From the standard (4.7) it looks like the conversion from int to unsigned int, when they both use the same number of bits, is purely conceptual:
If the de
"Destination type" refers to the type you're assigning/casting to.
The whole paragraph means a 32 bit unsigned int
converted to a 32 bit signed int
will stay as-is, given the value fits into the signed int
. If they don't fit, it depends on the implementation on what it does (e.g. truncate). That means it really depends on the implementation whether the bits stay or whether they're changed (there's no guarantee).
Or in other words: If the unsigned int
uses its most significant bit, the result is no longer predictable. Otherwise there's no change (other than changing the "name on the box").