[PATCH 0/1] TPM2 engine support for openssl

[PATCH 0/1] TPM2 engine support for openssl

[James Bottomley]

This is a completed version of the original RFC.  It's working now both
on the TPM2 simulator and on real hardware (I've converted my laptop to
TPM2).  I've updated it to use the latest version of the ASN.1 for the
key format (still using a TCG OID).

I have it building here (it's what I'm currently using for my laptop
VPNs):

https://build.opensuse.org/package/show/home:jejb1:Tumbleweed/openssl_tpm_engine

But note that this version also has experimental patches to activate
the in-kernel TPM Resource Manager because for multiple applications
TPM2 really doesn't work well without one.  Since the patch for the RM
is currently not upstream (yet), it's not going to work unless you have
a patched kernel.

James

---

James Bottomley (1):
  add TPM2 version of create_tpm2_key and libtpm2.so engine

 Makefile.am       |  12 +-
 create_tpm2_key.c | 451 +++++++++++++++++++++++++++++++++++++++++++
 e_tpm2.c          | 559 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
 tpm2-asn.h        |  59 ++++++
 tpm2-common.c     | 175 +++++++++++++++++
 tpm2-common.h     |  10 +
 6 files changed, 1264 insertions(+), 2 deletions(-)
 create mode 100644 create_tpm2_key.c
 create mode 100644 e_tpm2.c
 create mode 100644 tpm2-asn.h
 create mode 100644 tpm2-common.c
 create mode 100644 tpm2-common.h

-- 
2.6.6


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most 
engaging tech sites, SlashDot.org! http://sdm.link/slashdot

[PATCH 1/1] add TPM2 version of create_tpm2_key and libtpm2.so engine

[James Bottomley]

This patch adds RSA signing for TPM2 keys.  There's a limitation to the
way TPM2 does signing: it must recognise the OID for the signature. 
 That fails for the MD5-SHA1 signatures of the TLS/SSL certificate
verification protocol, so I'm using RSA_Decrypt for both signing
(encryption) and decryption ... meaning that this only works with TPM
decryption keys.  It is possible to use the prior code, which preserved
the distinction of signing and decryption keys, but only at the expense
of not being able to support SSL or TLS lower than 1.2

Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>

---
v2: - use TPM2_RSA_Decrypt for both decryption and signing operations
    - Add authority processing
    - Add TPM internal key creation
    - allow persistent parents
    - update to use transient connections to the TPM
---
 Makefile.am       |  12 +-
 create_tpm2_key.c | 451 +++++++++++++++++++++++++++++++++++++++++++
 e_tpm2.c          | 559 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
 tpm2-asn.h        |  59 ++++++
 tpm2-common.c     | 175 +++++++++++++++++
 tpm2-common.h     |  10 +
 6 files changed, 1264 insertions(+), 2 deletions(-)
 create mode 100644 create_tpm2_key.c
 create mode 100644 e_tpm2.c
 create mode 100644 tpm2-asn.h
 create mode 100644 tpm2-common.c
 create mode 100644 tpm2-common.h

diff --git a/Makefile.am b/Makefile.am
index 6695656..fb4f529 100644
--- a/Makefile.am
+++ b/Makefile.am
@@ -2,12 +2,20 @@ SUBDIRS=. test
 
 EXTRA_DIST = README  openssl.cnf.sample
 
-openssl_engine_LTLIBRARIES=libtpm.la
-bin_PROGRAMS=create_tpm_key
+openssl_engine_LTLIBRARIES=libtpm.la libtpm2.la
+bin_PROGRAMS=create_tpm_key create_tpm2_key
 openssl_enginedir=@libdir@/openssl/engines
 
 libtpm_la_LIBADD=-lcrypto -lc -ltspi
 libtpm_la_SOURCES=e_tpm.c e_tpm.h e_tpm_err.c
 
