Re: cgroup trace events acquire sleeping locks

[Steven Rostedt]

On Tue, 10 Jul 2018 11:26:35 +0200
Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote:

> > Hmm, good question. I could possibly make all the tracepoint code into
> > its own section. And then look to see if any spin locks exist in them.
> > That wouldn't be too trivial to implement though.  
> 
> pick a bit from the preempt_count (say right above NMI_MASK) and set it
> inside a trace-event and add in_trace().
> 
> Then make lockdep explode when in_trace().
> 
> Or something along those lines.

Sure. We have current->trace for special tracing flags. That could
easily be used.

-- Steve