Add UMULH()

This function/macro returns the high half of the unsigned 64-bit
product of two unsigned 32-bit operands.

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@21463 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

libavcodec/mathops.h

@@ -53,6 +53,12 @@ static av_always_inline int MULH(int a, int b){
 }
 #endif
 
+#ifndef UMULH
+static av_always_inline unsigned UMULH(unsigned a, unsigned b){
+    return ((uint64_t)(a) * (uint64_t)(b))>>32;
+}
+#endif
+
 #ifndef MUL64
 #   define MUL64(a,b) ((int64_t)(a) * (int64_t)(b))
 #endif