Miro Kropáček

Hmmm... I can imagine arguments pro and against, will check as well. CT2 TOS (7.04) is definitely patched on many places, so everything is possible. My guess would be 0x04000000...

So, let's start from the top: - all kernels (before your change, with your change and with your patch) boot OK with MP disabled - phystop/memtop is 0x00E00000 (as it...

I think you are on the right track there -- I was wrong with `Mxalloc`, it definitely returns the "classic" 0x01xxxxxx values (that's tested in TOS, mint+mp [the working one]...

Btw I'd love to know what was the reason for this change as well: https://github.com/freemint/freemint/commit/8a210bf42a469e5e79f48b377ce06b2ba04fac69?diff=unified#diff-b94a6c9a0a5b828ee7c42154ca9d11718ff1162f9907752395f3a10ecc20993cR108

I mean the change from `oldtc` to `oldcrp+8`, it just makes it, erm, less readable?

Oh, I think I found the source of the ST RAM magic: https://mikro.naprvyraz.sk/mint/199810/msg00078.html _So when installing mint, you shouldn't be in fastram while changing the PMMU and you need to...

PMMU is definitely set by TOS 7.04 but I guess we are resetting it again in FreeMiNT. The rest in the afternoon.

That's even scarier. :) @xdelatour can you try attached binary then? Compiled using gcc 7.5 and fairly recent gemlib (Oct 2022) / mintlib (Apr 2023): `m68k-atari-mint-gcc -O2 -fomit-frame-pointer -s -o...

@th-otto that really looks like what is happening. I'm wondering whether this is really desired behaviour... TosWin2 could refuse to buffer more or start overwriting the buffer from the beginning...

Very nice. If I had to pick one, I couldn't. ;-) Both ELF and threads are incredibly valuable to have. All right, maybe ELF would be a bit cooler to...