This document describes the basic PKCS#11 token interface
and token behavior.
The PKCS#11 standard specifies an application programming
interface (API), called “Cryptoki,” for devices that hold cryptographic
information and perform cryptographic functions. Cryptoki follows a simple object
based approach, addressing the goals of technology independence (any kind of
device) and resource sharing (multiple applications accessing multiple
devices), presenting to applications a common, logical view of the device
called a “cryptographic token”.
This document specifies the data types and functions
available to an application requiring cryptographic services using the ANSI C
programming language. The supplier of a Cryptoki library implementation
typically provides these data types and functions via ANSI C header files.
Generic ANSI C header files for Cryptoki are available from the PKCS#11 web
page. This document and up-to-date errata for Cryptoki will also be available
from the same place.
Additional documents may provide a generic, language-independent
Cryptoki interface and/or bindings between Cryptoki and other programming
languages.
Cryptoki isolates an application from the details of the
cryptographic device. The application does not have to change to interface to
a different type of device or to run in a different environment; thus, the
application is portable. How Cryptoki provides this isolation is beyond the
scope of this document, although some conventions for the support of multiple
types of device will be addressed here and possibly in a separate document.
Details of cryptographic mechanisms (algorithms) may be
found in the associated PKCS#11 Mechanisms documents.
For the purposes of this standard, the following definitions
apply:
AES Advanced
Encryption Standard, as defined in FIPS PUB 197.
API Application
programming interface.
Application Any
computer program that calls the Cryptoki interface.
ASN.1 Abstract
Syntax Notation One, as defined in X.680.
Attribute A
characteristic of an object.
BER Basic
Encoding Rules, as defined in X.690.
BLOWFISH The
Blowfish Encryption Algorithm of Bruce Schneier, www.schneier.com.
CAMELLIA The
Camellia encryption algorithm, as defined in RFC 3713.
CBC Cipher-Block
Chaining mode, as defined in FIPS PUB 81.
Certificate A
signed message binding a subject name and a public key, or a subject name and a
set of attributes.
CDMF Commercial
Data Masking Facility, a block encipherment method specified by International
Business Machines Corporation and based on DES.
CMAC Cipher-based
Message Authenticate Code as defined in [NIST sp800-38b] and [RFC 4493].
CMS Cryptographic
Message Syntax (see RFC 5652)
Cryptographic Device A device
storing cryptographic information and possibly performing cryptographic
functions. May be implemented as a smart card, smart disk, PCMCIA card, or
with some other technology, including software-only.
Cryptoki The
Cryptographic Token Interface defined in this standard.
Cryptoki library A
library that implements the functions specified in this standard.
CT-KIP Cryptographic
Token Key Initialization Protocol (as defined in [CT-KIP])
DER Distinguished
Encoding Rules, as defined in X.690.
DES Data
Encryption Standard, as defined in FIPS PUB 46-3.
DSA Digital
Signature Algorithm, as defined in FIPS PUB 186-4.
EC Elliptic
Curve
ECB Electronic
Codebook mode, as defined in FIPS PUB 81.
ECDH Elliptic
Curve Diffie-Hellman.
ECDSA Elliptic
Curve DSA, as in ANSI X9.62.
ECMQV Elliptic
Curve Menezes-Qu-Vanstone
GOST
28147-89 The encryption algorithm, as defined in Part 2 [GOST
28147-89] and [RFC 4357] [RFC 4490], and RFC [4491].
GOST R 34.11-94 Hash
algorithm, as defined in [GOST R 34.11-94] and [RFC 4357], [RFC 4490], and [RFC
4491].
GOST R 34.10-2001 The
digital signature algorithm, as defined in [GOST R 34.10-2001] and [RFC 4357],
[RFC 4490], and [RFC 4491].
IV Initialization
Vector.
MAC Message
Authentication Code.
Mechanism A
process for implementing a cryptographic operation.
MQV Menezes-Qu-Vanstone
OAEP Optimal
Asymmetric Encryption Padding for RSA.
Object An
item that is stored on a token. May be data, a certificate, or a key.
PIN Personal
Identification Number.
PKCS Public-Key
Cryptography Standards.
PRF Pseudo
random function.
PTD Personal
Trusted Device, as defined in MeT-PTD
RSA The
RSA public-key cryptosystem.
Reader The
means by which information is exchanged with a device.
Session A
logical connection between an application and a token.
SHA-1 The
(revised) Secure Hash Algorithm with a 160-bit message digest, as defined in
FIPS PUB 180-2.
SHA-224 The
Secure Hash Algorithm with a 224-bit message digest, as defined in RFC 3874.
Also defined in FIPS PUB 180-2 with Change Notice 1.
SHA-256 The
Secure Hash Algorithm with a 256-bit message digest, as defined in FIPS PUB
180-2.
SHA-384 The
Secure Hash Algorithm with a 384-bit message digest, as defined in FIPS PUB
180-2.
SHA-512 The
Secure Hash Algorithm with a 512-bit message digest, as defined in FIPS PUB
180-2.
Slot A
logical reader that potentially contains a token.
SSL The
Secure Sockets Layer 3.0 protocol.
Subject Name The
X.500 distinguished name of the entity to which a key is assigned.
SO A
Security Officer user.
TLS Transport
Layer Security.
Token The
logical view of a cryptographic device defined by Cryptoki.
User The
person using an application that interfaces to Cryptoki.
UTF-8 Universal
Character Set (UCS) transformation format (UTF) that represents ISO 10646 and
UNICODE strings with a variable number of octets.
WTLS Wireless
Transport Layer Security.
The following symbols are
used in this standard:
Table 1, Symbols
|
Symbol
|
Definition
|
|
N/A
|
Not applicable
|
|
R/O
|
Read-only
|
|
R/W
|
Read/write
|
The
following prefixes are used in this standard:
Table 2, Prefixes
|
Prefix
|
Description
|
|
C_
|
Function
|
|
CK_
|
Data
type or general constant
|
|
CKA_
|
Attribute
|
|
CKC_
|
Certificate
type
|
|
CKD_
|
Key
derivation function
|
|
CKF_
|
Bit
flag
|
|
CKG_
|
Mask
generation function
|
|
CKH_
|
Hardware
feature type
|
|
CKK_
|
Key
type
|
|
CKM_
|
Mechanism
type
|
|
CKN_
|
Notification
|
|
CKO_
|
Object
class
|
|
CKP_
|
Pseudo-random
function
|
|
CKS_
|
Session
state
|
|
CKR_
|
Return
value
|
|
CKU_
|
User
type
|
|
CKZ_
|
Salt/Encoding
parameter source
|
|
h
|
a
handle
|
|
ul
|
a
CK_ULONG
|
|
p
|
a
pointer
|
|
pb
|
a
pointer to a CK_BYTE
|
|
ph
|
a
pointer to a handle
|
|
pul
|
a
pointer to a CK_ULONG
|
Cryptoki is based on ANSI C types, and defines the following
data types:
/* an unsigned 8-bit value */
typedef unsigned char CK_BYTE;
/* an unsigned 8-bit character */
typedef CK_BYTE CK_CHAR;
/* an 8-bit UTF-8 character */
typedef CK_BYTE CK_UTF8CHAR;
/* a BYTE-sized Boolean flag */
typedef CK_BYTE CK_BBOOL;
/* an unsigned value, at least 32 bits long */
typedef unsigned long int CK_ULONG;
/* a signed value, the same size as a CK_ULONG */
typedef long int CK_LONG;
/* at least 32 bits; each bit is a Boolean flag */
typedef CK_ULONG CK_FLAGS;
Cryptoki also uses pointers to some of these data types, as
well as to the type void, which are implementation-dependent. These pointer
types are:
CK_BYTE_PTR /* Pointer to a CK_BYTE */
CK_CHAR_PTR /* Pointer to a CK_CHAR */
CK_UTF8CHAR_PTR /* Pointer to a CK_UTF8CHAR */
CK_ULONG_PTR /* Pointer to a CK_ULONG */
CK_VOID_PTR /* Pointer to a void */
Cryptoki also defines a pointer to a CK_VOID_PTR, which is
implementation-dependent:
CK_VOID_PTR_PTR /* Pointer to a CK_VOID_PTR */
In addition, Cryptoki defines a C-style NULL pointer, which
is distinct from any valid pointer:
NULL_PTR /* A NULL pointer */
It follows that many of the data and pointer types will vary
somewhat from one environment to another (e.g., a CK_ULONG will
sometimes be 32 bits, and sometimes perhaps 64 bits). However, these details
should not affect an application, assuming it is compiled with Cryptoki header
files consistent with the Cryptoki library to which the application is linked.
