Round the lower double-precision (64-bit) floating-point element in b up to an integer value, store the result as a double-precision floating-point element in the lower element of result, and copy the upper element from a to the upper element of dst.