Re: [Gems-users] Statistics

["Dave Z." <zhu_dave@xxxxxxxxx>]

Dan Gibson <degibson@xxxxxxxx> wrote:
> >  - Ruby statistics provide cumulative results for all processors. Is it possible to obtain statistics per chip or processor? For example, cc transitions on each processor?
> >  
>  Some statistics are already broken-down on a per-processor basis (eg total misses). Others can be trivially added by modifying ruby/profiler/Profiler.[Ch]
>
> I don't see how some statistics are stated on a per-processor basis. In my configuration, there are two processors but I only see total_misses: 39269 [39269 0]. Does each number in brackets refer to a processor? If so, it means the benchmark ran only on one of the processors. Is there a way to assign threads to processors?
>  Like processor affinity in Linux?  
>
> > - My benchmarks take a long time to simulate with Simics + Ruby. I would like to run a certain section of my benchmarks (e.g. initialization phase) in Simics only, and then load Ruby to get statistics of the memory system. Would this reduce the simulation time? How can this be done?
> >  
>  The best way to do this is to use a Simics "magic" instruction to break the simulation. Refer to the simics documentation on how to use a magic instruction -- your simics script will look something like this:
>  magic-break-enable
>  c       # run until you get a "magic" break
>  load-module ruby
>  ruby0.init
>  c       # run until you get another "magic" break, signalling the end of execution
>  ruby0.dump-stats filename
>
> I used
>  the magic instruction in my benchmark, MAGIC(1) and MAGIC(2) as mentioned in simics user guide. But when I ran the simulation, it didn't stop at the magic breakpoint but continued until the end. (I'm using sarek with solaris 10.) I didn't get any compilation errors. What could be the problem? 
Thank you in advance.

---

Get your own web address for just $1.99/1st yr. We'll help. Yahoo! Small Business.