Re: [patch 0/4] timer/nohz: Fix timer/nohz woes

[Paul E. McKenney]

On Sat, Jan 06, 2018 at 10:18:40PM +0100, Thomas Gleixner wrote:
> On Fri, 5 Jan 2018, Paul E. McKenney wrote:
> > But after more than 1,000 hours of test runs, split roughly evenly
> > among the above three scenarios, there is no statistically significant
> > difference in error rate among them.  This means that there is some
> > other bug lurking somewhere, and having the same appearance (lost timer).
> > Were you guys ever able to reproduce this via rcutorture?
> 
> No.

I was afraid of that...  ;-)

> We'll setup more testing on Monday. Which of the tests fails or at least
> exposes the highest failure rate?

TREE01, as in:

bash tools/testing/selftests/rcutorture/bin/kvm.sh --cpus 48 --duration 180 --configs "42*TREE01"

This results in 42 runs of TREE01 consuming about 21 hours of wall-clock
time.  (Each run of TREE01 uses 8 CPUs.)

							Thanx, Paul