Thread IDs

[Stefan Kalkowski]

Hi Peter,

of course you can use the former explained functionality of the cpu-session 
interface to obtain the thread id of a newly created one, as all threads are 
created in fact via the cpu-session interface.
But as long as you do not 'hand code' the whole thread creation process, but 
use the 'Thread_base' class resp. your own derived form of it, there is a 
handier way to do this.
The 'Genode::Thread_base' class and its derivates contain a 'Native_thread' 
member. This is the platform-specific representation of the thread within the 
generic this class. So you might get the thread id by the following snippet:

  // Assuming your thread object is called 'th'
  PDBG("thread id: %lx", th.tid().l4id.raw);

The value of 'tid()' will be valid only, after the thread to be created, 
already executed its bootstrapping code (at least under OKL4 this 
bootstrapping code copies the thread id to the thread object). So if you want 
to print out the thread id immediately after you created the thread, you 
should synchronize the thread creator's context with the new thread's context 
e.g. by using a lock:

  // creation context
  My_thread_class th("thread_xy");
  th.start();
  _startup_lock.lock();
  PDBG("id of new the new thread: %lx", th.tid().l4id.raw);

  ...

  // entry code of 'thread_xy'
  _startup_lock.unlock();

I hope this is what you were looking for.

Regards
Stefan

On Wednesday, 20. January 2010 12:11:41 Peter Nguyen wrote:
> I was more wondering how to get the numerical ID of the thread after
> i've created one. Is it possible to obtain that at all?
>
> Peter
>
> On 19/01/2010, at 8:06 PM, Stefan Kalkowski wrote:
> > Hi Peter,
> >
> > On Tuesday, 19. January 2010 06:30:22 Peter Nguyen wrote:
> >> Hi,
> >>
> >> So I've done some more messing around with genode (running on top
> >> of OKL4),
> >> more specifically creating some threads, and testing the
> >> functionality in
> >> general. I'm wondering how i can go about obtaining the thread's ID
> >> and
> >> printing it out. I've noticed that there are functions that are
> >> within the
> >> Thread class that can get me a Native_thread_id, but currently I
> >> can't seem
> >> to do much with it.
> >>
> >> Speaking of the native thread, is that type related to L4? Can I
> >> actually
> >> get a numerical value for that ID?
> >
> > that's right, 'Native_thread_id' is the platform-specific thread
> > representation, so in the OKL4-case it is 'L4_ThreadId_t' indeed. The
> > kernel-specific thread id in Genode/OKL4 is a member of
> > 'Thread_state' a
> > thread property, that can be obtained via the cpu-session interface.
> >
> > Although, it is recommended to use only generic Genode abstractions
> > to stay
> > platform-independent, if you've to deal with the kernel-api
> > directly, you
> > might use the cpu-session interface to gain some platform-specific
> > thread
> > data within 'Thread_state'.
> >
> > Here is some example code, printing all OKL4 thread IDs of one cpu-
> > session:
> >
> >  int i;
> >  Genode::Thread_state state;
> >  Genode::Thread_capability tcap = Genode::env()->cpu_session()-
> >
> > >first();
> >
> >  while(tcap.valid())
> >  {
> >     Genode::env()->cpu_session()->state(tcap, &state);
> >     PDBG("tid of thread no. %d is %lx", i, state.tid.raw);
> >     tcap = Genode::env()->cpu_session()->next(tcap);
> >     ++i;
> >  }
> >
> > Thereby, the order of the threads given by cpu-session's iterator
> > isn't
> > necessarily the order of the thread's creation time!
> >
> > Regards
> > Stefan
> >
> >> Peter
> >>
> >> ------------------------------------------------------------------------
> >>--- --- Throughout its 18-year history, RSA Conference consistently
> >> attracts the
> >> world's best and brightest in the field, creating opportunities for
> >> Conference attendees to learn about information security's most
> >> important
> >> issues through interactions with peers, luminaries and emerging and
> >> established companies. http://p.sf.net/sfu/rsaconf-dev2dev
> >> _______________________________________________
> >> Genode-main mailing list
> >> Genode-main at lists.sourceforge.net
> >> https://lists.sourceforge.net/lists/listinfo/genode-main
> >
> > --
> > Stefan Kalkowski
> > Genode Labs Developer
> > http://genode-labs.com
> >
> > -------------------------------------------------------------------------
> >----- Throughout its 18-year history, RSA Conference consistently attracts
> > the
> > world's best and brightest in the field, creating opportunities for
> > Conference
> > attendees to learn about information security's most important
> > issues through
> > interactions with peers, luminaries and emerging and established
> > companies.
> > http://p.sf.net/sfu/rsaconf-dev2dev
> > _______________________________________________
> > Genode-main mailing list
> > Genode-main at lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/genode-main
>
> ---------------------------------------------------------------------------
>--- Throughout its 18-year history, RSA Conference consistently attracts the
> world's best and brightest in the field, creating opportunities for
> Conference attendees to learn about information security's most important
> issues through interactions with peers, luminaries and emerging and
> established companies. http://p.sf.net/sfu/rsaconf-dev2dev
> _______________________________________________
> Genode-main mailing list
> Genode-main at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/genode-main



-- 
Stefan Kalkowski
Genode Labs Developer
http://genode-labs.com