@@ -54,9 +54,6 @@

 (   - mul32 memory, 12 bytes )

 (   - _divmod32 memory, 16 bytes )

-%DEBUG { #ff #0e DEO }

-%RTN  { JMP2r }

-%TOR { ROT ROT } ( a b c -> c a b )

 %COMPLEMENT32 { SWP2 #ffff EOR2 SWP2 #ffff EOR2 }

 %DUP4 { OVR2 OVR2 }

 %POP4 { POP2 POP2 }

@@ -74,7 +71,7 @@

     ,&loop JMP

     &done

     POP ( n )

-    RTN

+    JMP2r

 @bitcount16 ( x* -> n^ )

     SWP ( xlo xhi )

@@ -82,40 +79,40 @@

     DUP #00 NEQ ( xlo nhi nhi!=0 )

     ,&hi-set JCN ( xlo nhi )

     SWP ;bitcount8 JSR2 ADD ( nhi+nlo )

-    RTN 

+    JMP2r 

     &hi-set

     SWP POP #08 ADD ( nhi+8 )

-    RTN

+    JMP2r

 @bitcount32 ( x** -> n^ )

     SWP2 ( xlo* xhi* )

     ;bitcount16 JSR2 ( xlo* nhi )

     DUP #00 NEQ ( xlo* nhi nhi!=0 )

     ,&hi-set JCN ( xlo* nhi )

-    TOR ;bitcount16 JSR2 ADD RTN ( nhi+nlo )

+    ROT ROT ;bitcount16 JSR2 ADD JMP2r ( nhi+nlo )

     &hi-set

-    TOR POP2 #10 ADD ( nhi+16 )    

-    RTN

+    ROT ROT POP2 #10 ADD ( nhi+16 )    

+    JMP2r

 ( equality )

 ( x == y )

 @eq32 ( xhi* xlo* yhi* ylo* -> bool^ )

     ROT2 EQU2 STH

-    EQU2 STHr AND RTN

+    EQU2 STHr AND JMP2r

 ( x != y )

 @ne32 ( xhi* xlo* yhi* ylo* -> bool^ )

     ROT2 NEQ2 STH

-    NEQ2 STHr ORA RTN

+    NEQ2 STHr ORA JMP2r

 ( x == 0 )

 @is-zero32 ( x** -> bool^ )

-    ORA2 #0000 EQU2 RTN

+    ORA2 #0000 EQU2 JMP2r

 ( x != 0 )

 @non-zero32 ( x** -> bool^ )

-    ORA2 #0000 NEQ2 RTN

+    ORA2 #0000 NEQ2 JMP2r

 ( comparisons )

@@ -123,58 +120,58 @@

 @lt32 ( x** y** -> bool^ )

     ROT2 SWP2 ( xhi yhi xlo ylo )

     LTH2 ,&lt-lo JCN ( xhi yhi )

-    LTH2 RTN

+    LTH2 JMP2r

     &lt-lo

-    GTH2 #00 EQU RTN

+    GTH2 #00 EQU JMP2r

 ( x <= y )

 @lteq32 ( x** y** -> bool^ )

     ROT2 SWP2 ( xhi yhi xlo ylo )

     GTH2 ,&gt-lo JCN ( xhi yhi )

-    GTH2 #00 EQU RTN

+    GTH2 #00 EQU JMP2r

     &gt-lo

-    LTH2 RTN

+    LTH2 JMP2r

 ( x > y )

 @gt32 ( x** y** -> bool^ )

     ROT2 SWP2 ( xhi yhi xlo ylo )

     GTH2 ,&gt-lo JCN ( xhi yhi )

-    GTH2 RTN

+    GTH2 JMP2r

     &gt-lo

-    LTH2 #00 EQU RTN

+    LTH2 #00 EQU JMP2r

 ( x > y )

 @gteq32 ( x** y** -> bool^ )

     ROT2 SWP2 ( xhi yhi xlo ylo )

     LTH2 ,&lt-lo JCN ( xhi yhi )

-    LTH2 #00 EQU RTN

+    LTH2 #00 EQU JMP2r

     &lt-lo

-    GTH2 RTN

+    GTH2 JMP2r

 ( bitwise operations )