+libtpm2_la_LIBADD=-lcrypto -lc -ltss
+libtpm2_la_SOURCES=e_tpm2.c tpm2-common.c
+libtpm2_la_CFLAGS=-g -Werror
+
 create_tpm_key_SOURCES=create_tpm_key.c
 create_tpm_key_LDADD=-ltspi
+
+create_tpm2_key_SOURCES=create_tpm2_key.c tpm2-common.c
+create_tpm2_key_LDADD=-lcrypto -ltss
+create_tpm2_key_CFLAGS=-Werror
diff --git a/create_tpm2_key.c b/create_tpm2_key.c
new file mode 100644
index 0000000..ca3b38f
--- /dev/null
+++ b/create_tpm2_key.c
@@ -0,0 +1,451 @@
+/*
+ *
+ *   Copyright (C) 2016 James Bottomley <James.Bottomley@HansenPartnership.com>
+ *
+ *   GPLv2
+ */
+
+
+#include <stdio.h>
+#include <getopt.h>
+#include <string.h>
+#include <strings.h>
+#include <errno.h>
+
+#include <openssl/rsa.h>
+#include <openssl/pem.h>
+#include <openssl/evp.h>
+#include <openssl/err.h>
+#include <openssl/rand.h>
+
+#include <tss2/tss.h>
+#include <tss2/tssutils.h>
+#include <tss2/tssmarshal.h>
+
+#include "tpm2-asn.h"
+#include "tpm2-common.h"
+
+static struct option long_options[] = {
+	{"auth", 0, 0, 'a'},
+	{"help", 0, 0, 'h'},
+	{"key-size", 1, 0, 's'},
+	{"name-scheme", 1, 0, 'n'},
+	{"parent-handle", 1, 0, 'p'},
+	{"wrap", 1, 0, 'w'},
+	{0, 0, 0, 0}
+};
+
+static TPM_ALG_ID name_alg = TPM_ALG_SHA256;
+static int name_alg_size = SHA256_DIGEST_SIZE;
+
+void
+usage(char *argv0)
+{
+	fprintf(stderr, "\t%s: create a TPM key and write it to disk\n"
+		"\tusage: %s [options] <filename>\n\n"
+		"\tOptions:\n"
+		"\t\t-a|--auth          require a password for the key [NO]\n"
+		"\t\t-h|--help          print this help message\n"
+		"\t\t-s|--key-size      key size in bits [2048]\n"
+		"\t\t-n|--name-scheme   name algorithm to use sha1 [sha256] sha384 sha512\n"
+		"\t\t-p|--parent-handle persistent handle of parent key\n"
+		"\t\t-w|--wrap [file]   wrap an existing openssl PEM key\n"
+		"\nReport bugs to %s\n",
+		argv0, argv0, PACKAGE_BUGREPORT);
+	exit(-1);
+}
+
+void
+openssl_print_errors()
+{
+	ERR_load_ERR_strings();
+	ERR_load_crypto_strings();
+	ERR_print_errors_fp(stderr);
+}
+
+int
+openssl_write_tpmfile(const char *file, BYTE *pubkey, int pubkey_len,
+		      BYTE *privkey, int privkey_len, int empty_auth,
+		      TPM_HANDLE parent)
+{
+	TSSLOADABLE tssl;
+	BIO *outb;
+
+	/* clear structure so as not to have to set optional parameters */
+	memset(&tssl, 0, sizeof(tssl));
+	if ((outb = BIO_new_file(file, "w")) == NULL) {
+                fprintf(stderr, "Error opening file for write: %s\n", file);
+		return 1;
+	}
+	tssl.type = OBJ_txt2obj(OID_loadableKey, 1);
+	tssl.emptyAuth = empty_auth;
+	if ((parent & 0xff000000) == 0x81000000) {
+		tssl.parent = ASN1_INTEGER_new();
+		ASN1_INTEGER_set(tssl.parent, parent);
+	}
+	tssl.pubkey = ASN1_OCTET_STRING_new();
+	ASN1_STRING_set(tssl.pubkey, pubkey, pubkey_len);
+	tssl.privkey = ASN1_OCTET_STRING_new();
+	ASN1_STRING_set(tssl.privkey, privkey, privkey_len);
+
+	PEM_write_bio_TSSLOADABLE(outb, &tssl);
+	BIO_free(outb);
+	return 0;
+}
+
+EVP_PKEY *
+openssl_read_key(char *filename)
+{
+        BIO *b = NULL;
+	EVP_PKEY *pkey;
+
+        b = BIO_new_file(filename, "r");
+        if (b == NULL) {
+                fprintf(stderr, "Error opening file for read: %s\n", filename);
+                return NULL;
+        }
+
+        if ((pkey = PEM_read_bio_PrivateKey(b, NULL, PEM_def_callback, NULL)) == NULL) {
+                fprintf(stderr, "Reading key %s from disk failed.\n", filename);
+                openssl_print_errors();
+        }
+	BIO_free(b);
+
+        return pkey;
+}
+
+void tpm2_public_template_rsa(TPMT_PUBLIC *pub)
+{
+	pub->type = TPM_ALG_RSA;
+	pub->nameAlg = name_alg;
+	/* note: all our keys are decrypt only.  This is because
+	 * we use the TPM2_RSA_Decrypt operation for both signing
+	 * and decryption (see e_tpm2.c for details) */
+	pub->objectAttributes.val = TPMA_OBJECT_NODA |
+		TPMA_OBJECT_DECRYPT |
+		TPMA_OBJECT_USERWITHAUTH;
+	pub->authPolicy.t.size = 0;
+	pub->parameters.rsaDetail.symmetric.algorithm = TPM_ALG_NULL;
+	pub->parameters.rsaDetail.scheme.scheme = TPM_ALG_NULL;
+}
+
+TPM_RC openssl_to_tpm_public_rsa(TPMT_PUBLIC *pub, EVP_PKEY *pkey)
+{
+	RSA *rsa = EVP_PKEY_get1_RSA(pkey);
+	BIGNUM *n, *e;
+	int size = RSA_size(rsa);
+	unsigned long exp;
+
+	if (size > MAX_RSA_KEY_BYTES)
+		return TPM_RC_KEY_SIZE;
+
+#if OPENSSL_VERSION_NUMBER < 0x10100000
+	n = rsa->n;
+	e = rsa->e;
+#else
+	RSA_get0_key(&n, &e, NULL);
+#endif
+	exp = BN_get_word(e);
+	/* TPM limitations means exponents must be under a word in size */
+	if (exp == 0xffffffffL)
+		return TPM_RC_KEY_SIZE;
+	tpm2_public_template_rsa(pub);
+	pub->parameters.rsaDetail.keyBits = size*8;
+	if (exp == 0x10001)
+		pub->parameters.rsaDetail.exponent = 0;
+	else
+		pub->parameters.rsaDetail.exponent = exp;
+
+	pub->unique.rsa.t.size = BN_bn2bin(n, pub->unique.rsa.t.buffer);
+
+	return 0;
+}
+
+TPM_RC openssl_to_tpm_public(TPM2B_PUBLIC *pub, EVP_PKEY *pkey)
+{
+	TPMT_PUBLIC *tpub = &pub->publicArea;
+	pub->size = sizeof(*pub);
+
+	switch (EVP_PKEY_type(pkey->type)) {
+	case EVP_PKEY_RSA:
+		return openssl_to_tpm_public_rsa(tpub, pkey);
+	default:
+		break;
+	}
+	return TPM_RC_ASYMMETRIC;
+}
+
+TPM_RC openssl_to_tpm_private_rsa(TPMT_SENSITIVE *s, EVP_PKEY *pkey)
+{
+	BIGNUM *q;
+	TPM2B_PRIVATE_KEY_RSA *t2brsa = &s->sensitive.rsa;
+	RSA *rsa = EVP_PKEY_get1_RSA(pkey);
+
+#if OPENSSL_VERSION_NUMBER < 0x10100000
+	q = rsa->q;
+#else
+	BIGNUM *p;
+
+	RSA_get0_factors(rsa, &p, &q);
+#endif
+
+	if (!q)
+		return TPM_RC_ASYMMETRIC;
+
+	s->sensitiveType = TPM_ALG_RSA;
+	s->seedValue.b.size = 0;
+
+	t2brsa->t.size = BN_bn2bin(q, t2brsa->t.buffer);
+	return 0;
+}
+
+TPM_RC openssl_to_tpm_private(TPMT_SENSITIVE *priv, EVP_PKEY *pkey)
+{
+	switch (EVP_PKEY_type(pkey->type)) {
+	case EVP_PKEY_RSA:
+		return openssl_to_tpm_private_rsa(priv, pkey);
+	default:
+		break;
+	}
+	return TPM_RC_ASYMMETRIC;
+}
+
+TPM_RC wrap_key(TPM2B_PRIVATE *priv, const char *password, EVP_PKEY *pkey)
+{
+	TPMT_SENSITIVE s;
+	TPM2B_SENSITIVE b;
+	BYTE *buf;
+	int32_t size;
+	TPM_RC rc;
+
+	memset(&b, 0, sizeof(b));
+	memset(&s, 0, sizeof(s));
+
+	openssl_to_tpm_private(&s, pkey);
+
+	if (password) {
+		int len = strlen(password);
+
+		memcpy(s.authValue.b.buffer, password, len);
+		s.authValue.b.size = len;
+	} else {
+		s.authValue.b.size = 0;
+	}
+	size = sizeof(s);
+	buf = b.b.buffer;
+	rc = TSS_TPMT_SENSITIVE_Marshal(&s, &b.b.size, &buf, &size);
+	if (rc)
+		tpm2_error(rc, "TSS_TPMT_SENSITIVE_Marshal");
+
+	size = sizeof(*priv);
+	buf = priv->b.buffer;
+	priv->b.size = 0;
+	/* no encryption means innerIntegrity and outerIntegrity are
+	* absent, so the TPM2B_PRIVATE is a TPMT_SENSITIVE*/
+	rc = TSS_TPM2B_PRIVATE_Marshal((TPM2B_PRIVATE *)&b, &priv->b.size, &buf, &size);
+	if (rc)
+		tpm2_error(rc, "TSS_TPM2B_PRIVATE_Marshal");
+
+	return TPM_RC_ASYMMETRIC;
+}
+
+int main(int argc, char **argv)
+{
+	char *filename, c, *wrap = NULL, *auth = NULL;
+	int option_index;
+	const char *reason;
+	TSS_CONTEXT *tssContext = NULL;
+	TPM_HANDLE parent = 0;
+	TPM_RC rc = 0;
+	BYTE pubkey[sizeof(TPM2B_PUBLIC)],privkey[sizeof(TPM2B_PRIVATE)], *buffer;
+	uint16_t pubkey_len, privkey_len;
+	int32_t size, key_size = 0;
+	TPM2B_PUBLIC *pub;
+	TPM2B_PRIVATE *priv;
+
+
+	while (1) {
+		option_index = 0;
+		c = getopt_long(argc, argv, "n:s:ap:hw:",
+				long_options, &option_index);
+		if (c == -1)
+			break;
+
+		switch (c) {
+			case 'a':
+				auth = malloc(128);
+				break;
+			case 'h':
+				usage(argv[0]);
+				break;
+			case 'n':
+				if (!strcasecmp("sha1", optarg)) {
+					name_alg = TPM_ALG_SHA1;
+					name_alg_size = SHA1_DIGEST_SIZE;
+				} else if (strcasecmp("sha256", optarg)) {
+					/* default, do nothing */
+				} else if (strcasecmp("sha384", optarg)) {
+					name_alg = TPM_ALG_SHA384;
+					name_alg_size = SHA384_DIGEST_SIZE;
+#ifdef TPM_ALG_SHA512
+				} else if (strcasecmp("sha512", optarg)) {
