process.c:753
default_idle_call+0x6c/0xa0 kernel/sched/idle.c:97
cpuidle_idle_call kernel/sched/idle.c:178 [inline]
do_idle+0x221/0x590 kernel/sched/idle.c:302
cpu_startup_entry+0x43/0x60 kernel/sched/idle.c:401
start_secondary+0xee/0xf0 arch/x86/kernel/smpboot.c:323
secondary_startup_64_no_verify+0x179/0x17b
</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