 ( x & y )

 @and32 ( xhi* xlo* yhi* ylo* -> xhi|yhi* xlo|ylo* )

-    ROT2 AND2 STH2 AND2 STH2r RTN

+    ROT2 AND2 STH2 AND2 STH2r JMP2r

 ( x | y )

 @or32 ( xhi* xlo* yhi* ylo* -> xhi|yhi* xlo|ylo* )

-    ROT2 ORA2 STH2 ORA2 STH2r RTN

+    ROT2 ORA2 STH2 ORA2 STH2r JMP2r

 ( x ^ y )

 @xor32 ( xhi* xlo* yhi* ylo* -> xhi|yhi* xlo|ylo* )

-    ROT2 EOR2 STH2 EOR2 STH2r RTN

+    ROT2 EOR2 STH2 EOR2 STH2r JMP2r

 ( ~x )

 @complement32 ( x** -> ~x** )

-    COMPLEMENT32 RTN

+    COMPLEMENT32 JMP2r

 ( temporary registers )

 ( shared by most operations, except mul32 and div32 )

-[ @x0 $1 @x1 $1 @x2 $1 @x3 $1

-  @y0 $1 @y1 $1 @y2 $1 @y3 $1

-  @z0 $1 @z1 $1 @z2 $1 @z3 $1

-  @w0 $1 @w1 $1 @w2 $2 ]

+@m32 [ &x0 $1 &x1 $1 &x2 $1 &x3 $1

+       &y0 $1 &y1 $1 &y2 $1 &y3 $1

+       &z0 $1 &z1 $1 &z2 $1 &z3 $1

+       &w0 $1 &w1 $1 &w2 $2 ]

 ( bit shifting )

@@ -184,40 +181,40 @@

     DUP #10 LTH ;rshift32-1 JCN2 ( x n )

     DUP #18 LTH ;rshift32-2 JCN2 ( x n )

     ;rshift32-3 JMP2 ( x n )

-    RTN

+    JMP2r

 ( shift right by 0-7 bits )

 @rshift32-0 ( x** n^ -> x<<n )

-        STHk  SFT                      ;z3 STA  ( write z3 )

-    #00 STHkr SFT2 #00 ;z3 LDA ORA2 ;z2 STA2 ( write z2,z3 )

-    #00 STHkr SFT2 #00 ;z2 LDA ORA2 ;z1 STA2 ( write z1,z2 )

-    #00 STHr  SFT2 #00 ;z1 LDA ORA2             ( compute z0,z1 )

-    ;z2 LDA2

-    RTN

+        STHk  SFT                      ;m32/z3 STA  ( write z3 )

+    #00 STHkr SFT2 #00 ;m32/z3 LDA ORA2 ;m32/z2 STA2 ( write z2,z3 )

+    #00 STHkr SFT2 #00 ;m32/z2 LDA ORA2 ;m32/z1 STA2 ( write z1,z2 )

+    #00 STHr  SFT2 #00 ;m32/z1 LDA ORA2             ( compute z0,z1 )

+    ;m32/z2 LDA2

+    JMP2r

 ( shift right by 8-15 bits )

 @rshift32-1 ( x** n^ -> x<<n )

     #08 SUB STH POP 

-        STHkr SFT                      ;z3 STA  ( write z3 )

-    #00 STHkr SFT2 #00 ;z3 LDA ORA2 ;z2 STA2 ( write z2,z3 )

-    #00 STHr  SFT2 #00 ;z2 LDA ORA2             ( compute z1,z2 )

-    #00 TOR ;z3 LDA

-    RTN

+        STHkr SFT                      ;m32/z3 STA  ( write z3 )

+    #00 STHkr SFT2 #00 ;m32/z3 LDA ORA2 ;m32/z2 STA2 ( write z2,z3 )

+    #00 STHr  SFT2 #00 ;m32/z2 LDA ORA2             ( compute z1,z2 )

+    #00 ROT ROT ;m32/z3 LDA

+    JMP2r

 ( shift right by 16-23 bits )

