resolve #6; general clean-up

asm/forth-dictionary.s

@@ -3102,7 +3102,7 @@ getbuf:   lda   [SYSVARS],y         ; get buffer pointer
 setbuf1:  ldy   #SV_SBUF1         ; select buf 1
 setbuf:   pla                     ; update pointers
           sta   YR                ; in YR
-          sta   [SYSVARS],y         ; in the appropriate system var
+          sta   [SYSVARS],y       ; in the appropriate system var
           sta   STACKBASE+4,x     ; in the parameter stack
           iny
           iny
@@ -3110,7 +3110,8 @@ setbuf:   pla                     ; update pointers
           sta   YR+2
           sta   [SYSVARS],y
           sta   STACKBASE+6,x
-          jsr   _move
+          sec                     ; move down is faster
+          jsr   _memmove
           NEXT
 nomem:    ldy   #.loword(-18)
           lda   #.hiword(-18)
@@ -3779,7 +3780,7 @@ dword     MOVE,"MOVE"
           jsr   _popxr
           jsr   _popyr
           jsr   _popwr
-          jsr   _move
+          jsr   _memmove
           NEXT
 eword
 
@@ -3793,7 +3794,8 @@ dword     CMOVEUP,"CMOVE>"
           jsr   _popxr
           jsr   _popyr
           jsr   _popwr
-          jsr   _moveup
+          clc
+          jsr   _memmove_c
           NEXT
 eword
 
@@ -4483,7 +4485,8 @@ dword     PACK,"PACK"
           inc   YR
           bne   :+
           inc   YR+2
-:         jsr   _move
+:         sec                     ; move down is faster
+          jsr   _memmove_c
           NEXT
 bad:      ldy   #.loword(-18)
           lda   #.hiword(-18)
@@ -5396,7 +5399,8 @@ hword     SCONCAT,"SCONCAT"
           lda   XR+2
           adc   STACKBASE+2,x
           sta   STACKBASE+2,x
-          jsr   _move             ; move the string
+          sec                     ; move down is faster
+          jsr   _memmove_c        ; move the string
           NEXT
 eword
 

asm/interpreter.s

@@ -899,72 +899,6 @@ good:     sec
 .endproc
 .a16
 
-; Move XR bytes from [WR] to [YR], starting at the bottom
-; trashes WR, YR, and XR
-; could be optimized to move words, excepting the last one if odd number of bytes
-; use for moving data downward, but that adds two comparison instructions which
-; are slower than the SEP/REP, maybe
-.proc     _move
-.if 1 ; fast move in memmgr.s
-          sec
-          jmp   _memmove_c
-.else ; slower but smaller move
-lp:       lda   XR+2                ; see if zero bytes
-          ora   XR
-          bne   :+
-          rts
-:         jsr   _decxr              ; pre-decrement XR
-          sep   #SHORT_A
-          lda   [WR]
-          sta   [YR]
-          rep   #SHORT_A
-          jsr   _incwr              ; post increment WR
-          inc   YR                  ; and YR
-          bne   lp
-          inc   YR+2
-          bra   lp
-.endif
-.endproc
-
-; Move XR bytes from [WR] to [YR], starting at the top
-; trashes YR and XR
-; could be optimized to move words, excepting the last one if odd number of bytes
-; use for moving data upward
-.proc     _moveup
-.if 1 ; fast move in memmgr.s
-          clc
-          jmp   _memmove_c
-.else ; slower but smaller move
-lp:       jsr   _wrplusxr
-          lda   WR                  ; move WR to 1 past the end of the block
-          clc
-          adc   XR
-          sta   WR
-          lda   WR+2
-          adc   XR+2
-          sta   WR+2
-          lda   YR                  ; move YR to 1 past the end of the block
-          clc
-          adc   XR
-          sta   YR
-          lda   YR+2
-          adc   XR+2
-          sta   YR+2
-          lda   XR+2
-          ora   XR
-          bne   :+
-          rts
-:         jsr   _decxr              ; decrement XR
-          jsr   _decwr              ; and WR
-          jsr   _decyr              ; and YR
-          sep   #SHORT_A
-          lda   [WR]
-          sta   [YR]
-          rep   #SHORT_A
-          bra   lp
-.endif
-.endproc
-
 ; With word header address in YR, set YR to previous dictionary entry header
 ; return with Z flag set if the new address is zero
 .proc     _prevword

asm/mathlib.s

@@ -20,6 +20,7 @@ samesign: lda   STACKBASE+6,x
           cmp   STACKBASE+0,x
 :         rts
 .endproc
+_test32   = _stest32::samesign
 
