Re: make: dynamic scaling of maxJobs ?

[David Young <dyoung%pobox.com@localhost>]

On Fri, Aug 16, 2013 at 08:09:33AM +0100, David Laight wrote:
> On Fri, Aug 16, 2013 at 06:55:48AM +0000, David Holland wrote:
> > On Mon, Aug 05, 2013 at 12:56:27PM -0400, Greg Troxel wrote:
> >  > It's almost like each parallel build should take its fair shares, where
> >  > if you have 12 cpus there are 24 slots available, so each one can
> >  > basically register that it's doing something piggy and then 24/N can be
> >  > used for each.  (I haven't tried this.)
> > 
> > ...gosh, it's scheduling. Perhaps it should be in the kernel with the
> > other pieces of the scheduler...
> 
> Like some features that shares out cpu time between users and not
> processes (or lwps).

Or a feature that bundles a "coalition" of threads (such as make -j N
fires off) into one for the purpose of scheduling disk access.

Dave

-- 
David Young
dyoung%pobox.com@localhost    Urbana, IL    (217) 721-9981