Rakudo spectest time and parrot GCs

[Vasily Chekalkin]

Hello.

On Wed, May 4, 2011 at 6:29 AM, Patrick R. Michaud <pmichaud at pobox.com> wrote:
> Version                    T1     T2     T3     T4    Fastest   vs 2011.01
> --------------------------------------------------------------------------
> bench-s1 -- core.pm:
>  2011.01/3.0 (ms2)       254    279    278    275      254       100.0%
>  2011.02/3.1 (ms2)       320    321    320    321      320       126.0%
>  2011.03/3.2 (gms)       264    263    263    264      263       103.5%
>  2011.04/3.3 (gms)       291    292    293    290      290       114.2%

I don't understand this at all... Results on one of my boxes (Debian,
i386, 32 bits, 4GB):

Testing 2011.01
2011.01 run 1: 133 secs
2011.01 run 2: 139 secs
2011.01 run 3: 139 secs
Testing 2011.02
2011.02 run 1: 147 secs
2011.02 run 2: 147 secs
2011.02 run 3: 150 secs
Testing 2011.03
2011.03 run 1: 81 secs
2011.03 run 2: 82 secs
2011.03 run 3: 81 secs
Testing 2011.04
2011.04 run 1: 82 secs
2011.04 run 2: 82 secs
2011.04 run 3: 82 secs

This results obtained in same way as jnthn++ in https://gist.github.com/953615

Looks like we have major regression on amd64.

-- 
Bacek