[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <4af7b73c-2ed0-9787-c07c-6a577c4c3c04@amd.com>
Date: Mon, 2 Oct 2017 06:52:01 -0500
From: Brijesh Singh <brijesh.singh@....com>
To: Borislav Petkov <bp@...e.de>
Cc: brijesh.singh@....com, linux-kernel@...r.kernel.org,
kvm@...r.kernel.org, Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...hat.com>,
"H. Peter Anvin" <hpa@...or.com>,
Paolo Bonzini <pbonzini@...hat.com>,
Radim Krčmář <rkrcmar@...hat.com>,
Jonathan Corbet <corbet@....net>,
Tom Lendacky <thomas.lendacky@....com>, x86@...nel.org
Subject: Re: [Part2 PATCH v4 01/29] Documentation/virtual/kvm: Add AMD Secure
Encrypted Virtualization (SEV)
On 9/29/17 1:54 AM, Borislav Petkov wrote:
---
>> Documentation/virtual/kvm/00-INDEX | 3 +
>> .../virtual/kvm/amd-memory-encryption.txt | 210 +++++++++++++++++++++
>> 2 files changed, 213 insertions(+)
>> create mode 100644 Documentation/virtual/kvm/amd-memory-encryption.txt
> Here's a diff which applies ontop of this one, it moves the KVM_SEV_*
> commands to Documentation/virtual/kvm/api.txt where they're all together
> in one place for obvious advantages.
Thank you for the patch. One of feedback in RFC patches was to put the
command id and their definitions in amd-memory-encryption.txt hence I
was trying to follow that recommendation. Most of ioctls I have seen in
api.txt are either common or architecture specific. These commands are
vendor specific and I was not sure if they belong to api.txt. I am okay
with moving the definition to api.txt in next patch and we can see if
kvm folks have any objections.
Are you okay if I move the command id definition documentation later in
the patch -- basically include it as part of 'KVM: Define SEV key
management command id' [1]
[1] https://marc.info/?l=kvm&m=150585445223846&w=2
> Also I did some small cleanups while at it.
> Notable is that the commands are a smaller number now and
> KVM_SEV_RECEIVE_UPDATE_DATA, KVM_SEV_RECEIVE_START and a couple more are
> missing.
>
> On purpose?
The series does not implement the migration commands yet hence I took it
off.
Powered by blists - more mailing lists