]> Intel Corporation

james.d.beaney@intel.com

ARM Corporation

yogesh.deshpande@arm.com

Altera Corporation

andrew.draper@altera.com

Intel Corporation

vincent.r.scarlata@intel.com

Intel Corporation

ned.smith@intel.com

Security Remote ATtestation ProcedureS attestation profile corim This document is a profile of various IETF and TCG standards that support remote attestation. The profile supports Intel-specific adaptations and extensions for Evidence, Endorsements and Reference Values. This profile describes apticulareplication of CoRIM, EAT, CMW, TCG concise evidence, and TCG DICE specifications. In particular, CoRIM is extended to define measurement types that are unique to Intel and defines Reference Values types that support matching Evidence based on range and subset comparison. Multiple Evidence formats are anticipated, based on IETF and TCG specifications. Evidence formats are mapped to Reference Values expressions based on CoRIM and CoRIM extensions found in this profile. The Evidence to Reference Values mappings are either documented by industry specifications or by this profile. Reference Value Providers and Endorsers may use this profile to author mainifests containing Reference Values and Endorsements that require Intel profile support from parser implementations. Parser implementations can recognize the Intel profile by profile identifier values contained within attestation conceptual mmessages and from profile parameters to media types or profile specific content format identifiers. About This Document The latest revision of this draft can be found at . Status information for this document may be found at . Discussion of this document takes place on the Remote ATtestation ProcedureS Working Group mailing list (), which is archived at . Subscribe at . Source for this draft and an issue tracker can be found at .

Introduction This profile describes extensions and restrictions placed on Reference Values, Endorsements, and Evidence that support attestation capabilities of Intel products containing Intel(R) SGX(TM) or Intel(R) TDX(TM) technology, or Intel(R) products that contain DICE root of trust, DICE layers , or modules that implement SPDM . The CoRIM specifications and define a baseline schema for Reference Values and Endorsements that form the foundation for the extensions defined in this profile. contains a foundational representation for Attester actual state that Evidence, as specified by , , and , Reference Values, and Endorsements can map to that helps ensure compatibility with a broad spectrum of Verifier implementations. The Reference Values extensions in this profile describe reference state that can be expressed as set membership or value ranges. This profile defines extensions to CoRIM that support appraisal matching that is based on set membership, masked values, and numeric ranges. The baseline CoRIM, as defined by is a subset of the Intel profile. Intel products that implement exclusively to the baseline CoRIM may not rely upon this profile. The Intel profile extensions may be generally useful. Implementations based on the Intel profile does not necessarily imply an implementation is an Intel product. This profile extends CoMID measurement-values-map, as defined by (see also ), with measurement types that are unique to Intel products. Some measurement types are specific to Reference Values where multiple reference states may be included in reference manifests. Intel profile extensions use a CBOR tagged value that defines a comparison operator and operands that instruct Verifiers regarding subset, range, and masked values matching semantics. For example, a numeric operator 'greater-than' instructs the Verifier to match a numeric Evidence value if it is greater than a numeric range operand. This profile follows the Verifier behavior defined by and extends Verifier behavior to include operator-operand matching. If no operator is specified by Reference Values statements, the Verifier defaults to baseline matching semantics. If Evidence matches Reference Values and Endorsements apply, Endorsed Values may be added to the accetped claims set. When all Evidence and Endorsements are processed, the Verifier's set of accepted claims is available for Attestation Results computations. This profile doesn't define Attestation Results. Rather, an Attestation Results profile, such as may be referenced instead. This profile is compatible with multiple Evidence formats, as defined by , , and . It describes considerations when mapping Evidence formats to CoRIM that a Verifier may use when performig appraisals.

Conventions and Definitions The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 when, and only when, they appear in all capitals, as shown here. The reader is assumed to be familiar with the terms defined in Section 4 of and .

Background Complex platforms may contain a variety of hardware components, several of which may contain a hardware root of trust. Each root of trust may anchor one or more layers resulting in multiple instances of attestation Evidence. Evidence may be integrity protected by digital signatures, such as certificates , tokens or by a secure transport . For example, a system bus may allow dynamically configured peripheral devices that have attestation capabilities. Confidential computing environments, such as SGX, may extend an initial boundary to include a peripheral, or a peer enclave, that together forms a network of trustworthy nodes that a remote attestation Verifier may need to appraise. Multiple Evidence blocks may be combined into a composite Evidence block that is more easily conveyed. Complex platforms may have one or more lead Attester endpoints that communicate with a remote Verifier to convey composite Evidence. The composition of the complex platform is partially represented in the composite Evidence. However, composite Evidence may not fully describe platform composition. A complex platform may consist of multiple subsystems, such as network adapters, storage controllers, memory controllers, special purpose processors, etc. The various sub-subsystem components vendors may create hardware bills of material (HBOM) that describe sub-system composition. A complex platform vendor may assemble various sub-system components whose composition is described by a platform HBOM. Although CoRIM may be used to create HBOMs, use of this profile for HBOM creation is unanticipated. Nevertheless, a complex system may contain multiple identical instances of sub-sytem components that produce identical Evidence blocks. Additionally, dynamic insertion or removal of a component may result in composite Evidence blocks that reflect this dynamism.

Profile Identifier This profile applies to Reference Values from a CoRIM manifest that a Verifier uses to process Evidence. Profile identifier structures are defined by CoRIM , EAT and Concise Evidence .

