[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <51437ee4-e3cc-461b-8317-f20a4711a06c@amd.com>
Date: Mon, 7 Apr 2025 21:47:33 +0530
From: Neeraj Upadhyay <Neeraj.Upadhyay@....com>
To: Borislav Petkov <bp@...en8.de>
Cc: linux-kernel@...r.kernel.org, tglx@...utronix.de, mingo@...hat.com,
dave.hansen@...ux.intel.com, Thomas.Lendacky@....com, nikunj@....com,
Santosh.Shukla@....com, Vasant.Hegde@....com, Suravee.Suthikulpanit@....com,
David.Kaplan@....com, x86@...nel.org, hpa@...or.com, peterz@...radead.org,
seanjc@...gle.com, pbonzini@...hat.com, kvm@...r.kernel.org,
kirill.shutemov@...ux.intel.com, huibo.wang@....com, naveen.rao@....com
Subject: Re: [RFC v2 01/17] x86/apic: Add new driver for Secure AVIC
On 4/7/2025 6:47 PM, Borislav Petkov wrote:
> On Wed, Apr 02, 2025 at 04:04:34PM +0530, Neeraj Upadhyay wrote:
>> - snp_get_unsupported_features() looks like below.
>> It checks that, for the feature bits which are part of SNP_FEATURES_IMPL_REQ,
>> if they are enabled in hypervisor (and so reported in sev_status),
>> guest need to implement/enable those features. SAVIC also falls in that category
>> of SNP features.
>>
>> So, if CONFIG_AMD_SECURE_AVIC is disabled, guest would run with SAVIC feature
>> disabled in guest. This would cause undefined behavior for that guest if SAVIC
>> feature is active for that guest in hypervisor.
>
> Ok, so SNP_FEATURES_IMPL_REQ will contain the SAVIC bit (unconditionally,
> without the ifdeffery) and SNP_FEATURES_PRESENT will contain that thing you
> had suggested with the ifdeffery to denote whether SAVIC support has been
> enabled at build time:
>
> https://lore.kernel.org/r/e0362a96-4b3a-44b1-8d54-806a6b045799@amd.com
>
> or not.
>
> Right?
>
Yes, that is the intent here.
- Neeraj
Powered by blists - more mailing lists