Re: Setting memory-chunk-size claimed when free-memory approacheszero?

[Martin Egholm Nielsen <martin at egholm-nielsen dot dk>]

> When GC_collect_or_expand decides to expand the heap, it first
> tries to expand the heap by it's desired amount.  If that fails,
> it tries to expand by what it things it needs to satisfy the
> current allocation request.  If both of those fail, it tries
> to garbage collect GC_max_retries times and then gives up.
Well, it seems that my application is terminated before gc even gets the 
chance to chance to try the second approach.

> The current default behavior is to set GC_max_retries to zero,
> unless GC_MAXIMUM_HEAP_SIZE is set, in which case it's two by
> default.  This makes a bit of sense, since if GC_MAXIMUM_HEAP_SIZE
> is set, presumably it's set to avoid paging disasters.  (On an
> embedded platform, paging is probably not an issue, but you do
> want to set GC_MAXIMUM_HEAP_SIZE anyway, which should
> give you reasonable behavior.)
That did work, yes... But I think it's only the second-best solution :-) 
It would be better if I didn't have to specify the amount of memory 
available. But ok, it's an embedded system, so I know more or less how 
much is available for my application. But I still have to be rather 
conservative, for otherwise I may shoot above the target.
So, I guess the "correct" way of doing it is to specify 
GC_INITIAL_HEAP_SIZE to be the same as GC_MAXIMUM_HEAP_SIZE, as well. 
Hence, then I'm sure that amount of memory is available all the time...

> If you are really trying to adjust how aggressively the GC grows the
> heap, then setting GC_free_space_divisor instead or in addition
> is a good idea.
This can only be adjusted from within some C-code, right? I browsed misc.c
(http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/boehm-gc/misc.c?rev=1.30), 
and didn't find any environment-variable that suited...

Regards,
 Martin Egholm