Re: Nick's scheduler v17

[Nick Piggin <piggin@cyberone.com.au>]

Andrew Theurer wrote:

>>>>    *imbalance = min(this_load - load_avg, load_avg - max_load)
>>>>
>>>>That way you take just enough to either have busiest_queue or
>>>>this_rq's length be the load_avg.  I suppose you could take even
>>>>less, but IMO, the /=2 is what I really don't like.  Perhaps:
>>>>
>>>That is _exactly_ what I had before! Thats probably the way to go. Thanks
>>>for having a look at it.
>>>
>>>
>>>>*imbalance = min(this_load - load_avg, load_avg - max_load);
>>>>*imbalance = (*imbalance + FPT - 1) / FPT;
>>>>
>>>>This should work well for intranode balances, internode balances may
>>>>need a little optimization, since the load_avg really does not really
>>>>represent the load avg of the two nodes in question, just one cpu
>>>>from one of them and all the cpus from another.
>>>>
>>Oh, actually, after my path, load_avg represents the load average of _all_
>>the nodes. Have a look at find_busiest_node. Which jogs my memory of why
>>its not always a good idea to do your *imbalance min(...) thing (I actually
>>saw this happening).
>>
>
>Oops, I meant avg_load, which you calculate in find_busiest_queue on the fly.  
>

OK

>
>>5 CPUs, 4 processes running on one cpu. load_avg would be 0.8 for all cpus.
>>balancing doesn't happen. I have to think about this a bit more...
>>
>
>Actually, if we use avg_load, I guess it would be 0, since this is an unsigned 
>long.  Maybe avg_load needs to have a min value of 1.  Then if we apply:
>

Well its got a fixed point scaling factor.

>
>*imbalance = min(max_load - avg_load, avg_load - this_load)
>	     min(4 - 1, 1 - 0)
>	     	
>

I think you want:

*imbalance = min(max_load - avg_load, avg_load - this_load)
if ( (*imbalance < 1*FPT) &&
        (max_load - this_load) > 1*FPT )
    *imbalance = 1*FPT;

So if there is a total imbalance of more than 1 task, at least one
will be moved.



>
>And imbalance looks a lot better.  Only concern would be an idle cpu stealing 
>from another, leaving the other cpu idle.  I guess a check could be put 
>there.	
>	     
>

pull_task won't pull a running task, so you get some protection there.