Re: [hatari-devel] Code execution discontinuities and detecting them?

[ Thread Index | Date Index | More lists.tuxfamily.org/hatari-devel Archives ]



Interrupts are quite easy to detect in the DSP, as the first $40
locations of DSP56001 on-chip P:memory are used for interrupt vectors.
(These adresses work 2 by 2)

P:$0000 3  Hardware RESET (External)
P:$0002 3 Stack Error
P:$0004 3 Trace
P:$0006 3 SWI (Software Interrupt)
...
P:$003C 0-2 Available for Host Command
P:$003E P:$0000 or P:$E000 0-2 Illegal Instruction


When a program interrupts, the PC goes to one of these location (eg
:$002E and P:$002F),
So there's no separate exception handler invocation instruction,
like TRAP on m68k?


There's only the vector table at the beginning of the P: memory


does execute the 1 or 2 instructions there and then
returns to the original code.
Do you mean that exception handler can be only 1-2 instructions...?

No, the DSP PC will run to the vector address as described in the previous mail, and execute the 1 or 2 instructions there (2 if each instruction is 1 DSP word size, one else). One of these 2 instructions can be a JSR, ... and the PC will continue until it reachs a RTI.

Example :

Vector $26 contains (that's wrong, of course, as the JSR is a 2 word instruction because of the jump address, but it's just for the example)
P:$0026 JSR $2000
p:$0027 NOP



; We're starting here in the main program

P:$1000 NOP
P:$1001 NOP
P:$1002 NOP

Interrupt here at vector $26

P:$0026 JSR P:$02000        ; Jump to the interruption subroutine

P:$2000 ADD X1,Y1
P:$2001 RTI                        ; end of the interruption subroutine

P:$0027 NOP                        ; second vector interrupt

P:$1003 NOP                    ; main program again
P:$1004 NOP
....


Careful, if the first instruction is a jump, the PC may not go to the
second vector address
...or just that the code can be elsewhere? :-)

Yes, it can be elsewhere, or directly in the 2 words addresses directly into the vector.


Is RTI instruction involved in both cases?

Yes



Can there be exceptions within exception handler, and
in which case RTI returns just to previous handler?

? Sorry, I don't understand what you're asking here.

Regards

Laurent





Mail converted by MHonArc 2.6.19+ http://listengine.tuxfamily.org/