[MODERATED] Re: [PATCH 3/9] TAA 3

On Thu, Oct 24, 2019 at 09:43:56PM +0100, speck for Andrew Cooper wrote:
> The first TSX problem in Haswell/Broadwell lead to chaos when a
> microcode update took feature bits out.

Yeap, not worky with our current non-reentrant feature detection OSes.
:-P

> When the second problem came around, I understand from IRL conversations
> that several vendors firmly expressed their opinion about whether
> feature bits should disappear in microcode.
> 
> As HLE is a hint bit at best, it was left visible despite being disabled
> behind the scenes.

Yeah, and it being a hint bit only, I don't know whether we should care
about a box where RTM=0 but HLE=1.

I guess the relevant use case is big fat hypervisor. But perhaps the
easiest thing would be if the migration code gets told to ignore HLE.

-- 
Regards/Gruss,
    Boris.

SUSE Software Solutions Germany GmbH, GF: Felix Imendörffer, HRB 36809, AG Nürnberg
--