From owner-freebsd-security@freebsd.org  Fri Jan  5 08:16:50 2018
Return-Path: <owner-freebsd-security@freebsd.org>
Delivered-To: freebsd-security@mailman.ysv.freebsd.org
Received: from mx1.freebsd.org (mx1.freebsd.org
 [IPv6:2001:1900:2254:206a::19:1])
 by mailman.ysv.freebsd.org (Postfix) with ESMTP id 04690EBC1C3
 for <freebsd-security@mailman.ysv.freebsd.org>;
 Fri,  5 Jan 2018 08:16:50 +0000 (UTC) (envelope-from des@des.no)
Received: from smtp.des.no (smtp.des.no [194.63.250.102])
 by mx1.freebsd.org (Postfix) with ESMTP id C3EB17CF60
 for <freebsd-security@freebsd.org>; Fri,  5 Jan 2018 08:16:49 +0000 (UTC)
 (envelope-from des@des.no)
Received: from desk.des.no (smtp.des.no [194.63.250.102])
 by smtp.des.no (Postfix) with ESMTP id DEABA10397;
 Fri,  5 Jan 2018 08:16:48 +0000 (UTC)
Received: by desk.des.no (Postfix, from userid 1001)
 id A52495CE86; Fri,  5 Jan 2018 08:15:16 +0000 (UTC)
From: =?utf-8?Q?Dag-Erling_Sm=C3=B8rgrav?= <des@des.no>
To: Eric McCorkle <eric@metricspace.net>
Cc: "freebsd-security\@freebsd.org" <freebsd-security@freebsd.org>
Subject: Re: A more general possible meltdown/spectre countermeasure
References: <c98b7ac3-26f0-81ee-2769-432697f876e5@metricspace.net>
Date: Fri, 05 Jan 2018 09:15:16 +0100
In-Reply-To: <c98b7ac3-26f0-81ee-2769-432697f876e5@metricspace.net> (Eric
 McCorkle's message of "Thu, 4 Jan 2018 23:05:40 -0500")
Message-ID: <86efn4u3fv.fsf@desk.des.no>
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/25.3 (berkeley-unix)
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
X-BeenThere: freebsd-security@freebsd.org
X-Mailman-Version: 2.1.25
Precedence: list
List-Id: "Security issues \[members-only posting\]"
 <freebsd-security.freebsd.org>
List-Unsubscribe: <https://lists.freebsd.org/mailman/options/freebsd-security>, 
 <mailto:freebsd-security-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-security/>
List-Post: <mailto:freebsd-security@freebsd.org>
List-Help: <mailto:freebsd-security-request@freebsd.org?subject=help>
List-Subscribe: <https://lists.freebsd.org/mailman/listinfo/freebsd-security>, 
 <mailto:freebsd-security-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Fri, 05 Jan 2018 08:16:50 -0000

Eric McCorkle <eric@metricspace.net> writes:
> The obvious downside is that you take a performance hit storing things
> in non-cacheable locations, especially if you plan on doing heavy
> computation in that memory (say, encryption/decryption).  However, this
> is almost certainly going to be less than the projected 30-50%
> performance hit from other mitigations.

Where did you get those numbers?  Because the worst documented case for
KPTI is ~20% for I/O-intensive workloads, and PCID is likely to bring
this down to single digits if used correctly.  The KAISER paper claims a
slowdown of < 1%, but that may have been the result of undisclosed
features of the specific CPU they tested on.

DES
--=20
Dag-Erling Sm=C3=B8rgrav - des@des.no