Re: [RFC] #MC mess

From: Borislav Petkov
Date: Tue Feb 18 2020 - 15:53:07 EST

On Tue, Feb 18, 2020 at 03:08:50PM -0500, Steven Rostedt wrote:
> You could have the msr_tracepoint_active() check this per cpu variable?
> msr reading and writing is rather slow, and I'm sure reading a per_cpu
> variable is going to be in the noise of it.

Yeah, I was worrying about using the tracing MSR variants in NMI context
but Peter says tracing should do in_nmi() if it isn't doing so.

Same with #MC: we hold a subsequent #MC from getting raised with MCIP -
thanks Tony - but we can have other exceptions raised while in the #MC
handler. That too should be taken care of with the in_nmi() thing.