From:   Takahiro Itazuri <itazur@amazon.com>
To:     <bp@alien8.de>
CC:     <dave.hansen@linux.intel.com>, <itazur@amazon.com>,
        <kvm@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
        <mingo@redhat.com>, <pbonzini@redhat.com>, <seanjc@google.com>,
        <tglx@linutronix.de>, <x86@kernel.org>, <zulinx86@gmail.com>
Subject: Re: [PATCH 0/2] KVM: x86: Propagate AMD-specific IBRS bits to guests
Date:   Tue, 28 Feb 2023 19:41:53 +0000
Message-ID: <20230228194153.46995-1-itazur@amazon.com>
In-Reply-To: <Y/5U3N0XfRaZ2KkX@zn.tnic>
References: <Y/5U3N0XfRaZ2KkX@zn.tnic>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Content-Type: text/plain
Precedence: bulk

Date:   Tue, 28 Feb 2023 20:24:12 +0100
From:   Borislav Petkov <bp@alien8.de>
> I'd prefer if VMMs did supply whatever they prefer to the guests
> instead. None of those bits are used in the kernel for mitigations, as
> you've realized.

It is true that the kernel does not use those bits at all, but any
codes could be run inside guests.

One of examples is the following spectre/meltdown checker scipt used as
de facto standard.
https://github.com/speed47/spectre-meltdown-checker/blob/master/spectre-meltdown-checker.sh#L2768

Best regards,
Takahiro Itazuri