Re: [hatari-devel] Possible problem with Pexec emulation

[Eero Tamminen <oak@xxxxxxxxxxxxxx>]

Hi,

On 19.11.2022 8.36, Thorsten Otto wrote:
> On Samstag, 19. November 2022 06:27:28 CET Roger Burrows wrote:
> > I would assume the bus error occurs due to the blitter hardware trying to
> > write to $e00000, and the PC just happens to be at $e12924 in the standard
> > "wait for blitter" loop.
>
> Yes, i assume that too. Using --trace-vdi indead shows that vr_recfl() was
> called immediately before that. Unfortunately it does not tell me the used
> parameters. Is there some way to add them to the output?

Add VDI info breakpoint for VDI traps:
-------------------------------
b VdiOpcode ! 0xffff :quiet :info vdi
-------------------------------

Hatari VDI trap output will then look like this:
-------------------------------
Latest VDI Parameter block:
- Opcode/Subcode: 0x21/0x00 (vsin_mode)
- Device handle: 1
- Control: 0x0000d360
- Ptsin:   0x0000d238, 0 coordinate word pairs
- Ptsout:  0x0000d446, 0 coordinate word pairs
- Intin:   0x0000d260, 2 words
- Intout:  0x0000d45e, 1 words
VDI 0x21/0x00 (vsin_mode)
-------------------------------

If you need more information, write a debugger script that provides the 
required info, and use ":file <path/to/script>" breakpoint option instead.

(Script path can be skipped if you cd to script dir first.  If you do 
that 'cd' from a debugger script, remember to add '-f' option after the 
path.)


	- Eero

> Would also be interesting to know whether that crash also happens on real
> hardware.