From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from smtp1.linuxfoundation.org (smtp1.linux-foundation.org [172.17.192.35]) by mail.linuxfoundation.org (Postfix) with ESMTPS id C86868F0 for ; Wed, 12 Aug 2015 22:51:39 +0000 (UTC) Received: from mail-oi0-f51.google.com (mail-oi0-f51.google.com [209.85.218.51]) by smtp1.linuxfoundation.org (Postfix) with ESMTPS id 521E5258 for ; Wed, 12 Aug 2015 22:51:39 +0000 (UTC) Received: by oip136 with SMTP id 136so17499451oip.1 for ; Wed, 12 Aug 2015 15:51:38 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: <13275.1439419600@warthog.procyon.org.uk> References: <20436.1438090619@warthog.procyon.org.uk> <1438096213.5441.147.camel@HansenPartnership.com> <1438097471.5441.152.camel@HansenPartnership.com> <1438099839.5441.165.camel@HansenPartnership.com> <1438100102.26913.183.camel@infradead.org> <30361.1438101879@warthog.procyon.org.uk> <1438111168.26913.189.camel@infradead.org> <1438121016.5441.233.camel@HansenPartnership.com> <16035.1439324695@warthog.procyon.org.uk> <11239.1439403720@warthog.procyon.org.uk> <1439405139.3100.147.camel@infradead.org> <13275.1439419600@warthog.procyon.org.uk> From: Andy Lutomirski Date: Wed, 12 Aug 2015 15:51:18 -0700 Message-ID: To: David Howells Content-Type: text/plain; charset=UTF-8 Cc: James Bottomley , Luis Rodriguez , "ksummit-discuss@lists.linuxfoundation.org" , Kyle McMartin Subject: Re: [Ksummit-discuss] [TECH TOPIC] Firmware signing List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Wed, Aug 12, 2015 at 3:46 PM, David Howells wrote: > Andy Lutomirski wrote: > >> Once we're talking real, modern public keys, there's no point in even >> hashing them. A good cryptosystem will have 32-byte public keys, and >> a sufficiently strong hash will be 32 bytes. > > And likely non-compliant with various security certifications. Humor me: what security certification would not be okay with 65 bytes of ECDSA/P-256 public key? And what security certification would not be okay with the SHA-256 (or SHA3-256) hash of something appropriate? Even with compression, the standards should accept any representation whatsoever that is *decompressed* to the correct 65-byte octet string prior to calling into any crypto code. Of course, there are (or were or something) those who claim to have or have had patents on that, from my extremely vague memory. --Andy