Mercurial > hg > index.cgi
diff src/int.s @ 82:9a4e2364a966
Fix logic in int32_mul and overflow integer multiply to floating point
It seems logical to allow integer multiplication to overflow to floating
point. If this turns out to be unfortunate, it can be changed. In this
update, 32 bit integer multiplication will overflow to floating point.
| author | William Astle <lost@l-w.ca> |
|---|---|
| date | Sat, 07 Oct 2023 13:39:25 -0600 |
| parents | fbc14509955a |
| children | a492441bfc56 |
line wrap: on
line diff
--- a/src/int.s Sat Oct 07 12:59:43 2023 -0600 +++ b/src/int.s Sat Oct 07 13:39:25 2023 -0600 @@ -68,18 +68,18 @@ orb fpa0extra1 orb fpa0extra2 orb fpa0extra3 - bne OVERROR2 ; brif not - overflowed + bne int32_mul4 ; brif not - overflow to floating point ldb fpa0extra4 ; is bit 31 set? bpl int32_mul2 ; brif not - no overflow lda ,s ; negative result wanted? - bpl OVERROR2 ; brif not - we overflowed + bpl int32_mul4 ; brif not - overflow to floating point andb #0x7f ; lose extra sign bit - orb fpa0extra5 ; "or" in other bytes to see if all but bit 31 are zero + orb fpa0extra2 ; "or" in other bytes to see if all but bit 31 are zero orb fpa0extra6 orb fpa0extra7 - bne OVERROR2 ; brif any nonzero bits - we overflowed maximum negative number - ldb ,s+ ; do we want a negative result? - bpl int32_mul2 ; brif not + bne int32_mul4 ; brif any nonzero bits - we overflowed maximum negative number +int32_mul2 ldb ,s+ ; do we want a negative result? + bpl int32_mul3 ; brif not - don't negate result ldd zero ; negate result subd fpa0extra6 std fpa0extra6 @@ -87,11 +87,15 @@ sbcb fpa0extra5 sbca fpa0extra4 std fpa0extra4 -int32_mul2 ldd fpa0extra4 ; copy result to destination +int32_mul3 ldd fpa0extra4 ; copy result to destination std val.int,y ldd fpa0extra6 std val.int+2,y rts +int32_mul4 puls b ; get back desired sign + sex ; set proper sign for floating point result + ldx #fpa0extra ; point to 64 bit unsigned result + jmp fps_fromuint64s ; go convert to floating point using the sign in A ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; ; 32 bit multiply. ;