Re: [Linux kernel bug] INFO: task hung in blk_mq_get_tag

[Tetsuo Handa]

On 2024/05/14 11:05, Sam Sun wrote:
> Pointed out by Tetsuo, this kernel panic might be caused by sending
> NMI between cpus. As dump log shows:

You can do

  # echo 0 > /proc/sys/kernel/hung_task_all_cpu_backtrace

before starting the reproducer in order to disable
trigger_all_cpu_backtrace() from hung task.

If trigger_all_cpu_backtrace() from hung task can reproduce
the kernel panic, I think that trigger_all_cpu_backtrace() from
SysRq-l should be as well able to reproduce the kernel panic.
That is, you can build with CONFIG_MAGIC_SYSRQ=y and try

  # echo l > /proc/sysrq-trigger

for bisecting and reporting as a separate bug.