Intel Profile The profile identifier for the Intel Profile is the OID: {joint-iso-itu-t(2) country(16) us(840) organization(1) intel(113741) (1) intel-comid(16) profile(1)} 2.16.840.1.113741.1.16.1

Profile Specific Media and Content Types This profile uses the following media types:

• "application/eat+cwt"
• "application/eat+cwt; eat_profile=2.16.840.1.113741.1.16.1"
• "application/rim+cbor" (TBA)
• "application/rim+cbor" (TBA); profile=2.16.840.1.113741.1.16.1"

This profile uses the following content formats:

Content Type	C-F ID	TN Function
"application/eat+cwt"	263	1668547081
"application/eat+cwt; eat_profile=2.16.840.1.113741.1.16.1"	10005	1668556861
"application/toc+cbor"	10570	1668557428
"application/ce+cbor"	10571	1668557429

This profile uses the following top level CBOR tags (not already listed):

• 501, 570, 571

Attester Anatomy Attesters implement DICE layering using an initial Attesting Environment, also called a Root of Trust (RoT), that collection claims about one or more Target Environments. A Target Environment may become an Attesting Environment for a subsequent Target Environment, and so forth. There may be more than one RoT in the same Attester. Attesting Environments generate Evidence by signing collected claims using an Attestation Key. Environments may have other keys besides attestation keys. Keys can be regarded as claims that are collected and reported as Evidence. Keys can also be regarded as Target Environments that have measurements that are specific to the key. Confidential computing environments are Target Environments that can dynamically request Evidence from an Attesting Environment agent. Such Evidence may also be referred to as a 'Quote'. Each DICE layer may produce signed Evidence. Evidence formats include both signature and measurements formats. Signature formats may include a mix of X.509 certificates and EAT CWTs. Evidence measurements formats may include a mix of ASN.1 and CBOR, where ASN.1 uses DiceTcbInfo and related varients and CBOR uses concise evidence, and CMW. Multiple Evidence blocks may be bundled using CMW collections. Target Environments (other than cryptographic keys) are primarily identified using OIDs from Intel's OID arc (2.16.840.1.113741). Keys are identified using key identifiers, public key, or certificate digests as defined by $crypto-key-type-choice .

Evidence Profile Evidence may be integrity protected in various ways including: certificates , SPDM transcript , and CBOR web token (CWT) . Evidence contained in a certificate may be encoded using DiceTcbInfo and DiceTcbInfoSeq . Evidence contained in an SPDM payload may be encoded using the SPDM Measurement Block . Evidence may be formatted as concise-evidence and included in an alias certificate or an SPDM Measurement Manifest. The DiceTcbInfo and SPDM Evidence formats can be translated to CoMID. The concise evidence format is native to CoMID. This profile documents evidence mapping from DiceTcbInfo and SPDM Measurement Block to CoMID, as defined by . The CoMID extensions defined by this profile are applied to concise-evidence so that Verifiers that support this profile can consistently apply a common schema across Evidence, Reference Values, and Endorsements.

Evidence Hierarchy Evidence hierarchy refers to DICE layering where the platform bootstrap components double as Attesting Environments that collect measurements of the other bootstrap components (as Target Environments) until the quoting agent (e.g., SGX Quoting Enclave (QE), TDX Quoting TD (QTD)) is initialized. Tenant trusted execution environment (TEE) components can be dynamically loaded then request Evidence from its quoting agent. Quoting agents locally verify then sign measurments using the QTD / QE attestation key. A hierarchy of Evidence consisting of all the Evidence from a RoT to the tenant environment describes the Attester. A complex device may have multiple roots of trust, such as , each contributing an evidence hierarchy that results in several Evidence "chains", that together, constitute a complete Evidence hierarchy for the Attester device. The Evidence hierarchy should form a spanning tree that contains all Attester Evidence. All Attesting Environments within the device produce the spanning tree. CoRIM manifests contain Reference Values for the spanning tree so that Verifiers do not assume the spanning tree is defined by Evidence. Note that a failure or comporomise within the Attester device could result in a portion of the spanning tree being omitted. Evidence examples:

• A DICE certificate chain with a DiceTcbInfo extension, a DiceTcbInfoSeq extension, and a ConceptualMessageWrapper (CMW) extension containing a CBOR-encoded tagged-concise-evidence.
• An SPDM alias intermediate certification chain containing a CMW extension, and an SPDM measurement manifest containing tagged-concise-evidence.

Concise Evidence Concise evidence is a CDDL representation of Evidence that uses expressions from CoMID, which are a subset of CoRIM. See and . Evidence describes the actual state of the Attester. tagged-concise-evidence uses a CBOR tag (571) to identify concise-evidence . This profile uses concise-evicence in conceptual message wrappers and EAT tokens to encode Evidence. This profile extends concise-evidence by extending measurement-values-map.

Reference Values and Endorsements Profile The CoRIM specifications and define a baseline schema for Reference Values and Endorsements in this profile. The profile defines extensions to CoRIM for measurement types that are not representable by CoRIM or are more conveniently represented. This profile doesn't require use of extensions when base capabilities will suffice.

Concise Module ID Tag (CoMID) This profile uses concise-mid-tag in conceptual message wrappers and CoRIMs. This profile extends concise-mid-tag by extending measurement-values-map. Several extensions define two forms, one for representing actual state which is used for Endorsements and Evidence. The other form is used to represent reference state which is used for Reference Values.

Raw Value Measurements Raw value measurements encode vendor-defined values opaquely. However, the mkey value can add vendor-specific semantics when used with raw-value and name measurement types. Additionally, specific environment-map values can supply vendor-specific semantics to raw-value and name measurement types. Environments that project vendor-specific semantics are as follows:

Envoronment Identifier	Value	Semantics
class-id:OID=2.16.840.1.113741.1.5.3.6.8	560(bytes)	device type

CoRIM Extensions The Intel Profile extends measurement-values-map which is used by Evidence, Reference Values, and Endorsed Values by defining code points from the negative integer range. Reference Values extensions define types that can have multiple Reference Values that "match" a singleton Evidence value called "non-exact match" matching. Reference state expressions define non-exact-match matching semantics in terms of numeric ranges, time, sets, and masks.

