Re: Real time kernel

[Christoph Lameter]

On Mon, 12 Oct 2009, Dragoslav Zaric wrote:

> This is definition from http://rt.wiki.kernel.org/
>
> -------------------------------------------------------------------------
> Real-time applications have operational deadlines between some
> triggering event and the application's response to that event. To meet
> these operational deadlines, programmers use real-time operating
> systems (RTOS) on which the maximum response time can be calculated or
> measured reliably for the given application and environment.
> -------------------------------------------------------------------------

What is the calculated maximum response time in this case?

> So for specific requirements I would not say, you get slower kernel,
> you get purpose-making kernel, because on these kind of systems you
> probably do not do anything. It is probably the system that listens
> and responds to some events.

Ahh..

> Do you want to say, when using preemption in "normal user mode", Linux
> works slower then without preemption ?
> It makes sense,

Preemption increases code and data size of the kernel which causes the
processors to have to process more instructions and touch more cachelines
to do the same work.

> "because the latency of a task only depends on the tasks running at
> equal or higher priorities, all other tasks can be ignored."
> (http://rt.wiki.kernel.org/index.php/Frequently_Asked_Questions#What_is_real-time.3F)

Latency depends on the number of cachelines touched and on the number of
instructions that have to be processed by a processor until a reaction to
an event can occur. Increasing the latency by make processing more complex
can avoid outliers. You trade speed for guarantees in realtime OSes.


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/