Re: [Qemu-devel] [PATCH] qmp: extend QMP to provide read/write access to physical memory

[Bryan D. Payne]

By "evidence", I mean actual numbers for actual QEMU code.  Nothing
sophisticated, just use your new interface in a way you consider
relevant for your own use case, then approximate this use with existing
interfaces.  The approximation can be very rough.  For instance, showing
that doing the whole job with your approach is a much faster than doing
a necessary part of the job with existing commands would do.

Sure, I can put together some numbers to help with this discussion.

 

The QMP overhead could be too high.  QMP is control plane, not data
plane.  How much data do you envisage flowing here?

In theory, even the character device overhead could be too high.
Measuring it shouldn't be too hard.

This discussion has me thinking about whether QMP would be viable.  I think I'll take a little time to explore that approach in a little more depth before proceeding here.  I can report back with what I find.

If you control a QMP monitor, you own the guest.

Good point.  So I'll explore the performance aspects and let that drive the decision.

 

I feel code comment is find for internal interfaces, but for external
interfaces, a separate interface document is more appropriate.

Sounds good.

Cheers,

-bryan