Re: make: dynamic scaling of maxJobs ?

[David Laight <david%l8s.co.uk@localhost>]

On Mon, Aug 05, 2013 at 05:12:40AM +1000, matthew green wrote:
> 
> load isn't going to great heuristic for this.  when a job is
> waiting on disk/nfs io, it is +1 to the load.  so "idle" things
> add to the load, espcially in a build.

There are other cases where idle processes end up incrementing the
'load average', and it is also possible to generate very high cpu load
with a low load average!

An example of the former is a device-driver 'helper' process that
always sleeps non-interruptably.

The latter can happen when processes to a moderate amount of work
on every timer tick - eg audio processing of (say) RTP streams.

        David

-- 
David Laight: david%l8s.co.uk@localhost