Re: [RFC v3 0/8] x86, xsave: rework of extended state handling, LWP support

[Ingo Molnar <mingo@elte.hu>]

* Hans Rosenfeld <hans.rosenfeld@amd.com> wrote:

> > Here are a couple of suggestions to LWP hardware designers:
> > 
> >  - the fact that LWP cannot count kernel events right now is unfortunate - 
> >    there's no reason not to allow privileged user-space to request ring 3
> >    events as well - hopefully this misfeature will be fixed in future 
> >    iterations of the hardware.
> > 
> >  - it would be nice to allow the per task masking/unmasking of LWP without
> >    having to modify the cr0 (which can be expensive). A third mode
> >    implemented in the LWP_CFG MSG would suffice: it would make the LWP
> >    instructions privileged, but would otherwise allow LWP event collection
> >    to occur even on sandboxed code.
> > 
> >  - it would be nice to also log the previous retired instruction in the
> >    trace entry, to ease decoding of the real instruction that generated
> >    an event. (Fused instructions can generate their RIP at the first
> >    instruction.)
> 
> I will forward this to our hardware designers, but I have my doubts about the 
> first two of your suggestions. They seem to be orthogonal to what LWP is 
> supposed to be.

Not sure why you think those two suggestions are 'orthogonal to LWP', they are 
not:

 - the second suggestion adds a third security model to the current 
   all-or-nothing nature of LWP instructions.

 - the first suggestion is a variation of its current security model as well:
   it allows LWP driven event collection in kernel mode, not just user mode.

There is nothing fundamentally ring-3-only about the concept of 'light weight 
profiling' - while ring-3-only event collection is understandably necessary for 
unprivileged user-space, it is not the only interesting mode of lightweight 
event collection.

Thanks,

	Ingo