All numbers and values expressed in this document are
decimal, unless they are preceded by “0x”, in which case they are hexadecimal
values.
The CK_CHAR data type holds characters from the
following table, taken from ANSI C:
Table 3, Character Set
|
Category
|
Characters
|
|
Letters
|
A B C D E F G H I J K L M N O P Q R S T U V W
X Y Z a b c d e f g h i j k l m n o p q r s t u v w x y z
|
|
Numbers
|
0 1 2 3 4 5 6 7 8 9
|
|
Graphic characters
|
! “ # % & ‘ ( ) * + , - . / : ; < =
> ? [ \ ] ^ _ { | } ~
|
|
Blank
character
|
‘
‘
|
The CK_UTF8CHAR data type holds UTF-8 encoded Unicode
characters as specified in RFC2279. UTF-8 allows internationalization while
maintaining backward compatibility with the Local String definition of PKCS #11
version 2.01.
In Cryptoki, the CK_BBOOL data type is a Boolean type
that can be true or false. A zero value means false, and a nonzero value means
true. Similarly, an individual bit flag, CKF_..., can also be set
(true) or unset (false). For convenience, Cryptoki defines the following macros
for use with values of type CK_BBOOL:
#define CK_FALSE 0
#define CK_TRUE 1
For backwards compatibility, header files for this version
of Cryptoki also define TRUE and FALSE as (CK_DISABLE_TRUE_FALSE may be set by
the application vendor):
#ifndef CK_DISABLE_TRUE_FALSE
#ifndef FALSE
#define FALSE CK_FALSE
#endif
#ifndef TRUE
#define TRUE CK_TRUE
#endif
#endif
[ARIA] National Security Research Institute, Korea, “Block
Cipher Algorithm ARIA”,
URL: https://www.ietf.org/rfc/rfc5794.txt
[BLOWFISH] B. Schneier. “Description of a New
Variable-Length Key, 64-Bit Block Cipher (Blowfish)”, December 1993.
URL: https://www.schneier.com/paper-blowfish-fse.html
[CAMELLIA] M. Matsui, J. Nakajima, S. Moriai. “A
Description of the Camellia Encryption Algorithm”, April 2004.
URL: http://www.ietf.org/rfc/rfc3713.txt
[CDMF] Johnson, D.B. “The Commercial Data Masking Facility
(CDMF) data privacy algorithm”, March 1994.
URL: http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=5389557
[CHACHA] D. Bernstein, “ChaCha, a variant of
Salsa20”, January 2008.
URL: http://cr.yp.to/chacha/chacha-20080128.pdf
[DH] W. Diffie, M. Hellman. “New Directions in
Cryptography”, November 1976.
URL: http://www-ee.stanford.edu/~hellman/publications/24.pdf
[FIPS PUB 46-3] NIST. “FIPS 46-3: Data Encryption
Standard”, October 1999.
URL: http://csrc.nist.gov/publications/fips/fips46-3/fips46-3.pdf
[FIPS PUB 81] NIST. “FIPS 81: DES Modes of Operation”, December
1980.
URL: http://csrc.nist.gov/publications/fips/fips81/fips81.htm
[FIPS PUB 186-4] NIST. “FIPS 186-4: Digital
Signature Standard”, July 2013.
URL: http://nvlpubs.nist.gov/nistpubs/FIPS/NIST.FIPS.186-4.pdf
[FIPS SP 800-56A] NIST. “Special Publication 800-56A
Revision 2: Recommendation for Pair-Wise Key Establishment Schemes Using
Discrete Logarithm Cryptography” May 2013.
URL: http://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-56Ar2.pdf
[FIPS SP 800-108] NIST. “Special Publication
800-108 (Revised): Recommendation for Key Derivation Using Pseudorandom
Functions”, October 2009.
URL: https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-108.pdf
[GOST] V. Dolmatov, A. Degtyarev. “GOST R. 34.11-2012:
Hash Function”, August 2013.
URL: https://datatracker.ietf.org/doc/html/rfc6986
[MD2] B. Kaliski. RSA Laboratories. “The MD2
Message-Digest Algorithm”, April 1992.
URL: https://www.ietf.org/rfc/rfc1319.txt
[MD5] RSA Data Security. R. Rivest. “The MD5
Message-Digest Algorithm”, April 1992.
URL: https://www.ietf.org/rfc/rfc1321.txt
[NIST 802-208] NIST “Special Publication 800-208: Recommendation
for Stateful Hash-Based Signature Schemes”, October 2020.
URL: https://csrc.nist.gov/publications/detail/sp/800-208/final
[OAEP] M. Bellare, P. Rogaway. “Optimal Asymmetric
Encryption – How to Encrypt with RSA”, November 1995.
[PKCS11-Hist] PKCS #11 Cryptographic Token
Interface Historical Mechanisms Specification Version 3.0. Edited by Chris
Zimman and Dieter Bong. Latest stage.
https://docs.oasis-open.org/pkcs11/pkcs11-hist/v3.0/pkcs11-hist-v3.0.html
[PKCS11-Prof] PKCS #11
Profiles Version 3.1. Edited by Tim Hudson. Latest stage: https://docs.oasis-open.org/pkcs11/pkcs11-profiles/v3.1/pkcs11-profiles-v3.1.html.
[PKCS #1] K. Moriarty, B. Kaliski, J. Jonsson, A.
Rusch. RFC 8017 “PKCS #1: RSA Cryptography Specifications Version 2.2”,
November 2016
URL: https://www.rfc-editor.org/rfc/pdfrfc/rfc8017.txt.pdf
[PKCS #3] RSA Laboratories. “Diffie-Hellman
Key-Agreement Standard. v1.4”, November 1993.
URL: https://www.teletrust.de/fileadmin/files/oid/oid_pkcs-3v1-4.pdf
[PKCS #5] K. Moriarty, B. Kaliski, A. Rusch. RFC
8018. “PKCS #5: Password-Based Cryptography Specification Version 2.1”, January
2017
URL: https://www.rfc-editor.org/rfc/pdfrfc/rfc8018.txt.pdf
[PKCS #7] B.Kaliski. “PKCS #7 Cryptographic Message
Syntax Version 1.5”, March 1998
URL: https://www.rfc-editor.org/rfc/pdfrfc/rfc2315.txt.pdf
[PKCS #8] B. Kaliski. RFC 5208 “Public-Key
Cryptography Standards (PKCS) #8: Private-Key Information Syntax Specification
Version 1.2”, May 2008, obsoleted by RFC 5258 S.Turner “Asymmetric Key Packages”,
August 2010
URL: https://www.rfc-editor.org/rfc/pdfrfc/rfc5958.txt.pdf
[PKCS #12] K. Moriarty, M. Nystrom, S. Parkinson, A.
Rusch, M. Scott. “PKCS #12 Personal Information Exchange Syntax v1.1”, July
2014.
URL: https://www.rfc-editor.org/rfc/pdfrfc/rfc7292.txt.pdf
[POLY1305] D.J. Bernstein. “The
Poly1305-AES message-authentication code”, January 2005.
URL: https://cr.yp.to/mac/poly1305-20050329.pdf
[RFC 2409] D. Harkins, D.Carrel. RFC 2409: “The
Internet Key Exchange (IKE)”, November 1998.
URL: https://tools.ietf.org/html/rfc2409
[RFC 2119] Bradner, S., “Key
words for use in RFCs to Indicate Requirement Levels”, BCP 14, RFC 2119, March
1997.
URL: http://www.ietf.org/rfc/rfc2119.txt.
[RFC 2279] F. Yergeau. RFC 2279: “UTF-8, a
transformation format of ISO 10646 Alis Technologies”, January 1998.
URL: http://www.ietf.org/rfc/rfc2279.txt
[RFC 2534] Masinter, L., Wing, D., Mutz, A., and K.
Holtman. RFC 2534: “Media Features for Display, Print, and Fax”. March 1999.
URL: http://www.ietf.org/rfc/rfc2534.txt
[RFC 5652] R. Housley. RFC 5652: “Cryptographic
Message Syntax”, September 2009. URL:
http://www.ietf.org/rfc/rfc5652.txt
[RFC 5707] Rescorla, E., “The Keying Material
Exporters for Transport Layer Security (TLS)”, RFC 5705, March 2010.
URL: http://www.ietf.org/rfc/rfc5705.txt
[RFC 5996] C.
Kaufman, P. Hoffman, Y. Nir, P. Eronen. RFC 5996: “Internet Key Exchange
Protocol Version 2 (IKEv2)”, September 2010.
