Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Message-ID: <3bfe66204ee84a0bbccaf7cd20af0d8300fb9f26.camel@kernel.org>
Subject: Re: [PATCH V2] x86/sgx: Add poison handling to reclaimer
From:   Jarkko Sakkinen <jarkko@kernel.org>
To:     Reinette Chatre <reinette.chatre@intel.com>, tony.luck@intel.com,
        dave.hansen@linux.intel.com, tglx@linutronix.de, bp@alien8.de,
        luto@kernel.org, mingo@redhat.com, linux-sgx@vger.kernel.org,
        x86@kernel.org
Cc:     linux-kernel@vger.kernel.org
Date:   Thu, 20 Jan 2022 15:09:03 +0200
In-Reply-To: <be5af586f667c7bcb8ef01286ce75675de5d100f.1642630582.git.reinette.chatre@intel.com>
References: <be5af586f667c7bcb8ef01286ce75675de5d100f.1642630582.git.reinette.chatre@intel.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
User-Agent: Evolution 3.42.3 
MIME-Version: 1.0
Precedence: bulk

On Wed, 2022-01-19 at 14:23 -0800, Reinette Chatre wrote:
> The SGX reclaimer code lacks page poison handling in its main
> free path. This can lead to avoidable machine checks if a
> poisoned page is freed and reallocated instead of being
> isolated.
>=20
> A troublesome scenario is:
> =C2=A01. Machine check (#MC) occurs (asynchronous, !MF_ACTION_REQUIRED)
> =C2=A02. arch_memory_failure() is eventually called
> =C2=A03. (SGX) page->poison set to 1
> =C2=A04. Page is reclaimed
> =C2=A05. Page added to normal free lists by sgx_reclaim_pages()
> =C2=A0=C2=A0=C2=A0 ^ This is the bug (poison pages should be isolated on =
the
> =C2=A0=C2=A0=C2=A0 sgx_poison_page_list instead)
> =C2=A06. Page is reallocated by some innocent enclave, a second
> (synchronous)
> =C2=A0=C2=A0=C2=A0 in-kernel #MC is induced, probably during EADD instruc=
tion.
> =C2=A0=C2=A0=C2=A0 ^ This is the fallout from the bug
>=20
> (6) is unfortunate and can be avoided by replacing the open coded
> enclave page freeing code in the reclaimer with sgx_free_epc_page()
> to obtain support for poison page handling that includes placing the
> poisoned page on the correct list.
>=20
> Fixes: d6d261bded8a ("x86/sgx: Add new sgx_epc_page flag bit to mark
> free pages")
> Fixes: 992801ae9243 ("x86/sgx: Initial poison handling for dirty and
> free pages")

Same comment as for the first version: remove the first fixes tag.

BR, Jarkko