Message-ID: <3682fe69-a2e8-3018-fb29-ca3db9bfffe7@...1227...> Date: Wed, 26 Jan 2022 12:50:48 +0100 MIME-Version: 1.0 Content-Language: en-GB References: <20220125025128.2117-1-selva.nair@...277...> <20220125025128.2117-3-selva.nair@...277...> From: Arne Schwabe In-Reply-To: <20220125025128.2117-3-selva.nair@...277...> Subject: Re: [Openvpn-devel] [PATCH 3/3] Support PSS signing using pkcs11-helper >= 1.28 List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit To: selva.nair@...277..., openvpn-devel@lists.sourceforge.net Am 25.01.22 um 03:51 schrieb selva.nair@...277...: > From: Selva Nair > > - Call pkcs11h_certificate_signAny_ex() when available > so that the signature mechanism parameters can be pased. > (Required for RSA-PSS signature). > > Signed-off-by: Selva Nair > --- > src/openvpn/pkcs11_openssl.c | 123 +++++++++++++++++++++++++++++++++-- > 1 file changed, 118 insertions(+), 5 deletions(-) > > diff --git a/src/openvpn/pkcs11_openssl.c b/src/openvpn/pkcs11_openssl.c > index 9cf46b2c..5d1a5de6 100644 > --- a/src/openvpn/pkcs11_openssl.c > +++ b/src/openvpn/pkcs11_openssl.c > @@ -45,10 +45,112 @@ > #ifdef HAVE_XKEY_PROVIDER > static XKEY_EXTERNAL_SIGN_fn xkey_pkcs11h_sign; > > +#if PKCS11H_VERSION > ((1<<16) | (27<<8)) /* version > 1.27 */ > + > +/* Table linking OpenSSL digest NID with CKM and CKG constants in PKCS#11 */ > +#define MD_TYPE(n) {NID_sha##n, CKM_SHA##n, CKG_MGF1_SHA##n} > +static const struct > +{ > + int nid; > + unsigned long ckm_id; > + unsigned long mgf_id; > +} mdtypes[] = {MD_TYPE(224), MD_TYPE(256), MD_TYPE(384), MD_TYPE(512), > + {NID_sha1, CKM_SHA_1, CKG_MGF1_SHA1}, /* SHA_1 naming is an oddity */ > + {NID_undef, 0, 0}}; > + > +/* From sigalg, derive parameters for pss signature and fill in pss_params. > + * Its of type CK_RSA_PKCS_PSS_PARAMS struct with three fields to be filled in: > + * {enum hashAlg, enum mgf, ulong sLen} > + * where hashAlg is CKM_SHA256 etc., mgf is CKG_MGF1_SHA256 etc. > + */ > +static int > +set_pss_params(CK_RSA_PKCS_PSS_PARAMS *pss_params, XKEY_SIGALG sigalg, > + pkcs11h_certificate_t cert) > +{ > + int ret = 0; > + X509 *x509 = NULL; > + EVP_PKEY *pubkey = NULL; > + > + if ((x509 = pkcs11h_openssl_getX509(cert)) == NULL > + || (pubkey = X509_get0_pubkey(x509)) == NULL) > + { > + msg(M_WARN, "PKCS#11: Unable get public key"); > + goto cleanup; > + } > + > + /* map mdname to CKM and CKG constants for hash and mgf algorithms */ > + int i = 0; > + int nid = OBJ_sn2nid(sigalg.mdname); > + while (mdtypes[i].nid != NID_undef && mdtypes[i].nid != nid) > + { > + i++; > + } > + pss_params->hashAlg = mdtypes[i].ckm_id; > + pss_params->mgf = mdtypes[i].mgf_id; > + > + /* determine salt length */ > + int mdsize = EVP_MD_size(EVP_get_digestbyname(sigalg.mdname)); This will break for newer hashes since it relies on nids but we have a fixed table anyway, it will break before that. But maybe we should bail out if we cannot find an entry in the translation table? Or is EVP_MD_size(NID_undef) well defined? Arne