[Cs-dev] ParCS Trapped
| Date | 2010-08-25 08:09 |
| From | John ff |
| Subject | [Cs-dev] ParCS Trapped |
In cs_par_dispatch function dag_consume
if (UNLIKELY(dag->first_root == -1)) {
// csp_dag_print(csound, dag);
/* csound->Die(csound, */
/* Str("Expected a root to perform. Found none (%i
remaining)\
"), */
/* dag->remaining); */
RELS_LOCK(&(dag->consume_spinlock));
/* RELS_LOCK(&(dag->spinlock)); */
*node = NULL;
*update_hdl = -1;
{ struct timespec tt = {0, 100};
// nanosleep(&tt, NULL);
}
/* Really ought to wait until someone leaves comsume_dag_update
*/
return;
}
==John ffitch
------------------------------------------------------------------------------
Sell apps to millions through the Intel(R) Atom(Tm) Developer Program
Be part of this innovative community and reach millions of netbook users
worldwide. Take advantage of special opportunities to increase revenue and
speed time-to-market. Join now, and jumpstart your future.
http://p.sf.net/sfu/intel-atom-d2d
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net |
| Date | 2010-08-25 10:21 |
| From | Victor Lazzarini |
| Subject | Re: [Cs-dev] ParCS Trapped |
yes, now it runs. There is no speedup, though, as with xanadu &
sheppard.
Interesting that Felipe says ParCS does not run in more than one Core.
How can check this up, by the way?
It must be running in 2 here because of the speedups.
Victor
On 25 Aug 2010, at 08:09, John ff wrote:
> In cs_par_dispatch function dag_consume
> if (UNLIKELY(dag->first_root == -1)) {
>
> // csp_dag_print(csound, dag);
>
> /* csound->Die(csound, */
> /* Str("Expected a root to perform. Found none (%i
> remaining)\
> "), */
> /* dag->remaining); */
> RELS_LOCK(&(dag->consume_spinlock));
> /* RELS_LOCK(&(dag->spinlock)); */
> *node = NULL;
> *update_hdl = -1;
> { struct timespec tt = {0, 100};
> // nanosleep(&tt, NULL);
> }
> /* Really ought to wait until someone leaves comsume_dag_update
> */
> return;
> }
>
>
> ==John ffitch
>
> ------------------------------------------------------------------------------
> Sell apps to millions through the Intel(R) Atom(Tm) Developer Program
> Be part of this innovative community and reach millions of netbook
> users
> worldwide. Take advantage of special opportunities to increase
> revenue and
> speed time-to-market. Join now, and jumpstart your future.
> http://p.sf.net/sfu/intel-atom-d2d
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel
------------------------------------------------------------------------------
Sell apps to millions through the Intel(R) Atom(Tm) Developer Program
Be part of this innovative community and reach millions of netbook users
worldwide. Take advantage of special opportunities to increase revenue and
speed time-to-market. Join now, and jumpstart your future.
http://p.sf.net/sfu/intel-atom-d2d
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net |
| Date | 2010-08-25 10:25 |
| From | Victor Lazzarini |
| Subject | Re: [Cs-dev] ParCS Trapped |
There is a slight speed up with bigger ksmps. With -k 96 -k 96000,
it's slightly faster with -j 2
Victor
On 25 Aug 2010, at 10:21, Victor Lazzarini wrote:
> yes, now it runs. There is no speedup, though, as with xanadu &
> sheppard.
>
> Interesting that Felipe says ParCS does not run in more than one Core.
> How can check this up, by the way?
> It must be running in 2 here because of the speedups.
>
> Victor
> On 25 Aug 2010, at 08:09, John ff wrote:
>
>> In cs_par_dispatch function dag_consume
>> if (UNLIKELY(dag->first_root == -1)) {
>>
>> // csp_dag_print(csound, dag);
>>
>> /* csound->Die(csound, */
>> /* Str("Expected a root to perform. Found none (%i
>> remaining)\
>> "), */
>> /* dag->remaining); */
>> RELS_LOCK(&(dag->consume_spinlock));
>> /* RELS_LOCK(&(dag->spinlock)); */
>> *node = NULL;
>> *update_hdl = -1;
>> { struct timespec tt = {0, 100};
>> // nanosleep(&tt, NULL);
>> }
>> /* Really ought to wait until someone leaves comsume_dag_update
>> */
>> return;
>> }
>>
>>
>> ==John ffitch
>>
>> ------------------------------------------------------------------------------
>> Sell apps to millions through the Intel(R) Atom(Tm) Developer Program
>> Be part of this innovative community and reach millions of netbook
>> users
>> worldwide. Take advantage of special opportunities to increase
>> revenue and
>> speed time-to-market. Join now, and jumpstart your future.
>> http://p.sf.net/sfu/intel-atom-d2d
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
>
> ------------------------------------------------------------------------------
> Sell apps to millions through the Intel(R) Atom(Tm) Developer Program
> Be part of this innovative community and reach millions of netbook
> users
> worldwide. Take advantage of special opportunities to increase
> revenue and
> speed time-to-market. Join now, and jumpstart your future.
> http://p.sf.net/sfu/intel-atom-d2d
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel
------------------------------------------------------------------------------
Sell apps to millions through the Intel(R) Atom(Tm) Developer Program
Be part of this innovative community and reach millions of netbook users
worldwide. Take advantage of special opportunities to increase revenue and
speed time-to-market. Join now, and jumpstart your future.
http://p.sf.net/sfu/intel-atom-d2d
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net |
| Date | 2010-08-25 16:07 |
| From | Felipe Sateler |
| Subject | Re: [Cs-dev] ParCS Trapped |
On 25/08/10 05:21, Victor Lazzarini wrote: > yes, now it runs. There is no speedup, though, as with xanadu & > sheppard. > > Interesting that Felipe says ParCS does not run in more than one Core. > How can check this up, by the way? I have a CPU graphing monitor that has the peculiarity that it splits the available space in the number of cores/CPUs. So, if a process is eating 100% CPU but only on one of them, it will show up as a constant 50% in my display. -- Saludos, Felipe Sateler ------------------------------------------------------------------------------ Sell apps to millions through the Intel(R) Atom(Tm) Developer Program Be part of this innovative community and reach millions of netbook users worldwide. Take advantage of special opportunities to increase revenue and speed time-to-market. Join now, and jumpstart your future. http://p.sf.net/sfu/intel-atom-d2d _______________________________________________ Csound-devel mailing list Csound-devel@lists.sourceforge.net |