[LITMUS^RT] [Litmus^RT]in Global EDF how to get the best cpu available
sijili s
sijili.mec at gmail.com
Wed Dec 13 06:11:24 CET 2017
Hi Björn,
Thank you for your answer. I read a paper about "On the Design and
Implementation of a Cache-Aware Multicore Real-Time Scheduler" by John M.
Calandrino and James H. Anderson.
Is the details available in git hub for ca-sched scheduler.I wanted to know
how the same is implemented.
Sincerely,
Sijili
On Tue, Dec 12, 2017 at 11:49 PM, Björn Brandenburg <bbb at mpi-sws.org> wrote:
>
> > On 12. Dec 2017, at 07:20, sijili s <sijili.mec at gmail.com> wrote:
> >
> > I want to search all the available CPUs and find the best one among them
> based on a criteria. According to my understanding, at present GSN_EDF uses
> the most recent used CPU for scheduling the released task.
> >
> > If we need to search all the CPUs with their current status, where or in
> which module should I implement this operation. As I am new to Linux Kernel
> coding I am really stuck here.
>
> LITMUS^RT’s GSN-EDF implementation can be found in litmus/sched_gsn_edf.c.
>
> https://github.com/LITMUS-RT/litmus-rt/blob/linux-4.9-
> litmus/litmus/sched_gsn_edf.c
>
> The state maintained by the scheduler can be seen here:
>
> https://github.com/LITMUS-RT/litmus-rt/blob/
> 4acc377593580e7d04ad8b42b258e8c2b39030ee/litmus/sched_gsn_edf.c#L108-L114
>
> Understanding how the scheduler works and how one best goes about
> modifying it will take some time, but in the end it’s just a concurrent C
> program. So with some time spent studying it and the surrounding kernel you
> should be able to make sense of it. Unfortunately, there’s not much we can
> do to help here; you’ll just have to study the code yourself.
>
> Regards,
> Björn
>
>
> _______________________________________________
> litmus-dev mailing list
> litmus-dev at lists.litmus-rt.org
> https://lists.litmus-rt.org/listinfo/litmus-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.litmus-rt.org/pipermail/litmus-dev/attachments/20171213/3da57e04/attachment.html>
More information about the litmus-dev
mailing list