linux-kernel - Re: [PATCH] x86/sev: Make SEV

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <Z88NueKqfCPbpzQm@example.org>
Date: Mon, 10 Mar 2025 17:05:13 +0100
From: Alexey Gladkov <legion@...nel.org>
To: Juergen Gross <jgross@...e.com>
Cc: Borislav Petkov <bp@...en8.de>, Joerg Roedel <joro@...tes.org>,
	"Alexey Gladkov (Intel)" <alexey.gladkov@...el.com>,
	"Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>,
	Dave Hansen <dave.hansen@...el.com>, Joerg Roedel <jroedel@...e.de>,
	Ingo Molnar <mingo@...nel.org>, x86@...nel.org, hpa@...or.com,
	Tom Lendacky <thomas.lendacky@....com>,
	Nikunj A Dadhania <nikunj@....com>, linux-kernel@...r.kernel.org,
	Larry.Dewey@....com
Subject: Re: [PATCH] x86/sev: Make SEV_STATUS available via SYSFS

On Mon, Mar 10, 2025 at 04:52:37PM +0100, Juergen Gross wrote:
> On 10.03.25 16:43, Alexey Gladkov wrote:
> > On Mon, Mar 10, 2025 at 04:33:08PM +0100, Jürgen Groß wrote:
> >> On 10.03.25 16:11, Borislav Petkov wrote:
> >>> On Mon, Mar 10, 2025 at 03:50:09PM +0100, Alexey Gladkov wrote:
> >>>> Am I understand correctly that you and Joerg are proposing
> >>>>
> >>>> /sys/guest/tdx/...
> >>>> /sys/guest/sev/...
> >>>>
> >>>> ?
> >>>>
> >>>> Which path to use for the host side ?
> >>>>
> >>>> For guest: /sys/coco/guest/{tdx,sev}/...
> >>>> For host:  /sys/coco/host/{tdx,sev}/...
> >>>>
> >>>> Maybe it would be better to add the "coco" subdirectory or something like
> >>>> that ?
> >>>
> >>> Hmm, so we can do
> >>>
> >>> /sys/guest
> >>>
> >>> and extend
> >>>
> >>> /sys/hypervisor
> >>>
> >>> Or we can do what you're suggesting.
> >>>
> >>> If we do /sys/coco/host, then we'll have two different places to read HV info.
> >>>
> >>> Or we can stick *everything* coco needs in
> >>>
> >>> /sys/coco/{sev,tdx}
> >>>
> >>> but then it is coco-specific and if other guest types want to put stuff in
> >>> sysfs, it'll get ugly.
> >>>
> >>> So I guess having
> >>>
> >>> /sys/guest
> >>> and
> >>> /sys/hypervisor
> >>>
> >>> kinda keeps it all clean, hierarchy-wise...
> >>>
> >>> Right?
> >>
> >> Kind of.
> >>
> >> /sys/hypervisor is meant to provide data for running under a hypervisor.
> >>
> >> It is NOT meant to provide data for running as a hypervisor.
> >>
> >> So far when running either under Xen or under z/VM /sys/hypervisor is being
> >> populated.
> >>
> >> I'm not feeling really strong here. I just want to state the status quo.
> > 
> > OK, so I misunderstood.
> > 
> > If in the /sys/hypervisor we have information for guest (for running under
> > a hypervisor), where do you propose to put the information for the
> > host-side (for running as a hypervisor) ?
> > 
> 
> /sys/kvm/ might be a good fit when using KVM?

As far as I can see /sys/hypervisor should contain information about both
host-side and guest-side. If of course we would use the same files as
described in the ABI xen team.

Why do we need /sys/kvm/{tdx,sev}/ ?

I keep mentioning tdx and sev because they will have a different set of
files.

-- 
Rgrds, legion