[Avrora] Strange behavior by simulating CPU energy consumption
felodine22 at hotmail.com
Fri Jan 9 09:31:27 PST 2009
I test the algorithm with : x similar bytes followed by y different bytes with x+y=28
When x=4, 14,15, the CPU is active at 100%
For all other cases, it is 3%, which is the correct values.
I just checked with the call monitor and it appears that the last call is not returned if x=4, which explain why the CPU is active at 100%..
But I don't understand, why it is only the case for this case and not for the previous and following...
Thanks a lot for your help
> Date: Fri, 9 Jan 2009 08:35:02 -0800> From: zainul at ee.ucla.edu> To: avrora at lists.ucla.edu> Subject: Re: [Avrora] Strange behavior by simulating CPU energy consumption> > Is the 3% CPU is the correct one and the 100% the wrong one? Is the> error repeatable?> I usually use the call monitor to compare the call trace the CPU went> through for the two cases. This may reveal why the CPU energy became> higher.> > Let us know how that works out.> > Zainul.> > > On Fri, Jan 9, 2009 at 7:03 AM, delphine christin> <felodine22 at hotmail.com> wrote:> > Hello,> >> > I would like to evaluate the CPU energy consumption of my algorithm, which I> > call in the boot.booted event. My algorithm can be tested with different> > data sets, which differ only by one byte. By simulating all data sets, I can> > observe that the CPU is active 3% of the simulation time> > but in some inexplicable cases, the CPU is on for 100% of the time. In one> > case, the simulator indicates that the LEDs are on but I have no LEDs in> > my files . Moreover, I have tested this algorithm with a radio transmission> > and such behaviors were not observeable.> > Do you see any explainations to such very strange behaviors?> >> > Thanks a lot in advance for your answer.> > Delphine> >> > ________________________________> > Discutez sur Messenger où que vous soyez ! Mettez Messenger sur votre mobile> > !> > _______________________________________________> > Avrora mailing list> > Avrora at lists.ucla.edu> > http://lists.ucla.edu/cgi-bin/mailman/listinfo/avrora> >> >> _______________________________________________> Avrora mailing list> Avrora at lists.ucla.edu> http://lists.ucla.edu/cgi-bin/mailman/listinfo/avrora
Découvrez toutes les possibilités de communication avec vos proches
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Avrora