Minor FPEmulator bug

SYS&4048F,"Bug:Where's R0?"

Introduced in RO3.80, so I’m a bit late.

And for those of us that don’t have a machine to try this on, what is the bug and what is &4048F in textual form? What result are you expecting?

&4048F is an invalid SWI. The code goes through the usual pattern of looking up the appropriate error, but then does a LDMFD R13!, {R0.... Spotted by code inspection.

I did say minor.

If I understand correctly, you’re trying to say UnknownSWIError in FPEmulator should be changed from:

  131: UnknownSWIError
  132:         STMFD   sp!,{r0-r2,r4,lr}
  133:         ADR     r0, BadSWIToken
  134:         MOV     r1, #0
  135:         MOV     r2, #0
  136:         ADR     r4, FPETitle
  137:         SWI     XMessageTrans_ErrorLookup
  138:         LDMFD   sp!,{r0-r2,r4,pc}

to:

  131: UnknownSWIError
  132:         STMFD   sp!,{r1-r2,r4,lr}
  133:         ADR     r0, BadSWIToken
  134:         MOV     r1, #0
  135:         MOV     r2, #0
  136:         ADR     r4, FPETitle
  137:         SWI     XMessageTrans_ErrorLookup
  138:         LDMFD   sp!,{r1-r2,r4,pc}

Good spot.

MessageTrans_ErrorLookup always sets V. The SETV is superfluous.

MessageTrans_ErrorLookup always sets V

I’ve updated the MessageTrans_ErrorLookup wiki page, as it didn’t state V is set on exit.

👍