Data Types

Masked Values Masked values are a string of bytes (e.g., bstr) that may have a companion mask value. The mask indicates which bits in the value are ignored when doing bit-wise equivalency comparisons. Verifier matching applies the equivalency test, allowing dissimilar Evidence and Reference values to be considered equivalent even if the two values (Evidence and Reference) are dissimilar. Evidence typically does not supply a mask. A Reference Value may omit the mask if bit-wise equivalency is desired. The $masked-value-type type choice can be either ~tagged-bytes or $raw-value-type-choice. Evidence might be encoded as ~tagged-bytes or tagged-bytes which omits a mask value, while Reference Values of type tagged-masked-raw-value includes the mask value. The Verifier MUST ensure the lengths of values and mask are equivalent. If the mask is shorter than the longest value, the mask is appended with zeros (0) until it is the same length as the longest value, either Evidence or Reference Value. If the mask is longer than the longest value, the mask is truncated to the length of the longest value. All values are evaluated from left to right (big-endian) applying bit-wise comparisons. The masked value data types are as follows:

Expressions Expressions are used to specify richer Reference Values measurements that expect non-exact matching semantics. The Reference Value expression instructs the Verifier regarding matching parameters, such as greater-than or less-than, ranges, sets, etc. Typically, the Evidence value is one operand of an expression and the Reference Value contains the operator and additional operands. The operator and remaining operands are contained in an array. Expression arrays have an operator followed by zero or more operands. The operator definition identifies the additional operands and their data types. A Verifier forms an expression using Evidence as the first operand, obtains the operator from the first entry in the expression array, and any remaining array entries are operands. This document describes operations using infix notation where the first operand, operand_1, is obtained from Evidence, followed by the operator, followed by any remaining operands: operand_2, operand_3, etc... For example:

• From Evidence: operand_1, from Reference Values: [ operator, operand_2, operand_3, ... ].

Expression records are CBOR tagged to indicate the following CBOR is to be evaluated as an expression. Expression statements found in Reference Values informs the Verifier that Evidence is needed to complete the expression equation. Appraisal processing MUST evaluate the expression. This profile anticipates use of the CBOR tag #6.600XX to identify expression arrays. See . For example:

• #6.600XX([ operator, operand_2, operand_3, ... ]).

Expression Operators There are two classes of operators as follows:

1. Numeric: The numeric operand can be an integer, unsigned integer, or floading point value.
2. Set: The set operand can be an set (array) of any primitive value or a set of arrays containing any primitive value. The position of the items in a set is unordered, while the position of items in an array within a set is ordered.

Equivalence Operator By default, exact match rules are assumed. Consequently, no operator artifact is needed when Evidence values are identical to Reference Values.

Numeric Expressions Numeric operators apply to values that are integers, unsigned integers or floating point numbers. There are four numeric operators:

1. greater-than (gt),
2. greater-than-or-equal (ge),
3. less-than (lt),
4. less-than-or-equal (le).

The equals operator is not defined because an exact match rule is the default rule when an Evidence value is identical to a Reference Value. The numeric operator data type definitions are as follows: A numeric expression is an array containing a numeric operator and a numeric operand. The operand contains a numeric Reference Value that is matched with a numeric Evidence value. Evidence and Reference Values MUST be the same numeric type. For example, if a Reference Value numeric type is integer, then the Evidence numeric value must also be integer. This profile defines four macro numeric expressions, one for each numeric operator:

• tagged-numeric-gt,
• tagged-numeric-ge,
• tagged-numeric-lt,
• tagged-numeric-le.

In each case, the numeric operator is used to evaluate a Reference Value operand against an Evidence value operand that is obtained from Evidence. If the expression is read using infix notation, the first operand is Evidence, followed by the operator, followed by the Reference Value operand. Example:

• <evidence_numeric> <le> <reference_numeric>

The numeric expression definitions are as follows:

Set Expressions Set operators allow Reference Values, that are expressed as a set, to be compared with Evidence that is expressed as either a primitive value or a set. Set expression statements have two forms:

1. A binary relation between a primitive object 'O' and a set 'S', and
2. A binary relation between two sets, an Evidence set 'S1' and a Reference Values set 'S2'.

The first form, relation between an object and a set, has two operators:

• member and
• not-member.

The Evidence object 'O' is evaluated with the Reference Values set 'S'. The op.member and op.not-member operators expect a Reverence Value set as operand_2 and a primitive Evidence value as operand_1. Evaluation tests whether operand_1 is a member of the set operand_2. Example:

• <evidence-value> <set-operator> <reference-set>

The set data type definitions are defined as follows: A set expression array contins a set operator followed by a set of Reference Values. The set is defined by set-type. The set expression type definitions are as follows: Examples:

• <evidence_object> <member> <reference_set>
• <evidence_object> <not-member> <reference_set>

The Evidence object MUST NOT be nil. The Reference Values set may be the empty set. The second form, a relation between two sets, has three operators:

• subset,
• superset,
• disjoint.

The fist set, S1 is Evidence and set, S2 is the Reference Values set. The subset, superset, and disjoint operators test whether an Evidence set value satisfies a set operation, given a Reverence Value set. Examples:

• <evidence_set> <subset> <reference_set>
• <evidence_set> <superset> <reference_set>
• <evidence_set> <disjoint> <reference_set>

The Reference Values and Evidence sets may be the empty set. The set of sets data type definitions are as follows: For subset, every member in the Evidence set 'S1', MUST map to a member in the Reference Values set 'S2'. The Evidence set, 'S1', MAY be a proper subset of 'S2'. For example, if every member in set 'S1' maps to every member in set 'S2' then matching is successful. A successful result produces the set 'S1'. For superset, every member in the Reference Values set 'S2', MUST map to a member in the Evidence set 'S1'. A successful result produces the set 'S1'. For disjoint, every member in the Evidence set 'S1' MUST NOT map to any member in the Reference Values set 'S2'. A successful result produces the empty set. The set of sets expression definitions are as follows:

Measurement Extensions This profile extends the CoMID measurement-values-map with additional code point definitions, that accommodate Intel SGX and similar architectures. Measurement extensions don't change Verifier behavior. An extension enables the Verifier to validate the profile compliance of the input evidence and reference values, as it defines the acceptable data types in evidence and the expression operator that is explicitly supplied with the Reference Values, see . In cases where Evidence does not exactly match Reference Values, the operator definition determines the expected data types of the operands. Expected Verifier behavior is defined in The measurement extensions that follow are assumed to be appraised according to the appriasal steps described in .

