Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S934662Ab2J3VOp (ORCPT ); Tue, 30 Oct 2012 17:14:45 -0400 Received: from mail-oa0-f46.google.com ([209.85.219.46]:38290 "EHLO mail-oa0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S934577Ab2J3VOn (ORCPT ); Tue, 30 Oct 2012 17:14:43 -0400 MIME-Version: 1.0 In-Reply-To: <20121030192157.11000.12874.stgit@warthog.procyon.org.uk> References: <20121030191927.11000.68420.stgit@warthog.procyon.org.uk> <20121030192157.11000.12874.stgit@warthog.procyon.org.uk> Date: Tue, 30 Oct 2012 14:14:42 -0700 X-Google-Sender-Auth: ehAvL8JOMJscy4hQiiFpDsMF7K8 Message-ID: Subject: Re: [PATCH 17/23] pefile: Strip the wrapper off of the cert data block From: Kees Cook To: David Howells Cc: rusty@rustcorp.com.au, pjones@redhat.com, jwboyer@redhat.com, mjg@redhat.com, dmitry.kasatkin@intel.com, zohar@linux.vnet.ibm.com, keyrings@linux-nfs.org, linux-kernel@vger.kernel.org Content-Type: text/plain; charset=ISO-8859-1 X-System-Of-Record: true Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 3900 Lines: 118 On Tue, Oct 30, 2012 at 12:21 PM, David Howells wrote: > The certificate data block in a PE binary has a wrapper around the PKCS#7 > signature we actually want to get at. Strip this off and check that we've got > something that appears to be a PKCS#7 signature. > > Signed-off-by: David Howells > --- > > crypto/asymmetric_keys/pefile_parser.c | 60 ++++++++++++++++++++++++++++++++ > 1 file changed, 60 insertions(+) > > > diff --git a/crypto/asymmetric_keys/pefile_parser.c b/crypto/asymmetric_keys/pefile_parser.c > index efae278..24c117e 100644 > --- a/crypto/asymmetric_keys/pefile_parser.c > +++ b/crypto/asymmetric_keys/pefile_parser.c > @@ -15,6 +15,7 @@ > #include > #include > #include > +#include > #include > #include > #include > @@ -138,6 +139,61 @@ static int pefile_parse_binary(struct key_preparsed_payload *prep, > } > > /* > + * Check and strip the PE wrapper from around the signature and check that the > + * remnant looks something like PKCS#7. > + */ > +static int pefile_strip_sig_wrapper(struct key_preparsed_payload *prep, > + struct pefile_context *ctx) > +{ > + struct win_certificate wrapper; > + const u8 *pkcs7; > + > + if (ctx->sig_len < 8) { > + pr_devel("Signature wrapper too short\n"); > + return -ELIBBAD; > + } > + > + memcpy(&wrapper, prep->data + ctx->sig_offset, 8); Instead of the literal 8, sizeof(wrapper)? > + pr_devel("sig wrapper = { %x, %x, %x }\n", > + wrapper.length, wrapper.revision, wrapper.cert_type); > + if (wrapper.length != ctx->sig_len) { > + pr_devel("Signature wrapper len wrong\n"); > + return -ELIBBAD; > + } > + if (wrapper.revision != WIN_CERT_REVISION_2_0) { > + pr_devel("Signature is not revision 2.0\n"); > + return -ENOTSUPP; > + } > + if (wrapper.cert_type != WIN_CERT_TYPE_PKCS_SIGNED_DATA) { > + pr_devel("Signature certificate type is not PKCS\n"); > + return -ENOTSUPP; > + } > + > + ctx->sig_offset += 8; > + ctx->sig_len -= 8; Same for these? > + if (ctx->sig_len == 0) { > + pr_devel("Signature data missing\n"); > + return -EKEYREJECTED; > + } > + > + /* What's left should a PKCS#7 cert */ > + pkcs7 = prep->data + ctx->sig_offset; > + if (pkcs7[0] == (ASN1_CONS_BIT | ASN1_SEQ)) { > + if (pkcs7[1] == 0x82 && > + pkcs7[2] == (((ctx->sig_len - 4) >> 8) & 0xff) && > + pkcs7[3] == ((ctx->sig_len - 4) & 0xff)) > + return 0; > + if (pkcs7[1] == 0x80) > + return 0; > + if (pkcs7[1] > 0x82) > + return -EMSGSIZE; Can these literals be replaced with something more meaningful? > + } > + > + pr_devel("Signature data not PKCS#7\n"); > + return -ELIBBAD; > +} > + > +/* > * Parse a PE binary. > */ > static int pefile_key_preparse(struct key_preparsed_payload *prep) > @@ -152,6 +208,10 @@ static int pefile_key_preparse(struct key_preparsed_payload *prep) > if (ret < 0) > return ret; > > + ret = pefile_strip_sig_wrapper(prep, &ctx); > + if (ret < 0) > + return ret; > + > return -ENOANO; // Not yet complete > } > > -- Kees Cook Chrome OS Security -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/