Am 23.11.2005, 12:40 Uhr, schrieb felix winkelmann <[EMAIL PROTECTED]>:
On 11/23/05, Thomas Chust <[EMAIL PROTECTED]> wrote:
By the way, could someone perhaps explain me what exactly C_mutate does?
I played around with it a little and found that I have apparently
totally
misunderstood the documentation -- I thought that it would cause stack
allocated objects to be copied into heap when the mutated slot was in
the
heap itself.
It does so, but lazily - it schedules this slot to be marked at the next
minor
garbage collection.
[...]
Ah! But the only way to trigger a minor garbage collection is to longjmp
back to the trampoline, right?
Of course stack allocation is preferrable because it is really fast, but
out of curiosity, is there any way to explicitly allocate a garbage
collected block on the heap?
Thanks for the info,
Thomas
_______________________________________________
Chicken-users mailing list
Chicken-users@nongnu.org
http://lists.nongnu.org/mailman/listinfo/chicken-users
