diff --git a/libsrc/common/cc65_imul8x8r16.s b/libsrc/common/cc65_imul8x8r16.s
new file mode 100644
index 000000000..0e7d5479b
--- /dev/null
+++ b/libsrc/common/cc65_imul8x8r16.s
@@ -0,0 +1,22 @@
+;
+; 2014-03-27, Oliver Schmidt
+; 2014-05-08, Greg King
+;
+; CC65 library: 8x8 => 16 signed multiplication
+;
+
+        .export         _cc65_imul8x8r16
+        .import         imul8x8r16, popa, ptr1:zp
+
+
+;---------------------------------------------------------------------------
+; 8x8 => 16 signed multiplication routine.
+
+
+.proc   _cc65_imul8x8r16
+
+        sta     ptr1
+        jsr     popa
+        jmp     imul8x8r16
+
+.endproc
diff --git a/libsrc/common/cc65_umul8x8r16.s b/libsrc/common/cc65_umul8x8r16.s
index 5cbb4f8b0..cf3b26bb1 100644
--- a/libsrc/common/cc65_umul8x8r16.s
+++ b/libsrc/common/cc65_umul8x8r16.s
@@ -5,9 +5,7 @@
 ;
 
         .export         _cc65_umul8x8r16
-        .import         umul8x8r16, popa
-
-        .include        "zeropage.inc"
+        .import         umul8x8r16, popa, ptr1:zp
 
 
 ;---------------------------------------------------------------------------
diff --git a/libsrc/runtime/imul16x16r32.s b/libsrc/runtime/imul16x16r32.s
index f41d13602..6fad9abf0 100644
--- a/libsrc/runtime/imul16x16r32.s
+++ b/libsrc/runtime/imul16x16r32.s
@@ -13,9 +13,9 @@
 ;---------------------------------------------------------------------------
 ; 16x16 => 32 signed multiplication routine.
 ;
-;   lhs         rhs           result          result also in
+;   LHS         RHS           result          result also in
 ; -------------------------------------------------------------
-;   ptr1        ax            ax:sreg
+;   ax          ptr1          ax:sreg
 ;
 ; There is probably a faster way to do this.
 ;
diff --git a/libsrc/runtime/imul8x8r16.s b/libsrc/runtime/imul8x8r16.s
new file mode 100644
index 000000000..017297e4b
--- /dev/null
+++ b/libsrc/runtime/imul8x8r16.s
@@ -0,0 +1,75 @@
+;
+; 2010-11-02, Ullrich von Bassewitz
+; 2014-05-10, Greg King
+;
+; CC65 runtime: 8x8 => 16 signed multiplication
+;
+
+        .export         imul8x8r16, imul8x8r16m
+        .importzp       ptr1, ptr3, tmp1
+
+        .macpack        generic
+
+;---------------------------------------------------------------------------
+; 8x8 => 16 signed multiplication routine.
+;
+;   multiplicand  multiplier   product
+;   LHS             RHS        result
+; -------------------------------------------------------------
+;   .A (ptr3-low)   ptr1-low    .XA
+;
+
+imul8x8r16:
+        sta     ptr3
+
+imul8x8r16m:
+        ldx     #>0
+        bit     ptr3
+        bpl     @L7
+        dex
+@L7:    stx     ptr3+1          ; Extend sign of Left-Hand Side
+        ldy     #<0             ; Clear .XY accumulator
+        ldx     #>0
+        lda     ptr1
+        bmi     NegMult
+        bpl     @L2             ; Branch always
+
+@L0:    tya                     ; Add current multiplicand
+        add     ptr3
+        tay
+        txa
+        adc     ptr3+1
+        tax
+
+@L1:    asl     ptr3
+        rol     ptr3+1
+@L2:    lsr     ptr1            ; Get next bit of Right-Hand Side into carry
+        bcs     @L0
+        bnz     @L1             ; Loop if more one-bits in multiplier
+
+        tya                     ; Put result into cc65's accumulator
+        rts
+
+; The multiplier is negative.
+; Therefore, make it positive; and, subtract when multiplying.
+NegMult:
+        eor     #%11111111
+        sta     ptr1
+        inc     ptr1
+        bnz     @L2             ; Branch always
+
+@L0:    tya                     ; Subtract current multiplicand
+        sub     ptr3
+        tay
+        txa
+        sbc     ptr3+1
+        tax
+
+@L1:    asl     ptr3
+        rol     ptr3+1
+@L2:    lsr     ptr1            ; Get next bit of Right-Hand Side into carry
+        bcs     @L0
+        bnz     @L1             ; Loop if more one-bits in multiplier
+
+        tya                     ; Put result into cc65's accumulator
+        rts
diff --git a/libsrc/runtime/umul16x16r32.s b/libsrc/runtime/umul16x16r32.s
index c02cf8bb2..9ecd1596e 100644
--- a/libsrc/runtime/umul16x16r32.s
+++ b/libsrc/runtime/umul16x16r32.s
@@ -13,14 +13,14 @@
 ;---------------------------------------------------------------------------
 ; 16x16 => 32 unsigned multiplication routine. Because the overhead for a
 ; 16x16 => 16 unsigned multiplication routine is small, we will tag it with 
-; the matching labels as well.
+; the matching labels, as well.
 ;
-;  routine         lhs         rhs        result          result also in
+;  routine         LHS         RHS        result          result also in
 ; -----------------------------------------------------------------------
-;  umul16x16r32    ptr1        ax         ax:sreg          ptr1:sreg
-;  umul16x16r32m   ptr1        ptr3       ax:sreg          ptr1:sreg
-;  umul16x16r16    ptr1        ax         ax               ptr1
-;  umul16x16r16m   ptr1        ptr3       ax               ptr1
+;  umul16x16r32    ax          ptr1       ax:sreg          ptr1:sreg
+;  umul16x16r32m   ptr3        ptr1       ax:sreg          ptr1:sreg
+;  umul16x16r16    ax          ptr1       ax               ptr1
+;  umul16x16r16m   ptr3        ptr1       ax               ptr1
 ;
 ; ptr3 is left intact by the routine.
 ;
diff --git a/libsrc/runtime/umul8x16r24.s b/libsrc/runtime/umul8x16r24.s
index 64653f5f4..ff7d0bae6 100644
--- a/libsrc/runtime/umul8x16r24.s
+++ b/libsrc/runtime/umul8x16r24.s
@@ -13,14 +13,14 @@
 ;---------------------------------------------------------------------------
 ; 8x16 => 24 unsigned multiplication routine. Because the overhead for a
 ; 8x16 => 16 unsigned multiplication routine is small, we will tag it with
-; the matching labels as well.
+; the matching labels, as well.
 ;
-;  routine         lhs         rhs        result          result also in
+;  routine         LHS         RHS        result          result also in
 ; -----------------------------------------------------------------------
-;  umul8x16r24     ptr1-lo     ax         ax:sreg-lo      ptr1:sreg-lo
-;  umul8x16r24m    ptr1-lo     ptr3       ax:sreg-lo      ptr1:sreg-lo
+;  umul8x16r24     ax          ptr1-low   ax:sreg-low     ptr1:sreg-low
+;  umul8x16r24m    ptr3        ptr1-low   ax:sreg-low     ptr1:sreg-low
 ;
-; ptr3 ist left intact by the routine.
+; ptr3 is left intact by the routine.
 ;
 
