[nptl] Re: SCHED_RR issue

["Loic Domaigne" <loic-dev@gmx.net>]

> > BTW. Perhaps it would be interesting to dump the thread info, even 
> > if the test passed (?)
> 
> I cannot predict what we should see... maybe some kind of linear results
> reflecting the age of the threads... ?

First, it is additional "confirmation" that test indeed PASSED... 

Second, you are right: it reflects something like the age of the thread.
This help sometimes to infer some patterns about the underlying OS 
scheduler.

For instance, on TRU64 4.0F it seems that thread 1, then 2, ... is 
started-- that seems "natural", but SUS do not impose any scheme 
like this. 

BTW. Just started 1024 threads are prio 30 on 2xEV56 system / TRU64. 
Everything's OK so far. 


Cheers,
Loic.

-- 
--
// Sender address goes to /dev/null (!!) 
// Use my 32/64 bits, ANSI C89, compliant email-address instead:   

unsigned y[]=
{0,34432,26811,16721,41866,63119,61007,48155,26147,10986};
void x(z){putchar(z);}; unsigned t; 
main(i){if(i<10){t=(y[i]*47560)%65521;x(t>>8);x(t&255);main(++i);}}

Supergünstige DSL-Tarife + WLAN-Router für 0,- EUR*
Jetzt zu GMX wechseln und sparen http://www.gmx.net/de/go/dsl

Liste de diffusion nptl
Pour se désinscrire : mailto:nptl_request@bullopensource.org?subject=unsubscribe