Mail Archives: djgpp/1993/01/02/01:57:45
>It's not the malloc() that causes the swap space to fill. Remember, the
>memory is never actually allocated until ACCESSED.
I guess now I understand the problem. How about calloc? That does access the
memory (set it to zero).
>I thought os/2 also provided a similiar functionality (this is demand paging,
>right?). If so, does os/2 set a limit on how much memory can be requested
>by each task?
For a DOS program running under OS/2, one of the "DOS settings" is the
maximum amount of DPMI memory it can request. I guess I don't know what
happens if this is set larger than the available swap space and you try to
use all of the memory.
Some DPMI environments fix the disk swap size. For example, Quarterdeck's
QDPMI (which actually comes from Ergo) does this. That probably makes a
coreleft function much easier to write.
- Raw text -