Rough evaluation of execution time

[Thomas Doerfler]

Hello Fabio,

I am not a ERC32 specialist, but I had a short look at the 
simulator. I found out two thing:

If I simply enter 
$ sparc-rtems-sis --help

I get the following output:

zerberus:/home/thomas # /opt/rtems/bin/sparc-rtems-sis --help

 SIS - SPARC intruction simulator 2.7.5,  copyright Jiri Gaisler 1995
 Bug-reports to jgais at wd.estec.esa.nl

unknown option --help
usage: sis [-uart1 uart_device1] [-uart2 uart_device2]
[-nfp] [-freq frequency] [-c batch_file] [files]
[-sparclite] [-dumbio]


I would assume that you can specify the system frequency that 
you want to simulate with option "-freq"

If I start the simulator, and enter the "perf" command, I get 
the following line in output:

 ERC32 performance (14.0 MHz):   nan MOPS (  nan MIPS,   nan 
MFLOPS)


so I would assume, that the ERC32 is simulated by default with 
a 14 MHz clock. Does this make sense?

I general, good simulators will simulate the behaviour of the 
chip on a system clock per clock basis. So the events ocurring 
in the simulated chip really do not depend on the speed of the 
host computer, that simulates the chip. Assuming a 14MHz 
clock, the RTEMS time-of-day should be incremented by one 
second, when 14 million clock cycles have been simulated. I 
treally doesn't matter in this situation, whether your host 
computer has taken 100 Milliseconds or 10 Minutes to simulate 
these 14 Million clock events. 

Does this help you a bit?

wkr,
	Thomas.


> Dear Sirs,
> 	In the last days I sent some questions about a rough evaluation of 
> execution time using gdb/sis simulator of ERC32. Some of my questions 
> didn't find an answer. I kindly ask anyone of you that could help me to 
> reply this message when you find time.
> 
> Could anyone told me where I can find what erc32 is simulated by sis/gdb
> (a 10, 14, or 15 Mhz or what)?
> 
> Joel Sherrill told me that the simulator's notion of
> time is independent of the host computer's
> speed. What about the time resolution? I presume that this is host
> computer dipendent. In which way? When I use the 
> CONFIGURE_MICROSECONDS_PER_TICK for configuring time resolution in RTEMS 
> do I affect the time resolution of the simulator or is it host computer 
> dependent?
> 
> Thank you very much for help
> 
> 
> Fabio Degiovanni
> 
> 
> 
> -- 
> Dott. Ing. Degiovanni Fabio
> Eicas Automazione
> Via Vincenzo Vela, 27 10128 Torino (ITALIA)
> Telefoni +39-11-562.37.98/562.3088 Fax +39-11-436.06.79
> 
> 
> 

--------------------------------------------
IMD Ingenieurbuero fuer Microcomputertechnik
Thomas Doerfler           Herbststrasse 8
D-82178 Puchheim          Germany
email:    Thomas.Doerfler at imd-systems.de
PGP public key available at: http://www.imd-
systems.de/pgp_keys.htm