Re: [v2 1/3] x86/microcode/intel: Check microcode revision before updating sibling threads

[Borislav Petkov]

On Thu, Feb 22, 2018 at 05:19:18AM -0800, Raj, Ashok wrote:
> This isn't a simple WRMSR like others. Microcode engine needs to do 
> a bunch of validation.

So this is slowly starting to resemble a real reason why not to. That
should be part of the code comment.

> We now have new guidance that the sibling must be spinning and not
> doing other things that can introduce instability around loading
> microcode.

Why isn't *this* in the code comment?

> I think its safer...

That's not good enough - it should be:

"It is not safe because... <insert reasons here>".

So that you can justify the overhead.

-- 
Regards/Gruss,
    Boris.

SUSE Linux GmbH, GF: Felix ImendÃrffer, Jane Smithard, Graham Norton, HRB 21284 (AG NÃrnberg)
--