Re: [PATCH 2/2] x86/mce: Add mce=panic_on_ce_count to panic on a corrected error flood

[Breno Leitao]

On Tue, Aug 25, 2026 at 03:52:02PM -0700, Borislav Petkov wrote:
> On Tue, Aug 25, 2026 at 08:01:23PM +0000, Luck, Tony wrote:
> > L3 on Intel is shared by the whole socket. So you'd lose 50% of cores for an L3 cache
> > issue on a typical two socket system
>
> Would panicking the whole system be better?

That's basically the approach I was going for too (just more generic).

When things start to go south, panic clearly. That helps people like
me who watch for kernel health across the fleet and know how to handle
a clean panic caused by ETOOMANY MCI and MCEs, instead of something that
requires deeper investigation because the host is degraded.

In other words, I'm more interested in flagging that a host is a lemon
than in keeping it limping along half-alive, thus my approach in this
patchset.