 @rshift32-2 ( x** n^ -> x<<n )

     #10 SUB STH POP2

-        STHkr SFT                      ;z3 STA ( write z3 )

-    #00 STHr  SFT2 #00 ;z3 LDA ORA2            ( compute z2,z3 )

+        STHkr SFT                      ;m32/z3 STA ( write z3 )

+    #00 STHr  SFT2 #00 ;m32/z3 LDA ORA2            ( compute z2,z3 )

     #0000 SWP2

-    RTN

+    JMP2r

 ( shift right by 16-23 bits )

 @rshift32-3 ( x** n^ -> x<<n )

     #18 SUB STH POP2 POP ( x0 )

     #00 SWP #0000 SWP2 ( 00 00 00 x0 )

     STHr SFT

-    RTN

+    JMP2r

 ( x << n )

 @lshift32 ( x** n^ -> x<<n )

@@ -225,70 +222,70 @@

     DUP #10 LTH ;lshift32-1 JCN2 ( x n )

     DUP #18 LTH ;lshift32-2 JCN2 ( x n )

     ;lshift32-3 JMP2 ( x n )

-    RTN

+    JMP2r

 ( shift left by 0-7 bits )

 @lshift32-0 ( x** n^ -> x<<n )

     #40 SFT STH ( stash n<<4 )

-    #00 SWP STHkr SFT2                     ;z2 STA2 ( store z2,z3 )

-    #00 SWP STHkr SFT2 #00 ;z2 LDA ORA2 ;z1 STA2 ( store z1,z2 )

-    #00 SWP STHkr SFT2 #00 ;z1 LDA ORA2 ;z0 STA2 ( store z0,z1 )

-            STHr  SFT      ;z0 LDA ORA              ( calculate z0 )

-    ;z1 LDA ;z2 LDA2

-    RTN

+    #00 SWP STHkr SFT2                     ;m32/z2 STA2 ( store z2,z3 )

+    #00 SWP STHkr SFT2 #00 ;m32/z2 LDA ORA2 ;m32/z1 STA2 ( store z1,z2 )

+    #00 SWP STHkr SFT2 #00 ;m32/z1 LDA ORA2 ;m32/z0 STA2 ( store z0,z1 )

+            STHr  SFT      ;m32/z0 LDA ORA              ( calculate z0 )

+    ;m32/z1 LDA ;m32/z2 LDA2

+    JMP2r

 ( shift left by 8-15 bits )

 @lshift32-1 ( x** n^ -> x<<n )

     #08 SUB #40 SFT STH ( stash [n-8]<<4 )

-    #00 SWP STHkr SFT2                     ;z1 STA2 ( store z1,z2 )

-    #00 SWP STHkr SFT2 #00 ;z1 LDA ORA2 ;z0 STA2 ( store z0,z1 )

-            STHr  SFT      ;z0 LDA ORA              ( calculate z0 )

+    #00 SWP STHkr SFT2                     ;m32/z1 STA2 ( store z1,z2 )

+    #00 SWP STHkr SFT2 #00 ;m32/z1 LDA ORA2 ;m32/z0 STA2 ( store z0,z1 )

+            STHr  SFT      ;m32/z0 LDA ORA              ( calculate z0 )

     SWP POP ( x0 unused )

-    ;z1 LDA2 #00

-    RTN

+    ;m32/z1 LDA2 #00

+    JMP2r

 ( shift left by 16-23 bits )

 @lshift32-2 ( x** n^ -> x<<n )

     #10 SUB #40 SFT STH ( stash [n-16]<<4 )

-    #00 SWP STHkr SFT2                ;z0 STA2 ( store z0,z1 )

-            STHr  SFT  ;z0 LDA ORA             ( calculate z0 )

+    #00 SWP STHkr SFT2                ;m32/z0 STA2 ( store z0,z1 )

+            STHr  SFT  ;m32/z0 LDA ORA             ( calculate z0 )

     STH POP2 STHr

-    ;z1 LDA #0000

-    RTN

+    ;m32/z1 LDA #0000

+    JMP2r

 ( shift left by 24-31 bits )

 @lshift32-3 ( x** n^ -> x<<n )