+					name_alg = TPM_ALG_SHA512;
+					name_alg_size = SHA512_DIGEST_SIZE;
+#endif
+				} else {
+					usage(argv[0]);
+				}
+				break;
+			case 'p':
+				parent = strtol(optarg, NULL, 16);
+				break;
+			case's':
+				key_size = atoi(optarg);
+				break;
+			case 'w':
+				wrap = optarg;
+				break;
+			default:
+				usage(argv[0]);
+				break;
+		}
+	}
+
+	filename = argv[argc - 1];
+
+	if (argc < 2)
+		usage(argv[0]);
+
+	if (key_size && wrap) {
+		fprintf(stderr, "key-size and wrap are mutually exclusive\n");
+		usage(argv[0]);
+	} else if (!key_size && !wrap) {
+		/* for internal create, use default key size */
+		key_size = 2048;
+	}
+
+	if (parent && (parent & 0xff000000) != 0x81000000) {
+		fprintf(stderr, "you must specify a persistent parent handle\n");
+		usage(argv[0]);
+	}
+
+	if (auth) {
+		if (EVP_read_pw_string(auth, 128, "Enter TPM key authority: ", 1)) {
+			fprintf(stderr, "Passwords do not match\n");
+			exit(1);
+		}
+	}
+
+	rc = TSS_Create(&tssContext);
+	if (rc) {
+		reason = "TSS_Create";
+		goto out_err;
+	}
+
+	if (parent == 0) {
+		rc = tpm2_load_srk(tssContext, &parent, NULL, NULL);
+		if (rc) {
+			reason = "tpm2_load_srk";
+			goto out_delete;
+		}
+	}
+
+	if (wrap) {
+		Import_In iin;
+		Import_Out iout;
+		EVP_PKEY *pkey;
+
+		/* may be needed to decrypt the key */
+		OpenSSL_add_all_ciphers();
+		pkey = openssl_read_key(wrap);
+		if (!pkey) {
+			reason = "unable to read key";
+			goto out_delete;
+		}
+
+		iin.parentHandle = parent;
+		iin.encryptionKey.t.size = 0;
+		openssl_to_tpm_public(&iin.objectPublic, pkey);
+		/* set random iin.symSeed */
+		iin.inSymSeed.t.size = 0;
+		iin.symmetricAlg.algorithm = TPM_ALG_NULL;
+		wrap_key(&iin.duplicate, auth, pkey);
+		openssl_to_tpm_public(&iin.objectPublic, pkey);
+		rc = TSS_Execute(tssContext,
+				 (RESPONSE_PARAMETERS *)&iout,
+				 (COMMAND_PARAMETERS *)&iin,
+				 NULL,
+				 TPM_CC_Import,
+				 TPM_RS_PW, NULL, 0,
+				 TPM_RH_NULL, NULL, 0,
+				 TPM_RH_NULL, NULL, 0,
+				 TPM_RH_NULL, NULL, 0);
+		if (rc) {
+			reason = "TPM2_Import";
+			goto out_flush;
+		}
+		pub = &iin.objectPublic;
+		priv = &iout.outPrivate;
+	} else {
+		/* create a TPM resident key */
+		Create_In cin;
+		Create_Out cout;
+
+		tpm2_public_template_rsa(&cin.inPublic.publicArea);
+		cin.inPublic.publicArea.objectAttributes.val |=
+			TPMA_OBJECT_SENSITIVEDATAORIGIN;
+		if (auth) {
+			int len = strlen(auth);
+			memcpy(&cin.inSensitive.sensitive.userAuth.b,
+			       auth, len);
+			cin.inSensitive.sensitive.userAuth.b.size = len;
+		} else {
+			cin.inSensitive.sensitive.userAuth.b.size = 0;
+		}
+		cin.inSensitive.sensitive.data.t.size = 0;
+		cin.parentHandle = parent;
+		cin.outsideInfo.t.size = 0;
+		cin.creationPCR.count = 0;
+		cin.inPublic.publicArea.parameters.rsaDetail.keyBits = key_size;
+		cin.inPublic.publicArea.parameters.rsaDetail.exponent = 0;
+		cin.inPublic.publicArea.unique.rsa.t.size = 0;
+
+		rc = TSS_Execute(tssContext,
+				 (RESPONSE_PARAMETERS *)&cout,
+				 (COMMAND_PARAMETERS *)&cin,
+				 NULL,
+				 TPM_CC_Create,
+				 TPM_RS_PW, NULL, 0,
+				 TPM_RH_NULL, NULL, 0,
+				 TPM_RH_NULL, NULL, 0,
+				 TPM_RH_NULL, NULL, 0);
+		if (rc) {
+			reason = "TPM2_Create";
+			goto out_flush;
+		}
+
+		pub = &cout.outPublic;
+		priv = &cout.outPrivate;
+	}
+	tpm2_flush_srk(tssContext);
+	buffer = pubkey;
+	pubkey_len = 0;
+	size = sizeof(pubkey);
+	TSS_TPM2B_PUBLIC_Marshal(pub, &pubkey_len, &buffer, &size);
+	buffer = privkey;
+	privkey_len = 0;
+	size = sizeof(privkey);
+	TSS_TPM2B_PRIVATE_Marshal(priv, &privkey_len, &buffer, &size);
+	openssl_write_tpmfile(filename, pubkey, pubkey_len, privkey, privkey_len, auth == NULL, parent);
+	TSS_Delete(tssContext);
+	exit(0);
+
+ out_flush:
+	tpm2_flush_srk(tssContext);
+ out_delete:
+	TSS_Delete(tssContext);
+ out_err:
+	tpm2_error(rc, reason);
+
+	exit(1);
+}
diff --git a/e_tpm2.c b/e_tpm2.c
new file mode 100644
index 0000000..18e94fe
--- /dev/null
+++ b/e_tpm2.c
@@ -0,0 +1,559 @@
+
+/*
+ * Copyright (C) 2016 James.Bottomley@HansenPartnership.com
+ *
+ * GPLv2
+ *
+ */
+
+#include <stdio.h>
+#include <string.h>
+
+#include <openssl/crypto.h>
+#include <openssl/dso.h>
+#include <openssl/engine.h>
+#include <openssl/evp.h>
+#include <openssl/objects.h>
+#include <openssl/sha.h>
+#include <openssl/bn.h>
+#include <openssl/pem.h>
+#include <openssl/x509.h>
+
+#include <tss2/tss.h>
+#include <tss2/tssutils.h>
+#include <tss2/tssmarshal.h>
+#include <tss2/tssresponsecode.h>
+#include <tss2/Unmarshal_fp.h>
+
+#include "tpm2-asn.h"
+#include "tpm2-common.h"
+
+#define TPM2_ENGINE_EX_DATA_UNINIT		-1
+
+/* structure pointed to by the RSA object's app_data pointer */
+struct rsa_app_data
+{
+	TSS_CONTEXT *tssContext;
+	TPM_HANDLE parent;
+	TPM_HANDLE key;
+	char *auth;
+};
+
+static char *srk_auth;
+
+static int tpm2_engine_init(ENGINE * e)
+{
+	return 1;
+}
+
+static int tpm2_engine_finish(ENGINE * e)
+{
+	return 1;
+}
+
+static int tpm2_create_srk_policy(char *secret)
+{
+	int len;
+
+	if (!secret) {
+		OPENSSL_free(srk_auth);
+		srk_auth = NULL;
+	} else {
+		len = strlen(secret);
+		srk_auth = OPENSSL_malloc(len);
+		strcpy(srk_auth, secret);
+	}
+	return 1;
+}
+
+#define TPM_CMD_PIN ENGINE_CMD_BASE
+
+static int tpm2_engine_ctrl(ENGINE * e, int cmd, long i, void *p, void (*f) ())
+{
+	switch (cmd) {
+		case TPM_CMD_PIN:
+			return tpm2_create_srk_policy(p);
+		default:
+			break;
+	}
+	fprintf(stderr, "tpm2: engine command not implemented\n");
+
+	return 0;
+}
+
+
+#ifndef OPENSSL_NO_RSA
+/* rsa functions */
+static int tpm2_rsa_init(RSA *rsa);
+static int tpm2_rsa_finish(RSA *rsa);
+static int tpm2_rsa_pub_dec(int, const unsigned char *, unsigned char *, RSA *, int);
+static int tpm2_rsa_pub_enc(int, const unsigned char *, unsigned char *, RSA *, int);
+static int tpm2_rsa_priv_dec(int, const unsigned char *, unsigned char *, RSA *, int);
+static int tpm2_rsa_priv_enc(int, const unsigned char *, unsigned char *, RSA *, int);
+//static int tpm2_rsa_sign(int, const unsigned char *, unsigned int, unsigned char *, unsigned int *, const RSA *);
+#endif
+
+/* The definitions for control commands specific to this engine */
+#define TPM2_CMD_PIN		ENGINE_CMD_BASE
+static const ENGINE_CMD_DEFN tpm2_cmd_defns[] = {
+	{TPM2_CMD_PIN,
+	 "PIN",
+	 "Specifies the secret for the SRK (default is plaintext, else set SECRET_MODE)",
+	 ENGINE_CMD_FLAG_STRING},
+	/* end */
+	{0, NULL, NULL, 0}
+};
+
+#ifndef OPENSSL_NO_RSA
+static RSA_METHOD tpm2_rsa = {
+	"TPM2 RSA method",
+	tpm2_rsa_pub_enc,
+	tpm2_rsa_pub_dec,
+	tpm2_rsa_priv_enc,
+	tpm2_rsa_priv_dec,
+	NULL, /* set in tpm2_engine_init */
+	BN_mod_exp_mont,
+	tpm2_rsa_init,
+	tpm2_rsa_finish,
+	(RSA_FLAG_SIGN_VER | RSA_FLAG_NO_BLINDING),
+	NULL,
+	NULL, /* sign */
+	NULL, /* verify */
+	NULL, /* keygen */
+};
+#endif
+
+/* varibles used to get/set CRYPTO_EX_DATA values */
+static int ex_app_data = TPM2_ENGINE_EX_DATA_UNINIT;
+
+static TPM_HANDLE tpm2_load_key_from_rsa(RSA *rsa, TSS_CONTEXT **tssContext, char **auth)
+{
+	struct rsa_app_data *app_data = RSA_get_ex_data(rsa, ex_app_data);
+
+	if (!app_data)
+		return 0;
+
+	*auth = app_data->auth;
+	*tssContext = app_data->tssContext;
+
+	return app_data->key;
+}
+
+static char *tpm2_get_auth(UI_METHOD *ui_method, char *prompt, void *cb_data)
+{
+	UI *ui = UI_new();
+	/* Max auth size is name algorithm hash length, so this
+	 * is way bigger than necessary */
+	char auth[256], *ret = NULL;
+	int len;
+
+	if (ui_method)
+		UI_set_method(ui, ui_method);
+
+	UI_add_user_data(ui, cb_data);
+
+	if (UI_add_input_string(ui, prompt, 0, auth, 0, sizeof(auth)) == 0) {
+		fprintf(stderr, "UI_add_input_string failed\n");
+		goto out;
+	}
+
+	if (UI_process(ui)) {
+		fprintf(stderr, "UI_process failed\n");
+		goto out;
+	}
+
+	len = strlen(auth);
+	ret = OPENSSL_malloc(len + 1);
