Two's complement represents signed integers by giving the high-order bit negative weight while the remaining bits retain positive weights. A zero sign bit denotes a nonnegative value; a one sign bit places the representation in the negative half of the corresponding unsigned range. C23 permits only this signed representation, replacing the historical alternatives of sign-and-magnitude and one's complement. section 5.7.5
Table 5.7 Negation for 16-bit unsigned integer types
| Op | Value | b15 … b0 |
A |
240 |
0b00000000’11110000 |
~A |
65,295 |
0b11111111’00001111 |
+1 |
1 |
0b00000000’00000001 |
-A |
65,296 |
0b11111111’00010000 |
Constructs the two's-complement negative of 240 by complementing every bit and adding one: ~A + 1. section 5.7.5
The representation uses the same bit operations as unsigned arithmetic; interpretation changes when the high-order bit is one. The supplied comparison functions classify that upper unsigned half as negative, then reverse the unsigned ordering only when the operands have different signs. section 5.7.5
TAKEAWAY
- Positive signed values have the same representation as corresponding unsigned values. section 5.7.5
- For an unsigned value A, compute its two's-complement negative as ~A + 1. section 5.7.5
