Re: [PATCH v2] x86: Fix MCE error handing when kdump is enabled

[Corey Minyard <cminyard@mvista.com>]

On Thu, Oct 01, 2020 at 04:29:49PM +0000, Luck, Tony wrote:
> >> I was thinking about this some yesterday.  It seems to me that enabling
> >> IRQS in an MCE handler is just a bad idea, but it's really a bad idea
> >> for kdump.
> >
> > I don't think this code ever thought about kdump.
> 
> How useful is kdump after a machine check induced crash anyway?
> 
> kdump is useful for debugging software problems.  There are very
> few ways that a software bug can result in a machine check. There
> are many ways that a hardware problem can trigger a machine check
> and crash.
> 
> So it would seem (statistically) that the analysis of almost every kdump
> after a machine check just says "h/w issue".

I don't really know.  It seems like having an idea of what the software
was doing when the hardware died might be useful for the hardware
engineers.  I really don't know much about what triggers MCEs, though,
besides memory errors the hardware couldn't correct.

You could say that the regs don't matter, I suppose, and that's
probabaly fine.  But if it's easy enough to do, and the interfaces are
already there and work, and it speeds up the crash process a bit, why
not do it?

-corey