archive mirror
 help / color / mirror / Atom feed
From: James Bottomley <>
Cc:, Mimi Zohar <>,
	Jarkko Sakkinen <>,
	David Woodhouse <>,, David Howells <>
Subject: [PATCH v2 1/1] doc: add draft RFC for TPM Key format
Date: Mon, 24 May 2021 15:20:11 -0700	[thread overview]
Message-ID: <> (raw)
In-Reply-To: <>

Adds the xml file for the draft RFC and builds text and html versions
if the xml2rfc program is found.

Signed-off-by: James Bottomley <>


v2: Add missing sections plus minor updates
---                       |   2 +-                      |   4 +-
 doc/                   |  15 +
 doc/draft-bottomley-tpm2-keys.xml | 465 ++++++++++++++++++++++++++++++
 4 files changed, 484 insertions(+), 2 deletions(-)
 create mode 100644 doc/
 create mode 100644 doc/draft-bottomley-tpm2-keys.xml

diff --git a/ b/
index 33de0d9..787ba29 100644
--- a/
+++ b/
@@ -41,4 +41,4 @@ $(builddir)/%.1: $(srcdir)/ $(top_builddir)/%
 	cd $(DESTDIR)$(openssl_enginedir) && $(LN_S) -f libtpm2@SHREXT@ tpm2@SHREXT@
-SUBDIRS = tests
+SUBDIRS = tests doc
diff --git a/ b/
index 6efa7a5..e102dd2 100644
--- a/
+++ b/
@@ -128,6 +128,8 @@ fi
 AC_PATH_PROG(TPMSERVER, tpm_server,,/bin:/usr/bin:/usr/lib/ibmtss:/usr/libexec/ibmtss)
 AC_PATH_PROG(SWTPM, swtpm,,/bin:/usr/bin:/usr/lib/ibmtss:/usr/libexec/ibmtss)
 AC_PATH_PROG(SWTPM_IOCTL, swtpm_ioctl,,/bin:/usr/bin:/usr/lib/ibmtss:/usr/libexec/ibmtss)
+AC_CHECK_PROG(XML2RFC, xml2rfc, xml2rfc)
@@ -147,7 +149,7 @@ fi
-AC_OUTPUT([Makefile tests/Makefile])
+AC_OUTPUT([Makefile tests/Makefile doc/Makefile])
 cat <<EOF
diff --git a/doc/ b/doc/
new file mode 100644
index 0000000..0c24ce0
--- /dev/null
+++ b/doc/
@@ -0,0 +1,15 @@
+XML2RFC_TARGETS = draft-bottomley-tpm2-keys.txt \
+		draft-bottomley-tpm2-keys.html
+	rm -fr $(XML2RFC_TARGETS)
+$(builddir)/%.txt: $(srcdir)/%.xml
+	$(XML2RFC) --text -o $@ $<
+$(builddir)/%.html: $(srcdir)/%.xml
+	$(XML2RFC) --html -o $@ $<
diff --git a/doc/draft-bottomley-tpm2-keys.xml b/doc/draft-bottomley-tpm2-keys.xml
new file mode 100644
index 0000000..d2e9235
--- /dev/null
+++ b/doc/draft-bottomley-tpm2-keys.xml
@@ -0,0 +1,465 @@
+<?xml version="1.0"?>
+<!DOCTYPE rfc SYSTEM "rfc2629.dtd" [
+<!-- One method to get references from the online citation libraries.
+There has to be one entity for each item to be referenced.
+An alternate method (rfc include) is described in the references.
+<?rfc toc="yes" ?>
+<rfc ipr="trust200902" category="info" docName="draft-bottomley-tpm-keys-00">
+  <front>
+    <title abbrev="TPM 2 Key Format">ASN.1 Specification for TPM 2.0 Key Files</title>
+    <author initials="J." surname="Bottomley" fullname="James E.J. Bottomley">
+      <organization>Linux Kernel</organization>
+      <address>
+        <postal>
+          <street/>
+          <city/>
+          <region/>
+          <country>USA</country>
+        </postal>
+        <email></email>
+      </address>
+    </author>
+    <date month="May" year="2021"/>
+    <area>Security</area>
+    <keyword>I-D</keyword>
+    <keyword>Internet-Draft</keyword>
+    <keyword>X.509</keyword>
+    <abstract>
+      <t>
+	This specification is designed to be an extension to the ASN.1
+	(defined in <xref target="X.680"/>) specification of PKCS #1
+	<xref target="RFC8017"/> to define the file format of private
+	keys that need to be loaded into a TPM 2 device to operate.