 .proc     _invertay
           pha

asm/memmgr.s

@@ -419,17 +419,16 @@ loop:     jsr   _chk_himem
 done:     rts
 .endproc
 
-.if 1 ; include fast memory move routines
 ; Memory move routines
 ; move XR bytes of memory from [WR] to [YR]
 
-; Move appropriately based
+; Move appropriately based on source and destination
 .proc     _memmove
           lda   WR
           cmp   YR
           lda   WR+2
           sbc   YR+2
-          ; now carry is set if WR >= XR, move down in that case, otherwise move up
+          ; now carry is set if WR >= YR, move down in that case, otherwise move up
           ; fall-through
 .endproc
 
@@ -509,7 +508,7 @@ md6:      pha
           jsl   f:_callzr
           plb  
           pla  
-          eor   #$ffff            ; a xor $ffff = $ffff - a = -1 - a
+          eor   #$FFFF            ; a xor $FFFF = $FFFF - a = -1 - a
           adc   sizel
           sta   sizel             ; sizel = sizel - 1 - a
           sep   #$20
@@ -517,10 +516,9 @@ md6:      pha
           lda   sizeh             ; update high byte of size
           sbc   #$00
           bcs   md1
-          rep   #$30
+          plp
           .a16
           .i16
-          plp
           plx
           rts
 .endproc
@@ -533,11 +531,27 @@ tol       = YR
 sizeh     = XR+2
 sizel     = XR
 mu7       = ZR
+          ; first convert start addresses to end addresses
+          lda   WR
+          clc
+          adc   XR
+          sta   WR
+          lda   WR+2
+          adc   XR+2
+          sta   WR+2
+          lda   YR
+          clc
+          adc   XR
+          sta   YR
+          lda   YR+2
+          adc   XR+2
+          sta   YR+2
+          ; now start the move
           phx
           php
           lda   #$6B00            ; RTL
           sta   mu7+2
-          lda   #$0054            ; MVN
+          lda   #$0044            ; MVP
           sta   mu7
           sep   #$21              ; 8-bit accumulator, set carry
           .a8
@@ -557,25 +571,25 @@ mu7       = ZR
           cmp   froml
           bcs   mu3
           bra   mu4
-          .a8
+          .a8                     ; a is 8 bits when we branch to mu1!
 mu1:      sta   sizeh
-          eor   #$80   ; set v if size is zero, clear v otherwise
+          eor   #$80              ; set v if size is zero, clear v otherwise
           sbc   #$01
           cpx   #$FFFF
-          bne   mu2    ; if x is not $FFFF, then y must be
+          bne   mu2               ; if x is not $FFFF, then y must be
           dec   mu7+2
           rep   #$20
           .a16
           tya  
           cpy   #$FFFF
           bne   mu4
-          .a8
           sep   #$20
+          .a8
 mu2:      dec   mu7+1
           rep   #$20
           .a16
 mu3:      txa
-mu4:      bvc   mu5    ; branch if sizeh is nonzero
+mu4:      bvc   mu5               ; branch if sizeh is nonzero
           cmp   sizel
           bcc   mu6
           lda   sizel
@@ -585,18 +599,17 @@ mu6:      pha
           jsl   f:_callzr
           plb  
           pla  
-          eor   #$FFFF ; a xor $FFFF = $FFFF - a = -1 - a
+          eor   #$FFFF            ; a xor $FFFF = $FFFF - a = -1 - a
           adc   sizel
-          sta   sizel  ; sizel = sizel - 1 - a
+          sta   sizel             ; sizel = sizel - 1 - a
           sep   #$20
           .a8
-          lda   sizeh  ; update high byte of size
+          lda   sizeh             ; update high byte of size
           sbc   #$00
           bcs   mu1
-          rep   #$30
+          plp
           .a16
           .i16
-          plp
           plx
           rts
 .endproc
@@ -623,4 +636,3 @@ mu6:      pha
           pla                     ; ( 0 zrh zrl ah al -- 0 zrh zrl )
           rtl                     ; ( 0 zrh zrl -- )
 .endproc
-.endif