Profiling parallel tests

[Andy Armstrong]

On 22 Aug 2008, at 13:58, Nicholas Clark wrote:
>> Correct. But I think it's worth clearing out the structures that
>> relate to tests that have completed too - otherwise get_any() still
>> has a lot of corpses to step over.
>
> There seems to already be some pruning in _find_next_job():
>
>        # Prune any exhausted items.
>        shift @$seq while @$seq && _is_empty( $seq->[0] );
>
> That's doing something different from what you're suggesting?


Dunno :)

My plan was to set up a little test rig to benchmark just the  
scheduler and then proceed on the basis of whatever that throws up.

-- 
Andy Armstrong, Hexten