Tim, The parser assembly code is running but produces no image. I've fixed several segmentation faults, including one that was occurring when a fractal type has no parameters. Increasing the iteration count increases the time the non-image takes to generate. It appears the formula is being processed more or less correctly. I know there are still some problems with certain op codes, but I can't check them until I can see an image. It looks like the pointer to Z is set correctly and the new value is getting back to the C code. There may be a problem with eax as it is passed back to C, but that should make the code bailout all the time. Have you run into anything like this before? I'm hoping the solution occurs to me shortly, it usually does within hours of sending one of these emails. 8-)) Jonathan
Tim,
Have you run into anything like this before?
No but I'll bet you are close! :-)
Yes, I just have no idea what to look for. That leaves me making changes blindly. Not too efficient. I'll try saving CPU registers tonight. I needed to do that with the mandelbrot code. Jonathan
Jonathan wrote:
Yes, I just have no idea what to look for. That leaves me making changes blindly. Not too efficient. I'll try saving CPU registers tonight. I needed to do that with the mandelbrot code.
Isn't there a debugger for that environment? You could step through the code and see where it screws up. Tim
On Tuesday 12 February 2002 21:03, Tim wrote:
Jonathan wrote:
Yes, I just have no idea what to look for. That leaves me making changes blindly. Not too efficient. I'll try saving CPU registers tonight. I needed to do that with the mandelbrot code.
Isn't there a debugger for that environment? You could step through the code and see where it screws up.
Not that I know of. I should check the web site. Jonathan
participants (2)
-
Jonathan Osuch -
Tim Wegner