+      </t>
+    </abstract>
+  </front>
+  <middle>
+    <section anchor="intro" title="Introduction">
+      <t>
+	The Security of private keys has long been a concern and the
+	ability of ubiquitous devices like TPMs has made it useful to
+	use them for secure private key storage.  With the advent of
+	TPM 2.0, private key storage inside the TPM (acting as a token
+	which could be referred to by PKCS #11) has been discouraged,
+	and instead key files which are loaded and evicted as
+	necessary is the encouraged format.  This standard defines an
+	interoperable ASN.1 representation for such key files, so that
+	a key created by one tool should be loadable by a different
+	one.
+      </t>
+    </section>
+    <section anchor="terms" title="Terminology">
+      <t>
+        The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL
+        NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED",  "MAY", and
+        "OPTIONAL" in this document are to be interpreted as described in
+        <xref target="RFC2119"/>.
+      </t>
+      <section title="Notation">
+	<dl>
+	  <dt>ASN.1</dt>
+	  <dd>Abstract Syntax Notation defined in
+	  <xref target="X.680"/></dd>
+	  <dt>DER</dt>
+	  <dd>Distinguished Encoding Rules. Basically a defined binary
+	  representation for ASN.1</dd>
+	  <dt>MSO</dt>
+	  <dd>Most Significant Octet (the highest order
+	  byte of an integer)</dd>
+	  <dt>PEM</dt>
+	  <dd>Privacy enhanced Electronic Mail.  An ASCII compatible
+	  representation of DER</dd>
+	  <dt>TCG</dt>
+	  <dd>Trusted Computing Group</dd>
+	  <dt>TPM</dt>
+	  <dd>Trusted Platform Module</dd>
+	</dl>
+      </section>
+    </section>
+    <section anchor="keyrep" title="Key Representation">
+      <t>
+	All TPM 2.0 keys consist of two binary pieces, a public part,
+	which can be parsed according to the TPM specification for
+	TPM2B_PUBLIC <xref target="TPM2.0"/> and a private part, which
+	is cryptographically sealed in such a way as to be only
+	readable on the TPM that created it.  The purpose of this
+	specification is to specify a format by which the public and
+	private pieces of a TPM key can be loaded.
+      </t>
+      <t>
+	The design of the TPMkey ASN.1 format is that it should have a
+	distinguishing OID at the beginning so the DER form of the
+	key can be easily recognized.  In PEM form, the key MUST have
+	"-----BEGIN TSS2 PRIVATE KEY-----" and "-----END TSS2 PRIVATE
+	KEY-----" as the PEM guards. All additional information that
+	may be needed to load the key is specified as optional
+	explicit elements, which can be extended by later
+	specifications, which is why the TPMkey is not versioned.
+      </t>
+      <section anchor="tpmkey" title="TPMkey Syntax">
+	<figure><artwork>
+    type        OBJECT IDENTIFIER
+    policy      [1] EXPLICIT SEQUENCE OF TPMPolicy OPTIONAL
+    parent      INTEGER
+    pubkey      OCTET STRING
+    privkey     OCTET STRING
+  }
+	</artwork></figure>
+	<t>
+	  The fields of type TPMKey have the following meanings:
+	</t>
+	<section title="type">
+	  <t>
+	    A unique OID specifying the key type.  This standard
+	    currently defines three types of keys: a loadable key,
+	    specified by id-loadablekey, (to be loaded with
+	    TPM2_Load), an importable key, specified by
+	    id-importablekey, (to be loaded with TPM2_Import) and a
+	    sealed data key, specified by id-sealedkey, (to be
+	    extracted with TPM2_Unseal).  The TCG has reserved the
+	    following OID prefix for this:
+	  </t>
+	  <figure><artwork>
+  id-tpmkey OBJECT IDENTIFIER ::=
+    {joint-iso-itu-t(2) international-organizations(23) 133 10}
+	  </artwork></figure>
+	  <t>
+	    And the three key types are:
+	  </t>
+	  <figure><artwork>
+  id-loadablekey OBJECT IDENTIFIER ::=
+    {id-tpmkey 3}
+	  </artwork></figure>
+	  <figure><artwork>
+  id-importablekey OBJECT IDENTIFIER ::=
+    {id-tpmkey 4}
+	  </artwork></figure>
+	  <figure><artwork>
+  id-sealedkey OBJECT IDENTIFIER ::=
+    {id-tpmkey 5}
+	  </artwork></figure>
+	</section>
+	<section title="emptyAuth">
+	  <t>
+	    An implementation needs to know as it formulates the
+	    TPM2_Load/Import/Unseal command whether it must also send
+	    down an authorization, so this parameter gives that
+	    indication.  emptyAuth MUST be true if authorization is
+	    NOT required and MUST BE either false or absent if
+	    authorization is required.  Since this element has
+	    three states (one representing true and two representing
+	    false) it is RECOMMENDED that implementations emitting
+	    TPMkey representations use absence of the tag to represent
+	    false.  However, implementations reading TPMKey MUST
+	    be able to process all three possible states.
