[lttng-dev] Using lttng-ust with xenomai

[Jan Kiszka]

On 22.11.19 18:44, Norbert Lange wrote:
> Am Fr., 22. Nov. 2019 um 16:52 Uhr schrieb Jan Kiszka <jan.kiszka at siemens.com>:
>>
>> On 22.11.19 16:42, Mathieu Desnoyers wrote:
>>> ----- On Nov 22, 2019, at 4:14 AM, Norbert Lange nolange79 at gmail.com wrote:
>>>
>>>> Hello,
>>>>
>>>> I already started a thread over at xenomai.org [1], but I guess its
>>>> more efficient to ask here aswell.
>>>> The basic concept is that xenomai thread run *below* Linux (threads
>>>> and irg handlers), which means that xenomai threads must not use any
>>>
>>> I guess you mean "irq handlers" here.
>>>
>>>> linux services like the futex syscall or socket communication.
>>>>
>>>> ## tracepoints
>>>>
>>>> expecting that tracepoints are the only thing that should be used from
>>>> the xenomai threads, is there anything using linux services.
>>>> the "bulletproof" urcu apparently does not need anything for the
>>>> reader lock (aslong as the thread is already registered),
>>>
>>> Indeed the first time the urcu-bp read-lock is encountered by a thread,
>>> the thread registration is performed, which requires locks, memory allocation,
>>> and so on. After that, the thread can use urcu-bp read-side lock without
>>> requiring any system call.
>>
>> So, we will probably want to perform such a registration unconditionally
>> (in case lttng usage is enabled) for our RT threads during their setup.
> 
> Who is we? Do you plan to add automatic support at xenomai mainline?
> 
> But yes, some setup is likely needed if one wants to use lttng

I wouldn't refuse patches to make this happen in mainline. If patches 
are best applied there. We could use a deterministic and fast 
application tracing frame work people can build upon, and that they can 
smoothly combine with system level traces.

> 
> 
>>>
>>> That's indeed a good point. I suspect membarrier may not send any IPI
>>> to Xenomai threads (that would have to be confirmed). I suspect the
>>> latency introduced by this IPI would be unwanted.
>>
>> Is an "IPI" a POSIX signal here? Or are real IPI that delivers an
>> interrupt to Linux on another CPU? The latter would still be possible,
>> but it would be delayed until all Xenomai threads on that core eventual
>> took a break (which should happen a couple of times per second under
>> normal conditions - 100% RT load is an illegal application state).
> 
> Not POSIX, some inter-thread interrupts. point is the syscall waits
> for the set of
> registered *running* Linux threads. I doubt Xenomai threads can be reached that
> way, the shadow Linux thread will be idle and it won't block.
> I dont think its worth extending this syscall (seems rather dangerous actually,
> given that I had some deadlocks with other "lazy schemes", see below)

Ack. It sounds like this will become messy at best, fragile at worst.

> 
>>
>>>
>>>> liburcu has configure options allow forcing the usage of this syscall
>>>> but not disabling it, which likely is necessary for Xenomai.
>>>
>>> I suspect what you'd need there is a way to allow a process to tell
>>> liburcu-bp (or liburcu) to always use the fall-back mechanism which does
>>> not rely on sys_membarrier. This could be allowed before the first use of
>>> the library. I think extending the liburcu APIs to allow this should be
>>> straightforward enough. This approach would be more flexible than requiring
>>> liburcu to be specialized at configure time. This new API would return an error
>>> if invoked with a liburcu library compiled with --disable-sys-membarrier-fallback.
>>>
>>> If you have control over your entire system's kernel, you may want to try
>>> just configuring the kernel within CONFIG_MEMBARRIER=n in the meantime.
>>>
>>> Another thing to make sure is to have a glibc and Linux kernel which perform
>>> clock_gettime() as vDSO for the monotonic clock, because you don't want a
>>> system call there. If that does not work for you, you can alternatively
>>> implement your own lttng-ust and lttng-modules clock plugin .so/.ko to override
>>> the clock used by lttng, and for instance use TSC directly. See for instance
>>> the lttng-ust(3) LTTNG_UST_CLOCK_PLUGIN environment variable.
>>
>> clock_gettime & Co for a Xenomai application is syscall-free as well.
> 
> Yes, and that gave me a deadlock already, if a library us not compiled
> for Xenomai,
> it will either use the syscall (and you detect that immediatly) or it
> will work most of the time,
> and lock up once in a while if a Linux thread took the "writer lock"
> of the VDSO structures
> and your high priority xenomai thread is busy waiting infinitely.
> 
> Only sane approach would be to use either the xenomai function directly,
> or recreate the function (rdtsc + interpolation on x86).

rdtsc is not portable, thus a no-go.

> Either compiling/patching lttng for Cobalt (which I really would not
> want to do) or using a
> clock plugin.

I suspect you will want to have at least a plugin that was built against 
Xenomai libs.

> If the later is supposed to be minimal, then that would mean I would
> have to get the
> interpolation factors cobalt uses (without bringing in libcobalt).
> 
> Btw. the Xenomai and Linux monotonic clocks arent synchronised at all
> AFAIK, so timestamps will
> be different to the rest of Linux.

CLOCK_HOST_REALTIME is synchronized.

> On my last plattform I did some tracing using internal stamp and
> regulary wrote a
> block with internal and external timestamps so those could be
> converted "offline".

Sounds not like something we want to promote.

Jan

> Anything similar with lttng or tools handling the traces?
> 
> regards, Norbert
> 

-- 
Siemens AG, Corporate Technology, CT RDA IOT SES-DE
Corporate Competence Center Embedded Linux