|Re: [hatari-devel] Hatari WinUAE vs oldUAE information, and MiNT with MMU?|
[ Thread Index |
| More lists.tuxfamily.org/hatari-devel Archives
On 14 Jan 2021 at 16:03, Thorsten Otto wrote:
> On Donnerstag, 14. Januar 2021 15:31:06 CET Christian Zietz wrote:
> > Do you know if an unimplemented FPU instruction on a 68060 causes a
> > dedicated trap?
> The 060 can generate several of them, due to some instructions being removed:
> line-f trap, unimplemented effective address, and unimplemented integer
> instruction at least (may be more, i have to check that).
> Edit: the FPSP has entry points for:
> 0x000: _060_fpsp_snan
> 0x008: _060_fpsp_operr
> 0x010: _060_fpsp_ovfl
> 0x018: _060_fpsp_unfl
> 0x020: _060_fpsp_dz
> 0x028: _060_fpsp_inex
> 0x030: _060_fpsp_fline
> 0x038: _060_fpsp_unsupp
> 0x040: _060_fpsp_effadd
> So i think the OS should catch at least those vectors. The first 6 could also
> be generated by other FPUs, IIRC, and the linef could just mean there is no
> FPU at all.
If we want to talk about EmuTOS, let's take this to the EMuTOS mailing list