+	if (!ret)
+		goto out;
+
+	strcpy(ret, auth);
+
+ out:
+	UI_free(ui);
+
+	return ret;
+}
+
+static EVP_PKEY *tpm2_engine_load_key(ENGINE *e, const char *key_id,
+				      UI_METHOD *ui, void *cb_data)
+{
+	Load_In in;
+	Load_Out out;
+	TSS_CONTEXT *tssContext;
+	TPM_RC rc;
+	EVP_PKEY *pkey;
+	RSA *rsa;
+	BIO *bf;
+	TSSLOADABLE *tssl;
+	BYTE *buffer;
+	INT32 size;
+	struct rsa_app_data *app_data;
+	char oid[128];
+	int empty_auth;
+
+	if (!key_id) {
+		fprintf(stderr, "key_id is NULL\n");
+		return NULL;
+	}
+
+	bf = BIO_new_file(key_id, "r");
+	if (!bf) {
+		fprintf(stderr, "File %s does not exist or cannot be read\n", key_id);
+		return NULL;
+	}
+
+	tssl = PEM_read_bio_TSSLOADABLE(bf, NULL, NULL, NULL);
+
+	BIO_free(bf);
+
+	if (!tssl) {
+		fprintf(stderr, "Failed to parse file %s\n", key_id);
+		return NULL;
+	}
+
+	if (OBJ_obj2txt(oid, sizeof(oid), tssl->type, 1) == 0) {
+		fprintf(stderr, "Failed to parse object type\n");
+		goto err;
+	}
+
+	if (strcmp(OID_loadableKey, oid) == 0) {
+		;
+	} else if (strcmp(OID_12Key, oid) == 0) {
+		fprintf(stderr, "TPM1.2 key is not importable by TPM2.0\n");
+		goto err;
+	} else if (strcmp(OID_importableKey, oid) == 0) {
+		fprintf(stderr, "Importable keys currently unsupported\n");
+		goto err;
+	} else {
+		fprintf(stderr, "Unrecognised object type\n");
+		goto err;
+	}
+
+	app_data = OPENSSL_malloc(sizeof(struct rsa_app_data));
+
+	if (!app_data) {
+		fprintf(stderr, "Failed to allocate app_data\n");
+		goto err;
+	}
+
+	rc = TSS_Create(&tssContext);
+	if (rc) {
+		tpm2_error(rc, "TSS_Create");
+		goto err_free;
+	}
+
+	app_data->tssContext = tssContext;
+
+	app_data->parent = 0;
+	if (tssl->parent)
+		app_data->parent = ASN1_INTEGER_get(tssl->parent);
+
+	if (app_data->parent)
+		in.parentHandle = app_data->parent;
+	else
+		tpm2_load_srk(tssContext, &in.parentHandle, srk_auth, NULL);
+
+	empty_auth = tssl->emptyAuth;
+
+	buffer = tssl->privkey->data;
+	size = tssl->privkey->length;
+	TPM2B_PRIVATE_Unmarshal(&in.inPrivate, &buffer, &size);
+	buffer = tssl->pubkey->data;
+	size = tssl->pubkey->length;
+	TPM2B_PUBLIC_Unmarshal(&in.inPublic, &buffer, &size, FALSE);
+
+	/* create the new objects to return */
+	pkey = tpm2_to_openssl_public(&in.inPublic.publicArea);
+	if (!pkey) {
+		fprintf(stderr, "Failed to allocate a new EVP_KEY\n");
+		goto err_free_del;
+	}
+
+	rc = TSS_Execute(tssContext,
+			 (RESPONSE_PARAMETERS *)&out,
+			 (COMMAND_PARAMETERS *)&in,
+			 NULL,
+			 TPM_CC_Load,
+			 TPM_RS_PW, NULL, 0,
+			 TPM_RH_NULL, NULL, 0,
+			 TPM_RH_NULL, NULL, 0,
+			 TPM_RH_NULL, NULL, 0);
+
+	if (rc) {
+		tpm2_error(rc, "TPM2_Load");
+		goto err_free_key;
+	}
+
+	app_data->key = out.objectHandle;
+
+	app_data->auth = NULL;
+	if (empty_auth == 0) {
+		app_data->auth = tpm2_get_auth(ui, "TPM Key Password: ", cb_data);
+		if (!app_data->auth)
+			goto err_unload;
+	}
+
+	TSSLOADABLE_free(tssl);
+	rsa = EVP_PKEY_get1_RSA(pkey);
+	rsa->meth = &tpm2_rsa;
+	/* call our local init function here */
+	rsa->meth->init(rsa);
+
+	RSA_set_ex_data(rsa, ex_app_data, app_data);
+
+	/* release the reference EVP_PKEY_get1_RSA obtained */
+	RSA_free(rsa);
+	return pkey;
+
+ err_unload:
+	tpm2_flush_handle(tssContext, app_data->key);
+ err_free_key:
+	EVP_PKEY_free(pkey);
+ err_free_del:
+	TSS_Delete(tssContext);
+ err_free:
+	OPENSSL_free(app_data);
+	tpm2_flush_srk(tssContext);
+ err:
+	TSSLOADABLE_free(tssl);
+
+	return NULL;
+}
+
+/* Constants used when creating the ENGINE */
+static const char *engine_tpm2_id = "tpm2";
+static const char *engine_tpm2_name = "TPM2 hardware engine support";
+
+/* This internal function is used by ENGINE_tpm() and possibly by the
+ * "dynamic" ENGINE support too */
+static int tpm2_bind_helper(ENGINE * e)
+{
+	if (!ENGINE_set_id(e, engine_tpm2_id) ||
+	    !ENGINE_set_name(e, engine_tpm2_name) ||
+#ifndef OPENSSL_NO_RSA
+	    !ENGINE_set_RSA(e, &tpm2_rsa) ||
+#endif
+	    !ENGINE_set_init_function(e, tpm2_engine_init) ||
+	    !ENGINE_set_finish_function(e, tpm2_engine_finish) ||
+	    !ENGINE_set_ctrl_function(e, tpm2_engine_ctrl) ||
+	    !ENGINE_set_load_pubkey_function(e, tpm2_engine_load_key) ||
+	    !ENGINE_set_load_privkey_function(e, tpm2_engine_load_key) ||
+	    !ENGINE_set_cmd_defns(e, tpm2_cmd_defns))
+		return 0;
+
+	return 1;
+}
+
+
+#ifndef OPENSSL_NO_RSA
+static int tpm2_rsa_init(RSA *rsa)
+{
+	if (ex_app_data == TPM2_ENGINE_EX_DATA_UNINIT)
+		ex_app_data = RSA_get_ex_new_index(0, NULL, NULL, NULL, NULL);
+
+	if (ex_app_data == TPM2_ENGINE_EX_DATA_UNINIT) {
+		fprintf(stderr, "Failed to get memory for external data\n");
+		return 0;
+	}
+
+	return 1;
+}
+
+static int tpm2_rsa_finish(RSA *rsa)
+{
+	struct rsa_app_data *app_data = RSA_get_ex_data(rsa, ex_app_data);
+	TSS_CONTEXT *tssContext;
+
+	if (!app_data)
+		return 1;
+
+	tssContext = app_data->tssContext;
+
+	tpm2_flush_handle(tssContext, app_data->key);
+	if (app_data->parent == 0)
+		tpm2_flush_srk(tssContext);
+
+	OPENSSL_free(app_data);
+
+	TSS_Delete(tssContext);
+
+	return 1;
+}
+
+static int tpm2_rsa_pub_dec(int flen,
+			   const unsigned char *from,
+			   unsigned char *to,
+			   RSA *rsa,
+			   int padding)
+{
+	int rv;
+
+	rv = RSA_PKCS1_SSLeay()->rsa_pub_dec(flen, from, to, rsa,
+					     padding);
+	if (rv < 0) {
+		fprintf(stderr, "rsa_pub_dec failed\n");
+		return 0;
+	}
+
+	return rv;
+}
+
+static int tpm2_rsa_priv_dec(int flen,
+			    const unsigned char *from,
+			    unsigned char *to,
+			    RSA *rsa,
+			    int padding)
+{
+	TPM_RC rc;
+	int rv;
+	RSA_Decrypt_In in;
+	RSA_Decrypt_Out out;
+	TSS_CONTEXT *tssContext;
+	char *auth;
+
+	in.keyHandle = tpm2_load_key_from_rsa(rsa, &tssContext, &auth);
+
+	if (in.keyHandle == 0) {
+		rv = RSA_PKCS1_SSLeay()->rsa_priv_dec(flen, from, to, rsa,
+						      padding);
+		if (rv < 0)
+			fprintf(stderr, "rsa_priv_dec failed\n");
+
+		return rv;
+	}
+
+	rv = -1;
+	if (padding != RSA_PKCS1_PADDING) {
+		fprintf(stderr, "Non PKCS1 padding asked for\n");
+		return rv;
+	}
+
+	in.inScheme.scheme = TPM_ALG_RSAES;
+	in.cipherText.t.size = flen;
+	memcpy(in.cipherText.t.buffer, from, flen);
+	in.label.t.size = 0;
+
+	rc = TSS_Execute(tssContext,
+			 (RESPONSE_PARAMETERS *)&out,
+			 (COMMAND_PARAMETERS *)&in,
+			 NULL,
+			 TPM_CC_RSA_Decrypt,
+			 TPM_RS_PW, auth, 0,
+			 TPM_RH_NULL, NULL, 0,
+			 TPM_RH_NULL, NULL, 0,
+			 TPM_RH_NULL, NULL, 0);
+	if (rc) {
+		tpm2_error(rc, "TPM2_RSA_Decrypt");
+		return rv;
+	}
+
+	memcpy(to, out.message.t.buffer,
+	       out.message.t.size);
+
+	rv = out.message.t.size;
+
+	return rv;
+}
+
+static int tpm2_rsa_pub_enc(int flen,
+			   const unsigned char *from,
+			   unsigned char *to,
+			   RSA *rsa,
+			   int padding)
+{
+	int rv;
+
+	rv = RSA_PKCS1_SSLeay()->rsa_pub_enc(flen, from, to, rsa,
+					     padding);
+	if (rv < 0)
+		fprintf(stderr, "rsa_pub_enc failed\n");
+
+	return rv;
+}
+
+static int tpm2_rsa_priv_enc(int flen,
+			    const unsigned char *from,
+			    unsigned char *to,
+			    RSA *rsa,
+			    int padding)
+{
+	TPM_RC rc;
+	int rv, size;
+	RSA_Decrypt_In in;
+	RSA_Decrypt_Out out;
+	TSS_CONTEXT *tssContext;
+	char *auth;
+
+	in.keyHandle = tpm2_load_key_from_rsa(rsa, &tssContext, &auth);
+
+	if (in.keyHandle == 0) {
+		rv = RSA_PKCS1_SSLeay()->rsa_priv_enc(flen, from, to, rsa,
+						      padding);
+		if (rv < 0)
+			fprintf(stderr, "pass through signing failed\n");
+
+		return rv;
+	}
+
+	rv = -1;
+	if (padding != RSA_PKCS1_PADDING) {
+		fprintf(stderr, "Non PKCS1 padding asked for\n");
+		return rv;
+	}
+
+	/* this is slightly paradoxical that we're doing a Decrypt
+	 * operation: the only material difference between decrypt and
+	 * encrypt is where the padding is applied or checked, so if
+	 * you apply your own padding up to the RSA block size and use
+	 * TPM_ALG_NULL, which means no padding check, a decrypt