URL: https://tools.ietf.org/html/rfc5996
[RFC 8554] D. McGrew, m. Curcio, S. Fluhrer. RFC
8554: “Leighton-Micali Hash-Based Signatures”, April 2019.
URL: https://tools.ietf.org/html/rfc8554
[RIPEMD] H. Dobbertin, A. Bosselaers, B. Preneel. “The hash
function RIPEMD-160”, February 2012.
URL: http://homes.esat.kuleuven.be/~bosselae/ripemd160.html
[SALSA] D. Bernstein, “ChaCha, a variant of
Salsa20”, January 2008.
URL: http://cr.yp.to/chacha/chacha-20080128.pdf
[SHA-1] NIST. FIPS 180-4: “Secure Hash
Standard”, March 2012.
URL: http://csrc.nist.gov/publications/fips/fips180-4/fips-180-4.pdf
[SHA-2] NIST. FIPS
180-4: “Secure Hash Standard”, March 2012.
URL: http://csrc.nist.gov/publications/fips/fips180-4/fips-180-4.pdf
[TLS] [RFC2246] Dierks, T. and C. Allen, "The TLS
Protocol Version 1.0", RFC 2246, January 1999. URL: http://www.ietf.org/rfc/rfc2246.txt
, superseded by [RFC4346] Dierks, T. and E. Rescorla, "The Transport Layer
Security (TLS) Protocol Version 1.1", RFC 4346, April 2006.
URL: http://www.ietf.org/rfc/rfc4346.txt
, which was superseded by [TLS12].
[TLS12] [RFC5246] Dierks, T. and E. Rescorla,
"The Transport Layer Security (TLS) Protocol Version 1.2", RFC 5246,
August 2008.
URL: http://www.ietf.org/rfc/rfc5246.txt
[TWOFISH] B. Schneier, J. Kelsey, D. Whiting, C. Hall, N.
Ferguson. “Twofish: A 128-Bit Block Cipher”, June 1998.
URL: https://www.schneier.com/academic/twofish/
[X.500] ITU-T. Information Technology — Open Systems
Interconnection — The Directory: Overview of Concepts, Models and Services.
February 2001. Identical to ISO/IEC 9594-1
[X.509] ITU-T. Information Technology — Open Systems
Interconnection — The Directory: Public-key and Attribute Certificate
Frameworks. March 2000. Identical to ISO/IEC 9594-8
[X.680] ITU-T. Information Technology — Abstract Syntax
Notation One (ASN.1): Specification of Basic Notation. July 2002. Identical to
ISO/IEC 8824-1
[X.690] ITU-T. Information Technology — ASN.1 Encoding
Rules: Specification of Basic Encoding Rules (BER), Canonical Encoding Rules
(CER), and Distinguished Encoding Rules (DER). July 2002. Identical to ISO/IEC
8825-1
[CAP-1.2] Common
Alerting Protocol Version 1.2. 01 July 2010. OASIS Standard.
URL: http://docs.oasis-open.org/emergency/cap/v1.2/CAP-v1.2-os.html
[AES KEYWRAP] National
Institute of Standards and Technology, NIST Special Publication 800-38F,
Recommendation for Block Cipher Modes of Operation: Methods for Key Wrapping,
December 2012, http://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-38F.pdf
[ANSI C] ANSI/ISO. American National Standard
for Programming Languages – C. 1990.
[ANSI X9.31] Accredited Standards Committee X9.
Digital Signatures Using Reversible Public Key Cryptography for the Financial
Services Industry (rDSA). 1998.
[ANSI X9.42] Accredited Standards Committee X9. Public
Key Cryptography for the Financial Services Industry: Agreement of Symmetric
Keys Using Discrete Logarithm Cryptography. 2003.
[ANSI X9.62] Accredited Standards Committee X9. Public
Key Cryptography for the Financial Services Industry: The Elliptic Curve
Digital Signature Algorithm (ECDSA). 1998.
[ANSI X9.63] Accredited Standards Committee X9. Public
Key Cryptography for the Financial Services Industry: Key Agreement and Key
Transport Using Elliptic Curve Cryptography. 2001.
URL: http://webstore.ansi.org/RecordDetail.aspx?sku=X9.63-2011
[BRAINPOOL] ECC Brainpool Standard Curves and Curve
Generation, v1.0, 19.10.2005
URL: http://www.ecc-brainpool.org
[CC/PP] W3C. Composite Capability/Preference
Profiles (CC/PP): Structure and Vocabularies. World Wide Web Consortium,
January 2004.
URL: http://www.w3.org/TR/CCPP-struct-vocab/
[CDPD] Ameritech Mobile Communications et al. Cellular
Digital Packet Data System Specifications: Part 406: Airlink Security. 1993.
[CT-KIP] RSA Laboratories. Cryptographic Token
Key Initialization Protocol. Version 1.0, December 2005.
[GCS-API] X/Open Company Ltd. Generic
Cryptographic Service API (GCS-API), Base - Draft 2. February 14, 1995.
[ISO/IEC 7816-1] ISO. Information Technology —
Identification Cards — Integrated Circuit(s) with Contacts — Part 1: Physical
Characteristics. 1998.
[ISO/IEC 7816-4] ISO. Information Technology —
Identification Cards — Integrated Circuit(s) with Contacts — Part 4:
Interindustry Commands for Interchange. 1995.
[ISO/IEC 8824-1] ISO. Information Technology--
Abstract Syntax Notation One (ASN.1): Specification of Basic Notation. 2002.
[ISO/IEC 8825-1] ISO. Information
Technology—ASN.1 Encoding Rules: Specification of Basic Encoding Rules (BER),
Canonical Encoding Rules (CER), and Distinguished Encoding Rules (DER). 2002.
[ISO/IEC 9594-1] ISO. Information Technology —
Open Systems Interconnection — The Directory: Overview of Concepts, Models and
Services. 2001.
[ISO/IEC 9594-8] ISO. Information Technology —
Open Systems Interconnection — The Directory: Public-key and Attribute
Certificate Frameworks. 2001
[ISO/IEC 9796-2] ISO. Information Technology —
Security Techniques — Digital Signature Scheme Giving Message Recovery — Part
2: Integer factorization based mechanisms. 2002.
[Java MIDP] Java Community Process. Mobile Information
Device Profile for Java 2 Micro Edition. November 2002.
URL: http://jcp.org/jsr/detail/118.jsp
[LEGIFRANCE] Avis relatif aux paramètres de courbes
elliptiques définis par l'Etat français (Publication of Elliptic Curve
parameters by the French state)
URL:
https://www.legifrance.gouv.fr/affichTexte.do?cidTexte=JORFTEXT000024668816
[MeT-PTD] MeT. MeT PTD Definition – Personal Trusted
Device Definition, Version 1.0, February 2003.
URL: http://www.mobiletransaction.org
[NIST AES CTS] National Institute of Standards
and Technology, Addendum to NIST Special Publication 800-38A, “Recommendation
for Block Cipher Modes of Operation: Three Variants of Ciphertext Stealing for
CBC Mode”
URL: http://csrc.nist.gov/publications/nistpubs/800-38a/addendum-to-nist_sp800-38A.pdf
[PCMCIA] Personal Computer Memory Card
International Association. PC Card Standard, Release 2.1,. July 1993.
[RFC 2865] Rigney
et al, “Remote Authentication Dial In User Service (RADIUS)”, IETF RFC2865,
June 2000.
URL: http://www.ietf.org/rfc/rfc2865.txt.
[RFC 3686] Housley, “Using
Advanced Encryption Standard (AES) Counter Mode With IPsec Encapsulating
Security Payload (ESP),” IETF RFC 3686, January 2004.
URL: http://www.ietf.org/rfc/rfc3686.txt.
[RFC 3717] Matsui, et al, ”A
Description of the Camellia Encryption Algorithm,” IETF RFC 3717, April 2004.
URL: http://www.ietf.org/rfc/rfc3713.txt.
[RFC 3610] Whiting, D.,
Housley, R., and N. Ferguson, “Counter with CBC-MAC (CCM)", IETF RFC 3610,
September 2003.
URL: http://www.ietf.org/rfc/rfc3610.txt
[RFC 3874] Smit et al, “A 224-bit One-way Hash
Function: SHA-224,” IETF RFC 3874, June 2004.
URL: http://www.ietf.org/rfc/rfc3874.txt.
[RFC 3748] Aboba et al,
“Extensible Authentication Protocol (EAP)”, IETF RFC 3748, June 2004.
URL: http://www.ietf.org/rfc/rfc3748.txt.
[RFC 4269] South Korean Information Security
Agency (KISA) “The SEED Encryption Algorithm”, December 2005.
