[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <9e7b7406-ec24-2991-3577-ce7da61a61ca@amd.com>
Date:   Tue, 25 May 2021 09:47:24 -0500
From:   Brijesh Singh <brijesh.singh@....com>
To:     Borislav Petkov <bp@...en8.de>
Cc:     brijesh.singh@....com, x86@...nel.org,
        linux-kernel@...r.kernel.org, kvm@...r.kernel.org,
        tglx@...utronix.de, jroedel@...e.de, thomas.lendacky@....com,
        pbonzini@...hat.com, mingo@...hat.com, dave.hansen@...el.com,
        rientjes@...gle.com, seanjc@...gle.com, peterz@...radead.org,
        hpa@...or.com, tony.luck@...el.com
Subject: Re: [PATCH Part1 RFC v2 13/20] x86/sev: Register GHCB memory when
 SEV-SNP is active
On 5/25/21 9:35 AM, Borislav Petkov wrote:
>> In this particular case, the snp_register_ghcb() is shared between the
>> decompress and main kernel. The variable data->snp_ghcb_registered is
>> not visible in the decompressed path,
> Why is it not visible?
Maybe I should have said, its not applicable in the decompressed path.
The snp_ghcb_register is defined in the per-CPU structure, and used in
the per-CPU #VC handler. We don't have the per-CPU #VC handler in the
decompression code path.
Please see the arch/x86/kernel/sev.c
/* #VC handler runtime per-CPU data */
struct sev_es_runtime_data {
 ...
 ...
 bool snp_ghcb_registered;
}
Powered by blists - more mailing lists
 
