New images up with fixes

[Lee Salzman]

It's possible that some structure is the wrong size due to differences
in alignments on the platform. I would check that and see.

Lee

Lendvai Attila wrote:

>:: This image set should have all the latest bug-fixes so /may/ 
>:: provide  
>:: some help to those experiencing the current problem on Windows  
>:: (sorry, I don't have a definite fix yet until some deeper  
>:: investigation is done by Windows users or myself if I can get an  
>:: emulation or real XP installation going).
>
>I've added some output what messages are sent:
>
>Console
>atSlotNamed:
>write:to:from:startingAt:
>Bootstrapping libraries... (this may take a while. Save the image when
>done).
>ByteArray
>Slate: Growing heap to 4786844 bytes.
>
>This is not much...
>
>Then I stopped the process and it was looping in
>PSObjectHeap_findNextFree(), because the PSObject_totalSize() call
>returns zero for an object at address 4934080 (heap starts at around
>4300000). The object is a FormatObject type. Its map is at 0x004300a0
>(early on the heap...)
>
>So the hang is at the first time the gc is triggered and there's a zero
>sized object somewhere in the bootstrapped image. This is strange
>because it should be the same on unices, too.
>
>Novo, if you have any questions I'll be able to look at it early today
>your time, or at monday.
>
>- 101
>
>
>  
>