+	 * operation effectively becomes an encrypt */
+	size = RSA_size(rsa);
+	in.inScheme.scheme = TPM_ALG_NULL;
+	in.cipherText.t.size = size;
+	RSA_padding_add_PKCS1_type_1(in.cipherText.t.buffer, size, from, flen);
+	in.label.t.size = 0;
+
+	rc = TSS_Execute(tssContext,
+			 (RESPONSE_PARAMETERS *)&out,
+			 (COMMAND_PARAMETERS *)&in,
+			 NULL,
+			 TPM_CC_RSA_Decrypt,
+			 TPM_RS_PW, auth, 0,
+			 TPM_RH_NULL, NULL, 0,
+			 TPM_RH_NULL, NULL, 0,
+			 TPM_RH_NULL, NULL, 0);
+
+	if (rc) {
+		tpm2_error(rc, "TPM2_RSA_Decrypt");
+		return rv;
+	}
+
+	memcpy(to, out.message.t.buffer,
+	       out.message.t.size);
+
+	rv = out.message.t.size;
+
+	return rv;
+}
+
+#endif
+
+/* This stuff is needed if this ENGINE is being compiled into a self-contained
+ * shared-library. */
+static int tpm2_bind_fn(ENGINE * e, const char *id)
+{
+	if (id && (strcmp(id, engine_tpm2_id) != 0)) {
+		fprintf(stderr, "Called for id %s != my id %s\n",
+		       id, engine_tpm2_id);
+		return 0;
+	}
+	if (!tpm2_bind_helper(e)) {
+		fprintf(stderr, "tpm2_bind_helper failed\n");
+		return 0;
+	}
+	return 1;
+}
+
+IMPLEMENT_DYNAMIC_CHECK_FN()
+IMPLEMENT_DYNAMIC_BIND_FN(tpm2_bind_fn)
diff --git a/tpm2-asn.h b/tpm2-asn.h
new file mode 100644
index 0000000..2a08e3a
--- /dev/null
+++ b/tpm2-asn.h
@@ -0,0 +1,59 @@
+/* Copyright (C) 2016 James Bottomley <James.Bottomley@HansenPartnership.com>
+ *
+ * GPLv2
+ */
+#ifndef _TPM2_ASN_H
+#define _TPM2_ASN_H
+
+#include <openssl/asn1t.h>
+
+/*
+ * Define the format of a TPM key file.  The current format covers
+ * both TPM1.2 keys as well as symmetrically encrypted private keys
+ * produced by TSS2_Import and the TPM2 format public key which
+ * contains things like the policy but which is cryptographically tied
+ * to the private key.
+ *
+ * TPMKey ::= SEQUENCE {
+ *	type		OBJECT IDENTIFIER
+ *	emptyAuth	[0] EXPLICIT BOOLEAN OPTIONAL
+ *	parent		[1] EXPLICIT INTEGER OPTIONAL
+ *	pubkey		[2] EXPLICIT OCTET STRING OPTIONAL
+ *	privkey		OCTET STRING
+ * }
+ */
+
+typedef struct {
+	ASN1_OBJECT *type;
+	ASN1_BOOLEAN emptyAuth;
+	ASN1_INTEGER *parent;
+	ASN1_OCTET_STRING *pubkey;
+	ASN1_OCTET_STRING *privkey;
+} TSSLOADABLE;
+
+/* the two type oids are in the TCG namespace 2.23.133; we choose an
+ *  unoccupied child (10) for keytype file and two values:
+ *    1 : Key that is directly loadable
+ *    2 : Key that must first be imported then loaded
+ */
+#define OID_12Key		"2.23.133.10.1"
+#define OID_loadableKey		"2.23.133.10.2"
+#define OID_importableKey	"2.23.133.10.3"
+
+ASN1_SEQUENCE(TSSLOADABLE) = {
+	ASN1_SIMPLE(TSSLOADABLE, type, ASN1_OBJECT),
+	ASN1_EXP_OPT(TSSLOADABLE, emptyAuth, ASN1_BOOLEAN, 0),
+	ASN1_EXP_OPT(TSSLOADABLE, parent, ASN1_INTEGER, 1),
+	ASN1_EXP_OPT(TSSLOADABLE, pubkey, ASN1_OCTET_STRING, 2),
+	ASN1_SIMPLE(TSSLOADABLE, privkey, ASN1_OCTET_STRING)
+} ASN1_SEQUENCE_END(TSSLOADABLE)
+
+IMPLEMENT_ASN1_FUNCTIONS(TSSLOADABLE);
+
+/* This is the PEM guard tag */
+#define TSSLOADABLE_PEM_STRING "TSS2 KEY BLOB"
+
+static IMPLEMENT_PEM_write_bio(TSSLOADABLE, TSSLOADABLE, TSSLOADABLE_PEM_STRING, TSSLOADABLE)
+static IMPLEMENT_PEM_read_bio(TSSLOADABLE, TSSLOADABLE, TSSLOADABLE_PEM_STRING, TSSLOADABLE)
+
+#endif
diff --git a/tpm2-common.c b/tpm2-common.c
new file mode 100644
index 0000000..1b6569c
--- /dev/null
+++ b/tpm2-common.c
@@ -0,0 +1,175 @@
+/*
+ * Copyright (C) 2016 James Bottomley <James.Bottomley@HansenPartnership.com>
+ *
+ * GPLv2
+ */
+
+#include <stdio.h>
+
+#include <openssl/evp.h>
+#include <openssl/rsa.h>
+
+#include <tss2/tss.h>
+#include <tss2/tssresponsecode.h>
+
+#include "tpm2-common.h"
+
+void tpm2_error(TPM_RC rc, const char *reason)
+{
+	const char *msg, *submsg, *num;
+
+	fprintf(stderr, "%s failed with %d\n", reason, rc);
+	TSS_ResponseCode_toString(&msg, &submsg, &num, rc);
+	fprintf(stderr, "%s%s%s\n", msg, submsg, num);
+}
+
+
+static TPM_HANDLE hSRK = 0;
+
+TPM_RC tpm2_load_srk(TSS_CONTEXT *tssContext, TPM_HANDLE *h, const char *auth,TPM2B_PUBLIC *pub)
+{
+	static TPM2B_PUBLIC srk_pub;
+	TPM_RC rc;
+	CreatePrimary_In in;
+	CreatePrimary_Out out;
+
+	if (hSRK)
+		goto out;
+
+	/* SPS owner */
+	in.primaryHandle = TPM_RH_OWNER;
+	/* assume no owner password */
+	in.inSensitive.sensitive.userAuth.t.size = 0;
+	/* no sensitive date for storage keys */
+	in.inSensitive.sensitive.data.t.size = 0;
+	/* no outside info */
+	in.outsideInfo.t.size = 0;
+	/* no PCR state */
+	in.creationPCR.count = 0;
+
+	/* public parameters for an RSA2048 key  */
+	in.inPublic.publicArea.type = TPM_ALG_RSA;
+	in.inPublic.publicArea.nameAlg = TPM_ALG_SHA256;
+	in.inPublic.publicArea.objectAttributes.val =
+		TPMA_OBJECT_NODA |
+		TPMA_OBJECT_SENSITIVEDATAORIGIN |
+		TPMA_OBJECT_USERWITHAUTH |
+		TPMA_OBJECT_DECRYPT |
+		TPMA_OBJECT_RESTRICTED;
+	in.inPublic.publicArea.parameters.rsaDetail.symmetric.algorithm = TPM_ALG_AES;
+	in.inPublic.publicArea.parameters.rsaDetail.symmetric.keyBits.aes = 128;
+	in.inPublic.publicArea.parameters.rsaDetail.symmetric.mode.aes = TPM_ALG_CFB;
+	in.inPublic.publicArea.parameters.rsaDetail.scheme.scheme = TPM_ALG_NULL;
+	in.inPublic.publicArea.parameters.rsaDetail.keyBits = 2048;
+	/* means conventional 2^16+1 */
+	in.inPublic.publicArea.parameters.rsaDetail.exponent = 0;
+	in.inPublic.publicArea.unique.rsa.t.size = 0;
+	in.inPublic.publicArea.authPolicy.t.size = 0;
+
+	rc = TSS_Execute(tssContext,
+			 (RESPONSE_PARAMETERS *)&out,
+			 (COMMAND_PARAMETERS *)&in,
+			 NULL,
+			 TPM_CC_CreatePrimary,
+			 TPM_RS_PW, NULL, 0,
+			 TPM_RH_NULL, NULL, 0,
+			 TPM_RH_NULL, NULL, 0,
+			 TPM_RH_NULL, NULL, 0);
+
+	if (rc) {
+		tpm2_error(rc, "TSS_CreatePrimarhy");
+		return rc;
+	}
+
+	hSRK = out.objectHandle;
+	srk_pub = out.outPublic;
+ out:
+	*h = hSRK;
+	if (pub)
+		*pub = srk_pub;
+
+	return 0;
+}
+
+void tpm2_flush_srk(TSS_CONTEXT *tssContext)
+{
+	if (hSRK)
+		tpm2_flush_handle(tssContext, hSRK);
+	hSRK = 0;
+}
+
+void tpm2_flush_handle(TSS_CONTEXT *tssContext, TPM_HANDLE h)
+{
+	FlushContext_In in;
+
+	if (!h)
+		return;
+
+	in.flushHandle = h;
+	TSS_Execute(tssContext, NULL,
+		    (COMMAND_PARAMETERS *)&in,
+		    NULL,
+		    TPM_CC_FlushContext,
+		    TPM_RH_NULL, NULL, 0);
+}
+
+static EVP_PKEY *tpm2_to_openssl_public_rsa(TPMT_PUBLIC *pub)
+{
+	RSA *rsa = RSA_new();
+	EVP_PKEY *pkey;
+	unsigned long exp;
+	BIGNUM *n, *e;
+
+	if (!rsa)
+		return NULL;
+	pkey = EVP_PKEY_new();
+	if (!pkey)
+		goto err_free_rsa;
+	e = BN_new();
+	if (!e)
+		goto err_free_pkey;
+	n = BN_new();
+	if (!n)
+		goto err_free_e;
+	if (pub->parameters.rsaDetail.exponent == 0)
+		exp = 0x10001;
+	else
+		exp = pub->parameters.rsaDetail.exponent;
+	if (!BN_set_word(e, exp))
+		goto err_free;
+	if (!BN_bin2bn(pub->unique.rsa.t.buffer, pub->unique.rsa.t.size, n))
+		goto err_free;
+#if OPENSSL_VERSION_NUMBER < 0x10100000
+	rsa->n = n;
+	rsa->e = e;
+#else
+	RSA_set0_key(rsa, n, e, NULL);
+#endif
+	if (!EVP_PKEY_assign_RSA(pkey, rsa))
+		goto err_free;
+
+	return pkey;
+
+ err_free:
+	BN_free(n);
+ err_free_e:
+	BN_free(e);
+ err_free_pkey:
+	EVP_PKEY_free(pkey);
+ err_free_rsa:
+	RSA_free(rsa);
+
+	return NULL;
+}
+
+EVP_PKEY *tpm2_to_openssl_public(TPMT_PUBLIC *pub)
+{
+	switch (pub->type) {
+	case TPM_ALG_RSA:
+		return tpm2_to_openssl_public_rsa(pub);
+	default:
+		break;
+	}
+	return NULL;
+}
+
diff --git a/tpm2-common.h b/tpm2-common.h
new file mode 100644
index 0000000..abd750b
--- /dev/null
+++ b/tpm2-common.h
@@ -0,0 +1,10 @@
+#ifndef _TPM2_COMMON_H
+#define _TPM2_COMMON_H
+
+void tpm2_error(TPM_RC rc, const char *reason);
+TPM_RC tpm2_load_srk(TSS_CONTEXT *tssContext, TPM_HANDLE *h, const char *auth, TPM2B_PUBLIC *pub);
+void tpm2_flush_handle(TSS_CONTEXT *tssContext, TPM_HANDLE h);
+EVP_PKEY *tpm2_to_openssl_public(TPMT_PUBLIC *pub);
+void tpm2_flush_srk(TSS_CONTEXT *tssContext);
+
+#endif
-- 
2.6.6

