Mail Archives: djgpp/1995/08/29/20:15:49
On Tue, 29 Aug 1995, Charles Sandmann wrote:
> > this really looks like a big bug somewhere: compile the following code and
> > you'll se how v2 eats memory. Nope, it is not a malloc trouble: you can
> > obtain the same using new/delete.
>
> This is a malloc issue and a misunderstanding of what go32_dpmi_get_mem ..
> type calls return. First, malloc allocates in 2**n chuncks minus an overhead,
> so allocating 4Mb of memory probably grabs an 8Mb chunk.
OK, fine for now...
> Second, free()
> just returns that memory to the internal malloc() pool, not to DPMI.
??? shouldn't malloc release this memory to DPMI too?
This way if I malloc() 10Mb and then I free() them the DPMI available memory
to "any program" is now 10Mb less? Sure?
OK, anyway I can live happily with this behavior too, I just wanted to
report that. Also: I'm not sure but maybe in the sample program I sent I was
reading DPMI informations before freeing memory. Just a mistype, in the
original code free() was before.
Another thing:
my viewer is almost ready to ship, anyone willing to give me a home for it
(I mean a ftp site).
The limited version (just few things are missing) is freeware.
Ciao,
Davide.
- Raw text -