The benchmark output of is more than likely running at a coarser
granularity then your actual simulation's CPUs.

For instance, the actual code may be calling some "gettime"-like functions
that ask for time in seconds but your CPUs are running say 2GHz and zooming
through the parallel sections in less then a second of "simulated time".

Thus you would get zeros in the benchmark output but the benchmark still
did finish correctly.

And yes, you should use gem5-dev in my opinion.

On Fri, Jan 27, 2012 at 9:34 AM, Sangamesh K <[email protected]> wrote:

> Hi All,
>
> I am gem5 beginner and I am trying to run splash benchmarks in ALPHA_SE
> for an in-order CPU. The results obtained were all zeros.
>
> I am using gem-stable and command is as follows-
> build/ALPHA_SE/gem5.opt configs/example/se.py --inorder --caches -c
> ./../splash2/codes/kernels/fft/FFT -o "-t -p1"
>
> -----------------------------------------------------------------------------------------------------------------------------
>
> FFT with Blocking Transpose
>    1024 Complex Doubles
>    1 Processors
>    65536 Cache lines
>    16 Byte line size
>    4096 Bytes per page
>
>
>                  PROCESS STATISTICS
>             Computation      Transpose     Transpose
>  Proc          Time            Time        Fraction
>     0                 0              0          NaNQ
>
>                  TIMING INFORMATION
> Start time                        :                0
> Initialization finish time        :                0
> Overall finish time               :                0
> Total time with initialization    :                0
> Total time without initialization :                0
> Overall transpose time            :                0
> Overall transpose fraction        :             NaNQ
>
>               INVERSE FFT TEST RESULTS
> Checksum difference is 0.000 (1033.228, 1033.228)
> TEST PASSED
> warn: ignoring syscall sigprocmask(18446744073709547831, 1, ...)
> Exiting @ tick 2181887000 because target called exit()
>
> -----------------------------------------------------------------------------------------------------------------------------
>
> I am able to successfully run the splash benchmarks in SE mode using
> splash2/run.py. Editing the run.py script to use InOrderCPU() did not do
> any good and  results were same as above (zeros). Has anyone tried running
> the splash2 benchmarks using inorder CPU?
>
> Thanks,
> Sangam
>
> _______________________________________________
> gem5-users mailing list
> [email protected]
> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
>



-- 
- Korey
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Reply via email to