On Tue, Jul 06, 1999 at 04:36:29PM -0700, Phil Daly wrote:
> My development system presently has 2 CPUs but the end product may, in fact,
> have only 1. Obviously the code should not be written with 2 CPUs in mind
> but I was wondering what sort of a performance hit I would take by going
> back to 1 CPU? In the 2 CPU system, I would only be using 1 processor for
> real-time anyway so it's not a simple factor of two. I am trying to figure
> out if there is a performance hit running (for example) SMP aware real-time
> Linux under the 2.2.x kernel on a uniprocessor machine as opposed to the 2.0.36
> kernel.
>
> Not sure I phrased that very well but I hope you get the drift.
You can always compile your kernel without SMP support, and it will
only run on one CPU. Would be useful for testing, especially if you plan
on delivering to run on one CPU.
As for the performance hit, it depends on what you are doing.
dave...
--- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR
echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED]
----
For more information on Real-Time Linux see:
http://www.rtlinux.org/~rtlinux/
