[Mono-devel-list] Language Shootout - problems with Mono times

[Paolo Molaro]

On 08/31/04 Isaac Gouy wrote:
> We're having problems getting consistent times for the Mono C# programs
> in  "The Great Computer Language Shootout" 
> (Mono C# has been pulled from the shootout once already because of this
> problem - it would be good to fix it this time.)
[...]
> Even Linux time gives inconsistent results:
[...]
>    root at lin mono_shootout # time mono sieve.exe 1000
>    Count: 1028
> 
>    real    0m0.977s
>    user    0m0.030s
>    sys     0m0.020s
[...]
> Explanations and suggestions appreciated.
> (The shootout Perl script grabs user & sys times, forks, execs the mono
> program, and grabs user & sys times.)

I'm not sure how we could help, here: this looks like a libc/kernel bug
in resource usage accounting. Maybe mono's usage of threads confuses it.
I'm sure the kernel guys would love to hear about how a user process may
defeat resource constraints (if it is indeed a kernel bug:-).
Could you try with different kernel versions (which version do you
have?). On my 2.6.8.1 I always get user+sys much smaller than real time.
Note that, for example, time on MacOSX seems to report sensible time
results when running the same test with mono.

lupus

-- 
-----------------------------------------------------------------
lupus at debian.org                                     debian/rules
lupus at ximian.com                             Monkeys do it better