Re: Coroutines with call/cc

["Peder Chr. Norgaard" <pcn@tbit.dk>]

> 
> >>>>> " " == Peder Chr Norgaard <pcn@tbit.dk> writes:
> 
>      > (Except that my need is not threads, but co-routines. Guile
>      > have threads, working fine for many purposes, thank you, but
>      > they are too heavyweight for this kind of work.  Shortly, I
>      > want to "use call/cc to capture the part of the walking state
>      > that is on the stack"; if I can't get that to work, my
>      > alternative is not to use threads, but to capture the state
>      > explicitly - which means that I need to recode a piece of fine
>      > walking code to avoid use of recursion).
> 
> The explanation which brought me closest to understand call/cc is
> Dorai Sitaram's "Teach Yourself Scheme in Fixnum Days"
>   http://www.cs.rice.edu/~dorai/t-y-scheme/t-y-scheme.html
> coroutines are among the examples presented there.

Sure it looks like a well-written piece of tutorial.  I'll give it a try.
Thanks for the reference.

> 
> Are you sure that call/cc is less ``heavyweight'' than threads in
> Guile?  I think Guile needs to copy the entire stack each time a
> continuation is captured.

Yes, correct, but if the stack is not deep (mine isn't) I judge this is no
big deal.  The cost of the guile thread is in terms of memory:  each
thread needs a large piece of RAM that to allow for any possible stack,
both C and Scheme.  That is fine for heavier things like multiplexing
network traffic, for instance, but too expensive for an arbitrary data
access tool.

Whereas the "stack" copy of a call/cc (I think it is not truly a "copy" of
the stack, just a datastructure that holds references to the bindings) is
stored in the heap.

				--peder chr.


Peder Chr. Nørgaard                   System Developer, M. Sc.
Telebit Communications A/S            tel: +45 86 28 81 77 - 49
Fabrikvej 11                          fax: +45 86 28 81 86
DK-8260 Viby J   Denmark              e-mail: pcn@tbit.dk