From owner-freebsd-security@freebsd.org Sun Mar 19 14:44:00 2017 Return-Path: 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 1DAF0D121DB; Sun, 19 Mar 2017 14:44:00 +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 D9CB410DE; Sun, 19 Mar 2017 14:43:59 +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 D37F4B846; Sun, 19 Mar 2017 14:43:51 +0000 (UTC) Received: by desk.des.no (Postfix, from userid 1001) id 9D544468C; Sun, 19 Mar 2017 15:43:46 +0100 (CET) From: =?utf-8?Q?Dag-Erling_Sm=C3=B8rgrav?= To: Andrey Chernov Cc: Xin LI , Steven Chamberlain , kostikbel@gmail.com, "freebsd-security\@freebsd.org" , freebsd Subject: Re: arc4random weakness References: <20170313220639.GB65190@pyro.eu.org> <20170315130615.GC25448@pyro.eu.org> <5160183b-9778-59aa-6cf9-118014a588eb@freebsd.org> <8677f9d8-b326-2526-47ce-f2e18421c074@freebsd.org> Date: Sun, 19 Mar 2017 15:43:46 +0100 In-Reply-To: <8677f9d8-b326-2526-47ce-f2e18421c074@freebsd.org> (Andrey Chernov's message of "Thu, 16 Mar 2017 22:26:09 +0300") Message-ID: <861sttpbrx.fsf@desk.des.no> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/25.1 (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: 2.1.23 Precedence: list List-Id: "Security issues \[members-only posting\]" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 19 Mar 2017 14:44:00 -0000 Andrey Chernov writes: > Theo kindly explained that zeroing whole page instead of single variable > suits to his newest arc4random better, since clears two structs at once > (including ChaCha state), making some form of backward secrecy. Yes, avoiding leaking key material to child processes would be useful for more than just arc4random. DES --=20 Dag-Erling Sm=C3=B8rgrav - des@des.no