From: =?ISO-8859-15?Q?Adrian-Ken_R=FCegsegger?= Subject: Re: tcrypt: hmac test with keys > blocksize fail Date: Sat, 03 May 2008 13:44:41 +0200 Message-ID: <481C5029.9040406@swiss-it.ch> References: <481BAD7A.3010003@swiss-it.ch> <20080503062113.GB30328@Chamillionaire.breakpoint.cc> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-15 Content-Transfer-Encoding: 7bit Cc: linux-crypto@vger.kernel.org To: Sebastian Siewior Return-path: Received: from zux006-004-203.adsl.green.ch ([81.6.4.203]:2364 "EHLO mailx.swiss-it.ch" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755404AbYECLot (ORCPT ); Sat, 3 May 2008 07:44:49 -0400 In-Reply-To: <20080503062113.GB30328@Chamillionaire.breakpoint.cc> Sender: linux-crypto-owner@vger.kernel.org List-ID: Sebastian Siewior wrote: > * Adrian-Ken R?egsegger | 2008-05-03 02:10:34 [+0200]: > >> Hi, > Hi Adrian-Ken, Hello Sebastian, >> using the cryptodev-2.6 tree I noticed that the hmac tests that have >> keys larger than blocksize for md5 and the various sha algorithms all >> fail (tcrypt mode=10[0-5]). The other tests seem to pass just fine. >> >> The issue seems to have come from commit >> de224c309b5631bdaae3fcd6880cfb93b52f5a53. > Is this a bisect result? No, this was a guess of mine, since I tested the two revisions just prior and after the three tcrypt-commits from you. I did a full bisect and the actual guilty commit is: [562954d5e01d08154cf15c7e12e6e9ec803f50f7] [CRYPTO] tcrypt: Change the usage of the test vectors >> I have tested 48c8949ea8460216783dd33640121187b9531b60 which does not >> contain this bug. It's just before the various tcrypt-changes from >> Sebastian Siewior. > I tested it with tcrypt mode=0 what covers 100-105 but it may slip > through. I will look into this later. As I mentioned, it's only the hmac tests with keys larger than blocksize that fail. Additionaly I just saw, that ecb(des) encryption test nr 5 passes but it seems that the setkey operation fails with: setkey() failed flags=100100 -Adrian