The compiler can generate smulbb perfectly well nowadays.

src/core/SkMathPriv.h

Index: src/core/SkMathPriv.h
diff --git a/src/core/SkMathPriv.h b/src/core/SkMathPriv.h
index 345815354cc68a887bc1ab76c5c543c90628a1b5..b9184a0726934fdb801280014a7c7f30773b6d8d 100644
--- a/src/core/SkMathPriv.h
+++ b/src/core/SkMathPriv.h
@@ -57,7 +57,7 @@ static inline unsigned SkClampUMax(unsigned value, unsigned max) {
 static inline U8CPU SkMulDiv255Trunc(U8CPU a, U8CPU b) {
     SkASSERT((uint8_t)a == a);
     SkASSERT((uint8_t)b == b);
-    unsigned prod = SkMulS16(a, b) + 1;
+    unsigned prod = a*b + 1;
     return (prod + (prod >> 8)) >> 8;
 }
 
@@ -67,7 +67,7 @@ static inline U8CPU SkMulDiv255Trunc(U8CPU a, U8CPU b) {
 static inline U8CPU SkMulDiv255Ceiling(U8CPU a, U8CPU b) {
     SkASSERT((uint8_t)a == a);
     SkASSERT((uint8_t)b == b);
-    unsigned prod = SkMulS16(a, b) + 255;
+    unsigned prod = a*b + 255;
     return (prod + (prod >> 8)) >> 8;
 }