[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] current not very current (vs curr_vcpu)



On Fri, 19 Feb 2010 21:34:39 +0000
Keir Fraser <keir.fraser@xxxxxxxxxxxxx> wrote:

> On 19/02/2010 19:23, "Mukesh Rathor" <mukesh.rathor@xxxxxxxxxx> wrote:
> 
> > Yes, but my point is it doesn't appear to be running idle vcpu as
> > indicated by regs->rsp and regs->rip. They both point to dom0
> > context. This from printk in ns16550_interrupt().
> > 
> > To rephrase the question, if regs->rip and regs->rsp show guest
> > context in do_IRQ(), then current must always point to guest vcpu,
> > correct?
> 
> Oh, so the serial interrupt interrupted a guest's execution? Then yes,
> current should point at that guest.
> 
>  -- Keir
> 

ah, I see what's going on. context_switch() is scheduling idle vcpu, and
calls continue_idle_domain() to reset_stack_and_jump(idle_loop).
well, reset_stack_and_jump() is setting rsp to guest_cpu_user_regs(),
and interrupt is coming right at that instant. so: 

diff = (char *)guest_cpu_user_regs() - (char *)(r)  is 0

and as a result, guest_mode(regs) == true.

It appears to me, that guest_mode() needs to check if current
rsp is already guest_cpu_user_regs() in which case a stack switch
has not happened, and guest_mode should be false. 
Or, perhaps, just not enable interrupts in context_switch and do 
it in idle_loop().

what do you think?

thanks,
Mukesh


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


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.