The tee.advisory-ids Measurement Extension The tee.advisory-ids extension enables Attesters to report known security advisories and for Reference Values Providers (RVP) to assert updated security advisories. It can also be used by Endorsers to assert security advisory information through conditional endorsement. The $tee-advisory-ids-type is used to specify a set of security advisories, where each identifier is represented using a string. Evidence may report a set of advisories the Attester believes are relevant. The set of advisories are constrained by the set-tstr-type structure. As a Reference Value expression, an empty set can be used to signify that no outstanding advisories are expected. If the Evidence also contains the empty set then the Reference corroborates the Evidence. The $tee-advisory-ids-type is a list of strings, each identifying a single security advisory. When used with Evidence the set-tstr-type type is used. When used with Reference Values or Endorsements the set-tstr-type, tagged-exp-tstr-member, or tagged-exp-tstr-not-member types can be used. $tee-advisory-ids-type ) $tee-advisory-ids-type /= set-tstr-type $tee-advisory-ids-type /= tagged-exp-tstr-not-member $tee-advisory-ids-type /= tagged-exp-tstr-member ]]>

The tee-advisory-ids-type Comparison Algorithm The comparison algorithm for tee-advisory-ids-type is used when Endorsement or Reference Values triples conditions are matched with an Environment Claims Tuple (ECT) in the Verifier's Accepted Claims Set (ACS). The triple condition containing a tee-advisory-ids-type Claim matches an ACS ECT according to the comparison algorithm for set of strings as defined in .

The tee.attributes Measurement Extension The tee.attributes extension enables the Attester to report TEE attributes and an RVP to assert a reference TEE attributes and mask. The $tee-attributes-type is used to specify TEE attributes in 8 or 16 byte words. If Evidence uses an 8 byte mask, then the Reference Values expression also uses an 8 byte value and mask. The $tee-attributes-type is a singleton value omitting the mask value when used as Endorsement or Evidence and a tuple containing the reference and mask when used as a Reference Value. $tee-attributes-type ) $tee-attributes-type /= $masked-value-type ]]> Alternatively, the TEE attributes may be encoded using mkey where mkey contains the non-negative tee.attributes and mval.raw-value contains the $tee-attributes-type.mask-type value.

The tee.cryptokeys Measurement Extension The tee.cryptokeys extension identifies cryptographic keys associated with a Target Environment. If multiple $crypto-key-type-choice measurements are supplied, array position disambiguates each entry. Appraisal compares values indexed by array position. [ + $tee-cryptokey-type ] ) $tee-cryptokey-type /= $crypto-key-type-choice ]]> Alternatively, the TEE cryptokeys may be encoded using mkey where mkey contains the non-negative tee.cryptokeys and mval.cryptokeys contains the $tee-cryptokey-type value.

The tee.tcbdate Measurement Extension The tee.tcbdate (code point -72) extension is used by Endorsers to assert validity of a TEE component. For example, a conditional endorsement might locate a component based on a few expected Claims, then augment them with a tee.tcbdate Claim. The $tee-date-type can be expressed in several ways:

• ISO 8601 strings of the form YYYY-MM-DDTHH:MM:SSZ.
• POSIX time which is the number of seconds since January 1, 1970 (midnight UTC).
• RFC9581 etime .

$tee-date-type ) $tee-date-type /= ~tdate $tee-date-type /= tdate $tee-date-type /= time $tee-date-type /= etime ; RFC9581 ]]> ~tdate strings must be converted to a numeric value (i.e.,~time) before operations involving time are applied. Alternatively, tee.tcbdate may be encoded using mkey where mkey contains the non-negative code point value and where mval.name contains the string representation $tee-date-type without the CBOR tag (i.e., ~tdate - see Section 3.7 ).

The tee.mrtee and tee.mrsigner Measurement Extension The tee.mrtee extension enables an Attester to report digests for the SGX enclave or TDX TD (e.g., MRENCLAVE, MRTD). The tee.mrsigner extension enables an Attester to report the signer of the TEE digest (e.g., MRSIGNER). The $tee-digest-type has multiple type structures involving digest values. A singleton digest has a hash algorithm identifier and the digest value. When used as Evidence, either a signleton digest or a set of digests can be reported. When used as Reference Values or Endorsements, a set of digests can be asserted signifying equivalence matching. Alternatively, matching may be expressed as set membership or set difference expressions. $tee-digest-type ) $$measurement-values-map-extension //= ( &(tee.mrsigner: -84) => $tee-digest-type ) $tee-digest-type /= digest ; see corim $tee-digest-type /= digests-type ; see corim $tee-digest-type /= set-digest-type $tee-digest-type /= tagged-exp-digest-member $tee-digest-type /= tagged-exp-digest-not-member ]]> Alternatively, the TEE digests may be encoded using mkey where mkey contains the non-negative tee.mrtee or tee.mrsigner and mval.digests contains a digests-type value.

The tee-digest-type Comparison Algorithm The comparison algorithm for tee-digest-type is used when the condition statement in an Endorsement or Reference Values triple is matched with an Environment Claim Tuple (ECT) from the Verifier's Accepted Claims Set (ACS). The comparison algorithm for set of digests is defined in .

The tee.instance-id Measurement Extension The tee.instance-id extension enables the Attester to report the TEE (TD or enclave) instance identifier as an Evidence value and the RVP to assert an exact-match Reference Value. The $tee-instance-id-type is an unsigned integer or bstr. $tee-instance-id-type ) $tee-instance-id-type /= uint $tee-instance-id-type /= bstr ]]> Alternatively, the TEE instance ID may be encoded using mkey where mkey contains the non-negative tee.instance-id and mval.raw-value contains the $tee-instance-id-type value. If the $tee-instance-id-type is an unsigned integer, the integer values is converted to a single byte bstr.

The tee.platform-instance-id Measurement Extension Platform Instance ID is a globally unique identifier generated by the platform during Platform Establishment. This value remains consistent across trusted computing base (TCB) recoveries, but is regenerated during Platform Establishment due to desire to reset keys or to add and remove hardware. See (Section 3.7 ). The tee.platform-instance-id extension enables the Attester to report the platform instance identifier as an Evidence value and the RVP to assert an exact-match Reference Value. The $tee-platform-instance-id-type is a bstr. $tee-platform-instance-id-type ) $tee-platform-instance-id-type /= bstr ]]> Alternatively, the platform instance ID may be encoded using mkey where mkey contains the non-negative tee.platform-instance-id and mval.raw-value contains the $tee-platform-instance-id-type value.