URL: https://ftp.rfc-editor.org/in-notes/rfc4269.txt
[RFC 4309] Housley, R., “Using Advanced Encryption
Standard (AES) CCM Mode with IPsec Encapsulating Security Payload (ESP),” IETF
RFC 4309, December 2005.
URL: http://www.ietf.org/rfc/rfc4309.txt
[RFC 4357] V. Popov, I. Kurepkin, S. Leontiev
“Additional Cryptographic Algorithms for Use with GOST 28147-89, GOST R
34.10-94, GOST R 34.10-2001, and GOST R 34.11-94 Algorithms”, January 2006.
URL: http://www.ietf.org/rfc/rfc4357.txt
[RFC 4490] S. Leontiev, Ed. G. Chudov, Ed.
“Using the GOST 28147-89, GOST R 34.11-94,GOST R 34.10-94, and GOST R
34.10-2001 Algorithms with Cryptographic Message Syntax (CMS)”, May 2006.
URL: http://www.ietf.org/rfc/rfc4490.txt
[RFC 4491] S. Leontiev, Ed., D. Shefanovski,
Ed., “Using the GOST R 34.10-94, GOST R 34.10-2001, and GOST R 34.11-94 Algorithms
with the Internet X.509 Public Key Infrastructure Certificate and CRL Profile”,
May 2006.
URL: http://www.ietf.org/rfc/rfc4491.txt
[RFC 4493] J. Song et al. RFC 4493: The AES-CMAC
Algorithm. June 2006.
URL: http://www.ietf.org/rfc/rfc4493.txt
[RFC 5705] Rescorla, E., “The Keying Material
Exporters for Transport Layer Security (TLS)”, RFC 5705, March 2010.
URL: http://www.ietf.org/rfc/rfc5705.txt
[RFC 5869] H. Krawczyk, P. Eronen, “HMAC-based
Extract-and-Expand Key Derivation Function (HKDF)“, May 2010
URL: http://www.ietf.org/rfc/rfc5869.txt
[RFC 7539] Y Nir, A. Langley. RFC 7539: ChaCha20
and Poly1305 for IETF Protocols, May 2015
URL: https://tools.ietf.org/rfc/rfc7539.txt
[RFC 7748] Aboba et al, “Elliptic Curves for
Security”, IETF RFC 7748, January 2016
URL: https://tools.ietf.org/html/rfc7748
[RFC 8032] Aboba et al, “Edwards-Curve Digital
Signature Algorithm (EdDSA)”, IETF RFC 8032, January 2017
URL: https://tools.ietf.org/html/rfc8032
[SEC 1] Standards for Efficient Cryptography Group (SECG).
Standards for Efficient Cryptography (SEC) 1: Elliptic Curve Cryptography.
Version 1.0, September 20, 2000.
[SEC 2] Standards for Efficient Cryptography Group (SECG).
Standards for Efficient Cryptography (SEC) 2: Recommended Elliptic Curve Domain
Parameters. Version 1.0, September 20, 2000.
[WTLS] WAP. Wireless Transport Layer Security Version —
WAP-261-WTLS-20010406-a. April 2001.
URL: http://openmobilealliance.org/tech/affiliates/wap/wap-261-wtls-20010406-a.pdf
[XEDDSA] The XEdDSA and VXEdDSA Signature Schemes -
Revision 1, 2016-10-20, Trevor Perrin (editor)
URL: https://signal.org/docs/specifications/xeddsa/
[X.500] ITU-T. Information Technology — Open Systems
Interconnection — The Directory: Overview of Concepts, Models and Services.
February 2001. Identical to ISO/IEC 9594-1
[X.509] ITU-T. Information Technology — Open Systems
Interconnection — The Directory: Public-key and Attribute Certificate
Frameworks. March 2000. Identical to ISO/IEC 9594-8
[X.680] ITU-T. Information Technology — Abstract Syntax
Notation One (ASN.1): Specification of Basic Notation. July 2002. Identical to
ISO/IEC 8824-1
[X.690] ITU-T. Information Technology — ASN.1 Encoding
Rules: Specification of Basic Encoding Rules (BER), Canonical Encoding Rules
(CER), and Distinguished Encoding Rules (DER). July 2002. Identical to ISO/IEC
8825-1
There is a large array of Cryptoki-related data types that
are defined in the Cryptoki header files. Certain packing and pointer-related
aspects of these types are platform and compiler-dependent; these aspects are
therefore resolved on a platform-by-platform (or compiler-by-compiler) basis
outside of the Cryptoki header files by means of preprocessor directives.
This means that when writing C or C++ code, certain
preprocessor directives MUST be issued before including a Cryptoki header file.
These directives are described in the remainder of this section.
Plattform specific implementation hints can be found in the
pkcs11.h header file.
Cryptoki structures are packed to occupy as little space as
is possible. Cryptoki structures SHALL be packed with 1-byte alignment.
Because different platforms and compilers have different
ways of dealing with different types of pointers, the following 6 macros SHALL
be set outside the scope of Cryptoki:
¨
CK_PTR
CK_PTR is the
“indirection string” a given platform and compiler uses to make a pointer to an
object. It is used in the following fashion:
typedef CK_BYTE CK_PTR CK_BYTE_PTR;
¨
CK_DECLARE_FUNCTION
CK_DECLARE_FUNCTION(returnType,
name), when followed by a
parentheses-enclosed list of arguments and a semicolon, declares a Cryptoki API
function in a Cryptoki library. returnType is the return type of the function,
and name is its name. It SHALL be used in the following fashion:
CK_DECLARE_FUNCTION(CK_RV, C_Initialize)(
CK_VOID_PTR pReserved
);
¨
CK_DECLARE_FUNCTION_POINTER
CK_DECLARE_FUNCTION_POINTER(returnType,
name), when followed by a parentheses-enclosed list of arguments and a
semicolon, declares a variable or type which is a pointer to a Cryptoki API
function in a Cryptoki library. returnType is the return type of the function,
and name is its name. It SHALL be used in either of the following fashions to
define a function pointer variable, myC_Initialize, which can point to a
C_Initialize function in a Cryptoki library (note that neither of the following
code snippets actually assigns a value to myC_Initialize):
CK_DECLARE_FUNCTION_POINTER(CK_RV, myC_Initialize)(
CK_VOID_PTR pReserved
);
or:
typedef CK_DECLARE_FUNCTION_POINTER(CK_RV,
myC_InitializeType)(
CK_VOID_PTR pReserved
);
myC_InitializeType myC_Initialize;
¨
CK_CALLBACK_FUNCTION
CK_CALLBACK_FUNCTION(returnType,
name), when followed by a
parentheses-enclosed list of arguments and a semicolon, declares a variable or
type which is a pointer to an application callback function that can be used by
a Cryptoki API function in a Cryptoki library. returnType is the return type
of the function, and name is its name. It SHALL be used in either of the following
fashions to define a function pointer variable, myCallback, which can point to
an application callback which takes arguments args and returns a CK_RV (note
that neither of the following code snippets actually assigns a value to
myCallback):
CK_CALLBACK_FUNCTION(CK_RV, myCallback)(args);
or:
typedef CK_CALLBACK_FUNCTION(CK_RV, myCallbackType)(args);
myCallbackType myCallback;
¨
NULL_PTR
NULL_PTR is the value of a NULL pointer. In any ANSI C
environment—and in many others as well—NULL_PTR
SHALL be defined simply as 0.
The general Cryptoki data types are described in the
following subsections. The data types for holding parameters for various
mechanisms, and the pointers to those parameters, are not described here; these
types are described with the information on the mechanisms themselves, in
Section 6.
A C or C++ source file in a Cryptoki application or library
can define all these types (the types described here and the types that are
specifically used for particular mechanism parameters) by including the
top-level Cryptoki include file, pkcs11.h. pkcs11.h, in turn, includes the
other Cryptoki include files, pkcs11t.h and pkcs11f.h. A source file can also
include just pkcs11t.h (instead of pkcs11.h); this defines most (but not all)
of the types specified here.
When including either of these header files, a source file
MUST specify the preprocessor directives indicated in Section 2.
Cryptoki represents general information with the following
types:
¨
CK_VERSION; CK_VERSION_PTR
CK_VERSION is a structure that describes the version
of a Cryptoki interface, a Cryptoki library, or an SSL or TLS implementation,
or the hardware or firmware version of a slot or token. It is defined as
follows:
typedef struct CK_VERSION {
CK_BYTE major;
CK_BYTE minor;
} CK_VERSION;
The fields of the structure have the following meanings:
major major version number (the
integer portion of the version)
minor minor version number (the
hundredths portion of the version)
Example: For version 1.0, major = 1 and minor
= 0. For version 2.10, major = 2 and minor = 10. Table 4 below lists the major and minor version values for the officially published
Cryptoki specifications.
