linux-kernel - Re: [PATCH 00/19] Refresh queued CET virtualization series

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <YqsB9upUystxvl+d@hirez.programming.kicks-ass.net>
Date:   Thu, 16 Jun 2022 12:12:06 +0200
From:   Peter Zijlstra <peterz@...radead.org>
To:     Yang Weijiang <weijiang.yang@...el.com>
Cc:     pbonzini@...hat.com, seanjc@...gle.com, x86@...nel.org,
        kvm@...r.kernel.org, linux-kernel@...r.kernel.org,
        rick.p.edgecombe@...el.com
Subject: Re: [PATCH 00/19] Refresh queued CET virtualization series

On Thu, Jun 16, 2022 at 04:46:24AM -0400, Yang Weijiang wrote:

> To minimize the impact to exiting kernel/KVM code, most of KVM patch
> code can be bypassed during runtime.Uncheck "CONFIG_X86_KERNEL_IBT"
> and "CONFIG_X86_SHADOW_STACK" in Kconfig before kernel build to get
> rid of CET featrures in KVM. If both of them are not enabled, KVM
> clears related feature bits as well as CET user bit in supported_xss,
> this makes CET related checks stop at the first points. Since most of
> the patch code runs on the none-hot path of KVM, it's expected to
> introduce little impact to existing code.

Do I understand this right in that a host without X86_KERNEL_IBT cannot
run a guest with X86_KERNEL_IBT on? That seems unfortunate, since that
was exactly what I did while developing the X86_KERNEL_IBT patches.

I'm thinking that if the hardware supports it, KVM should expose it,
irrespective of the host kernel using it.