[cgl_discussion] Question for TEMs/ISVs/OEMs regarding pthrea d requirements

Mika Kukkonen mika at osdl.org
Thu Jan 30 13:11:40 PST 2003


Not directly helping on the issue, but might be of some use:

About half a year ago there was an interesting debate between
two hard real-time Linux vendors (RTLinux and TimeSys) about
priority inheritance:
  http://www.linuxdevices.com/articles/AT7168794919.html
and
  http://www.linuxdevices.com/articles/AT5698775833.html

--MiKu

On to, 2003-01-30 at 12:52, Perez-Gonzalez, Inaky wrote:
> 
> Finally somebody answers!!!! Thanks a lot, Pradeep
> 
> > >More specifically, I need to know what are the realtime 
> > requirements of a
> > >pthreads package regarding mutex protocols. POSIX specifies 
> > three priority
> > >protocols, PRIO_NONE, PRIO_PROTECT and PRIO_INHERIT. Does 
> > anybody know or
> > >rely on the three of them and extensively require, for example,
> > >PRIO_INHERIT?
> >
> > Some of the applications that we are developing, rely on 
> > PRIO_INHERIT to
> > avoid priority inversion problem. 
> 
> Well, the point I am trying to make here if is it is worth to implement
> them, because they are a royal PITA; they create so many problems that just
> thinking about it gives me headaches.
> 
> > What tools will you provide to applications if 
> > this option is not supported.
> 
> It is hard to tell, because once you have an application relying on that,
> probably the only way to avoid using it is redesigning the application and I
> am sure you don't want to do it. Hmmm ... 
> 
> Can you give me a rough usage model of the applications that use them? I
> mean, how many threads/processes with different priorities, the
> interdependencies, consumer/producer, etc?
> 
> Thanks, I really appreciate your input
> 
> Inaky Perez-Gonzalez -- Not speaking for Intel - opinions are my own [or my
> fault]
> 
> 
> _______________________________________________
> cgl_discussion mailing list
> cgl_discussion at lists.osdl.org
> http://lists.osdl.org/mailman/listinfo/cgl_discussion
-- 
"Good ideas do not die, they just lie down and get recycled." -- me




More information about the cgl_discussion mailing list