-- 
openssl-dev mailing list
To unsubscribe: https://mta.openssl.org/mailman/listinfo/openssl-dev

Re: [tpmdd-devel] [PATCH 1/1] add TPM2 version of create_tpm2_key and libtpm2.so engine

[Jason Gunthorpe]

On Sat, Dec 31, 2016 at 02:52:43PM -0800, James Bottomley wrote:
> This patch adds RSA signing for TPM2 keys.  There's a limitation to the
> way TPM2 does signing: it must recognise the OID for the signature. 
>  That fails for the MD5-SHA1 signatures of the TLS/SSL certificate
> verification protocol, so I'm using RSA_Decrypt for both signing
> (encryption) and decryption ... meaning that this only works with TPM
> decryption keys.  It is possible to use the prior code, which preserved
> the distinction of signing and decryption keys, but only at the expense
> of not being able to support SSL or TLS lower than 1.2

[Note, I haven't looked closely at TPM2, but TPM1.2 has a concept of
 key usage, and I assume that is carried over in the below comments]

I think it is very important to natively support the sign-only key
usage restriction. TPM1.2 goes so far as to declare keys that can be
used for arbitary decrypt as 'legacy do not use'.

IMHO the best way to do this is to look at the sign operation openssl
is trying to do and see if it can be sent down the sign path to the
TPM. Only if that fails should the decrypt path be used.

For TPM1.2 you could create a sign-only key with the
TPM_SS_RSASSAPKCS1v15_DER mode and feed it arbitary NIDs - the TPM did
not check the data to sign, AFAIK.

Ideally the user should be able to setup a sign-only key and the
correct SSL negotiation parameters and have a working system, eg a
sign-only key used with TLS 1.3 and ephemeral keyx should work.

> +	/* this is slightly paradoxical that we're doing a Decrypt
> +	 * operation: the only material difference between decrypt and
> +	 * encrypt is where the padding is applied or checked, so if
> +	 * you apply your own padding up to the RSA block size and use
> +	 * TPM_ALG_NULL, which means no padding check, a decrypt
> +	 * operation effectively becomes an encrypt */

IIRC this duality is exactly why key usage exists, and why good crypto
practice has been to forbid decrypt/encrypt on the same key.

Jason

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most 
engaging tech sites, SlashDot.org! http://sdm.link/slashdot

Re: [PATCH 1/1] add TPM2 version of create_tpm2_key and libtpm2.so engine

[Andrey Pronin]

On Tue, Jan 03, 2017 at 04:11:26PM -0700, Jason Gunthorpe wrote:
> On Sat, Dec 31, 2016 at 02:52:43PM -0800, James Bottomley wrote:
> > This patch adds RSA signing for TPM2 keys.  There's a limitation to the
> > way TPM2 does signing: it must recognise the OID for the signature. 
> >  That fails for the MD5-SHA1 signatures of the TLS/SSL certificate
> > verification protocol, so I'm using RSA_Decrypt for both signing
> > (encryption) and decryption ... meaning that this only works with TPM
> > decryption keys.  It is possible to use the prior code, which preserved
> > the distinction of signing and decryption keys, but only at the expense
> > of not being able to support SSL or TLS lower than 1.2
> 
> [Note, I haven't looked closely at TPM2, but TPM1.2 has a concept of
>  key usage, and I assume that is carried over in the below comments]
> 
> I think it is very important to natively support the sign-only key
> usage restriction. TPM1.2 goes so far as to declare keys that can be
> used for arbitary decrypt as 'legacy do not use'.
> 
> IMHO the best way to do this is to look at the sign operation openssl
> is trying to do and see if it can be sent down the sign path to the
> TPM. Only if that fails should the decrypt path be used.
> 
> For TPM1.2 you could create a sign-only key with the
> TPM_SS_RSASSAPKCS1v15_DER mode and feed it arbitary NIDs - the TPM did
> not check the data to sign, AFAIK.

Unfortunately, it does check the data and even more than that. First,
it checks that the size of the passed data matches the size of the
digest of the specified hashing algorithm. And then it constructs the
DigestInfo with that algorithm ID on its side. So, yes, you have to use
a raw Decrypt to support cases like SHA1+MD5.

> 
> Ideally the user should be able to setup a sign-only key and the
> correct SSL negotiation parameters and have a working system, eg a
> sign-only key used with TLS 1.3 and ephemeral keyx should work.
> 
> > +	/* this is slightly paradoxical that we're doing a Decrypt
> > +	 * operation: the only material difference between decrypt and
> > +	 * encrypt is where the padding is applied or checked, so if
> > +	 * you apply your own padding up to the RSA block size and use
> > +	 * TPM_ALG_NULL, which means no padding check, a decrypt
> > +	 * operation effectively becomes an encrypt */
> 
> IIRC this duality is exactly why key usage exists, and why good crypto
> practice has been to forbid decrypt/encrypt on the same key.
> 
> Jason
> 
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most 
> engaging tech sites, SlashDot.org! http://sdm.link/slashdot
> _______________________________________________
> tpmdd-devel mailing list
> tpmdd-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
> https://lists.sourceforge.net/lists/listinfo/tpmdd-devel

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most 
engaging tech sites, SlashDot.org! http://sdm.link/slashdot

Re: [TrouSerS-tech] [PATCH 1/1] add TPM2 version of create_tpm2_key and libtpm2.so engine

[James Bottomley]

On Tue, 2017-01-03 at 16:11 -0700, Jason Gunthorpe wrote:
> On Sat, Dec 31, 2016 at 02:52:43PM -0800, James Bottomley wrote:
> > This patch adds RSA signing for TPM2 keys.  There's a limitation to 
> > the way TPM2 does signing: it must recognise the OID for the 
> > signature.  That fails for the MD5-SHA1 signatures of the TLS/SSL 
> > certificate verification protocol, so I'm using RSA_Decrypt for 
> > both signing (encryption) and decryption ... meaning that this only 
> > works with TPM decryption keys.  It is possible to use the prior 
> > code, which preserved the distinction of signing and decryption 
> > keys, but only at the expense of not being able to support SSL or
> > TLS lower than 1.2
> 
> [Note, I haven't looked closely at TPM2, but TPM1.2 has a concept of
>  key usage, and I assume that is carried over in the below comments]

The TPM1.2 all uses the correct signing functions, the problem is only
with 2.0.

> I think it is very important to natively support the sign-only key
> usage restriction. TPM1.2 goes so far as to declare keys that can be
> used for arbitary decrypt as 'legacy do not use'.
> 
> IMHO the best way to do this is to look at the sign operation openssl
> is trying to do and see if it can be sent down the sign path to the
> TPM. Only if that fails should the decrypt path be used.