+	  </t>
+	</section>
+	<section title="policy">
+	  <t>
+	    This MUST be present if the TPM key has a policy hash
+	    because it describes to the implementation how to
+	    construct the policy.  The forms of the policy statement
+	    are described in section <xref target="policy"/>.
+	  </t>
+	</section>
+	<section title="secret">
+	  <t>
+	    This section describes the additional cryptographic
+	    secret used to specify the outer wrapping of an
+	    importable key.  It MUST be present for key type
+	    id-importablekey and MUST NOT be present for any other
+	    key type.
+	  </t>
+	  <t>
+	    Importable keys (designed to be processed by TPM2_Import)
+	    MUST have an unencrypted inner wrapper (symmetricAlg MUST
+	    be TPM_ALG_NULL and encryptionKey MUST be empty) and an
+	    outer wrapper encrypted to the parent key using
+	    inSymSeed. The secret parameter is the fully marshalled
+	    TPM2B_ENCRYPTED_SECRET form of inSymSeed.
+	  </t>
+	</section>
+	<section title="parent">
+	  <t>
+	    This MUST be present for all keys and specifies the handle
+	    of the parent key.  The parent key SHOULD be either a
+	    persistent handle (MSO 0x81) or a permanent handle (MSO
+	    0x40).  Since volatile handle numbering can change
+	    unexpectedly depending on key load order, the parent
+	    SHOULD NOT be a volatile handle (MSO 0x80). The parent MUST
+	    NOT have any other MSO.
+	  </t>
+	  <t>
+	     If a permanent handle (MSO 0x40) is specified then the
+	     implementation MUST run TPM2_CreatePrimary on the handle
+	     using the TCG specified Elliptic Curve template for the
+	     NIST P-256 curve and use the primary key so generated as
+	     the parent.
+	  </t>
+	</section>
+	<section title="pubkey">
+	  <t>
+	    This MUST be present and MUST correspond to the fully
+	    marshalled TPM2B_PUBLIC structure of the TPM Key.
+	  </t>
+	</section>
+	<section title="privkey">
+	  <t>
+	    This MUST be present and MUST correspond to the fully
+	    marshalled TPM2B_PRIVATE structure of the TPM Key.  For
+	    importable keys, this must be the duplicate parameter that
+	    would be input to TPM2_Import.
+	  </t>
+	</section>
+      </section>
+    </section>
+    <section anchor="policy" title="Key Policy Specification">
+      <t>
+	Policy is constructed on a TPM by executing a sequence of
+	policy statements.  This specification currently only defines
+	a limited subset of the allowed policy statements.  The policy
+	is specified by a hash, which the execution of the policy
+	statements must reach in order for the policy to be validated
+	(See <xref target="TPM2.0"/> Part 1 for a detailed description.
+      </t>
+      <t>
+	The TPMPolicy ASN.1 MUST be a sequence of policy statements
+	which correspond exactly to TPM policy instructions in the
+	order they should be executed and additionally from which the
+	ultimate policy hash can be constructed.
+      </t>
+      <t>
+	The current policy specification is strictly for AND based
+	policy only and may be extended at a later date with OR
+	policy.  However, the ASN.1 for policy is formulated as CONS
+	elements, leaving the possibility of adding additional but
+	optional elements for policy statements which are not
+	supported by this standard (such as TPM2_PolicyAuthorize).
