From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from smtp.codeaurora.org by pdx-caf-mail.web.codeaurora.org (Dovecot) with LMTP id PwG+EwJEGltcAQAAmS7hNA ; Fri, 08 Jun 2018 08:53:27 +0000 Received: by smtp.codeaurora.org (Postfix, from userid 1000) id 5FF9F608BF; Fri, 8 Jun 2018 08:53:27 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on pdx-caf-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-2.9 required=2.0 tests=BAYES_00,MAILING_LIST_MULTI autolearn=ham autolearn_force=no version=3.4.0 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by smtp.codeaurora.org (Postfix) with ESMTP id 1B283608BF; Fri, 8 Jun 2018 08:53:22 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 smtp.codeaurora.org 1B283608BF Authentication-Results: pdx-caf-mail.web.codeaurora.org; dmarc=fail (p=none dis=none) header.from=redhat.com Authentication-Results: pdx-caf-mail.web.codeaurora.org; spf=none smtp.mailfrom=linux-kernel-owner@vger.kernel.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751141AbeFHIxT (ORCPT + 25 others); Fri, 8 Jun 2018 04:53:19 -0400 Received: from mx3-rdu2.redhat.com ([66.187.233.73]:54392 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1750922AbeFHIxQ (ORCPT ); Fri, 8 Jun 2018 04:53:16 -0400 Received: from smtp.corp.redhat.com (int-mx03.intmail.prod.int.rdu2.redhat.com [10.11.54.3]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 7DE5340252E9; Fri, 8 Jun 2018 08:53:15 +0000 (UTC) Received: from dhcp-128-65.nay.redhat.com (ovpn-12-137.pek2.redhat.com [10.72.12.137]) by smtp.corp.redhat.com (Postfix) with ESMTPS id 2BAE01116705; Fri, 8 Jun 2018 08:53:08 +0000 (UTC) Date: Fri, 8 Jun 2018 16:53:05 +0800 From: Dave Young To: David Howells Cc: keyrings@vger.kernel.org, linux-kernel@vger.kernel.org, kexec@lists.infradead.org, David Woodhouse , mathew.j.martineau@linux.intel.com, Laura Abbott , jwboyer@redhat.com Subject: Re: [PATCH] certs: always use secondary keyring first if possible Message-ID: <20180608085305.GA8594@dhcp-128-65.nay.redhat.com> References: <20171118044711.GA7352@dhcp-128-65.nay.redhat.com> <20171214102513.GA2371@dhcp-128-65.nay.redhat.com> <20180608072848.GA7773@dhcp-128-65.nay.redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20180608072848.GA7773@dhcp-128-65.nay.redhat.com> User-Agent: Mutt/1.9.5 (2018-04-13) X-Scanned-By: MIMEDefang 2.78 on 10.11.54.3 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.11.55.6]); Fri, 08 Jun 2018 08:53:15 +0000 (UTC) X-Greylist: inspected by milter-greylist-4.5.16 (mx1.redhat.com [10.11.55.6]); Fri, 08 Jun 2018 08:53:15 +0000 (UTC) for IP:'10.11.54.3' DOMAIN:'int-mx03.intmail.prod.int.rdu2.redhat.com' HELO:'smtp.corp.redhat.com' FROM:'dyoung@redhat.com' RCPT:'' Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 06/08/18 at 03:28pm, Dave Young wrote: > On 12/14/17 at 06:25pm, Dave Young wrote: > > On 11/18/17 at 12:47pm, Dave Young wrote: > > > Commit d3bfe84129f6 introduced secondary_trusted_keys keyring, current > > > users of verify_pkcs7_signature are below: > > > net/wireless/reg.c : uses its own trusted_keys > > > kernel/module_signing.c : pass NULL trusted_keys > > > crypto/asymmetric_keys/verify_pefile.c : pass NULL trusted_keys > > > > > > For both module and pefile verification, there is no reason to use builtin > > > keys only. Actually in Fedora kernel module signing code passes 1UL, but > > > kexec code does not pass 1UL for pefile verification thus we have below bug > > > https://bugzilla.redhat.com/show_bug.cgi?id=1470995 > > > > > > Drop the hard code 1UL checking so that pefile verification can use > > > secondary keyring as well. > > > > > > Signed-off-by: Dave Young > > > --- > > > certs/system_keyring.c | 2 -- > > > 1 file changed, 2 deletions(-) > > > > > > --- linux-x86.orig/certs/system_keyring.c > > > +++ linux-x86/certs/system_keyring.c > > > @@ -229,8 +229,6 @@ int verify_pkcs7_signature(const void *d > > > goto error; > > > > > > if (!trusted_keys) { > > > - trusted_keys = builtin_trusted_keys; > > > - } else if (trusted_keys == (void *)1UL) { > > > #ifdef CONFIG_SECONDARY_TRUSTED_KEYRING > > > trusted_keys = secondary_trusted_keys; > > > #else > > > > Another ping. > > > > If the (-1UL) is really needed, below file need update to use it > > But I think it is ugly.. > > crypto/asymmetric_keys/verify_pefile.c > > Ping again. Can anyone response to this issue? > > Let me describe more details about the problem: > > In Fedora kernel there is a patch below which is not upstreamed: > https://src.fedoraproject.org/rpms/kernel/blob/master/f/Fix-for-module-sig-verification.patch > > It has below changes: > > --- > diff --git a/kernel/module_signing.c b/kernel/module_signing.c > index 937c844..d3d6f95 100644 > --- a/kernel/module_signing.c > +++ b/kernel/module_signing.c > @@ -81,6 +81,6 @@ int mod_verify_sig(const void *mod, unsigned long *_modlen) > } > > return verify_pkcs7_signature(mod, modlen, mod + modlen, sig_len, > - NULL, VERIFYING_MODULE_SIGNATURE, > + (void *)1UL, VERIFYING_MODULE_SIGNATURE, > NULL, NULL); > } > --- > > Above change is needed because the verify_pkcs7_signature is doing > below: > --- > if (!trusted_keys) { > trusted_keys = builtin_trusted_keys; > } else if (trusted_keys == (void *)1UL) { > #ifdef CONFIG_SECONDARY_TRUSTED_KEYRING > trusted_keys = secondary_trusted_keys; > #else > trusted_keys = builtin_trusted_keys; > #endif > } > --- > > The trusted_keys is an argument passed to verify_pkcs7_signature > function. We can see that users of this function must pass "-1UL" > as trusted_keys to use secondary keyring. This "-1UL" is not I meant to say "1UL" instead of "-1UL".. > documented and it looks a hardcode api. Besides of the module > signing code, actually as I mentioned in the patch log kexec/kdump > also need passing "-1UL" to use the secondary keyring. > > But why do we need that hack? If I understand it correctly > if use secondary then builtin can still be used, see commit log > of d3bfe84129f65e0af2450743ebdab33d161d01c9: > If the secondary keyring is enabled, a link is created from that to > .builtin_trusted_keys so that the the latter will automatically be searched > too if the secondary keyring is searched. > > So why not directly use secondary in case trusted_keys == NULL? > > I'm not familar with the certs/keyring code, if I'm wrong please > correct me. > > -- > Thanks > Dave >