E162 - Implicit integer to float conversion may lose precision
An integer is converted implicitly to a floating point type that cannot represent every value of the
integer type. REAL represents integers up to 24 bits exactly, and LREAL up to 53 bits. Larger values
are rounded.
Code example:
PROGRAM mainProg
VAR
count : DINT := 16777217;
value : REAL;
END_VAR
value := count; // warning: 16777217 becomes 16777216.0
END_PROGRAM
To silence the warning, convert explicitly:
value := DINT_TO_REAL(count);
An integer passed to a generic function declared for ANY_REAL is converted the same way and warns as well. For the standard functions that are meant to take integers, such as SQRT and LN, this is a known false positive.