Re: [V6][PATCH 4/6] x86, nmi: add in logic to handle multiple eventsand unknown NMIs

[Avi Kivity]

On 10/02/2011 03:48 PM, Andi Kleen wrote:
> >  Second, I wanted to use NMIs as a way of waking up a vcpu sleeping with
> >  interrupts disabled (in the context of Jeremy's paravirt spinlock
> >  patches).  Looks like we'll have to use paravirtualization for that.
>
> Just use a mode of perf that doesn't use NMIs.
> Interrupt off profiling is rarely needed if at all on modern kernels.

I tend to agree.   But that's not my call.

(note 'perf kvm' will need adjustment if we make that change - likely 
need to capture the interrupt during vmexit and dispatch it manually; 
not sure if it's workable for svm at all)

--
error compiling committee.c: too many arguments to function

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/