That should probably read:

It looks like all tests and arithmetic, even on pointers, is 16 bit only, and unsigned arithmetic seems to be sort of signed. That's rather broken. Its not really broken. Its just that typeof(size_t) in this compiler is unsigned int (16bit). You have the same problem when using ancient DOS-compiler...

Uh... But the cookie jar, besides being resident in memory, it is represented in a file, right? No. Cookies like _CPU, _VDO, _AKP etc are set by the OS while booting, depending on detected hardware. The _IDT cookie is only set by mint and magic (or by some other tool), but not by TOS or EmuTOS. If ...

There should be quite some programs that can display the cookie jar (a CPX modules for example). But can't remember one that can change it. Does that make sense at all? Changes will be lost on the next reboot, anyway.

mjaap wrote: ↑Wed Sep 23, 2020 7:37 pm Well, if EmuTOS follows TOS, it will eventually arrive at TOS 4.92 (which could be turned into MultiTOS with MiNT)

Interpolating the time it took to advance from version 0.1 to version 1.0, that will happen in ~83 years

I don't think YAART works on Amigas :D It does if you are running EmuTOS ;) And i think there should be similar programs also for Amigas. Typically, you should be able to specify the number of iterations, so you can let them run for quite some time. And it will immediately tell you when there are a...

Memory testing programs like YAART may also put some pressure on the memory bus.

@thorsten.otto do you think the paj version of 3.06 should work as is then? No i don't think so. There are some other, PAK specific patches. But a large stuff to get it going are the already mentioned things. I don't know if it assumes DMA IIRC, they properly test for its presence. Blitter code is ...

Really what we need is a merged 2.06 and 3.06 build, sort of 2.56. :-) That is essentially what the PAK patches do, using TOS 3.06 (because it is 030 aware), but patch out the TT video modes. BTW the VDI of TOS 2/3 also uses self-modifying code (well not really "modifying", small routines...

That updated package works great Good ;) vdi_fix.S was in a different location from what system/GNUmakefile expected. Ah, yes. That is located in a subdirectory, and was copied to the wrong location when creating the archive. Should be fixed now. aes/winx.h is missing I've included that now, but as...