The tee.isvprodid Measurement Extension The tee.isvprodid extension enables the Attester to report the ISV product identifier Evidence value and the RVP to assert an exact-match Reference Value. The $tee-isvprodid-type is an unsigned integer. The $tee-isvprodid-type is an exact match measurement. $tee-isvprodid-type ) $tee-isvprodid-type /= uint $tee-isvprodid-type /= bstr ]]> Alternatively, the TEE product ID may be encoded using mkey where mkey contains the non-negative tee.isvprodid and mval.raw-value contains the $tee-isvprodid-type value.

The tee.miscselect Measurement Extension The tee.miscselect extension enables the Attester to report the (TBD:miscselect-description) Evidence value and the RVP to assert a Reference Value and mask. The $tee-miscselect-type is a 4 byte value and mask. The $tee-miscselect-type is a singleton mask-type value when used as Endorsement or Evidence and a tagged-masked-raw-value when used a Reference Value. $tee-miscselect-type ) $tee-miscselect-type /= $masked-value-type ]]> Alternatively, the TEE miscselect may be encoded using mkey where mkey contains the non-negative tee.miscselect and mval.raw-value contains the measurement value and mval.raw-value-mask` contains the mask value.

The tee.model Measurement Extension The tee.model extension enables the Attester to report the TEE model string as Evidence and the RVP to assert an exact-match Reference Value. The $tee-model-type is a string. The $tee-model-type is an exact match measurement. $tee-model-type ) $tee-model-type /= tstr ]]> Alternatively, the TEE model may be encoded using mkey where mkey contains the non-negative tee.model and mval.name contains the $tee-model-type value.

The tee.pceid Measurement Extension The tee.pceid extension enables the Attester to report the PCEID as Evidence and the RVP to assert an exact-match Reference Value. The $tee-pceid-type is a string. The $tee-pceid-type is an exact match measurement. $tee-pceid-type ) $tee-pceid-type /= tstr ]]> Alternatively, the PCEID may be encoded using mkey where mkey contains the non-negative tee.pceid and mval.name contains the $tee-pceid-type value.

The tee.isvsvn Measurement Extension The tee.isvsvn extension enables the Attester to report the SVN for the independent software vendor supplied component as Evidence and the RVP to assert a Reference Value that is greater-than-or-equal to the reported SVN. The $tee-svn-type is either an unsigned integer when reported as Evidence, or a tagged numeric expression that contains an SVN and a numeric greater-than-or-equal operator. The Verifier ensures the Evidence value is greater-that-or-equal to the Reference Value. The $tee-svn-type is a svn-type when used as Endorsement or Evidence and a tagged-numeric-expression when used as a Reference Value. $tee-svn-type ) $tee-svn-type /= svn-type .within numeric-type $tee-svn-type /= tagged-numeric-ge ]]> Alternatively, the TEE isvsvn may be encoded using mkey where mkey contains the non-negative tee.isvsvn and mval.svn contains the svn value as svn-type.

The tee.tcb-comp-svn Measurement Extension The tee.tcb-comp-svn extension enables the Attester to report an array of SVN values for the TCB when asserted as Evidence and an array of tagged-numeric-ge entries when asserted as a Reference Value. The $tee-tcb-comp-svn-type is an array containing 16 SVN values when reported as Evidence and an array of 16 expression records each containing the numeric ge operator and a reference SVN value. The Verifier evaluates each SVN in the Evidence array with the corresponding reference expression, by array position. If all Evidence values match their respective expressions, evaluation is successful. The array of SVN Evidence is accepted. $tee-tcb-comp-svn-type ) $tee-tcb-comp-svn-type /= [ 16*16 svn-type .within numeric-type ] $tee-tcb-comp-svn-type /= [ 16*16 tagged-numeric-ge ] ]]>

The tee.tcb-eval-num Measurement Extension The tee.tcb-eval-num extension enables the Attester to report a TCB evaluation number as Evidence and the RVP to assert a Reference Value expression that compares the tcb-eval-num Evidence with the Reference Value using the greater-than-or-equal operator. The $tee-tcb-eval-num-type is an unsigned integer when reported as Evidence and a tagged numeric expression when asserted as Reference Values. $tee-tcb-eval-num-type ) $tee-tcb-eval-num-type /= uint .within numeric-type $tee-tcb-eval-num-type /= tagged-numeric-ge ]]> Alternatively, the TEE tcb-eval-num Evidence may be encoded using mkey where mkey contains the non-negative tee.tcb-eval-num and mval.raw-value contains the tcb-eval-num encoded as 4-byte bstr value.

The tee.tcb-status Measurement Extension The tee.tcb-status extension enables Attesters to report the status of the TEE trusted computing base (TCB) and for Reference Value Providers (RVP) to assert expected TCB status. It can also be used by Endorsers to assert TCB status through conditional endorsement. The tee-tcbstatus-type is used to specify TCB status as a set of status strings or as an expression with a set membership operator. The $tee-tcbstatus-type is a status array containing strings describing TCB status values. When describing Evidence the set-tstr-type type is used. When describing Reference Values or Endorsements the set-tstr-type, tagged-exp-tstr-member, or tagged-exp-tstr-not-member types can be used. $tee-tcbstatus-type ) $tee-tcbstatus-type /= set-tstr-type $tee-tcbstatus-type /= tagged-exp-tstr-member $tee-tcbstatus-type /= tagged-exp-tstr-not-member ]]>

The tee-tcbstatus-type Comparison Algorithm The comparison algorithm for tee-tcbstatus-type is used when Endorsement or Reference Values triples conditions are matched with an Environment Claims Tuple (ECT) in the Verifier's Accepted Claims Set (ACS). The triple condition containing a tee-tcbstatuss-type Claim matches an ACS ECT according to the comparison algorithm for set of strings as defined in .

The tee.vendor Measurement Extension The tee.vendor extension enables the Attester to report the TEE vendor name as Evidence and for the RVP to assert the TEE vendor name. The $tee-vendor-type is a string containing the vendor name as a string. The vendor string in Evidence must exactly match the vendor string in Reference Values. The $tee-vendor-type is an exact match measurement. $tee-vendor-type ) $tee-vendor-type /= tstr ]]> Alternatively, the TEE vendor may be encoded using mkey where mkey contains the non-negative tee.vendor and mval.name contains the $tee-vendor-type value.

