undefined

points

[-]

I've always believed that they kept 32 bits ints on 64 bits CPU as a default because going full 64 bits would make the code and data structures bigger for "no reason" (it's not often that one hits the 4.10^9 limit in system code (if you don't count timestamps, that is)). For instance, a load-register-with-immediate instruction would normally take 5 bytes (opcode+value) on 32 bits, but 9 bytes on 64.

by Joker_vD3 hours ago|

parent|

[-]

> would make... data structures bigger for "no reason"

This was the main reason that in stayed 32 bits. Not because larger footprint of structs would be catastrophic — it wouldn't: the 64-bit migration was motivated mainly by the fact that the 32-bit architectures couldn't easily address all that actually existing physical memory past 4 GiB (and virtual memory as well) — but because lots and I mean lots of on-disk data structures were defined in terms of char/short/int, people routinely dumped/gulped their data structures as-is, with no marshalling, so changing the size of int would break literally everything that worked with on-disk data, starting with the filesystem implementations themselves.

by sparkie10 hours ago|

parent|

prev|

[-]

I think backward compatibility was the main aim. Intel tried redesigning the architecture as 64-bit native (Itanium), but AMD done a better job at backward compatibility - and intel eventually adopted it as x86-64.

The amd64 design could run most 32-bit code with minimal changes. All the 32-bit instructions had the same encoding, besides push/pop which instead acted on 64-bits. The 64-bit instructions were basically opt-in, though a few opcodes (0x40..0x4F) had to be deprecated for the REX prefix.