diff --git a/dda_maths.c b/dda_maths.c
index 1c6d8db..e5ea5b2 100644
--- a/dda_maths.c
+++ b/dda_maths.c
@@ -25,13 +25,19 @@
   Found on  http://stackoverflow.com/questions/4144232/
   how-to-calculate-a-times-b-divided-by-c-only-using-32-bit-integer-types-even-i
 */
-const uint32_t muldiv(uint32_t multiplicand, uint32_t multiplier,
-                      uint32_t divisor) {
+const int32_t muldiv(int32_t multiplicand, uint32_t multiplier,
+                     uint32_t divisor) {
   uint32_t quotient = 0;
   uint32_t remainder = 0;
+  uint8_t negative_flag = 0;
   uint32_t qn = multiplier / divisor;
   uint32_t rn = multiplier % divisor;
 
+  if (multiplicand < 0) {
+    negative_flag = 1;
+    multiplicand = -multiplicand;
+  }
+
   while(multiplicand) {
     if (multiplicand & 1) {
       quotient += qn;
@@ -55,6 +61,6 @@ const uint32_t muldiv(uint32_t multiplicand, uint32_t multiplier,
     quotient++;
 
   // remainder is valid here, but not returned
-  return quotient;
+  return negative_flag ? -((int32_t)quotient) : (int32_t)quotient;
 }
 
diff --git a/dda_maths.h b/dda_maths.h
index 9efcf7a..a47d6e1 100644
--- a/dda_maths.h
+++ b/dda_maths.h
@@ -5,7 +5,7 @@
 
 
 // return rounded result of multiplicand * multiplier / divisor
-const uint32_t muldiv(uint32_t multiplicand, uint32_t multiplier,
-                      uint32_t divisor);
+const int32_t muldiv(int32_t multiplicand, uint32_t multiplier,
+                     uint32_t divisor);
 
 #endif	/* _DDA_MATHS_H */