linux-kernel - Re: x86/sgx: uapi change proposal

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: <20181222063241.GA8895@linux.intel.com>
Date:   Sat, 22 Dec 2018 08:32:41 +0200
From:   Jarkko Sakkinen <jarkko.sakkinen@...ux.intel.com>
To:     Andy Lutomirski <luto@...nel.org>
Cc:     Sean Christopherson <sean.j.christopherson@...el.com>,
        Jethro Beekman <jethro@...tanix.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
        "x86@...nel.org" <x86@...nel.org>,
        Dave Hansen <dave.hansen@...ux.intel.com>,
        Peter Zijlstra <peterz@...radead.org>,
        "H. Peter Anvin" <hpa@...or.com>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "linux-sgx@...r.kernel.org" <linux-sgx@...r.kernel.org>,
        Josh Triplett <josh@...htriplett.org>,
        Haitao Huang <haitao.huang@...ux.intel.com>,
        "Dr . Greg Wettstein" <greg@...ellic.com>
Subject: Re: x86/sgx: uapi change proposal

On Fri, Dec 21, 2018 at 09:12:46AM -0800, Andy Lutomirski wrote:
> /dev/sgx/enclave: Each instance is an enclave.
> 
> /dev/sgx/epc: Used to get raw EPC for KVM. Might have different
> permissions, perhaps 0660 and group kvm.
> 
> /dev/sgx/something_else: For when SGX v3 adds something else :)

Responding again to this anyway now that I have had time think about
it.

Here is now I see it:

1. /dev/sgx/enclave should be /dev/sgx as it is now.
2. /dev/sgx/epc should be something that you'd reach through /dev/kvm.
   This essentially a circular dependency. KVM uapi should provide
   KVM services. Now you sprinkle KVM uapi to two subsystems.
3. "something else" is securityfs (e.g. provisioning). That is kind of
   stuff that it is meant for.

I'm sorry but from my perspective this does not look too good no
matter what glasses I put on...

/Jarkko