[LITMUS^RT] SCHED_FIFO priority vs Litmus priority

ayoade gbadebo gbaduz at gmail.com
Fri Feb 28 20:21:35 CET 2014


I have been able to find where litmus rt priorities is defined as shown
below but in the paper titled "A Comparison of Scheduling Latency in Linux,
PREEMPT RT, and LITMUSRT Felipe Cerqueira Bj¨orn B. Brandenburg" - the
paper says "LITMUSRT's fixed-priority plugin currently supports 512
distinct priorities; the priority field is ignored by EDF-based plugins.

That is EDF plugin does not use the priority value set by the the user
Litmusrt  application. So what priority does EDF based plugin use? Is it
based on priority boosting where a value of priority = 0 means the EDF
scheduler should use priority boosting? Can someone elaborate?
"
/include/litmus/rt_param.h

+#define LITMUS_MAX_PRIORITY     512
+#define LITMUS_HIGHEST_PRIORITY   1
+#define LITMUS_LOWEST_PRIORITY    (LITMUS_MAX_PRIORITY - 1)




On Mon, Feb 24, 2014 at 7:56 AM, Björn Brandenburg <bbb at mpi-sws.org> wrote:

>
> On 20 Feb 2014, at 20:56, ayoade gbadebo <gbaduz at gmail.com> wrote:
>
> > I observed the priority for SCHED_FIFO is 2 and LITMUS priority is 6 as
> shown in the header files.
> >
> > Knowing that priority values that are lower means a higher priority
> task, which means that in the case above, sched fifo tasks have higher
> priority than litmus task.
> > Can My assumption be verified ?. I always thought litmus tasks have the
> highest priority in the system.
>
> You are confusing scheduling priorities with the symbolic constants used
> to identify scheduling policies. The #defines that you are looking at have
> nothing to do with scheduling priorities. The LITMUS^RT scheduling class
> indeed takes precedence over SCHED_FIFO.
>
> - 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/20140228/2ea7a7d3/attachment.html>


More information about the litmus-dev mailing list