Problems with GC, trees and array concatenation

[Oskar Linde]

BCS skrev:
> Reply to Oskar,
> 
> 
>> I have no idea why the Individual->real substitution above would make
>> any difference, if not by pure chance.
>>
> 
> Would not real[] allocations be tagged as "does not contain pointers"?
> 

Yes, but I don't see how that would make any difference. There will, as 
far as I can tell, never be any dangling pointers or otherwise spurious 
pointers generated by the Individual class, unless the array 
concatenation code somehow generates non-zero-initialized data.

/Oskar