Re: Hang during smokey posix_cond test w/ cobalt and PREEMPT_RT enabled

On 22.07.26 08:59, Dominic Rath wrote:
> On 7/21/2026 10:03 AM, Jan Kiszka wrote:
>>> What I'm not so sure about is the following:
>>>
>>> - is running Xenomai 3 with cobalt on a kernel with PREEMPT_RT
>>> supported, i.e. should this work?
>>>
>>
>> Conceptually, yes. Practically, this combination is very likely
>> undertested (e.g. not part of our CI).
> 
> Understood. Thanks for the explanation.
>>> - is it conceivable that the main thread continues execution in the
>>> Linux domain before lostage_task_wakeup got called? Or does this
>>> indicate that there's something else wrong with our setup?
>>>
>>> I've enabled all of the Xenomai, dovetail and PREEMPT_RT debug checks
>>> that made sense to me and I'm not getting any errors, but of course I
>>> might have missed some useful CONFIG option.
>>>
>>
>> Did you already compare the event flow you are seeing with a reference
>> flow you could generated from qemu-arm64 image of xenomai-images? That
>> may help looking for the essential differences.
>>
>> Furthermore, you could patch xenomai-image to turn on PREEMPT_RT for
>> 6.12 on qemu-arm64 as well and possibly run your test in that setup
>> (provided its portable enough).
> Thanks, that's a great idea, I haven't thought of that. The test is the
> posix_cond test in smokey, i.e. nothing of our own, so I'd assume it is
> portable.
> 

That test is regularly run on arm64 as part of our CI, but only on
kernels without PRREMPT-RT.

> I'll make sure to give this a try and report back, it just might take a
> couple of days.
> 

Perfect.

Jan

-- 
Siemens AG, Foundational Technologies
Linux Expert Center