Re: [PATCH RESEND] x86/speculation: Fix user-mode spectre-v2 protection with KERNEL_IBRS

[KP Singh]

On Mon, Feb 20, 2023 at 10:52 AM Borislav Petkov <bp@xxxxxxxxx> wrote:
>
> On Mon, Feb 20, 2023 at 10:44:21AM -0800, KP Singh wrote:
> > No it cannot with IBRS which is really just KERNEL_IBRS enabled, we
>
> See my other reply. The intent is there to be able to do it. What needs
> to be figured out now is *why* we said no STIBP with IBRS? Was it an
> omission or was there some intent behind it.
>

Sure, it looks like an omission to me, we wrote a POC on Skylake that
was able to do cross-thread training with the current set of
mitigations.

STIBP with IBRS is still correct if spectre_v2=ibrs had really meant
IBRS everywhere, but just means KERNEL_IBRS, which means only kernel
is protected, userspace is still unprotected.

> --
> Regards/Gruss,
>     Boris.
>
> https://people.kernel.org/tglx/notes-about-netiquette