Stef wrote:Chilly Willy wrote:So close! I get to the very end and get this from the build logs:
m68k-elf-ld: address 0x4bc90 of bin/Release/gencube3d section .data is not within region ram
I can't find anything on it that seems to match. I spent the last DAY figuring out it wasn't compiler specific. I tried 3.4.0, 3.4.5, 4.1.0, and 4.1.1. I also tried the uClinux toolchains with the same results.
Generally this message comes from a bad md.ld definition file.
Are you using the last version (4 i believe) of the mini dev kit ? If yes, unfortunatly the gencube3D example is no more compatible (it can be modified though). Try with the included sample (partic or something like that) and keep me informed about your result, thanks

Yes, I'm using version 4. I also recompiled the .o files in case that was an issue (the ABI changes sometimes between compiler versions). I fixed the differences in gencube3D between what was there and what it is now. The reported error is the only one.
Using the partic sample that comes in the dev kit yields the same error:
m68k-elf-ld: address 0x28cbcc of bin/Debug/particle section .data is not within region ram
The only change I made to that sample was you had "#include "vangelisS.h"" when the name was capitalized (linux is case sensitive).
Given your dev kit was done for Windows based stuff, I wonder if there's a difference that needs to be accounted for in the ld file. I need to read up more on those. I found this page on the uClinux site about absolute linking. I noticed that they use --remove-section during the objcopy phase. I don't know if any of that applies here.
http://uldp.home.at/uclinux_doc_4.html
EDIT: Well, it's something about the ld file. I modified the one from the linked page and it goes through, but the rom doesn't work. Here's what I've done so far. I think I'm missing a section out of the rom part that the dev kit relies on to make the rom work. I'll keep hacking at the sections part until it breaks again, or it gets fixed.
EDIT 2: Well, the problem was the BLOCK (0x4) in the data line. Remove that and it doesn't give the error. However, the resulting bin file doesn't work. Now I have to find out why that is. A quick look shows the header to be fine (same as the bin that comes with the sample). Just a note, I'm using the original md.ld file again, with just the change made to the data section line to remove the BLOCK (0x4).