Catch-It <PICLIST@MITVMA.MIT.EDU> wrote:

> If I have got this wrong could somebody correct me please?

    Gladly.

> 1> MOVF   AL,W    ;low byte
> 2> ADDWF  BL,F
> 3> MOVF   AML,W   ;Middle low byte
> 4> SKPNC
> 5> INCFSZ AML,W
> 6> ADDWF  BML,F
> 7> MOVF   AMH,W   ;Middle high byte
> 8> SKPNC
> 9> INCFSZ AMH,W
>
> 1 Status affected: None
> 2 (OK, say we add to numbers and generate a carry )  Status
>   affected: C,DC,Z   >Carry bit set
> 3 Status affected: None >Carry bit set<
> 4 Status affected: None >Carry bit set<
> 5 OK, we add the carry to the higher byte and skip stuff at line 6
>   (0 in the next byte). Status affected: None   >Carry bit set<
> 6 THIS WOULD SET THE CARRY FLAG CORRECTLY
> 7 Status affected: None  >Carry bit set<
> 8 OH DEAR! Carry bit still set from the original add!!! BUG BUG!

    Sigh.... It's NOT a bug; it's SUPPOSED to work that way.

    Go through your example again, more carefully:

    1 & 2:      AL is added to BL and a carry is generated.

    3, 4, & 5:  The carry is propagated to AML.  The result happens
                to be 0, so we skip step 6.

                The carry remains set.  This is correct, since the
                result of step 5 could ONLY have been 0 if AML were
                incremented from FF to 0.

    6:          Skipped.

    7 & 8:      Increment AMH, since there's a carry from step 5
                (when we incremented AML from FF to 0).  No bug.

    -Andy

=== Andrew Warren - fastfwd@ix.netcom.com
=== Fast Forward Engineering - Vista, California
=== http://www.geocities.com/SiliconValley/2499