On Thu, Feb 17, 2011 at 07:04:51PM +0100, Jan Kiszka wrote: > >>> Should not manipulate mp_state of a running vcpu (should only do that > >>> for migration when vcpu is stopped), since its managed by the kernel, > >>> for irqchip case. > >> > >> Not for asynchronously injected MCEs. The target CPU would simply > >> oversleep them. MCEs are not in the scope of the in-kernel irqchip. > > > > Pending MCE exception could break out of in-kernel halt emulation. > > Can't follow. What do you mean? That the kernel already takes care? I > didn't find a trace, so I added that code.
Nevermind. This is rare and "halted -> running" transition in userspace is harmless.
