Re: [RFC patch 1/2] sched: dynamically adapt granularity withnr_running

[Mike Galbraith]

On Mon, 2010-09-13 at 11:55 +0200, Peter Zijlstra wrote:
> On Mon, 2010-09-13 at 11:50 +0200, Mike Galbraith wrote:
> > Perhaps lag should be negated if you've received a
> > reasonable chunk or something.. but what we really want is a service
> > deadline.
> 
> Hey, I've got a patch for that too :-)
> 
> Not rebased to anything current, but should be able to get frobbed onto
> the last zero-lag thingy without too much grief (I think).

A frobbing I shall go.

> +/*
> + * Entity is eligible once it received less service than it ought to have,
> + * eg. lag >= 0.
> + *
> + * lag_i = S_i - s_i = w_i*(V - w_i)
> + *
> + * lag_i >=0 -> V >= v_i
> + *
> + *     \Sum (v_i - v)*w_i
> + * V = ------------------ + v
> + *          \Sum w_i
> + *
> + * lag_i >= 0 -> \Sum (v_i - v)*w_i >= (v_i - v)*(\Sum w_i)
> + */

Ew, more of that icky stuff.

	-Mike

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/