Table 4, Major and minor version values for published Cryptoki specifications
|
Version
|
major
|
minor
|
|
1.0
|
0x01
|
0x00
|
|
2.01
|
0x02
|
0x01
|
|
2.10
|
0x02
|
0x0a
|
|
2.11
|
0x02
|
0x0b
|
|
2.20
|
0x02
|
0x14
|
|
2.30
|
0x02
|
0x1e
|
|
2.40
|
0x02
|
0x28
|
|
3.0
|
0x03
|
0x00
|
Minor revisions of the Cryptoki standard are always upwardly
compatible within the same major version number.
CK_VERSION_PTR is a pointer to a CK_VERSION.
¨
CK_INFO; CK_INFO_PTR
CK_INFO provides general information about Cryptoki.
It is defined as follows:
typedef struct CK_INFO {
CK_VERSION cryptokiVersion;
CK_UTF8CHAR manufacturerID[32];
CK_FLAGS flags;
CK_UTF8CHAR libraryDescription[32];
CK_VERSION libraryVersion;
} CK_INFO;
The fields of the structure have the following meanings:
cryptokiVersion Cryptoki
interface version number, for compatibility with future revisions of this
interface
manufacturerID ID
of the Cryptoki library manufacturer. MUST be padded with the blank character
(‘ ‘). Should not be null-terminated.
flags bit
flags reserved for future versions. MUST be zero for this version
libraryDescription character-string
description of the library. MUST be padded with the blank character (‘ ‘).
Should not be null-terminated.
libraryVersion Cryptoki
library version number
For libraries written to this document, the value of cryptokiVersion
should match the version of this specification; the value of libraryVersion
is the version number of the library software itself.
CK_INFO_PTR is a pointer to a CK_INFO.
¨
CK_NOTIFICATION
CK_NOTIFICATION holds the types of notifications that
Cryptoki provides to an application. It is defined as follows:
typedef CK_ULONG CK_NOTIFICATION;
For this version of Cryptoki, the following types of
notifications are defined:
The notifications have the following meanings:
CKN_SURRENDER Cryptoki is
surrendering the execution of a function executing in a session so that the
application may perform other operations. After performing any desired
operations, the application should indicate to Cryptoki whether to continue or
cancel the function (see Section 5.21.1).
Cryptoki represents slot and token information with the
following types:
¨
CK_SLOT_ID; CK_SLOT_ID_PTR
CK_SLOT_ID is a Cryptoki-assigned value that
identifies a slot. It is defined as follows:
typedef CK_ULONG CK_SLOT_ID;
A list of CK_SLOT_IDs is returned by C_GetSlotList.
A priori, any value of CK_SLOT_ID can be a valid slot
identifier—in particular, a system may have a slot identified by the value 0.
It need not have such a slot, however.
CK_SLOT_ID_PTR is a pointer to a CK_SLOT_ID.
¨
CK_SLOT_INFO; CK_SLOT_INFO_PTR
CK_SLOT_INFO provides information about a slot. It
is defined as follows:
typedef struct CK_SLOT_INFO {
CK_UTF8CHAR slotDescription[64];
CK_UTF8CHAR manufacturerID[32];
CK_FLAGS flags;
CK_VERSION hardwareVersion;
CK_VERSION
firmwareVersion;
} CK_SLOT_INFO;
The fields of the structure have the following meanings:
slotDescription character-string
description of the slot. MUST be padded with the blank character (‘ ‘). MUST
NOT be null-terminated.
manufacturerID ID
of the slot manufacturer. MUST be padded with the blank character (‘ ‘). MUST
NOT be null-terminated.
flags bits
flags that provide capabilities of the slot. The flags are defined below
hardwareVersion version
number of the slot’s hardware
firmwareVersion version
number of the slot’s firmware
The following table defines the flags field:
Table 5, Slot Information Flags
|
Bit Flag
|
Mask
|
Meaning
|
|
CKF_TOKEN_PRESENT
|
0x00000001
|
True if a token is present in the slot (e.g., a
device is in the reader)
|
|
CKF_REMOVABLE_DEVICE
|
0x00000002
|
True if the reader supports removable devices
|
|
CKF_HW_SLOT
|
0x00000004
|
True if the slot is a hardware slot, as opposed to a
software slot implementing a “soft token”
|
For a given slot, the value of the CKF_REMOVABLE_DEVICE
flag never changes. In addition, if this flag is not set for a given
slot, then the CKF_TOKEN_PRESENT flag for that slot is always
set. That is, if a slot does not support a removable device, then that slot
always has a token in it.
CK_SLOT_INFO_PTR is a pointer to a CK_SLOT_INFO.
¨
CK_TOKEN_INFO; CK_TOKEN_INFO_PTR
CK_TOKEN_INFO provides information about a token. It
is defined as follows:
typedef struct CK_TOKEN_INFO {
CK_UTF8CHAR label[32];
CK_UTF8CHAR manufacturerID[32];
CK_UTF8CHAR model[16];
CK_CHAR serialNumber[16];
CK_FLAGS flags;
CK_ULONG ulMaxSessionCount;
CK_ULONG ulSessionCount;
CK_ULONG ulMaxRwSessionCount;
CK_ULONG ulRwSessionCount;
CK_ULONG ulMaxPinLen;
CK_ULONG ulMinPinLen;
CK_ULONG ulTotalPublicMemory;
CK_ULONG ulFreePublicMemory;
CK_ULONG ulTotalPrivateMemory;
CK_ULONG ulFreePrivateMemory;
CK_VERSION hardwareVersion;
CK_VERSION firmwareVersion;
CK_CHAR utcTime[16];
} CK_TOKEN_INFO;
The fields of the structure have the following meanings:
label application-defined
label, assigned during token initialization. MUST be padded with the blank
character (‘ ‘). MUST NOT be null-terminated.
manufacturerID ID
of the device manufacturer. MUST be padded with the blank character (‘ ‘). MUST
NOT be null-terminated.
model model
of the device. MUST be padded with the blank character (‘ ‘). MUST NOT be
null-terminated.
serialNumber character-string
serial number of the device. MUST be padded with the blank character (‘ ‘).
MUST NOT be null-terminated.
flags bit
flags indicating capabilities and status of the device as defined below
ulMaxSessionCount maximum
number of sessions that can be opened with the token at one time by a single
application (see CK_TOKEN_INFO Note below)
ulSessionCount number
of sessions that this application currently has open with the token (see CK_TOKEN_INFO
Note below)
ulMaxRwSessionCount maximum
number of read/write sessions that can be opened with the token at one time by
a single application (see CK_TOKEN_INFO Note below)
ulRwSessionCount number
of read/write sessions that this application currently has open with the token
(see CK_TOKEN_INFO Note below)
ulMaxPinLen maximum
length in bytes of the PIN
ulMinPinLen minimum
length in bytes of the PIN
ulTotalPublicMemory the
total amount of memory on the token in bytes in which public objects may be
stored (see CK_TOKEN_INFO Note below)
ulFreePublicMemory the
amount of free (unused) memory on the token in bytes for public objects (see CK_TOKEN_INFO
Note below)
ulTotalPrivateMemory the
total amount of memory on the token in bytes in which private objects may be
stored (see CK_TOKEN_INFO Note below)
ulFreePrivateMemory the
amount of free (unused) memory on the token in bytes for private objects (see CK_TOKEN_INFO
Note below)
hardwareVersion version
number of hardware
firmwareVersion version
number of firmware
utcTime current
time as a character-string of length 16, represented in the format
YYYYMMDDhhmmssxx (4 characters for the year; 2 characters each for the month,
the day, the hour, the minute, and the second; and 2 additional reserved ‘0’
characters). The value of this field only makes sense for tokens equipped with
a clock, as indicated in the token information flags (see below)
The following table defines the flags field:
Table 6,
Token Information Flags
|
Bit
Flag
|
Mask
|
Meaning
|
|
CKF_RNG
|
0x00000001
|
True
if the token has its own random number generator
|
|
CKF_WRITE_PROTECTED
|
0x00000002
|
True
if the token is write-protected (see below)
|
|
CKF_LOGIN_REQUIRED
|
0x00000004
|
True
if there are some cryptographic functions that a user MUST be logged in to
perform
|
|
CKF_USER_PIN_INITIALIZED
|
0x00000008
|
True
if the normal user’s PIN has been initialized
|
|
CKF_RESTORE_KEY_NOT_NEEDED
|
0x00000020
|
True
if a successful save of a session’s cryptographic operations state always
contains all keys needed to restore the state of the session
|
|
CKF_CLOCK_ON_TOKEN
|
0x00000040
|
True
if token has its own hardware clock
|
|
CKF_PROTECTED_AUTHENTICATION_PATH
|
0x00000100
|
True
if token has a “protected authentication path”, whereby a user can log into
the token without passing a PIN through the Cryptoki library
|
|
CKF_DUAL_CRYPTO_OPERATIONS
|
0x00000200
|
True
if a single session with the token can perform dual cryptographic operations
(see Section 5.14)
|
|
CKF_TOKEN_INITIALIZED
|
0x00000400
|
True
if the token has been initialized using C_InitToken or an equivalent
mechanism outside the scope of this standard. Calling C_InitToken when this
flag is set will cause the token to be reinitialized.
|
|
CKF_SECONDARY_AUTHENTICATION
|
0x00000800
|
True
if the token supports secondary authentication for private key objects.
