[LITMUS^RT] question about rtspin -- Sisu

[Jonathan Herman]

The cputime() function is in liblitmus/bin/clocks.c, and calls
clock_gettime with CLOCK_THREAD_CPUTIME_ID. This returns the amount of CPU
time consumed by that thread only, not the current system time.


On Wed, Apr 17, 2013 at 5:00 AM, Björn Brandenburg <bbb at mpi-sws.org> wrote:

> On 17.04.2013, at 04:57, Sisu Xi <xisisu at gmail.com> wrote:
>
> > I am trying to understand the rtspin program and have some question
> about it.
> >
> > 1) Seems it will read execution time from a file. Is this the execution
> time for each individual job?
>
> Yes
>
> > Is there any sample file for this? How should I format the file?
>
> I don't have a sample file, but the format should be fairly simple. Chris
> Kenna implemented this; he might be able to tell you more about this.
>
> Anyway, the file support is entirely optional, you don't have to use it (I
> usually don't).
>
> >
> > 2) The budget enforcement, if configured as precise_enforcement, will
> force job to yield when it exceed wcet.
> > However, if I configure it as no_enforcement, the wcet parameter won't
> be used. cause the scheduling decision are made based on period/deadline,
> is this correct?
>
> That depends on which scheduling plugin you are using. The scheduling
> policy has nothing to do with rtspin.
>
> > 3) Is it possible to run multiple rtspin task at the same time? I see it
> records time using:
> >
> >       while (now + last_loop < start + exec_time) {
> >               loop_start = now;
> >               tmp += loop_once();
> >               now = cputime();
> >               last_loop = now - loop_start;
> >
> > If multiple tasks are running, would't the time be incorrect?
>
> No. Why would it?
>
> - Björn
>
>
> _______________________________________________
> litmus-dev mailing list
> litmus-dev at lists.litmus-rt.org
> https://lists.litmus-rt.org/listinfo/litmus-dev
>



-- 
Jonathan Herman
Department of Computer Science at UNC Chapel Hill
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.litmus-rt.org/pipermail/litmus-dev/attachments/20130417/29c380c5/attachment.html>