Bit-fields

“Field” for short, used when storage space is at a premium, kinda like IDs.

struct {
    unsigned int a: 1;
    unsigned int b: 1;
    unsigned int c: 1;
} fields;

The : 1 means a one bit field. Almost everything about fields are implementation-dependent. Fields are not so great (not very portable, specify signed for unsigned for portability) some are left to right, some are right to left (defined from). Fields must only be ints.

Features of bit-fields:

• Unnamed, with a number behind: Basically just making a padding of N bits, reserving that space for nothing, just for purposes of alignment.
• Unnamed, width of 0: specifying a boundary between different bit-fields.

The reason it can be used for alignment is because operating systems like reading memory in chunks, instead of a small unit like 1 byte. Say your OS reads in 4 byte chunks at a time, then even though a char would only fill in one of those 4 bytes, the OS will pad that 4 byte chunk to 4 bytes.

#C #C/features