(Deprecated; new implementations MUST NOT set this flag)
|
|
CKF_USER_PIN_COUNT_LOW
|
0x00010000
|
True
if an incorrect user login PIN has been entered at least once since the last
successful authentication.
|
|
CKF_USER_PIN_FINAL_TRY
|
0x00020000
|
True
if supplying an incorrect user PIN will cause it to become locked.
|
|
CKF_USER_PIN_LOCKED
|
0x00040000
|
True
if the user PIN has been locked. User login to the token is not possible.
|
|
CKF_USER_PIN_TO_BE_CHANGED
|
0x00080000
|
True
if the user PIN value is the default value set by token initialization or
manufacturing, or the PIN has been expired by the card.
|
|
CKF_SO_PIN_COUNT_LOW
|
0x00100000
|
True
if an incorrect SO login PIN has been entered at least once since the last
successful authentication.
|
|
CKF_SO_PIN_FINAL_TRY
|
0x00200000
|
True
if supplying an incorrect SO PIN will cause it to become locked.
|
|
CKF_SO_PIN_LOCKED
|
0x00400000
|
True
if the SO PIN has been locked. SO login to the token is not possible.
|
|
CKF_SO_PIN_TO_BE_CHANGED
|
0x00800000
|
True
if the SO PIN value is the default value set by token initialization or
manufacturing, or the PIN has been expired by the card.
|
|
CKF_ERROR_STATE
|
0x01000000
|
True
if the token failed a FIPS 140-2 self-test and entered an error state.
|
Exactly what the CKF_WRITE_PROTECTED flag means is
not specified in Cryptoki. An application may be unable to perform certain
actions on a write-protected token; these actions can include any of the
following, among others:
·
Creating/modifying/deleting any object on the token.
·
Creating/modifying/deleting a token object on the token.
·
Changing the SO’s PIN.
·
Changing the normal user’s PIN.
The token may change the value of the CKF_WRITE_PROTECTED
flag depending on the session state to implement its object management policy.
For instance, the token may set the CKF_WRITE_PROTECTED flag unless the
session state is R/W SO or R/W User to implement a policy that does not allow
any objects, public or private, to be created, modified, or deleted unless the
user has successfully called C_Login.
The CKF_USER_PIN_COUNT_LOW, CKF_USER_PIN_COUNT_LOW,
CKF_USER_PIN_FINAL_TRY, and CKF_SO_PIN_FINAL_TRY flags may always
be set to false if the token does not support the functionality or will not
reveal the information because of its security policy.
The CKF_USER_PIN_TO_BE_CHANGED and CKF_SO_PIN_TO_BE_CHANGED
flags may always be set to false if the token does not support the
functionality. If a PIN is set to the default value, or has expired, the
appropriate CKF_USER_PIN_TO_BE_CHANGED or CKF_SO_PIN_TO_BE_CHANGED
flag is set to true. When either of these flags are true, logging in with the
corresponding PIN will succeed, but only the C_SetPIN function can be called.
Calling any other function that required the user to be logged in will cause
CKR_PIN_EXPIRED to be returned until C_SetPIN is called successfully.
CK_TOKEN_INFO Note: The fields ulMaxSessionCount,
ulSessionCount, ulMaxRwSessionCount, ulRwSessionCount, ulTotalPublicMemory,
ulFreePublicMemory, ulTotalPrivateMemory, and ulFreePrivateMemory can have the
special value CK_UNAVAILABLE_INFORMATION, which means that the token and/or
library is unable or unwilling to provide that information. In addition, the
fields ulMaxSessionCount and ulMaxRwSessionCount can have the special value
CK_EFFECTIVELY_INFINITE, which means that there is no practical limit on the
number of sessions (resp. R/W sessions) an application can have open with the
token.
It is important to check these fields for these special
values. This is particularly true for CK_EFFECTIVELY_INFINITE, since an
application seeing this value in the ulMaxSessionCount or ulMaxRwSessionCount
field would otherwise conclude that it can’t open any sessions with the token,
which is far from being the case.
The upshot of all this is that the correct way to interpret
(for example) the ulMaxSessionCount field is something along the lines of the
following:
CK_TOKEN_INFO info;
.
.
if ((CK_LONG) info.ulMaxSessionCount
== CK_UNAVAILABLE_INFORMATION) {
/* Token refuses to give value of ulMaxSessionCount */
.
.
} else if (info.ulMaxSessionCount ==
CK_EFFECTIVELY_INFINITE) {
/* Application can open as many sessions as it wants */
.
.
} else {
/* ulMaxSessionCount really does contain what it should
*/
.
.
}
CK_TOKEN_INFO_PTR is a pointer to a CK_TOKEN_INFO.
Cryptoki represents session information with the following
types:
¨
CK_SESSION_HANDLE; CK_SESSION_HANDLE_PTR
CK_SESSION_HANDLE is a Cryptoki-assigned value that
identifies a session. It is defined as follows:
typedef CK_ULONG CK_SESSION_HANDLE;
Valid
session handles in Cryptoki always have nonzero values. For developers’
convenience, Cryptoki defines the following symbolic value:
CK_SESSION_HANDLE_PTR is a pointer to a CK_SESSION_HANDLE.
¨
CK_USER_TYPE
CK_USER_TYPE holds the types of Cryptoki users
described in [PKCS11-UG] and, in
addition, a context-specific type described in Section 4.9.
It is defined as follows:
typedef CK_ULONG CK_USER_TYPE;
For this version of Cryptoki, the following types of users
are defined:
CKU_SO
CKU_USER
CKU_CONTEXT_SPECIFIC
¨
CK_STATE
CK_STATE holds the session state, as described in [PKCS11-UG]. It is defined as follows:
typedef CK_ULONG CK_STATE;
For this version of Cryptoki, the following session states
are defined:
CKS_RO_PUBLIC_SESSION
CKS_RO_USER_FUNCTIONS
CKS_RW_PUBLIC_SESSION
CKS_RW_USER_FUNCTIONS
CKS_RW_SO_FUNCTIONS
¨ CK_SESSION_INFO; CK_SESSION_INFO_PTR
CK_SESSION_INFO provides information about a
session. It is defined as follows:
typedef struct CK_SESSION_INFO {
CK_SLOT_ID slotID;
CK_STATE state;
CK_FLAGS flags;
CK_ULONG ulDeviceError;
} CK_SESSION_INFO;
The fields of the structure have the following meanings:
slotID ID
of the slot that interfaces with the token
state the
state of the session
flags bit
flags that define the type of session; the flags are defined below
ulDeviceError an
error code defined by the cryptographic device. Used for errors not covered by
Cryptoki.
The following table defines the flags field:
Table 7, Session Information Flags
|
Bit Flag
|
Mask
|
Meaning
|
|
CKF_RW_SESSION
|
0x00000002
|
True if the session is read/write; false if
the session is read-only
|
|
CKF_SERIAL_SESSION
|
0x00000004
|
This flag is provided for backward
compatibility, and should always be set to true
|
CK_SESSION_INFO_PTR
is a pointer to a CK_SESSION_INFO.
Cryptoki represents object information with the following
types:
¨
CK_OBJECT_HANDLE; CK_OBJECT_HANDLE_PTR
CK_OBJECT_HANDLE is a token-specific identifier for
an object. It is defined as follows:
typedef CK_ULONG CK_OBJECT_HANDLE;
When an object is created or found on a token by an
application, Cryptoki assigns it an object handle for that application’s
sessions to use to access it. A particular object on a token does not
necessarily have a handle which is fixed for the lifetime of the object;
however, if a particular session can use a particular handle to access a
particular object, then that session will continue to be able to use that
handle to access that object as long as the session continues to exist, the
object continues to exist, and the object continues to be accessible to the
session.
