Re: [V4][PATCH 4/6] x86, nmi: add in logic to handle multiple events and unknown NMIs

[Robert Richter <robert.richter@amd.com>]

On 14.09.11 08:56:40, Don Zickus wrote:
> On Tue, Sep 13, 2011 at 04:58:27PM -0400, Don Zickus wrote:
> > 
> > V3:
> >   - redesigned the algorithm to utilize Avi's idea of detecting a back-to-back
> >     NMI with %rip.
> 
> Hi Robert,
> 
> I realized I added an optimization for executing the nmi handlers to help
> minimize the impact on the virt folks and realize it might break your IBS
> stuff.
> 
> > -static int notrace __kprobes nmi_handle(unsigned int type, struct pt_regs *regs)
> > +static int notrace __kprobes nmi_handle(unsigned int type, struct pt_regs *regs, bool b2b)
> >  {
> >  	struct nmi_desc *desc = nmi_to_desc(type);
> >  	struct nmiaction *next_a, *a, **ap = &desc->head;
> > @@ -87,6 +87,16 @@ static int notrace __kprobes nmi_handle(unsigned int type, struct pt_regs *regs)
> >  
> >  		handled += a->handler(type, regs);
> >  
> > +		/*
> > + 		 * Optimization: only loop once if this is not a 
> > + 		 * back-to-back NMI.  The idea is nothing is dropped
> > + 		 * on the first NMI, only on the second of a back-to-back
> > + 		 * NMI.  No need to waste cycles going through all the
> > + 		 * handlers.
> > + 		 */
> > +		if (!b2b && handled)
> > +			break;
> > +
> >  		a = next_a;
> >  	}
> >  	rcu_read_unlock();
> 
> The optimization is to run through the handlers until one of them claims
> the NMI but only for the first NMI.  Whereas on the second half of a
> back-to-back NMI, run through all the handlers regardless of how many
> claim they handled it.
> 
> Does your IBS stuff need to always run through two handlers?

As said in my previous answer, it might probably work, but I will test
it.

-Robert

-- 
Advanced Micro Devices, Inc.
Operating System Research Center