Re: [PATCH v4 03/36] x86/bugs: Restructure mmio mitigation

[Pawan Gupta]

On Mon, Mar 24, 2025 at 10:29:15AM +0100, Borislav Petkov wrote:
> On Thu, Mar 13, 2025 at 12:26:06PM -0700, Pawan Gupta wrote:
> > Hmm, that would not be straightforward, specially for sysfs status.
> 
> See below:
> 
> - the unknown thing is done only for this vuln and not for the others
> 
> - it doesn't do anything besides reporting things differently - it doesn't
>   apply any mitigations - it is simply causing unnecessary complications which
>   don't bring anything besides maintenance overhead. Unless I'm missing an
>   angle...

"Unknown" status reporting was requested by Andrew Cooper. I am not able to
find that conversation though. IIRC, the reason was out-of-service CPUs
were not tested for the presence of vulnerability. Adding Andrew to Cc.

> - all the currently unaffected CPUs can also be in "unknown" status so why is
>   this special?

Makes sense.

> IOW, just whack the thing.

I will let Andrew comment on this.