[LITMUS^RT] Right size for scheduling trace buffer

[Björn Brandenburg]

On Apr 19, 2013, at 12:45 PM, Youcef Amine Benabbas <s9yobena at stud.uni-saarland.de> wrote:

> 
> On Thu, 2013-04-18 at 11:55 -0400, Glenn Elliott wrote:
>> 3) Run the trace tasks as SCHED_FIFO task (check out the "sched_tool"
>> utility).  This will ensure that any idle time from Litmus is spend
>> flushing the ring buffers. (Note: Running tracers as SCHED_FIFO used
>> to crash Litmus, but I believe it has been fixed).
> 
> Indeed, LITMUS^RT crashes. I run my tracing task, then set its priority
> to (SCHED_FIFO, 50). 
> 
> When there are no litmus real-time tasks, everything is fine, however,
> as soon as launch a couple of rtspins, with a wait for synchronous
> release flag, the kernel crashes as I run release_ts, here is part of
> the call trace:
> 
> resched_task
> push_rt_task
> post_scheduler_rt
> schedule
> ? sub_preempt_count
> do_nanosleep
> hrtimer_nanosleep
> ? update_rmtp
> ? hrtimer_start_range_ns
> sys_nanosleep
> system_call_fastpath
> 
> RIP [.....] sched_state_will_scheduler
> 
> I am using litmus 2012.1. Does any one know if the problem was fixed
> before this version? 

It was  fixed by Jonathan and is already merged in master.

	https://github.com/LITMUS-RT/litmus-rt/commit/c158b5fbe42dacd34393082b6b8f9f6573ab8c5c

Version 2012.1 is quite outdated; please use a more recent kernel.

Thanks,
Björn