Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.36 as permitted sender) client-ip=23.128.96.36;
Date:   Tue, 19 Sep 2023 10:56:30 -0700
From:   Kees Cook <kees@kernel.org>
To:     Dave Hansen <dave.hansen@intel.com>,
        Matteo Rizzo <matteorizzo@google.com>,
        Linus Torvalds <torvalds@linux-foundation.org>
CC:     Ingo Molnar <mingo@kernel.org>,
        "Lameter, Christopher" <cl@os.amperecomputing.com>,
        penberg@kernel.org, rientjes@google.com, iamjoonsoo.kim@lge.com,
        akpm@linux-foundation.org, vbabka@suse.cz,
        roman.gushchin@linux.dev, 42.hyeyoo@gmail.com,
        keescook@chromium.org, linux-kernel@vger.kernel.org,
        linux-doc@vger.kernel.org, linux-mm@kvack.org,
        linux-hardening@vger.kernel.org, tglx@linutronix.de,
        mingo@redhat.com, bp@alien8.de, dave.hansen@linux.intel.com,
        x86@kernel.org, hpa@zytor.com, corbet@lwn.net, luto@kernel.org,
        peterz@infradead.org, jannh@google.com, evn@google.com,
        poprdi@google.com, jordyzomer@google.com
Subject: =?US-ASCII?Q?Re=3A_=5BRFC_PATCH_00/14=5D_Prevent_cross?= =?US-ASCII?Q?-cache_attacks_in_the_SLUB_allocator?=
User-Agent: K-9 Mail for Android
In-Reply-To: <782b131a-c3b9-7679-824a-70625c966def@intel.com>
References: <20230915105933.495735-1-matteorizzo@google.com> <7a4f5128-28fd-3c5f-34c2-1c34f4448174@intel.com> <1d7573c0-ebbc-6ed2-f152-1045eb0542f9@os.amperecomputing.com> <CAHKB1w+rVyww0UrHhzeGfPA7FM482Z-7ApzXvekVqLHvTDAV3Q@mail.gmail.com> <ZQiLX0W2Tcr+wdJT@gmail.com> <CAHk-=wgGzB4u-WZsDpdgjwX1w5=9CLE0gorhaNFD09P1FUGeuQ@mail.gmail.com> <CAHKB1w+9GgY_e6J+rZ4zDaXrPZab5xteTuDEH0Z2hWe6x-pT5g@mail.gmail.com> <782b131a-c3b9-7679-824a-70625c966def@intel.com>
Message-ID: <F5445DE2-CE12-49F8-BF8D-D63ECF6BCC33@kernel.org>
MIME-Version: 1.0
Content-Type: text/plain;
 charset=utf-8
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On September 19, 2023 9:02:07 AM PDT, Dave Hansen <dave=2Ehansen@intel=2Eco=
m> wrote:
>On 9/19/23 08:48, Matteo Rizzo wrote:
>>> I think the whole "make it one single compile-time option" model is
>>> completely and fundamentally broken=2E
>> Wouldn't making this toggleable at boot time or runtime make performanc=
e
>> even worse?
>
>Maybe=2E
>
>But you can tolerate even more of a performance impact from a feature if
>the people that don't care can actually disable it=2E
>
>There are also plenty of ways to minimize the overhead of switching it
>on and off at runtime=2E  Static branches are your best friend here=2E

Let's start with a boot time on/off toggle (no per-slab, no switch on out-=
of-space, etc)=2E That should be sufficient for initial ease of use for tes=
ting, etc=2E But yes, using static_branch will nicely DTRT here=2E


--=20
Kees Cook