Re: [PATCH 12/14] sched: make dl_bw a sub-quota of rt_bw

[Ingo Molnar]

* Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote:

> On Tue, Oct 15, 2013 at 03:11:33PM +0200, Ingo Molnar wrote:
> > The scheduler has over 800 #ifdefs or similar preprocessor directives:
> > 
> >   comet:~/tip> git grep '^#' kernel/sched/ | grep -v include | wc -l
> >   855
> > 
> > I'd like to see this simplified a bit, _especially_ before we add new 
> > complexity ...
> 
> I agree; however I also think its unfair to pile this on Juri.

Well, it's that particular ugly #ifdef that caught my eye.

> [...] Esp the avenue pushed here, which will increase UP text and 
> (runtime) data sizes, to which I think some people will still object.
> 
> There's this entire 'maker' community (aka. internet of things) now 
> using very small Linux devices. Think Raspberry-Pi, Intel Quark (however 
> much I hate Intel for shipping a new 32bit device) etc.
> 
> These people are wanting to run Linux on these tiny devices, with tiny 
> memories etc; we shouldn't make their life harder than it already is.
> 
> Also; you didn't object about the 29 new #ifdefs currently in tip.

It was the straw that broke the camel's back.

> How about we use some of the 'quality' conference time in EDI to do some 
> big code shuffles to get rid of some of them. I'm sure we can reduce the 
> number of ifdeffery by simply merging various similar blocks and maybe 
> write a few extra helper functions.
> 
> This is how kernel/sched/ got created in the first place; as a means to 
> keep the fingers busy at a conference ;-)

LOL, agreed, we can try that ;-)

Also, we can certainly merge SCHED_DEADLINE first, as long as there's a 
hard feature stop afterwards and as long as cleanups will eventually 
arrive.

Thanks,

	Ingo
--
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/