[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20230814103527.GD776869@hirez.programming.kicks-ass.net>
Date: Mon, 14 Aug 2023 12:35:27 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Borislav Petkov <bp@...en8.de>
Cc: x86@...nel.org, linux-kernel@...r.kernel.org, David.Kaplan@....com,
Andrew.Cooper3@...rix.com, jpoimboe@...nel.org,
gregkh@...uxfoundation.org
Subject: Re: [RFC][PATCH 07/17] x86/cpu/kvm: Provide UNTRAIN_RET_VM
On Sun, Aug 13, 2023 at 12:36:57PM +0200, Borislav Petkov wrote:
> On Wed, Aug 09, 2023 at 09:12:25AM +0200, Peter Zijlstra wrote:
> > With the difference being that UNTRAIN_RET_VM uses
> > X86_FEATURE_IBPB_ON_VMEXIT instead of X86_FEATURE_ENTRY_IBPB.
> >
> > This cures VMEXIT doing potentially unret+IBPB or double IBPB.
>
> Can't - I have a separate flag for that and I set it only when !IBPB:
>
> case SRSO_CMD_IBPB_ON_VMEXIT:
> if (IS_ENABLED(CONFIG_CPU_SRSO)) {
> if (!boot_cpu_has(X86_FEATURE_ENTRY_IBPB) && has_microcode) {
> setup_force_cpu_cap(X86_FEATURE_IBPB_ON_VMEXIT);
>
Of course you can, just also set it with regular IBPB :-)
Powered by blists - more mailing lists