Why we don't use ip as destination in assembly language?

Answer 1

We don't use ip as the destination register in assembly for the simple reason it is not directly accessible for writing. Since we cannot write to it directly, we cannot use it as a destination.

However, it is possible to indirectly change the ip register via a jmp so there is no actual need to be able to write directly.

Note that in 16-bit and 32-bit mode we cannot even read the ip directly, we must use an indirect copy, such as:

call next_line

next_line:

pop eax

Having copied the ip, we can now modify it before (indirectly) writing to the ip via a jmp:

jmp eax

Note that in 64-bit mode we can reference instructions relative to ip so there is much less need to copy it, but we still cannot write to it directly, we must use jmp instead.