Valid
object handles in Cryptoki always have nonzero values. For developers’
convenience, Cryptoki defines the following symbolic value:
CK_OBJECT_HANDLE_PTR is a pointer to a CK_OBJECT_HANDLE.
¨
CK_OBJECT_CLASS; CK_OBJECT_CLASS_PTR
CK_OBJECT_CLASS is a value
that identifies the classes (or types) of objects that Cryptoki recognizes. It
is defined as follows:
typedef CK_ULONG CK_OBJECT_CLASS;
Object classes are defined with the objects that use them.
The type is specified on an object through the CKA_CLASS attribute of the
object.
Vendor defined values for this type may also be specified.
Object classes CKO_VENDOR_DEFINED and above are
permanently reserved for token vendors. For interoperability, vendors should
register their object classes through the PKCS process.
CK_OBJECT_CLASS_PTR is a pointer to a CK_OBJECT_CLASS.
¨
CK_HW_FEATURE_TYPE
CK_HW_FEATURE_TYPE is a value that identifies a
hardware feature type of a device. It is defined as follows:
typedef CK_ULONG CK_HW_FEATURE_TYPE;
Hardware feature types are defined with the objects that use
them. The type is specified on an object through the CKA_HW_FEATURE_TYPE
attribute of the object.
Vendor defined values for this type may also be specified.
Feature types CKH_VENDOR_DEFINED and above are
permanently reserved for token vendors. For interoperability, vendors should
register their feature types through the PKCS process.
¨
CK_KEY_TYPE
CK_KEY_TYPE is a value that identifies a key type. It
is defined as follows:
typedef CK_ULONG
CK_KEY_TYPE;
Key types are defined with the objects and mechanisms that
use them. The key type is specified on an object through the CKA_KEY_TYPE
attribute of the object.
Vendor defined values for this type may also be specified.
Key types CKK_VENDOR_DEFINED and above are
permanently reserved for token vendors. For interoperability, vendors should
register their key types through the PKCS process.
¨
CK_CERTIFICATE_TYPE
CK_CERTIFICATE_TYPE is a value that identifies a
certificate type. It is defined as follows:
typedef CK_ULONG CK_CERTIFICATE_TYPE;
Certificate types are defined with the objects and
mechanisms that use them. The certificate type is specified on an object
through the CKA_CERTIFICATE_TYPE attribute of the object.
Vendor defined values for this type may also be specified.
Certificate types CKC_VENDOR_DEFINED
and above are permanently reserved for token vendors. For interoperability,
vendors should register their certificate types through the PKCS process.
¨
CK_CERTIFICATE_CATEGORY
CK_CERTIFICATE_CATEGORY is a value that identifies a
certificate category. It is defined as follows:
typedef CK_ULONG CK_CERTIFICATE_CATEGORY;
For this version of Cryptoki, the following certificate
categories are defined:
|
Constant
|
Value
|
Meaning
|
|
CK_CERTIFICATE_CATEGORY_UNSPECIFIED
|
0x00000000UL
|
No category specified
|
|
CK_CERTIFICATE_CATEGORY_TOKEN_USER
|
0x00000001UL
|
Certificate belongs to owner of the token
|
|
CK_CERTIFICATE_CATEGORY_AUTHORITY
|
0x00000002UL
|
Certificate belongs to a certificate
authority
|
|
CK_CERTIFICATE_CATEGORY_OTHER_ENTITY
|
0x00000003UL
|
Certificate belongs to an end entity (i.e.:
not a CA)
|
¨ CK_ATTRIBUTE_TYPE
CK_ATTRIBUTE_TYPE is a value that identifies an
attribute type. It is defined as follows:
typedef CK_ULONG CK_ATTRIBUTE_TYPE;
Attributes are defined with the objects and mechanisms that
use them. Attributes are specified on an object as a list of type, length value
items. These are often specified as an attribute template.
Vendor defined values for this type may also be specified.
Attribute types CKA_VENDOR_DEFINED and above are
permanently reserved for token vendors. For interoperability, vendors should
register their attribute types through the PKCS process.
¨
CK_ATTRIBUTE; CK_ATTRIBUTE_PTR
CK_ATTRIBUTE is a structure that includes the type,
value, and length of an attribute. It is defined as follows:
typedef struct CK_ATTRIBUTE {
CK_ATTRIBUTE_TYPE type;
CK_VOID_PTR pValue;
CK_ULONG ulValueLen;
} CK_ATTRIBUTE;
The fields of the structure have the following meanings:
type the
attribute type
pValue pointer
to the value of the attribute
ulValueLen length
in bytes of the value
If an attribute has no value, then ulValueLen = 0,
and the value of pValue is irrelevant. An array of CK_ATTRIBUTEs
is called a “template” and is used for creating, manipulating and searching for
objects. The order of the attributes in a template never matters, even
if the template contains vendor-specific attributes. Note that pValue
is a “void” pointer, facilitating the passing of arbitrary values. Both the
application and Cryptoki library MUST ensure that the pointer can be safely
cast to the expected type (i.e., without word-alignment errors).
The constant
CK_UNAVAILABLE_INFORMATION is used in the ulValueLen field to denote an invalid
or unavailable value. See C_GetAttributeValue for further details.
CK_ATTRIBUTE_PTR is a pointer to a CK_ATTRIBUTE.
¨ CK_DATE
CK_DATE is a structure that defines a date. It is
defined as follows:
typedef struct CK_DATE {
CK_CHAR year[4];
CK_CHAR month[2];
CK_CHAR day[2];
} CK_DATE;
The fields of the structure have the following meanings:
year the
year (“1900” - “9999”)
month the
month (“01” - “12”)
day the
day (“01” - “31”)
The fields hold numeric characters from the character set
in Table 3, not the literal byte values.
When a Cryptoki object carries an attribute of this type,
and the default value of the attribute is specified to be "empty,"
then Cryptoki libraries SHALL set the attribute's ulValueLen to 0.
Note that implementations of previous versions of Cryptoki
may have used other methods to identify an "empty" attribute of type
CK_DATE, and applications that needs to interoperate with these libraries
therefore have to be flexible in what they accept as an empty value.
¨
CK_PROFILE_ID;
CK_PROFILE_ID_PTR
CK_PROFILE_ID is an unsigend ulong value represting a
specific token profile. It is defined as follows:
typedef CK_ULONG CK_PROFILE_ID;
Profiles are defines in the PKCS #11 Cryptographic Token
Interface Profiles document. s. ID's greater than 0xffffffff may cause
compatibility issues on platforms that have CK_ULONG values of 32 bits, and
should be avoided.
Vendor defined values for this type may also be specified.
Profile IDs CKP_VENDOR_DEFINED and above are
permanently reserved for token vendors. For interoperability, vendors should
register their object classes through the PKCS process.
Valid Profile IDs in Cryptoki always have nonzero values.
For developers’ convenience, Cryptoki defines the following symbolic value:
CK_PROFILE_ID_PTR is a pointer to a CK_PROFILE_ID.
¨
CK_JAVA_MIDP_SECURITY_DOMAIN
CK_JAVA_MIDP_SECURITY_DOMAIN is a
value that identifies the Java MIDP security domain of a certificate. It is
defined as follows:
typedef CK_ULONG CK_JAVA_MIDP_SECURITY_DOMAIN;
For this version of
Cryptoki, the following security domains are defined. See the Java MIDP
specification for further information:
|
Constant
|
Value
|
Meaning
|
|
CK_SECURITY_DOMAIN_UNSPECIFIED
|
0x00000000UL
|
No domain specified
|
|
CK_SECURITY_DOMAIN_MANUFACTURER
|
0x00000001UL
|
Manufacturer protection domain
|
|
CK_SECURITY_DOMAIN_OPERATOR
|
0x00000002UL
|
Operator protection domain
|
|
CK_SECURITY_DOMAIN_THIRD_PARTY
|
0x00000003UL
|
Third party protection domain
|
Cryptoki supports the following types for describing
mechanisms and parameters to them:
¨
CK_MECHANISM_TYPE; CK_MECHANISM_TYPE_PTR
CK_MECHANISM_TYPE is a value that identifies a
mechanism type. It is defined as follows:
typedef CK_ULONG CK_MECHANISM_TYPE;
Mechanism types are defined with the objects and mechanism
descriptions that use them.
Vendor defined values for this type may also be specified.
Mechanism types CKM_VENDOR_DEFINED
and above are permanently reserved for token vendors. For interoperability,
vendors should register their mechanism types through the PKCS process.
CK_MECHANISM_TYPE_PTR is a pointer to a CK_MECHANISM_TYPE.