Appraisal Algorithm The Intel profile anticipates appraisal algorithms will be based on the appraisal algorithm defined in . This profile extends the appraisal algorithm to recognize profile extensions that form equations. An Evidence measurement forms one of the operands: (evidence operand). A Reference Value forms the operator and remaining operands:

• [expression operator, reference value operand, etc...]

For example, if a numeric Reference Value is 14, and the expressions operator is gt the Reference Value might contain the Claim: #6.60010([ 1, 14]). Given Evidence contains the value: 15. The in-fix construction of the equation would be: 15 gt 14. The Verifier evaluates whether 15 is greater-than 14.

Complex Expressions Complex expressions can be used to assess whether the Target Environment is in a particular state before certain Endorsement claims can be asserted. For example, if an SGX enclave has an svn value that is less than the prescribed minimum svn, the enclave status may be considered "OutOfDate" or may have a known security advisory. The CoMID conditional-endorsement-triples or conditional-endorsement-series-triples describe complex Endorsement expressions. This profile uses these triples with the reference measurement values extensions described in .

Comparison Algorithm for Sets The comparison algorithm for sets describes set equivalence, set membership, and set difference (not membership). The Verifier's Accepted Claims Set (ACS) contains a list of Environment Claims Tuples (ECT). The condition ECTs are compared to ACS ECTs based on this comparison algorithm. The set comparison algorithm processes sets of strings and sets of digests.

Comparison Algorithm for Set of Strings There are three string set representations: set-tstr-type, tagged-exp-tstr-member, and tagged-exp-tstr-not-member.

• set-tstr-type - Every string in the condition set-tstr-type MUST match a string in the ACS.ECT.element-map.element-claims.set-tstr-type set. The string position in the array is not significant. The ACS.ECT.element-map.element-claims.set-tstr-type set MUST be equivalent to the condition set-tstr-type set (i.e., the two sets have the same cardinality and the same set members).
• tagged-exp-tstr-member - The condition ECT set operator MUST equal member and every string in the condition set-tstr-type MUST match a string in the ACS.ECT.element-map.element-claims.set-tstr-type set. The string position in the array is not significant. The ACS.ECT.element-map.element-claims.set-tstr-type set MAY contain strings not found in the condition set-tstr-type.
• tagged-exp-tstr-not-member - The condition ECT set operator MUST equal not-member and every string in the condition set-tstr-type MUST NOT match a string in the ACS.ECT.element-map.element-claims.set-tstr-type set. The string position in the array is not significant.

Comparison Algorithm for Set of Digests There are five digest set representations: digest, digest-type, set-digest-type, tagged-exp-digest-member, and tagged-exp-digest-not-member.

• digest - The singleton digest in the condition MUST match at least one digest in the ACS.ECT.element-map.element-claims.set-digest-type set.
• digest-type and set-digest-type - Every digest in the condition digest-type or set-digest-type MUST match a digest in the ACS.ECT.element-map.element-claims.set-digest-type set. The digest position in the array is not significant. The ACS.ECT.element-map.element-claims.set-digest-type set MUST be equivalent to the condition set-digest-type set (i.e., the two sets have the same cardinality and the same set members). Matching based on the empty set is permitted when the set-digest-type is used.
• tagged-exp-digest-member - The condition ECT set operator MUST equal member and every digest in the condition set-digest-type MUST match a digest in the ACS.ECT.element-map.element-claims.set-digest-type set. The digest position in the array is not significant. The ACS.ECT.element-map.element-claims.set-digest-type set MAY contain digests not found in the condition set-digest-type.
• tagged-exp-digest-not-member - The condition ECT set operator MUST equal not-member and every digest in the condition set-digest-type MUST NOT match a digest in the ACS.ECT.element-map.element-claims.set-digest-type set. The digest position in the array is not significant.

Reporting Attestation Results Attestation verification can be performed by a pipeline consisting of multiple stages where each input manifest demarks a stage. The final stage prepares Attestation Results according to Relying Party specifications. This profile does not define an attestation results format. The Relying Party should specify suitable Attestation Results formats such as or . The precise Attestation Results format used, if negotiated by Verifier and Relying Party, should reference this profile to acknowledge that the Relying Party and Verifier both support the extensions defined in this document.

Security Considerations The security of this profile depends on the security considerations of the various normative references.

IANA Considerations IANA is requested to allocate the following tags in the CBOR Tags registry , preferably with the CBOR tag value requested.

Tag	Data Item	Semantics	Reference
60010	tag	Numeric Expressions, see	RFCthis
60020	tag	Set digest Expressions, see	RFCthis
60021	tag	Set tstr Expressions, see	RFCthis

