Re: [syzbot] [arm?] BUG: using __this_cpu_read() in preemptible code in __delay

[Will Deacon]

On Thu, Feb 26, 2026 at 07:05:25AM -0800, syzbot wrote:
> syzbot found the following issue on:
> 
> HEAD commit:    32a92f8c8932 Convert more 'alloc_obj' cases to default GFP..
> git tree:       upstream
> console output: https://syzkaller.appspot.com/x/log.txt?x=1019fc02580000
> kernel config:  https://syzkaller.appspot.com/x/.config?x=b77d736d2f4fe1b1
> dashboard link: https://syzkaller.appspot.com/bug?extid=b29d18a34746f34c94cc
> compiler:       aarch64-linux-gnu-gcc (Debian 14.2.0-19) 14.2.0, GNU ld (GNU Binutils for Debian) 2.44
> userspace arch: arm64
> 
> Unfortunately, I don't have any reproducer for this issue yet.
> 
> Downloadable assets:
> disk image (non-bootable): https://storage.googleapis.com/syzbot-assets/fa3fbcfdac58/non_bootable_disk-32a92f8c.raw.xz
> vmlinux: https://storage.googleapis.com/syzbot-assets/aabf4d7d9d6a/vmlinux-32a92f8c.xz
> kernel image: https://storage.googleapis.com/syzbot-assets/0367e274b699/Image-32a92f8c.gz.xz
> 
> IMPORTANT: if you fix the issue, please add the following tag to the commit:
> Reported-by: syzbot+b29d18a34746f34c94cc@xxxxxxxxxxxxxxxxxxxxxxxxx
> 
> BUG: using __this_cpu_read() in preemptible [00000000] code: syz.0.1012/5537
> caller is __this_cpu_preempt_check+0x20/0x40 lib/smp_processor_id.c:64
> CPU: 0 UID: 0 PID: 5537 Comm: syz.0.1012 Tainted: G             L      syzkaller #0 PREEMPT 
> Tainted: [L]=SOFTLOCKUP
> Hardware name: linux,dummy-virt (DT)
> Call trace:
>  show_stack+0x18/0x24 arch/arm64/kernel/stacktrace.c:499 (C)
>  __dump_stack lib/dump_stack.c:94 [inline]
>  dump_stack_lvl+0x7c/0xb0 lib/dump_stack.c:120
>  dump_stack+0x1c/0x28 lib/dump_stack.c:129
>  check_preemption_disabled+0xe0/0xe8 lib/smp_processor_id.c:47
>  __this_cpu_preempt_check+0x20/0x40 lib/smp_processor_id.c:64
>  __arch_counter_get_cntvct_stable arch/arm64/include/asm/arch_timer.h:195 [inline]
>  __delay+0x40/0x374 arch/arm64/lib/delay.c:39
>  __const_udelay+0x48/0x64 arch/arm64/lib/delay.c:66
>  udelay include/asm-generic/delay.h:62 [inline]

This should be fixed by the first patch in this series:

https://lore.kernel.org/r/20260226082234.26707-1-maz@xxxxxxxxxx

I'll queue that today.

Will

#syz fix: arm64: Fix sampling the "stable" virtual counter in preemptible section