Mark Duckworth wrote:
For the time being we are stuck with a one size stack. I was
wondering if anyone has contemplated the idea of setting a default
stack size other than 0 bytes?
I think a clean way to fix this problem would be to include a
parameter in MINT.CNF called DefaultStackSize. So when the kernel
loads an executable with extended MiNT header and the stacksize is set
to 0, it will dynamically patch it with the default size.
I think it would be easy to get such behavior by making a little patch
to Pexec(). Such patch would not break any compatibility.
Kernel experts, what do you thing about that ?