The problem is the MD5-SHA1 signature of SSL and TLS < v1.2.   This
cannot be performed by the TPM because it's not listed as a supported
signature, so the choice is either to deprecate these protocols (not
really viable since they're in wide use in old websites) or use decrypt
to do the signatures.  Once we get to the point of having to use
decrypt, there's no reason to preserve the signing distinction since we
never know when a key will be used to decrypt or sign.

Note that google took an alternative approach and modified their TSS to
work with a MD5-SHA1 signature:

https://chromium-review.googlesource.com/#/c/420811/

But this requires a modification to the TPM as well, which we can't do.

> For TPM1.2 you could create a sign-only key with the
> TPM_SS_RSASSAPKCS1v15_DER mode and feed it arbitary NIDs - the TPM 
> did not check the data to sign, AFAIK.
> 
> Ideally the user should be able to setup a sign-only key and the
> correct SSL negotiation parameters and have a working system, eg a
> sign-only key used with TLS 1.3 and ephemeral keyx should work.
> 
> > +	/* this is slightly paradoxical that we're doing a Decrypt
> > +	 * operation: the only material difference between decrypt
> > and
> > +	 * encrypt is where the padding is applied or checked, so
> > if
> > +	 * you apply your own padding up to the RSA block size and
> > use
> > +	 * TPM_ALG_NULL, which means no padding check, a decrypt
> > +	 * operation effectively becomes an encrypt */
> 
> IIRC this duality is exactly why key usage exists, and why good 
> crypto practice has been to forbid decrypt/encrypt on the same key.

Given the signature restrictions, TPM 2.0 just can't be made to work
with the older SSL/TLS protocols, so I'm opting to keep compatibility
and not benefit from the distinction between signing and decryption
keys.

James

> Jason
> 
> ---------------------------------------------------------------------
> ---------
> Check out the vibrant tech community on one of the world's most 
> engaging tech sites, SlashDot.org! http://sdm.link/slashdot
> _______________________________________________
> TrouSerS-tech mailing list
> TrouSerS-tech-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
> https://lists.sourceforge.net/lists/listinfo/trousers-tech
> 


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most 
engaging tech sites, SlashDot.org! http://sdm.link/slashdot

Re: [TrouSerS-tech] [PATCH 1/1] add TPM2 version of create_tpm2_key and libtpm2.so engine

[Jason Gunthorpe]

On Tue, Jan 03, 2017 at 03:22:56PM -0800, James Bottomley wrote:
> > I think it is very important to natively support the sign-only key
> > usage restriction. TPM1.2 goes so far as to declare keys that can be
> > used for arbitary decrypt as 'legacy do not use'.
> > 
> > IMHO the best way to do this is to look at the sign operation openssl
> > is trying to do and see if it can be sent down the sign path to the
> > TPM. Only if that fails should the decrypt path be used.
> 
> The problem is the MD5-SHA1 signature of SSL and TLS < v1.2.   This
> cannot be performed by the TPM because it's not listed as a supported
> signature, so the choice is either to deprecate these protocols (not
> really viable since they're in wide use in old websites) or use decrypt
> to do the signatures.  Once we get to the point of having to use
> decrypt, there's no reason to preserve the signing distinction since we
> never know when a key will be used to decrypt or sign.

I'm not disputing your analysis, just remarking that it seem very
undesirable to ban *all* sign-only keys just to support a single
legacy SSL configuration.

This is why I suggest looking at the sign being requested and using
the sign path if it is *possible*, otherwise requiring a key with the
broader key usage.

Not everything is SSL - openssh uses these routines too and it should
be able to use the sign only key type without a limitation?

Jason

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most 
engaging tech sites, SlashDot.org! http://sdm.link/slashdot

Re: [TrouSerS-tech] [PATCH 1/1] add TPM2 version of create_tpm2_key and libtpm2.so engine

[James Bottomley]

On Tue, 2017-01-03 at 16:40 -0700, Jason Gunthorpe wrote:
> On Tue, Jan 03, 2017 at 03:22:56PM -0800, James Bottomley wrote:
> > > I think it is very important to natively support the sign-only 
> > > key usage restriction. TPM1.2 goes so far as to declare keys that 
> > > can be used for arbitary decrypt as 'legacy do not use'.
> > > 
> > > IMHO the best way to do this is to look at the sign operation 
> > > openssl is trying to do and see if it can be sent down the sign 
> > > path to the TPM. Only if that fails should the decrypt path be
> > > used.
> > 
> > The problem is the MD5-SHA1 signature of SSL and TLS < v1.2.   This
> > cannot be performed by the TPM because it's not listed as a 
> > supported signature, so the choice is either to deprecate these 
> > protocols (not really viable since they're in wide use in old 
> > websites) or use decrypt to do the signatures.  Once we get to the 
> > point of having to use decrypt, there's no reason to preserve the 
> > signing distinction since we never know when a key will be used to
> > decrypt or sign.
> 
> I'm not disputing your analysis, just remarking that it seem very
> undesirable to ban *all* sign-only keys just to support a single
> legacy SSL configuration.

It's not just a single situation.  MD5-SHA1 is where it will fall apart
on backwards compatibility but my current TPM doesn't understand
anything other than sha1 or sha256, so it wouldn't allow more state of
the art algorithms like sha224, sha384 or sha512 either.

> This is why I suggest looking at the sign being requested and using
> the sign path if it is *possible*, otherwise requiring a key with the
> broader key usage.

It's a trade off: the user configuration nightmare: how does an
ordinary user know what type of TPM key they need ... they'll need to
know what their TPM is capable of and what signatures they're likely to
need.  VS the benefits of having single use keys.

I'm just not sure I see enough benefits to trying to preserve the
decrypt vs sign distinction, whereas I do see the floods of complaints
from users who got it wrong or think it should work as advertised.

> Not everything is SSL - openssh uses these routines too and it should
> be able to use the sign only key type without a limitation?

Agreed that openssh only has forward compatibility problems, but it
still has them thanks to the newer algorithm problem.

Actually, the next problem would be gnome-keyring.  It does openssl
signatures and fishing the algorithm type out by the time 

James


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most 
engaging tech sites, SlashDot.org! http://sdm.link/slashdot

Re: [TrouSerS-tech] [PATCH 1/1] add TPM2 version of create_tpm2_key and libtpm2.so engine

[Jason Gunthorpe]

On Tue, Jan 03, 2017 at 04:17:06PM -0800, James Bottomley wrote:
> On Tue, 2017-01-03 at 16:40 -0700, Jason Gunthorpe wrote:

> > I'm not disputing your analysis, just remarking that it seem very
> > undesirable to ban *all* sign-only keys just to support a single
> > legacy SSL configuration.
> 
> It's not just a single situation.  MD5-SHA1 is where it will fall apart
> on backwards compatibility but my current TPM doesn't understand
> anything other than sha1 or sha256, so it wouldn't allow more state of
> the art algorithms like sha224, sha384 or sha512 either.

Okay, yes, that is horrible :( If it is that bad it might not be worth
the effort..

> I'm just not sure I see enough benefits to trying to preserve the
> decrypt vs sign distinction, whereas I do see the floods of complaints
> from users who got it wrong or think it should work as advertised.

I probably wouldn't change the process for key generation - make the
tools default to decrypt keys and have an advanced option for
sign-only.

Jason

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most 
engaging tech sites, SlashDot.org! http://sdm.link/slashdot

add TPM2 version of create_tpm2_key and libtpm2.so engine -> Hash algoritms

[Kenneth Goldman]

[-- Attachment #1.1: Type: text/plain, Size: 791 bytes --]

Jason Gunthorpe <jgunthorpe-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org> wrote on 01/03/2017 
07:42:17 PM:

> > ... but my current TPM doesn't understand
> > anything other than sha1 or sha256, so it wouldn't allow more state of
> > the art algorithms like sha224, sha384 or sha512 either.
> 
> Okay, yes, that is horrible :( If it is that bad it might not be worth
> the effort.

The place to ask for new algorithms is the TCG's Device Driver WG.  It's
an odd WG name, but this is the WG where the TPM mandatory algorithms are
specified.

A real, commercial use case will likely be an effective argument, since
these are resource constrained and cost sensitive.

SHA-384 and SHA-512 are currently optional, which traditionally means 
they won't be implemented.



[-- Attachment #1.2: Type: text/html, Size: 1101 bytes --]

[-- Attachment #2: Type: text/plain, Size: 203 bytes --]

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most 
engaging tech sites, SlashDot.org! http://sdm.link/slashdot

[-- Attachment #3: Type: text/plain, Size: 192 bytes --]

_______________________________________________
tpmdd-devel mailing list
tpmdd-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
https://lists.sourceforge.net/lists/listinfo/tpmdd-devel

Re: add TPM2 version of create_tpm2_key and libtpm2.so engine -> Hash algoritms

[Jason Gunthorpe]

On Wed, Jan 04, 2017 at 01:48:44PM -0500, Kenneth Goldman wrote:
>    Jason Gunthorpe <jgunthorpe-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org> wrote on 01/03/2017
>    07:42:17 PM:
>    > > ... but my current TPM doesn't understand
>    > > anything other than sha1 or sha256, so it wouldn't allow more state
>    of
>    > > the art algorithms like sha224, sha384 or sha512 either.
>    >
>    > Okay, yes, that is horrible :( If it is that bad it might not be worth
>    > the effort.

>    The place to ask for new algorithms is the TCG's Device Driver
>    WG.  It's an odd WG name, but this is the WG where the TPM
>    mandatory algorithms are specified.  A real, commercial use case
>    will likely be an effective argument, since these are resource
>    constrained and cost sensitive.  SHA-384 and SHA-512 are
>    currently optional, which traditionally means they won't be
>    implemented.

