On 08/16/2017 02:59 PM, Steven Rostedt wrote:

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

quoted

I haven't been thinking about RT tasks. You are right that it can be a=

quoted

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

That is true too:-)

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

I am thinking about maybe letting a RT task to sleep a tiny amount of
time instead of calling schedule(). Hopefully, that will allow a
lower-priority task to make forward progress.

Cheers,
Longman