Re: [PATCH v2 1/3] KVM: x86: implement KVM_{GET|SET}_TSC_STATE
From: Peter Zijlstra
Date: Tue Dec 08 2020 - 04:49:34 EST
On Mon, Dec 07, 2020 at 06:41:41PM +0100, Thomas Gleixner wrote:
> Right this happens still occasionally, but for quite some time this is
> 100% firmware sillyness and not a fundamental property of the hardware
> anymore.
Ever since Nehalem (2008) TSC is synchronized on <= 2 sockets, and any
observed deviation is firmware fail. I don't remember exactly where 4
socket and up got reliable.
(there's the physical hotplug case, but let's not make this complicated)
AMD has had Constant TSC since Barcelona which is roughly the same
timeframe IIRC.
So basically every TSC fail in the last decase is due to firmware being
shit.