     #18 SUB #40 SFT ( x0 x1 x2 x3 r=[n-24]<<4 )

     SFT ( x0 x1 x2 x3<<r )

     SWP2 POP2 SWP POP #0000 #00

-    RTN

+    JMP2r

 ( arithmetic )

 ( x + y )

 @add32 ( xhi* xlo* yhi* ylo* -> zhi* zlo* )

-    ;y2 STA2 ;y0 STA2 ( save ylo, yhi )

-    ;x2 STA2 ;x0 STA2 ( save xlo, xhi )

-    #0000 #0000 ;z0 STA2 ;z2 STA2 ( reset zhi, zlo )

+    ;m32/y2 STA2 ;m32/y0 STA2 ( save ylo, yhi )

+    ;m32/x2 STA2 ;m32/x0 STA2 ( save xlo, xhi )

+    #0000 #0000 ;m32/z0 STA2 ;m32/z2 STA2 ( reset zhi, zlo )

     ( x3 + y3 => z2z3 )

-    #00 ;x3 LDA #00 ;y3 LDA ADD2 ;z2 STA2

+    #00 ;m32/x3 LDA #00 ;m32/y3 LDA ADD2 ;m32/z2 STA2

     ( x2 + y2 + z2 => z1z2 )

-    #00 ;x2 LDA ;z1 LDA2 ADD2 ;z1 STA2

-    #00 ;y2 LDA ;z1 LDA2 ADD2 ;z1 STA2

+    #00 ;m32/x2 LDA ;m32/z1 LDA2 ADD2 ;m32/z1 STA2

+    #00 ;m32/y2 LDA ;m32/z1 LDA2 ADD2 ;m32/z1 STA2

     ( x1 + y1 + z1 => z0z1 )

-    #00 ;x1 LDA ;z0 LDA2 ADD2 ;z0 STA2

-    #00 ;y1 LDA ;z0 LDA2 ADD2 ;z0 STA2

+    #00 ;m32/x1 LDA ;m32/z0 LDA2 ADD2 ;m32/z0 STA2

+    #00 ;m32/y1 LDA ;m32/z0 LDA2 ADD2 ;m32/z0 STA2

     ( x0 + y0 + z0 => z0 )

-    ;x0 LDA ;z0 LDA ADD ;z0 STA

-    ;y0 LDA ;z0 LDA ADD ;z0 STA

+    ;m32/x0 LDA ;m32/z0 LDA ADD ;m32/z0 STA

+    ;m32/y0 LDA ;m32/z0 LDA ADD ;m32/z0 STA

     ( load zhi,zlo )

-    ;z0 LDA2 ;z2 LDA2

-    RTN

+    ;m32/z0 LDA2 ;m32/z2 LDA2

+    JMP2r

 ( -x )

 @negate32 ( x** -> -x** )

@@ -298,36 +295,36 @@

     ,&done JCN ( xlo non-zero => don't inc hi )

     SWP2 INC2 SWP2 ( -xhi -xlo )

     &done

-    RTN

+    JMP2r

 ( x - y )

 @sub32 ( x** y** -> z** )

-    ;negate32 JSR2 ;add32 JSR2 RTN

+    ;negate32 JSR2 ;add32 JSR2 JMP2r

 ( 16-bit multiplication )

 @mul16 ( x* y* -> z** )

-    ;y1 STA ;y0 STA ( save ylo, yhi )

-    ;x1 STA ;x0 STA ( save xlo, xhi )

-    #0000 #00 ;z1 STA2 ;z3 STA ( reset z1,z2,z3 )

-    #0000 #00 ;w0 STA2 ;w2 STA ( reset w0,w1,w2 )

+    ;m32/y1 STA ;m32/y0 STA ( save ylo, yhi )

+    ;m32/x1 STA ;m32/x0 STA ( save xlo, xhi )

+    #0000 #00 ;m32/z1 STA2 ;m32/z3 STA ( reset z1,z2,z3 )

+    #0000 #00 ;m32/w0 STA2 ;m32/w2 STA ( reset w0,w1,w2 )

     ( x1 * y1 => z1z2 )