 umul8x16r24:
diff --git a/libsrc/runtime/umul8x8r16.s b/libsrc/runtime/umul8x8r16.s
index 34ff82872..aba758e1a 100644
--- a/libsrc/runtime/umul8x8r16.s
+++ b/libsrc/runtime/umul8x8r16.s
@@ -1,7 +1,7 @@
 ;
 ; Ullrich von Bassewitz, 2010-11-02
 ;
-; CC65 runtime: 8x8 => 16 multiplication
+; CC65 runtime: 8x8 => 16 unsigned multiplication
 ;
 
         .export         umul8x8r16, umul8x8r16m
@@ -9,11 +9,11 @@
 
 
 ;---------------------------------------------------------------------------
-; 8x8 => 16 multiplication routine.
+; 8x8 => 16 unsigned multiplication routine.
 ;
-;   lhs         rhs           result          result also in
+;   LHS            RHS          result      result in also
 ; -------------------------------------------------------------
-;   ptr1-lo     ptr3-lo         ax              ptr1
+;   .A (ptr3-low)  ptr1-low     .XA             ptr1
 ;
 
 umul8x8r16:
@@ -21,7 +21,7 @@ umul8x8r16:
 umul8x8r16m:
         lda     #0              ; Clear byte 1
         ldy     #8              ; Number of bits
-        lsr     ptr1            ; Get first bit of lhs into carry
+        lsr     ptr1            ; Get first bit of RHS into carry
 @L0:    bcc     @L1
         clc
         adc     ptr3
@@ -30,8 +30,6 @@ umul8x8r16m:
         dey
         bne     @L0
         tax
-        stx     ptr1+1          ; Result in a/x and ptr1
+        stx     ptr1+1          ; Result in .XA and ptr1
         lda     ptr1            ; Load the result
         rts                     ; Done
-
-