[LITMUS^RT] (Nearly) full tickless operation in 3.10

Mac Mollison mollison at cs.unc.edu
Sat May 11 07:06:17 CEST 2013


on Fri, 10 May 2013 16:26:12 -0400
Glenn Elliott <gelliott at cs.unc.edu> wrote:

> Neat!  Seems like CONFIG_NO_HZ_FULL could be compatible with Release
> Master.  Although, the release master CPU would still have to be
> managed by a periodic tick.

I don't think so, because according to the article, you can't
(currently) run more than one process on a CPU that has ticks disabled,
though they plan to rectify that in the future. Or am I missing
something?

I got lucky with this, since my userspace scheduler runs inside only
one process.

- Mac

on Fri, 10 May 2013 16:26:12 -0400
Glenn Elliott <gelliott at cs.unc.edu> wrote:

> Neat!  Seems like CONFIG_NO_HZ_FULL could be compatible with Release
> Master.  Although, the release master CPU would still have to be
> managed by a periodic tick.  I wonder if enabling CONFIG_NO_HZ_FULL
> makes the boot CPU run as CONFIG_NO_HZ_IDLE?  I suppose there's still
> a chance that we could force the boot CPU to run as CONFIG_NO_HZ_IDLE
> with Litmus patches.
> 
> -Glenn
> 
> 
> On May 10, 2013, at 4:10 PM, Christopher Kenna <cjk at cs.unc.edu> wrote:
> 
> > The article below is appearing on some news sites and I thought that
> > this list would be interested in it.
> > 
> > http://lwn.net/SubscriberLink/549580/82983789866b5fad/
> > 
> > _______________________________________________
> > litmus-dev mailing list
> > litmus-dev at lists.litmus-rt.org
> > https://lists.litmus-rt.org/listinfo/litmus-dev
> 
> 
> _______________________________________________
> litmus-dev mailing list
> litmus-dev at lists.litmus-rt.org
> https://lists.litmus-rt.org/listinfo/litmus-dev




More information about the litmus-dev mailing list