Re: Thoughts on infant mortality...

[Leopold Toetsch]

Nicholas Clark wrote:

On Thu, Jan 09, 2003 at 01:32:57PM +0100, Leopold Toetsch wrote:

But I still favor the combination of:
- code reordering, like done for pmc_new
- DOD/GC disabling (e.g. aggregate clone)
- active anchoring to the root set, where above is not applicable

Which combined means that there is no stackwalking?

Yep. This is the goal.

ie user code has to have the discipline either to have things linked to the
root set by the time it makes a parrot call or explicitly disable DOD across
such a call.

Yes. We need some XS guidelines anyway. One more rule or less or some ENTER/LEAVE or LOCAL_PMC macros, which do the Right Thing.

(or to explode in a cloud of feathers)

Explode is a harsh word for a seg fault, though somehow softened by the feathers :-)

Nicholas Clark

leo