did you try to count directly using the hardware timers 0,1,2, or 3?

On 5/28/07, Ixent Galpin <[EMAIL PROTECTED]> wrote:

Safraz,

Thanks for pointing me to Avrora, it looks a very promising tool for my
forthcoming research.

With regards to the CPU cycle count problem of zero I was getting last
week, I found the solution to the problem after much playing around.  I
thought it would be useful post the solution I found, as in the mailing
list archive there is another messages relating to the same problem
(with no reply).

Basically, the file bb_cycle_map was not being populated.  This file is
one of the outputs of Power Tossim and contains the mapping from basic
block identifier to number of CPU cycles in that basic block.  I
realised that this was due to the -g (debugging) option not being passed
properly in the nesc app makefile.

Note also, however, that Power Tossim does not seem to work with the
latest version of CIL (1.3.6).  I got it working with 1.3.4 though.

Ixent

Sarfraz Nawaz wrote:
> lxent,
>
> CPU cycle profiling in PowerTossim is not very accurate. You can have a
> look at Avrora or Atemu for more accurate cpu profiling.
>
> On 5/25/07, *Ixent Galpin * <[EMAIL PROTECTED]
> <mailto:[EMAIL PROTECTED]>> wrote:
>
>     Hi,
>
>     I'm trying to get count PowerTossim CPU cycles.  I've followed the
>     instructions at
>     http://www.eecs.harvard.edu/~shnayder/ptossim/install.html
>     <http://www.eecs.harvard.edu/~shnayder/ptossim/install.html> and
>     http://www2.uic.edu/~tcanli1/cpucyclecountingcode.htm, but keep on
>     getting zero values for my CPU cycle count.
>
>     I suspect that this may be related to the version of CIL i am using.
>     Does any one know what version of CIL I should use?
>
>     Unfortunately the CIL and SourceForge websites do not have older
version
>     s of CIL (I've tried with all three versions they have there).  When
>     PowerTossim was produced the most recent version of CIL would have
been
>     1.2.4 (released: February 20, 2004) but I cannot find this version.
>     Their SVN repository doesn't go back that far in time.
>
>     If anyone knows where I might be able to find them, or has any
>     suggestions, I'd be very grateful.
>
>     Ixent
>     _______________________________________________
>     Tinyos-help mailing list
>     Tinyos-help@Millennium.Berkeley.EDU
>     <mailto:Tinyos-help@Millennium.Berkeley.EDU>
>
https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
>
>
_______________________________________________
Tinyos-help mailing list
Tinyos-help@Millennium.Berkeley.EDU
https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help

_______________________________________________
Tinyos-help mailing list
Tinyos-help@Millennium.Berkeley.EDU
https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help

Reply via email to