Explanation about the high level design choices of Strong APA

[Richi Dubey]

Yes, This is a serious issue. I will work on finding this and documenting
each case to make sure that there are no loose ends. Meanwhile, please have
a look at the low-level description.

Thank you for the review.


On Thu, Aug 6, 2020 at 8:12 PM Gedare Bloom <gedare at rtems.org> wrote:

> Thanks for the detailed write-up. My only concern with your choice is
> whether these two function pointers, get_highest_ready and
> get_lowest_ready, will get called along other paths when there hasn't
> been a task arrival/departure? This shouldn't cause a correctness
> problem, but it may cause a performance problem.
>
>  A bit relatedly, are there other task arrivals/departures that you
> should understand to make sure you are covering things completely? Do
> they all traverse the Scheduler_Unblock/Scheduler_Block? If not, then
> it would be a correctness problem.
>
> Gedare
>
> On Tue, Aug 4, 2020 at 8:20 AM Richi Dubey <richidubey at gmail.com> wrote:
> >
> > Hi,
> >
> > Implementing the Strong APA scheduler requires finding places to put the
> two most important algorithms (pages 5 and 6 from the paper) in the current
> RTEMS SMP scheduler framework.
> >
> > I have chosen get_highest_ready and get_lowest_ready to implement these
> function and my reasoning behind doing so is documented in the blog post-
> https://rtemswithrichi.wordpress.com/adding-strong-apa-to-rtems-identifying-relevant-functions/
> .
> >
> > I would love to hear your insights about the post.
> > Thanks,
> > Richi.
> > _______________________________________________
> > devel mailing list
> > devel at rtems.org
> > http://lists.rtems.org/mailman/listinfo/devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rtems.org/pipermail/devel/attachments/20200806/570c1e0b/attachment.html>