[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [MiNT] libgem16: vs_color



Helmut Karlowski wrote:
It gets the stack from the kernel. I changed to the trunk-MiNT (ozk-MiNT does also) and everything is even. The ozk-XaAES now detects pixel-format also. So it's all up to using the right kernel (or a fixed fvdi).

Congratulations for this investigation.

Of course, as everyone says, an odd stack is totally unacceptable. It crashes on 68000 and ColdFire, and is slow on 68020+.

I would like to know an additional thing.
Does MiNT really provide XaAES an odd stack at the beginning, or does the stack become odd later ? It would be interesting to know what is the original XaAES stack value. I already said that the addresses of your local variables seem a bit high (not sure if it is a problem or not). A possible scenario could be a correct original stack, then some corruption which moves the stack pointer to a random address (high and odd), then everything will continue apparently well.

If the old FreeMiNT gives XaAES an odd stack, it is a very big issue and the cause must be found and fixed, because it may be random.

If XaAES corrupts the stack itself, of course it must be found, too.

We should not stop before finding the real cause of the original problem.

--
Vincent Rivière