Re: [hatari-devel] FPU detection, who is wrong?

[Thorsten Otto <halgara@xxxxxxxx>]

> I have no idea where those 2 values came from, they were there before I
> knew nothing about UAE and I assumed they were correct.

Seems to be clarified now. I can't see anything wrong with using 0x20 for the revision field, as long as the frame size field is set correctly. Looks like the detection code in MiNT is wrong. After all, it's a cosmetic change only, and i'm not sure wether there ever has been any Atari around with a 68881.

But thanks for answering.

Thorsten

Toni Wilen <twilen@xxxxxxxxxx> schrieb am 20:15 Montag, 16.Februar 2015:

> Looking at the 68881 user's manual ...
>
> https://archive.org/stream/bitsavers_motorola68anual1ed1987_15689294/68881_68882_Users_Manual_1ed_1987#page/n244/mode/1up
>
> ... it seems like both are wrong - 68881 can report 0x1f and 0x3f, and
> 68882 reports 0x1f. Newer versions of the FPUs could report even other
> version numbers. To distinguish both, the length field should be used
> instead, I think.
>
> I've cc-ed Toni Wilen ... maybe he knows where the 0x20 comes from...?

I have no idea where those 2 values came from, they were there before I
knew nothing about UAE and I assumed they were correct.

I just tested what my 68030 based Amiga 3000 with 68882 returns, version
number = 0x1f. Unfortunately I don't have any 68881s.