Re: [PATCH 4/4] x86, mce: Have MCE persistent event off by defaultfor now

[Ingo Molnar]

* Borislav Petkov <bp@xxxxxxxxx> wrote:

> > Probably make some sense to delay such a message (so that in the boot case 
> > we give the daemon a chance to get started before complaining that it 
> > hasn't shown up for work).
> 
> Yep, that and also I need to address the case for catching earlybird MCEs, 
> when perf hasn't been initialized yet. I'm thinking we could reuse the mcelog 
> buffer and feed those into the RAS daemon after init. Something like that.

I'd suggest that to solve that problem we initialize events earlier and allow 
something like a small static buffer backed event which does not need any 
dynamic memory so it can be enabled early on and can receive events, at the 
same time the MCE hw bits are enabled.

That event can get a larger buffer once the system is more fully up - without 
losing whatever events it is already holding.

Thanks,

	Ingo
--
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/