This is the mail archive of the guile@sources.redhat.com mailing list for the Guile project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: GUILE's GC - why we struggling to solve already solved problems ? (was Re: finding the top of the stack in linux)


28-Jul-00 03:06 you wrote:
> "Khimenko Victor" <guile@khim.sch57.msk.ru> writes:

>> But why the hell we are [...] not even using clearly
>> appropriate parts (like stack placement finding subroitine) from ALREADY
>> existing and TESTED solutions.

> [...]

>> we still can use parts of it to determine stack placement on
>> different systems.

> Would you be willing to cut out and prepare the stack finding code for
> Guile?

I will try to do it. It's not trivial, unfortunatelly. It's easy to cut this
code. It's MUCH harder to do so correct. As you can guess it's big pile of
#ifdef's (if it's Linux then we do it this way, if it's HP-UX then that way
and if it's Win32 then still another code is used) and so we have verification
problem :-/ Plus there are other question: how to do this - manual cut and
paste with tuning or script to convert from GC_ to scm_GC prefix and strip
out unneeded code from appropriate .h and .c files ? Manual way is easier
but then we'll need to repeat this process once every month or two (when new
version of BGC is out).

> You don't need special Guile knowledge to do that.  Only make sure
> that identifiers are either static or have the prefix scm_.

Yes, but there are compileability problem - see above :-/

> The interface for finding top of main stack simply needs to return a
> pointer.  We need to discuss the interface for finding the top for the
> rest of the threads.




Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]