+      </t>
+      <section title="TPMPolicy Syntax">
+	<figure><artwork>
+ TPMPolicy ::= SEQUENCE {
+    CommandCode   [0] EXPLICIT INTEGER
+    CommandPolicy [1] EXPLICIT OCTET STRING
+  }
+	</artwork></figure>
+	<t>
+	  The Fields of type TPMPolicy have the following meanings:
+	</t>
+	<section title="CommandCode">
+	  <t>
+	    This is the integer representation of the TPM command code
+	    for the policy statement.
+	  </t>
+	</section>
+	<section title="CommandPolicy">
+	  <t>
+	    This is a binary string representing a fully marshalled,
+	    TPM ordered, command body for the TPM policy command.
+	    Therefore to send the command, the implementation simply
+	    marshalls the command code and appends this octet string
+	    as the body.
+	  </t>
+	  <t>
+	    Commands which have no body, such as TPM2_AuthVal, MUST be
+	    specified as a zero length OCTET STRING
+	  </t>
+	</section>
+      </section>
+      <section title="Policy Implementation Considerations">
+	<t>
+	  The policy hash for AND based policies is constructed by extension of the prior policy hash
+	</t>
+	<figure><artwork>
+  newHash = HASH ( oldHash || policyHash )
+	</artwork></figure>
+	<t>
+	  where policyHash is usually simply the hash of the fully
+	  marshalled policy command (including the CommandCode).
+	  However, this isn't true for TPM2_PolicyCounterTimer() so
+	  always consult the <xref target="TPM2.0"/> specifications
+	  for how to construct the policyHash.
+	</t>
+	<t>
+	  The implementation should fail fast for policy problems, so
+	  if an individual policy command returns a failure (which
+	  usually indicates a particular policy requirement cannot be
+	  met), that failure should be reported in as much detail as
+	  possible and processing of the key should fail at that
+	  point.
+	</t>
+	<section title="Authorization Policy">
+	  <t>
+	    When Authorization (Passing in a password) is required,
+	    the emptyAuth parameter MUST be absent or set to false
+	    and additionally the TPM_CC_PolicyAuthValue MUST be
+	    specified as the command code for one entry in the
+	    TPMPolicy sequence.  However, the implementation MAY
+	    choose to execute either TPM2_PolicyPassword for TPM_RS_PW
+	    or TPM2_PolicyAuthValue for HMAC based authorization
+	    depending on whether the command being authorized is using
+	    sessions or not.  If the policy does not require an
+	    authorization then the emptyAuth parameter MUST be set to
+	    true.
+	  </t>
+	</section>
+      </section>
+    </section>
+    <section anchor="implementation" title="Implementation Considerations">
+      <t>
+	Implementations SHOULD support all TCG mandated algorithms,
+	but MAY omit those deemed insecure, such as the SHA1 hash.
+      </t>
+      <t>
+	TPM2_Import transforms the privKey into a TPM2B_PRIVATE which
+	can then be used as a source to TPM2_Load, making the loading
+	of importable keys is necessarily a two stage process, which
+	can be time consuming on some TPMs.  Since the TPM2B_PRIVATE
+	structure emitted by TPM2_Import is fully secure,
+	Implementations SHOULD minimize the number of TPM2_Import
+	operations by caching the emitted TPM2B_PRIVATE.
+      </t>
+    </section>
+    <section anchor="security" title="Security Considerations">
+      <t>
+	The TPM 2.0 supports a variety of algorithms, the most common
+	being SHA1 and SHA256 for hashing and RSA2048 and NIST P-256
+	for asymmetric keys.  Implementors SHOULD NOT use deprecated
+	algorithms, such as SHA1, for any TPM operation.  In
+	particular, the algorithm used for the policy hash SHOULD NOT
+	be SHA1 and this means that SHA1 SHOULD NOT be used as the
+	name algorithm hash for any TPM key.
+      </t>
+      <t>
+	TPM 2.0 supports a session mode (TPM_RS_PW) where
+	authorizations are passed to the TPM in clear text over the
+	TPM connection.  Implementations SHOULD consider the
+	possibility of snooping on the wire between the implementation
+	and the TPM, such as <xref target="TPM GENIE"/>, and SHOULD
+	use HMAC session authorizations as best practice for all TPM
+	keys.
+      </t>
+      <t>
+	In addition to snooping authorizations, snooping may also
+	occur when key material is being exchanged between the TPM and
+	the implementation, such as wrapping of private keys and the
+	sealing and unsealing operations for sealed keys.
