Re: A warning for Scheme implementations that derived the Garbage Col lector from SIOD

[Dirk Herrmann]

On Tue, 12 Jun 2001, Carrette, George J. wrote:

> The problem is that the implicit assumption about machine stack organization
> used in the gc_mark_and_sweep were never universally true, and could have
> failed on strange architectures that were highly unlikely to be encountered.
> 
> But now these "highly unlikely" situations of non-linear C call-stack
> management are becoming more likely, and the implementation assumptions are
> soon to hit-the-fan in a hard way as the next generation of 64-bit
> processors become mainstream.
> 
> I've had problems with some SPARC and AIX situations.

In what way do these stack organizations differ?  To me it seems, that a
change in the stack architecture may complicate things, but as long as one
is at all able to determine all data that can be considered part of the
stack, the conservative marking is still feasible.  And, there is _always_
a way to find all data on the stack, at least for the operating system -
it may be, however, that the application may not be allowed to extract
this informtion.  (This, however, also seems unlikely:  How should
debuggers work?)

Best regards,
Dirk Herrmann