Re: [RFC PATCH 00/11] sched: CFS low-latency features

[Mathieu Desnoyers <mathieu.desnoyers@efficios.com>]

* Peter Zijlstra (peterz@infradead.org) wrote:
> On Thu, 2010-08-26 at 19:09 -0400, Mathieu Desnoyers wrote:
> > > WTF can't the damned delivery thread not be created when timer_create
> > > is called and the signal be delivered to that very thread directly via
> > > SIGEV_THREAD_ID ?
> > 
> > Yeah, that sounds exactly like what I proposed about an hour ago on IRC ;) I'm
> > pretty sure that would work.
> > 
> > The only thing we might have to be careful about is what happens if the timer
> > re-fires before the thread completes its execution. We might want to let the
> > signal handler detect these overruns somehow.
> 
> Simply don't use SIGEV_THREAD and spawn you own thread and use
> SIGEV_THREAD_ID yourself, the programmer knows the semantics and knows
> if he cares about overlapping timers etc.

>From man timer_create:

       SIGEV_THREAD
              Upon  timer  expiration,  invoke  sigev_notify_function as if it
              were the start function of a new thread.  (Among the implementa‐
              tion  possibilities  here are that each timer notification could
              result in the creation of a new thread, or that a single  thread
              is  created  to  receive  all  notifications.)   The function is
              invoked   with   sigev_value   as   its   sole   argument.    If
              sigev_notify_attributes  is  not  NULL,  it  should  point  to a
              pthread_attr_t structure that defines  attributes  for  the  new
              thread (see pthread_attr_init(3).

So basically, it's the glibc implementation that is broken, not the standard.

The programmer should expect that thread execution can overlap though.

Thanks,

Mathieu


-- 
Mathieu Desnoyers
Operating System Efficiency R&D Consultant
EfficiOS Inc.
http://www.efficios.com