References Normative References Fraunhofer SIT Linaro arm Intel Huawei Technologies Remote Attestation Procedures (RATS) enable Relying Parties to assess the trustworthiness of a remote Attester and therefore to decide whether or not to engage in secure interactions with it. Evidence about trustworthiness can be rather complex and it is deemed unrealistic that every Relying Party is capable of the appraisal of Evidence. Therefore that burden is typically offloaded to a Verifier. In order to conduct Evidence appraisal, a Verifier requires not only fresh Evidence from an Attester, but also trusted Endorsements and Reference Values from Endorsers and Reference Value Providers, such as manufacturers, distributors, or device owners. This document specifies the information elements for representing Endorsements and Reference Values in CBOR format. Trusted Computing Group (TCG) Trusted Computing Group (TCG) Trusted Computing Group (TCG) Trusted Computing Group (TCG) Fraunhofer SIT Intel Linaro University of Applied Sciences Bonn-Rhein-Sieg Google LLC This document defines a conceptual message wrapper (CMW) format - an encapsulation method applicable to any RATS conceptual message, such as Evidence, Attestation Results, Endorsements, and Reference Values. It also describes a collection type that aggregates one or more CMWs into a single message. In addition, this document specifies a corresponding CBOR tag, JSON Web Tokens (JWT) and CBOR Web Tokens (CWT) claims, and an X.509 extension. These mechanisms enable the embedding of enveloped conceptual messages into CBOR-based protocols, web APIs, and PKIX protocols. Moreover, a Media Type and a CoAP Content-Format are defined for transporting CMWs over HTTP, MIME, CoAP, and other Internet protocols. Security Theory LLC Mediatek USA Qualcomm Technologies Inc. Red Hound Software, Inc. An Entity Attestation Token (EAT) provides an attested claims set that describes state and characteristics of an entity, a device like a smartphone, IoT device, network equipment or such. This claims set is used by a relying party, server or service to determine the type and degree of trust placed in the entity. An EAT is either a CBOR Web Token (CWT) or JSON Web Token (JWT) with attestation-oriented claims. The Concise Binary Object Representation (CBOR, RFC 8949) is a data format whose design goals include the possibility of extremely small code size, fairly small message size, and extensibility without the need for version negotiation. In CBOR, one point of extensibility is the definition of CBOR tags. RFC 8949 defines two tags for time: CBOR tag 0 (RFC 3339 time as a string) and tag 1 (POSIX time as int or float). Since then, additional requirements have become known. The present document defines a CBOR tag for time that allows a more elaborate representation of time, as well as related CBOR tags for duration and time period. This document is intended as the reference document for the IANA registration of the CBOR tags defined. In many standards track documents several words are used to signify the requirements in the specification. These words are often capitalized. This document defines these words as they should be interpreted in IETF documents. This document specifies an Internet Best Current Practices for the Internet Community, and requests discussion and suggestions for improvements. RFC 2119 specifies common key words that may be used in protocol specifications. This document aims to reduce the ambiguity by clarifying that only UPPERCASE usage of the key words have the defined special meanings. IANA Informative References In network protocol exchanges, it is often useful for one end of a communication to know whether the other end is in an intended operating state. This document provides an architectural overview of the entities involved that make such tests possible through the process of generating, conveying, and evaluating evidentiary Claims. It provides a model that is neutral toward processor architectures, the content of Claims, and protocols. This memo profiles the X.509 v3 certificate and X.509 v2 certificate revocation list (CRL) for use in the Internet. An overview of this approach and model is provided as an introduction. The X.509 v3 certificate format is described in detail, with additional information regarding the format and semantics of Internet name forms. Standard certificate extensions are described and two Internet-specific extensions are defined. A set of required certificate extensions is specified. The X.509 v2 CRL format is described in detail along with standard and Internet-specific extensions. An algorithm for X.509 certification path validation is described. An ASN.1 module and examples are provided in the appendices. [STANDARDS-TRACK] CBOR Web Token (CWT) is a compact means of representing claims to be transferred between two parties. The claims in a CWT are encoded in the Concise Binary Object Representation (CBOR), and CBOR Object Signing and Encryption (COSE) is used for added application-layer security protection. A claim is a piece of information asserted about a subject and is represented as a name/value pair consisting of a claim name and a claim value. CWT is derived from JSON Web Token (JWT) but uses CBOR rather than JSON. This document proposes a notational convention to express Concise Binary Object Representation (CBOR) data structures (RFC 7049). Its main goal is to provide an easy and unambiguous way to express structures for protocol messages and data formats that use CBOR or JSON. Cisco Systems Fraunhofer SIT MIT Linaro Intel This document defines reusable Attestation Result information elements. When these elements are offered to Relying Parties as Evidence, different aspects of Attester trustworthiness can be evaluated. Additionally, where the Relying Party is interfacing with a heterogeneous mix of Attesting Environment and Verifier types, consistent policies can be applied to subsequent information exchange between each Attester and the Relying Party. This document also defines two serialisations of the proposed information model, utilising CBOR and JSON. Distributed Managability Task Force (DMTF) Trusted Computing Group (TCG) Microsoft Microsoft Intel Intel Intel Intel® Trust Domain Extensions (TDX) introduce architectural elements designed for the deployment of hardware-isolated virtual machines (VMs) known as trust domains (TDs). TDX is designed to provide a secure and isolated environment for running sensitive workloads or applications. This Entity Attestation Token (EAT) profile outlines claims for an Intel TDX attestation result which facilitate the establishment of trust between a relying party and the environment. Armidale Consulting Fraunhofer SIT Linaro In the IETF Remote Attestation Procedures (RATS) architecture, a Verifier accepts Evidence and, using Appraisal Policy typically with additional input from Endorsements and Reference Values, generates Attestation Results in formats that are useful for Relying Parties. This document illustrates the purpose and role of Endorsements and discusses some considerations in the choice of message format for Endorsements in the scope of the RATS architecture. This document does not aim to define a conceptual message format for Endorsements and Reference Values. Instead, it extends RFC9334 to provide further details on Reference Values and Endorsements, as these topics were outside the scope of the RATS charter when RFC9334 was developed. Intel Corporation

Acknowledgments The authors wish to thank Shanwei Cen, Piotr Zmijewski, Francisco J. Chinchilla and Dionna Amalie Glaze for their valuable contributions.

