Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <06a9fef8579e880b9b031f03911739d4d902dbe0.camel@intel.com>
Subject: Re: [PATCH] [v2] x86/sgx: Allow enclaves to use Asynchrounous Exit
 Notification
From:   Kai Huang <kai.huang@intel.com>
To:     Dave Hansen <dave.hansen@intel.com>,
        Dave Hansen <dave.hansen@linux.intel.com>
Cc:     Jarkko Sakkinen <jarkko@kernel.org>,
        Andy Lutomirski <luto@kernel.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
        "H. Peter Anvin" <hpa@zytor.com>,
        Sean Christopherson <seanjc@google.com>,
        Haitao Huang <haitao.huang@linux.intel.com>, x86@kernel.org,
        linux-sgx@vger.kernel.org, linux-kernel@vger.kernel.org
Date:   Mon, 25 Jul 2022 22:36:17 +1200
In-Reply-To: <a82e840f-2030-7ab3-7160-f1b900ecdb7d@intel.com>
References: <20220720191347.1343986-1-dave.hansen@linux.intel.com>
         <ab467244dd03b5f94bafe9068b1c02790033c18c.camel@intel.com>
         <a82e840f-2030-7ab3-7160-f1b900ecdb7d@intel.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
User-Agent: Evolution 3.44.3 (3.44.3-1.fc36) 
MIME-Version: 1.0
Precedence: bulk

On Fri, 2022-07-22 at 08:21 -0700, Dave Hansen wrote:
> On 7/22/22 06:26, Kai Huang wrote:
> > Did a quick look at the spec.  It appears ENCLU[EDECCSSA] should be use=
d
> > together with AEX-notify.  So besides advertising the new
> > SGX_ATTR_ASYNC_EXIT_NOTIFY bit to the KVM guest, I think we should also
> > advertise the ENCLU[EDECCSSA] support in guest's CPUID, like below (unt=
ested)?
>=20
> Sounds like a great follow-on patch!  It doesn't seem truly functionally
> required from the spec:
>=20
> > EDECCSSA is a new Intel SGX user leaf function
> > (ENCLU[EDECCSSA]) that can facilitate AEX notification handling...
>=20
> If that's wrong or imprecise, I'd love to hear more about it and also
> about how the spec will be updated.
>=20

They are enumerated separately, but looks in practice the notify handler wi=
ll
use it to switch back to the correct/targeted CSSA to continue to run norma=
lly
after handling the exit notify.  This is my understanding of the "facilitat=
e"
mean in the spec.

Btw, in real hardware I think the two should come together, meaning no real
hardware will only support one.=20

Haitao, could you give us more information?

--=20
Thanks,
-Kai