RIP size on x86-64

[gz]

rip size is currently 4 bytes, but maybe it should be 8?

https://github.com/bminor/binutils-gdb/blob/master/gdb/features/i386/64bit-core.xml lists it as:

  <reg name="rip" bitsize="64" type="code_ptr"/>

[bet4it]

I will fix it in a following PR ( with another way ) , wait for me😅

[gz]

random thought: it might be nice to statically type-encode the size of the registers when when they're passed to the read/write_registers. One way could be to move the val: &[u8] and dst: &mut [u8] in some enum with the RegId and encode the length explicitly (specify as &[u8; 8] etc.)?

I'm thinking that might save some slice length checks when implementing these functions. But not sure if even possible just a thought :)

[daniel5151]

@gz I assume you meant read/write_register (as opposed to read/write_registers)?

If so, then you might find this discussion from the initial implementation PR interesting: #22 (comment)

Though now that I'm re-reading this idea a year later, I realize that there might be some easier approaches to pull off something similar... I certainly wouldn't be opposed to having stricter type-safety in these APIs, so I'm open to any ideas!

[daniel5151]

And as for the original issue: yes, rip having a size of 4 seems like an oversight in the initial #34 implementation.

Pinging @keiichiw, since this might translate to a subtle correctness issue in your implementation.

[gz]

Yes I meant the single register writes with the dynamic val and dst slices.

[daniel5151]

@gz if this is something you'd like fixed sooner, feel free to send a quick fixup PR. Otherwise, a fix for this will land alongside some other changes to read_register in #85

[gz]

Sent pull request, yes would be nice to have this fixed, thanks!