Execution Time of Whetstone on LEON3

[Joel Sherrill]

On Thu, May 4, 2017 at 2:27 PM, Jiri Gaisler <jiri at gaisler.se> wrote:

>
>
> On 05/04/2017 04:00 PM, Joel Sherrill wrote:
>
>
>
> On Thu, May 4, 2017 at 12:34 AM, Sebastian Huber <
> sebastian.huber at embedded-brains.de> wrote:
>
>> On 03/05/17 22:52, Joel Sherrill wrote:
>>
>>> Has anyone run the whetstone benchmark on real leon3 HW?
>>> I'm almost at 20 minutes of simulated CPU time on tsim so
>>> am suspicious something is wrong.
>>>
>>> Any ideas?
>>>
>>
>> Maybe soft-float plus a low CPU frequency? Do you have a function
>> profiling histogram of the execution?
>>
>>
> It is the default leon3 build and the leon3 is simulated at 50Mhz.  I have
> never used profiling on tsim. Any guidance?
>
> The test  did finally finish and here is the output of tsim's perf command:
>
> *** END OF TEST WHETSTONE ***
>
> Program exited normally.
> tsim> perf
>
>
>  Cycles       : 74462444473
>  Instructions : 45632185502
>  Overall CPI  :      1.63
>
>  CPU performance (50.0 MHz)  : 30.64 MOPS (30.64 MIPS,  0.00 MFLOPS)
>  Cache hit rate              : 98.0 % (inst: 97.9, data:99.7)
>  AHB bandwidth utilisation   : 34.8 % (inst: 14.1, data: 20.8)
>  Simulated time              : 1489.25 s
>  Processor utilisation       : 100.00 %
>  Real-time performance       : 122.02 %
>  Simulator performance       :  37.39 MIPS
>  Used time (sys + user)      : 1220.46 s
>
> So it just took a LONG time.
>
> --joel
>
>
> Note that TSIM reports 0.0 MFLOPS. This means that soft-float was used to
> compile the benchmark. Using hard-float should give about 100x speed-up ...
>
> Thanks. I wonder if we should have a leon3 variant with hardware FPU
enabled.

--joel


> Jiri.
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rtems.org/pipermail/devel/attachments/20170504/aa073f5d/attachment-0002.html>