Re: [PATCH v6 11/10] x86/retpoline: Avoid return buffer underflows on context switch II

[Woodhouse, David]

On Mon, 2018-01-08 at 17:21 -0800, Andi Kleen wrote:
> On Mon, Jan 08, 2018 at 05:16:02PM -0800, Andi Kleen wrote:
> > > If we clear the registers, what the hell are you going to put in the
> > > RSB that helps you?
> >Â
> > RSB allows you to control chains of gadgets.
> 
> I admit the gadget thing is a bit obscure.
> 
> There's another case we were actually more worried about:Â
> 
> On Skylake and Broadwell when the RSB underflows it will fall back to theÂ
> indirect branch predictor, which can be poisoned and we try to avoid
> using with retpoline. So we try to avoid underflows, and this filling
> helps us with that.

That's no longer true for Broadwell with the latest microcode, right?

And is why we *were* talking about using retpoline only for pre-
Skylake, and using IBRS on Skylake onwards (where IBRS isn't quite so
horribly slow anyway).

Attachment: smime.p7s
Description: S/MIME cryptographic signature