Full Intel Profile CDDL any ; at least one base time * (nint/text) => any ; elective supplementary information * uint => any ; critical supplementary information } etime-detailed = ({ $$ETIME-BASETIME ClockQuality-group * $$ETIME-ELECTIVE * $$ETIME-CRITICAL * ((nint/text) .feature "etime-elective-extension") => any * (uint .feature "etime-critical-extension") => any }) .within etime-framework $$ETIME-BASETIME //= (1: ~time) $$ETIME-BASETIME //= (4: ~decfrac) $$ETIME-BASETIME //= (5: ~bigfloat) $$ETIME-ELECTIVE //= (-3: uint) $$ETIME-ELECTIVE //= (-6: uint) $$ETIME-ELECTIVE //= (-9: uint) $$ETIME-ELECTIVE //= (-12: uint) $$ETIME-ELECTIVE //= (-15: uint) $$ETIME-ELECTIVE //= (-18: uint) $$ETIME-ELECTIVE //= (-1 => $ETIME-TIMESCALE) $$ETIME-ELECTIVE //= (-13 => $ETIME-TIMESCALE) $$ETIME-CRITICAL //= (13 => $ETIME-TIMESCALE) $ETIME-TIMESCALE /= &(etime-utc: 0) $ETIME-TIMESCALE /= &(etime-tai: 1) ClockQuality-group = ( ? &(ClockClass: -2) => uint .size 1 ; PTP/RFC8575 ? &(ClockAccuracy: -4) => uint .size 1 ; PTP/RFC8575 ? &(OffsetScaledLogVariance: -5) => uint .size 2 ; PTP/RFC8575 ? &(Uncertainty: -7) => ~time/~duration ? &(Guarantee: -8) => ~time/~duration ) Duration = #6.1002(etime-detailed) simple-Period = #6.1003([ start: ~Etime / null end: ~Etime / null ? duration: ~Duration ]) Period = #6.1003([ (start: ~Etime, ((end: ~Etime) // (end: null, duration: ~Duration))) // (start: null, end: ~Etime, duration: ~Duration) ]) etime = #6.1001({* (int/tstr) => any}) duration = #6.1002({* (int/tstr) => any}) period = #6.1003([~etime/null, ~etime/null, ?~duration]) member = 6 not-member = 7 set-operators /= op.mem set-operators /= op.nmem set-digest-type /= [ * digest ] set-digest-expression = [ set-operators, set-digest-type ] tagged-set-digest-expression = #6.60020( set-digest-expression ) set-tstr-type /= [ * tstr ] set-tstr-expression = [ set-operators, set-tstr-type ] tagged-set-tstr-expression = #6.60021( set-tstr-expression ) tagged-exp-digest-member = #6.60020([ op.mem .within set-operators, set-digest-type ]) tagged-exp-digest-not-member = #6.60020([ op.nmem .within set-operators, set-digest-type ]) tagged-exp-tstr-member = #6.60021([ op.mem .within set-operators, set-tstr-type ]) tagged-exp-tstr-not-member = #6.60021([ op.nmem .within set-operators, set-tstr-type ]) subset = 8 superset = 9 disjoint = 10 set-of-set-type = [ * [ + any ]] set-of-set-operator = subset / superset / disjoint set-of-set-expression = [ set-of-set-operator, set-of-set-type ] tagged-set-of-set-expression = #6.60020(set-of-set-expression) tagged-exp-subset = #6.60010([ subset .within set-of-set-operator, set-of-set-type ]) tagged-exp-superset = #6.60010([ superset .within set-of-set-operator, set-of-set-type ]) tagged-exp-disjoint = #6.60010([ disjoint .within set-of-set-operator, set-of-set-type ]) $masked-value-type /= ~tagged-bytes $masked-value-type /= $raw-value-type-choice $$measurement-values-map-extension //= ( &(tee.advisory-ids: -89) => $tee-advisory-ids-type ) $tee-advisory-ids-type /= set-tstr-type $tee-advisory-ids-type /= tagged-exp-tstr-not-member $tee-advisory-ids-type /= tagged-exp-tstr-member $$measurement-values-map-extension //= ( &(tee.attributes: -82) => $tee-attributes-type ) $tee-attributes-type /= $masked-value-type $$measurement-values-map-extension //= ( &(tee.cryptokeys: -91) => [ + $tee-cryptokey-type ] ) $tee-cryptokey-type /= $crypto-key-type-choice $$measurement-values-map-extension //= ( &(tee.tcbdate: -72) => $tee-date-type ) $tee-date-type /= ~tdate $tee-date-type /= tdate $tee-date-type /= time $tee-date-type /= etime ; RFC9581 $$measurement-values-map-extension //= ( &(tee.mrtee: -83) => $tee-digest-type ) $$measurement-values-map-extension //= ( &(tee.mrsigner: -84) => $tee-digest-type ) $tee-digest-type /= digest ; see corim $tee-digest-type /= digests-type ; see corim $tee-digest-type /= set-digest-type $tee-digest-type /= tagged-exp-digest-member $tee-digest-type /= tagged-exp-digest-not-member $$measurement-values-map-extension //= ( &(tee.instance-id: -77) => $tee-instance-id-type ) $tee-instance-id-type /= uint $tee-instance-id-type /= bstr $$measurement-values-map-extension //= ( &(tee.isvprodid: -85) => $tee-isvprodid-type ) $tee-isvprodid-type /= uint $tee-isvprodid-type /= bstr $$measurement-values-map-extension //= ( &(tee.miscselect: -81) => $tee-miscselect-type ) $tee-miscselect-type /= $masked-value-type $$measurement-values-map-extension //= ( &(tee.model: -71) => $tee-model-type ) $tee-model-type /= tstr op.eq=0 op.gt=1 op.ge=2 op.lt=3 op.le=4 op.mem=6 op.nmem=7 op.sub=8 op.sup=9 op.dif=10 $$measurement-values-map-extension //= ( &(tee.pceid: -80) => $tee-pceid-type ) $tee-pceid-type /= tstr $$measurement-values-map-extension //= ( &(tee.isvsvn: -73) => $tee-svn-type ) $tee-svn-type /= svn-type .within numeric-type $tee-svn-type /= tagged-numeric-ge $$measurement-values-map-extension //= ( &(tee.tcb-comp-svn: -125) => $tee-tcb-comp-svn-type ) $tee-tcb-comp-svn-type /= [ 16*16 svn-type .within numeric-type ] $tee-tcb-comp-svn-type /= [ 16*16 tagged-numeric-ge ] $$measurement-values-map-extension //= ( &(tee-tcb-eval-num: -86) => $tee-tcb-eval-num-type ) $tee-tcb-eval-num-type /= uint .within numeric-type $tee-tcb-eval-num-type /= tagged-numeric-ge $$measurement-values-map-extension //= ( &(tee.tcbstatus: -88) => $tee-tcbstatus-type ) $tee-tcbstatus-type /= set-tstr-type $tee-tcbstatus-type /= tagged-exp-tstr-member $tee-tcbstatus-type /= tagged-exp-tstr-not-member $$measurement-values-map-extension //= ( &(tee.vendor: -70) => $tee-vendor-type ) $tee-vendor-type /= tstr $$measurement-values-map-extension //= ( &(tee.platform-instance-id: -101) => $tee-platform-instance-id-type ) $tee-platform-instance-id-type /= bstr ]]>