From: John Haxby Subject: [PATCHv2 0/1] fips-allowed tests fail with non-FIPS ciphers Date: Thu, 24 Sep 2015 18:24:34 +0100 Message-ID: <1443115475-27398-1-git-send-email-john.haxby@oracle.com> Cc: Herbert Xu , "David S. Miller" , Stephan Mueller , John Haxby To: linux-crypto@vger.kernel.org Return-path: Received: from ukc1-proxy-mwg04-o.oracle.com ([144.24.20.229]:18615 "EHLO sheep.uk.oracle.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1754852AbbIXRZS (ORCPT ); Thu, 24 Sep 2015 13:25:18 -0400 Sender: linux-crypto-owner@vger.kernel.org List-ID: Hello All, "Make fips=1 work on 4.1", they said, wittily, "it'll be easy." I suppose it wasn't that complicated, although I seem to be unearthing other problems as I go along. The first problem was dracut (and I owe an upstream patch for that) and the second problem was tcrypt. The tcrypt module was failing on authenc ciphers that wrap non-FIPS ciphers and hashes. These ones in fact: authenc(hmac(md5),ecb(cipher_null)) authenc(hmac(sha1),cbc(des)) authenc(hmac(sha1),ecb(cipher_null)) authenc(hmac(sha224),cbc(des)) authenc(hmac(sha256),cbc(des)) authenc(hmac(sha384),cbc(des)) authenc(hmac(sha512),cbc(des)) Stepham Mueller pointed out that no authenc() ciphers are FIPS approved and that ecb(des) also managed to get .fips_approved set. The following patch removes fips_allowed for all those patches. Again, Stephan pointed out that ansi_cprng will need to be taken off the allowed list at the end of the year. This patch doesn't pre-empt that. jch John Haxby (1): Disable fips-allowed for authenc() and des() ciphers crypto/testmgr.c | 16 ---------------- 1 file changed, 16 deletions(-) -- 2.4.3