-    #00 ;x1 LDA #00 ;y1 LDA MUL2 ;z2 STA2

+    #00 ;m32/x1 LDA #00 ;m32/y1 LDA MUL2 ;m32/z2 STA2

     ( x0 * y1 => z0z1 )

-    #00 ;x0 LDA #00 ;y1 LDA MUL2 ;z1 LDA2 ADD2 ;z1 STA2

+    #00 ;m32/x0 LDA #00 ;m32/y1 LDA MUL2 ;m32/z1 LDA2 ADD2 ;m32/z1 STA2

     ( x1 * y0 => w1w2 )

-    #00 ;x1 LDA #00 ;y0 LDA MUL2 ;w1 STA2

+    #00 ;m32/x1 LDA #00 ;m32/y0 LDA MUL2 ;m32/w1 STA2

     ( x0 * y0 => w0w1 )

-    #00 ;x0 LDA #00 ;y0 LDA MUL2 ;w0 LDA2 ADD2 ;w0 STA2

+    #00 ;m32/x0 LDA #00 ;m32/y0 LDA MUL2 ;m32/w0 LDA2 ADD2 ;m32/w0 STA2

     ( add z and a<<8 )

-    #00 ;z1 LDA2 ;z3 LDA

-    ;w0 LDA2 ;w2 LDA #00

+    #00 ;m32/z1 LDA2 ;m32/z3 LDA

+    ;m32/w0 LDA2 ;m32/w2 LDA #00

     ;add32 JSR2

-    RTN

+    JMP2r

 ( x * y )

 @mul32 ( x** y** -> z** ) 

@@ -340,7 +337,7 @@

     ( [x0*y0]<<32 will completely overflow )

     ADD2 ,&z0 LDR2 ADD2 ( sum += x0*y1<<16 + x1*y0<<16 )

     ,&z1 LDR2

-    RTN

+    JMP2r

 [ &x0 $2 &x1 $2

   &y0 $2 &y1 $2

   &z0 $2 &z1 $2 ]

@@ -348,18 +345,18 @@

 @div32 ( x** y** -> q** )

     ;_divmod32 JSR2

     ;_divmod32/quo0 LDA2 ;_divmod32/quo1 LDA2

-    RTN

+    JMP2r

 @mod32 ( x** y** -> r** )

     ;_divmod32 JSR2

     ;_divmod32/rem0 LDA2 ;_divmod32/rem1 LDA2

-    RTN

+    JMP2r

 @divmod32 ( x** y** -> q** r** )

     ;_divmod32 JSR2

     ;_divmod32/quo0 LDA2 ;_divmod32/quo1 LDA2

     ;_divmod32/rem0 LDA2 ;_divmod32/rem1 LDA2

-    RTN

+    JMP2r

 ( calculate and store x / y and x % y )

 @_divmod32 ( x** y** -> )

@@ -372,7 +369,7 @@

     ,&div0 LDR2 ,&div1 LDR2

     ;lt32 JSR2 ,&is-zero JCN ,&not-zero JMP

     &is-zero

-    #0000 ,&quo0 STR2 #0000 ,&quo1 STR2 RTN

+    #0000 ,&quo0 STR2 #0000 ,&quo1 STR2 JMP2r

     ( x >= y so the answer is >= 1 )

     &not-zero

@@ -412,7 +409,7 @@

     ,&div0 LDR2 ,&div1 LDR2 #01 ;rshift32 JSR2 ,&div1 STR2 ,&div0 STR2 ( div >>= 1 )

     ,&cur0 LDR2 ,&cur1 LDR2 #01 ;rshift32 JSR2 ,&cur1 STR2 ,&cur0 STR2 ( cur >>= 1 )

     ,&cur0 LDR2 ,&cur1 LDR2 ;non-zero32 JSR2 ,&loop JCN ( if cur>0, loop. else we're done )

-    RTN

+    JMP2r

 ( greatest common divisor - euclidean algorithm )

 @gcd32 ( x** y** -> z** )

@@ -432,4 +429,4 @@

     ,&loop JMP

     &done

     POP4 ( x )

-    RTN

+    JMP2r