During my work on a "hobby-amiga-os" I was confound by an issue with enabling interrupts. What would happen was that when my code turned on interrupts the CPU would immediately jump to $0 (likely the value of the interrupt vector pointer at boot).
I confirmed that my code works on vAmiga, and running on a 68020 emulation with vbr on FS-UAE. This leads to the conclusion that there is a bug in 68000/010 emulation code, perhaps related to the issue of caching.
I have included a test-case in assembly language in the archive attached to the report, it also contains pre-compiled roms, config files and a Makefile for easy building..
Since this is still in early development it has not been tested on original amiga hardware.
Best regards.
bug-report.zip
During my work on a "hobby-amiga-os" I was confound by an issue with enabling interrupts. What would happen was that when my code turned on interrupts the CPU would immediately jump to $0 (likely the value of the interrupt vector pointer at boot).
I confirmed that my code works on vAmiga, and running on a 68020 emulation with vbr on FS-UAE. This leads to the conclusion that there is a bug in 68000/010 emulation code, perhaps related to the issue of caching.
I have included a test-case in assembly language in the archive attached to the report, it also contains pre-compiled roms, config files and a Makefile for easy building..
Since this is still in early development it has not been tested on original amiga hardware.
Best regards.
bug-report.zip