RE: [Xen-ia64-devel] SMP designs and discuss

["Magenheimer, Dan (HP Labs Fort Collins)" <dan.magenheimer@xxxxxx>]

Sorry, I am catching up after being gone most of the week.

> > > * Idle regs.
> > >   Currently, idle domains have no regs (the regs field is NULL).
> > >   [I am not sure it is true for idle0].
> > >   Is it a problem ?
> > >   I had to modify the heartbeat so that it doesn't reference regs.
> >
> > Personally I don't think idle should exist, but it definitely
> > shouldn't require state to be saved and restored.
> Maybe true for idle0 but what do other processors execute 
> when there is no 
> domain ?

I was proposing that idle is an anachronistic concept from before
processors could go to sleep.

This was argued at some length on xen-devel a few months ago but
for now we need to do whatever the core Xen scheduling code
requires to get SMP working and we can consider making context
switching more efficient later.

> > >   How many VHPT per system ?
> > >    1) Only one
> > >    2) One per LP  (current design)
> > >    3) One per VCPU (original Xen-VTI).
> > >   I think (1) is not scalable...
> >
> > I thought the current design is (1).  Perhaps I didn't
> > look closely enough at your SMP patch!
> Well, the VHPT is allocated in mm_init, which is called once per cpu.

I see you are correct.  This is probably a mistake.

Anybody know if the folks experimenting with LVHPT on Linux/ia64
have studied this?

Dan

_______________________________________________
Xen-ia64-devel mailing list
Xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-ia64-devel