On Wed, 16 Aug 2017 14:46:42 -0400
Waiman Long [off-list ref] wrote:

I haven't been thinking about RT tasks. You are right that it can be a
problem in this case. I think I will have to revert back to use
mutex_lock() if a RT task is running. Though in this case, the lock
inversion problem will still be there. However, it is highly unlikely
that a RT task will need to read write the block trace sysfs files.

And it is highly unlikely that the lock inversion will happen. But
let's not switch one bug with another. And with PREEMPT_RT coming, that
can boost tasks into being RT, it can make the likelihood of RT tasks
running normally non RT tasks higher.

-- Steve