+	Implementations SHOULD always use HMAC sessions with
+	TPMA_SESSION_DECRYPT when sensitive information is passed in
+	to the TPM and HMAC sessions with TPMA_SESSION_ENCRYPT when
+	sensitive information is received from the TPM.
+      </t>
+      <t>
+	The easiest way to get the TPM to wrap an external private key
+	is to use TPM2_Import.  However, since TPMA_SESSION_DECRYPT
+	only protects the first parameter (which is encryptionKey),
+	the duplicate should use inner symmetric encryption with a
+	randomly generated ephemeral key, which is then presented to
+	the TPM via the protected encryptionKey parameter.
+      </t>
+      <t>
+	The TPM has a mode where it can generate private key material
+	internally (using TPM2_Create) such that the private part of
+	the key can never leave the TPM.  Implementations SHOULD
+	support this mode but should be aware that while keys created
+	like this may be more secure than wrapped keys, they can also
+	be used only while access to the TPM that created them is
+	available, so implementations SHOULD also support wrapping for
+	keys that are expected to outlive the TPM that's using them.
+	Clients can then develop best practices around TPM wrapped
+	identity keys, possibly with TPM created sub keys, which can
+	only be used on the device they were wrapped for.
+      </t>
+      <t>
+	Since TPM keys can only be used by the specific TPM that
+	created them, which is usually embedded in a piece of
+	hardware, they are secure against exfiltration attacks.
+	However, consideration should be given to an attacker gaining
+	access to the system containing the TPM.  TPM keys are most
+	secure when used as part of an operating system that has
+	guaranteed trust properties, such as secure and measured boot.
+	Implementations SHOULD assist users in constructing key
+	policies that ensure the key can be used only when the
+	operating system is within its trusted parameters to minimize
+	threats from co-located attackers.
+      </t>
+    </section>
+    <section anchor="IANA" title="IANA Considerations">
+      <t>
+	None.
+      </t>
+    </section>
+    <section anchor="comments" title="Comments on and Enhancements to this Document">
+      <t>
+	Comments on this document should be addressed to the author
+	( but should also CC the
+	email lists of the two projects implementing this
+	specification:
+      </t>
+      <t>
+	The OpenSSL engine:
+      </t>
+      <t>
+	The Linux Kernel:
+      </t>
+      <t>
+	The OpenSSL TPM2 engine <xref target="OPENSSL TPM2 ENGINE"/>
+	is currently the only implementation of this full
+	specification, so enhancements should be proposed after
+	patches implementing the enhancement have been accepted by
+	openssl_tpm2_engine or another full specification
+	implementation.
+      </t>
+    </section>
+  </middle>
+  <back>
+    <references title="Normative References">
+      &RFC2119;
+      &RFC8017;
+      <reference anchor="TPM2.0" target="">
+	<front>
+          <title>TPM 2.0 Library Specification</title>
+          <author>
+            <organization>Trusted Computing Group</organization>
+          </author>
+          <date year="2013" month="March" day="15"/>
+	</front>
+      </reference>
+      <reference anchor="X.680" target="">
+	<front>
+	  <title>ITU-T Recommendation X.680,
+              Information technology - Abstract Syntax Notation One
+          (ASN.1):  Specification of basic notation.</title>
+	  <author><organization>International Telecommunication Union</organization></author>
+	  <date year="2015" month="August"/>
+	</front>
+      </reference>
+    </references>
+    <references title="Informative References">
+      <reference anchor="TPM GENIE" target="">
+	<front>
+	  <title>TPM Genie: Interposer Attacks Against the Trusted
+	  Platform Module Serial Bus</title>
+	  <author initials="J." surname="Boone" fullname="J. Boone">
+	    <organization>NCC Group</organization>
+	  </author>
+	  <date year="2018" month="March" day="9"/>
+	</front>
+      </reference>
+      <reference anchor="OPENSSL TPM2 ENGINE" target="">
+	<front>
+	  <title>OpenSSL TPM2 Engine</title>
+	  <author><organization>Open Source Project</organization></author>
+	</front>
+      </reference>
+    </references>
+  </back>

      reply	other threads:[~2021-05-24 22:20 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-24 22:20 [PATCH v2 0/1] draft RFC for TPM key format James Bottomley
2021-05-24 22:20 ` James Bottomley [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \ \ \ \ \ \ \ \ \ \

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).