We don't need the algorithm in the TPM. We just need to be able to RSA
sign an arbitary OID + externally computed hash like TPM 1.2 could.

What is the recommended way to create a key with a sign-only intent
that can be used with arbitary OID + computed hash?

James is proposing using the Decrypt op to do this job.

Jason

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most 
engaging tech sites, SlashDot.org! http://sdm.link/slashdot

Re: [Ibmtpm20tss-users] [tpmdd-devel] add TPM2 version of create_tpm2_key and libtpm2.so engine -> Hash algoritms

[James Bottomley]

On Wed, 2017-01-04 at 11:54 -0700, Jason Gunthorpe wrote:
> On Wed, Jan 04, 2017 at 01:48:44PM -0500, Kenneth Goldman wrote:
> >    Jason Gunthorpe <jgunthorpe@obsidianresearch.com> wrote on
> > 01/03/2017
> >    07:42:17 PM:
> >    > > ... but my current TPM doesn't understand
> >    > > anything other than sha1 or sha256, so it wouldn't allow
> > more state
> >    of
> >    > > the art algorithms like sha224, sha384 or sha512 either.
> >    >
> >    > Okay, yes, that is horrible :( If it is that bad it might not
> > be worth
> >    > the effort.
> 
> >    The place to ask for new algorithms is the TCG's Device Driver
> >    WG.  It's an odd WG name, but this is the WG where the TPM
> >    mandatory algorithms are specified.  A real, commercial use case
> >    will likely be an effective argument, since these are resource
> >    constrained and cost sensitive.  SHA-384 and SHA-512 are
> >    currently optional, which traditionally means they won't be
> >    implemented.
> 
> We don't need the algorithm in the TPM. We just need to be able to 
> RSA sign an arbitary OID + externally computed hash like TPM 1.2
> could.
> 
> What is the recommended way to create a key with a sign-only intent
> that can be used with arbitary OID + computed hash?

There isn't one.  TPM_ALG_NULL is illegal for sign operations.  The
Part 1 Architecture Guide (version 1.16 lists all the potentially
supported OIDs in section B.6).  The idea is that the TPM is supposed
to be able to validate that you're not causing it to generate a
nefarious signature, so you pass in the hash alone along with the
algorithm and it validates the legality of the hash and then returns
back the OID prepended hash signed by the key.

> James is proposing using the Decrypt op to do this job.

I believe the TCG has decided this is the only way to sign arbitrary
data.

James


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most 
engaging tech sites, SlashDot.org! http://sdm.link/slashdot

Re: [Ibmtpm20tss-users] add TPM2 version of create_tpm2_key and libtpm2.so engine -> Hash algoritms

[Kenneth Goldman]

[-- Attachment #1.1: Type: text/plain, Size: 419 bytes --]

James Bottomley <jejb-23VcF4HTsmIX0ybBhKVfKdBPR1lH4CV8@public.gmane.org> wrote on 01/04/2017 02:45:21 PM:

> > James is proposing using the Decrypt op to do this job.
> 
> I believe the TCG has decided this is the only way to sign arbitrary
> data.

My suspicion is that this was an omission, not a decision.  In fact,
Part 1 explicitly describes a caller provided OID.  It just wasn't
implemented - yet.



[-- Attachment #1.2: Type: text/html, Size: 589 bytes --]

[-- Attachment #2: Type: text/plain, Size: 203 bytes --]

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most 
engaging tech sites, SlashDot.org! http://sdm.link/slashdot

[-- Attachment #3: Type: text/plain, Size: 192 bytes --]

_______________________________________________
tpmdd-devel mailing list
tpmdd-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
https://lists.sourceforge.net/lists/listinfo/tpmdd-devel

Re: add TPM2 version of create_tpm2_key and libtpm2.so engine -> Hash algoritms

[Kenneth Goldman]

[-- Attachment #1.1: Type: text/plain, Size: 795 bytes --]

Jason Gunthorpe <jgunthorpe-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org> wrote on 01/04/2017 
01:54:34 PM:

> We don't need the algorithm in the TPM. We just need to be able to RSA
> sign an arbitary OID + externally computed hash like TPM 1.2 could.
> 
> What is the recommended way to create a key with a sign-only intent
> that can be used with arbitary OID + computed hash?

Probably the long-term correct way is to go to the TCG and ask for a 
new feature.  However, this is (often certified) hardware, so the turn 
around is likely to be a year.

> James is proposing using the Decrypt op to do this job.

That works.  I've coded it.  However, the key doesn't have
"sign only intent", in that the caller not only provides the OID
but also provides the padding.



[-- Attachment #1.2: Type: text/html, Size: 1115 bytes --]

[-- Attachment #2: Type: text/plain, Size: 203 bytes --]

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most 
engaging tech sites, SlashDot.org! http://sdm.link/slashdot

[-- Attachment #3: Type: text/plain, Size: 192 bytes --]

_______________________________________________
tpmdd-devel mailing list
tpmdd-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
https://lists.sourceforge.net/lists/listinfo/tpmdd-devel

Re: [tpmdd-devel] [PATCH 1/1] add TPM2 version of create_tpm2_key and libtpm2.so engine

[Jarkko Sakkinen]

On Tue, Jan 03, 2017 at 04:40:53PM -0700, Jason Gunthorpe wrote:
> On Tue, Jan 03, 2017 at 03:22:56PM -0800, James Bottomley wrote:
> > > I think it is very important to natively support the sign-only key
> > > usage restriction. TPM1.2 goes so far as to declare keys that can be
> > > used for arbitary decrypt as 'legacy do not use'.
> > > 
> > > IMHO the best way to do this is to look at the sign operation openssl
> > > is trying to do and see if it can be sent down the sign path to the
> > > TPM. Only if that fails should the decrypt path be used.
> > 
> > The problem is the MD5-SHA1 signature of SSL and TLS < v1.2.   This
> > cannot be performed by the TPM because it's not listed as a supported
> > signature, so the choice is either to deprecate these protocols (not
> > really viable since they're in wide use in old websites) or use decrypt
> > to do the signatures.  Once we get to the point of having to use
> > decrypt, there's no reason to preserve the signing distinction since we
> > never know when a key will be used to decrypt or sign.
> 
> I'm not disputing your analysis, just remarking that it seem very
> undesirable to ban *all* sign-only keys just to support a single
> legacy SSL configuration.

Could this configuration just marked as unsupported, or not?

> This is why I suggest looking at the sign being requested and using
> the sign path if it is *possible*, otherwise requiring a key with the
> broader key usage.
> 
> Not everything is SSL - openssh uses these routines too and it should
> be able to use the sign only key type without a limitation?
> 
> Jason

/Jarkko

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most 
engaging tech sites, SlashDot.org! http://sdm.link/slashdot

Re: [Ibmtpm20tss-users] [TrouSerS-tech] [PATCH 1/1] add TPM2 version of create_tpm2_key and libtpm2.so engine

[Kenneth Goldman]

[-- Attachment #1.1: Type: text/plain, Size: 1226 bytes --]

Trimmed the CC list a bit.  Where does this discussion really belong? 
Trousers is for TPM 1.2,
and it's not a TSS or TPM device driver issue.

If you're all TCG members, the TCG's TPM WG is the real place to go if you 
want to get 
something fixed.

James Bottomley <James.Bottomley-d9PhHud1JfjCXq6kfMZ53/egYHeGw8Jk@public.gmane.org> wrote on 
01/03/2017 06:22:56 PM:

> > 
> > [Note, I haven't looked closely at TPM2, but TPM1.2 has a concept of
> >  key usage, and I assume that is carried over in the below comments]
> 
> The TPM1.2 all uses the correct signing functions, the problem is only
> with 2.0.
> 
> > I think it is very important to natively support the sign-only key
> > usage restriction. TPM1.2 goes so far as to declare keys that can be
> > used for arbitary decrypt as 'legacy do not use'.

TPM 2.0 has several features to support this:

- Signing keys, a usage restriction that can only sign.
- Policy only, with a policy restricted to specific commands, e.g., Quote 
or Sign.
- Restricted keys, that have a fixed algorithm and can only sign TPM 
generated data.

What it doesn't have is the ability (for a signing key) to prepend a 
caller
specified OID and padding.



[-- Attachment #1.2: Type: text/html, Size: 1716 bytes --]

[-- Attachment #2: Type: text/plain, Size: 203 bytes --]

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most 
engaging tech sites, SlashDot.org! http://sdm.link/slashdot

[-- Attachment #3: Type: text/plain, Size: 192 bytes --]

_______________________________________________
tpmdd-devel mailing list
tpmdd-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
https://lists.sourceforge.net/lists/listinfo/tpmdd-devel

Re: [PATCH 1/1] add TPM2 version of create_tpm2_key and libtpm2.so engine

[Ken Goldman]

On 1/3/2017 6:22 PM, James Bottomley wrote:
>
> Note that google took an alternative approach and modified their TSS to
> work with a MD5-SHA1 signature:
>
> https://chromium-review.googlesource.com/#/c/420811/
>
> But this requires a modification to the TPM as well, which we can't do.

Right.  It's not a TSS issue.  Modification is futile.  It just passes 
parameters on to the TPM.  The place for this change is the TCG's TPM 
work group.



------------------------------------------------------------------------------
Developer Access Program for Intel Xeon Phi Processors
Access to Intel Xeon Phi processor-based developer platforms.
With one year of Intel Parallel Studio XE.
Training and support from Colfax.
Order your platform today. http://sdm.link/xeonphi