¨
CK_MECHANISM; CK_MECHANISM_PTR
CK_MECHANISM is a structure that specifies a
particular mechanism and any parameters it requires. It is defined as follows:
typedef struct CK_MECHANISM {
CK_MECHANISM_TYPE mechanism;
CK_VOID_PTR pParameter;
CK_ULONG ulParameterLen;
} CK_MECHANISM;
The fields of the structure have the following meanings:
mechanism the
type of mechanism
pParameter pointer
to the parameter if required by the mechanism
ulParameterLen length
in bytes of the parameter
Note that pParameter is a
“void” pointer, facilitating the passing of arbitrary values. Both the
application and the Cryptoki library MUST ensure that the pointer can be safely
cast to the expected type (i.e., without word-alignment errors).
CK_MECHANISM_PTR is a pointer to a CK_MECHANISM.
¨
CK_MECHANISM_INFO; CK_MECHANISM_INFO_PTR
CK_MECHANISM_INFO is a structure that provides
information about a particular mechanism. It is defined as follows:
typedef struct CK_MECHANISM_INFO {
CK_ULONG ulMinKeySize;
CK_ULONG ulMaxKeySize;
CK_FLAGS flags;
} CK_MECHANISM_INFO;
The fields of the structure have the following meanings:
ulMinKeySize the
minimum size of the key for the mechanism (whether this is measured in bits or
in bytes is mechanism-dependent)
ulMaxKeySize the
maximum size of the key for the mechanism (whether this is measured in bits or
in bytes is mechanism-dependent)
flags bit
flags specifying mechanism capabilities
For some mechanisms, the ulMinKeySize and ulMaxKeySize
fields have meaningless values.
The following table defines the flags field:
Table 8, Mechanism Information Flags
|
Bit
Flag
|
Mask
|
Meaning
|
|
CKF_HW
|
0x00000001
|
True
if the mechanism is performed by the device; false if the mechanism is
performed in software
|
|
CKF_MESSAGE_ENCRYPT
|
0x00000002
|
True
if the mechanism can be used with C_MessageEncryptInit
|
|
CKF_MESSAGE_DECRYPT
|
0x00000004
|
True
if the mechanism can be used with C_MessageDecryptInit
|
|
CKF_MESSAGE_SIGN
|
0x00000008
|
True
if the mechanism can be used with C_MessageSignInit
|
|
CKF_MESSAGE_VERIFY
|
0x00000010
|
True
if the mechanism can be used with C_MessageVerifyInit
|
|
CKF_MULTI_MESSAGE
|
0x00000020
|
True
if the mechanism can be used with C_*MessageBegin. One of
CKF_MESSAGE_* flag must also be set.
|
|
CKF_FIND_OBJECTS
|
0x00000040
|
This
flag can be passed in as a parameter to C_SessionCancel to cancel an
active object search operation. Any other use of this flag is outside the
scope of this standard.
|
|
CKF_ENCRYPT
|
0x00000100
|
True
if the mechanism can be used with C_EncryptInit
|
|
CKF_DECRYPT
|
0x00000200
|
True
if the mechanism can be used with C_DecryptInit
|
|
CKF_DIGEST
|
0x00000400
|
True
if the mechanism can be used with C_DigestInit
|
|
CKF_SIGN
|
0x00000800
|
True
if the mechanism can be used with C_SignInit
|
|
CKF_SIGN_RECOVER
|
0x00001000
|
True
if the mechanism can be used with C_SignRecoverInit
|
|
CKF_VERIFY
|
0x00002000
|
True
if the mechanism can be used with C_VerifyInit
|
|
CKF_VERIFY_RECOVER
|
0x00004000
|
True
if the mechanism can be used with C_VerifyRecoverInit
|
|
CKF_GENERATE
|
0x00008000
|
True
if the mechanism can be used with C_GenerateKey
|
|
CKF_GENERATE_KEY_PAIR
|
0x00010000
|
True
if the mechanism can be used with C_GenerateKeyPair
|
|
CKF_WRAP
|
0x00020000
|
True
if the mechanism can be used with C_WrapKey
|
|
CKF_UNWRAP
|
0x00040000
|
True
if the mechanism can be used with C_UnwrapKey
|
|
CKF_DERIVE
|
0x00080000
|
True
if the mechanism can be used with C_DeriveKey
|
|
CKF_EXTENSION
|
0x80000000
|
True
if there is an extension to the flags; false if no extensions. MUST be false
for this version.
|
CK_MECHANISM_INFO_PTR is a pointer to a CK_MECHANISM_INFO.
Cryptoki represents information about functions with the
following data types:
¨
CK_RV
CK_RV is a value that identifies the return value of
a Cryptoki function. It is defined as follows:
Vendor defined values for this type may also be specified.
Section 5.1 defines the meaning of each CK_RV value.
Return values CKR_VENDOR_DEFINED and above are permanently reserved for
token vendors. For interoperability, vendors should register their return
values through the PKCS process.
¨ CK_NOTIFY
CK_NOTIFY is the type of a pointer to a function used
by Cryptoki to perform notification callbacks. It is defined as follows:
typedef CK_CALLBACK_FUNCTION(CK_RV, CK_NOTIFY)(
CK_SESSION_HANDLE hSession,
CK_NOTIFICATION event,
CK_VOID_PTR pApplication
);
The arguments to a notification callback function have the
following meanings:
hSession The
handle of the session performing the callback
event The
type of notification callback
pApplication An
application-defined value. This is the same value as was passed to C_OpenSession
to open the session performing the callback
¨
CK_C_XXX
Cryptoki also defines an entire family of other function
pointer types. For each function C_XXX in the Cryptoki API (see Section
4.12 for detailed information about each of them), Cryptoki defines a type CK_C_XXX,
which is a pointer to a function with the same arguments and return value as C_XXX
has. An appropriately-set variable of type CK_C_XXX may be used by an
application to call the Cryptoki function C_XXX.
¨ CK_FUNCTION_LIST; CK_FUNCTION_LIST_PTR;
CK_FUNCTION_LIST_PTR_PTR
CK_FUNCTION_LIST is a structure which contains a
Cryptoki version and a function pointer to each function in the Cryptoki API.
It is defined as follows:
typedef struct CK_FUNCTION_LIST {
CK_VERSION version;
CK_C_Initialize C_Initialize;
CK_C_Finalize C_Finalize;
CK_C_GetInfo C_GetInfo;
CK_C_GetFunctionList C_GetFunctionList;
CK_C_GetSlotList
C_GetSlotList;
CK_C_GetSlotInfo
C_GetSlotInfo;
CK_C_GetTokenInfo C_GetTokenInfo;
CK_C_GetMechanismList
C_GetMechanismList;
CK_C_GetMechanismInfo C_GetMechanismInfo;
CK_C_InitToken C_InitToken;
CK_C_InitPIN C_InitPIN;
CK_C_SetPIN C_SetPIN;
CK_C_OpenSession
C_OpenSession;
CK_C_CloseSession C_CloseSession;
CK_C_CloseAllSessions C_CloseAllSessions;
CK_C_GetSessionInfo C_GetSessionInfo;
CK_C_GetOperationState C_GetOperationState;
CK_C_SetOperationState C_SetOperationState;
CK_C_Login C_Login;
CK_C_Logout C_Logout;
CK_C_CreateObject C_CreateObject;
CK_C_CopyObject C_CopyObject;
CK_C_DestroyObject C_DestroyObject;
CK_C_GetObjectSize C_GetObjectSize;
CK_C_GetAttributeValue C_GetAttributeValue;
CK_C_SetAttributeValue C_SetAttributeValue;
CK_C_FindObjectsInit C_FindObjectsInit;
CK_C_FindObjects C_FindObjects;
CK_C_FindObjectsFinal C_FindObjectsFinal;
CK_C_EncryptInit C_EncryptInit;
CK_C_Encrypt C_Encrypt;
CK_C_EncryptUpdate C_EncryptUpdate;
CK_C_EncryptFinal C_EncryptFinal;
CK_C_DecryptInit C_DecryptInit;
CK_C_Decrypt C_Decrypt;
CK_C_DecryptUpdate C_DecryptUpdate;
CK_C_DecryptFinal C_DecryptFinal;
CK_C_DigestInit
C_DigestInit;
CK_C_Digest C_Digest;
CK_C_DigestUpdate C_DigestUpdate;
CK_C_DigestKey C_DigestKey;
CK_C_DigestFinal C_DigestFinal;
CK_C_SignInit
C_SignInit;
CK_C_Sign C_Sign;
CK_C_SignUpdate C_SignUpdate;
CK_C_SignFinal C_SignFinal;
CK_C_SignRecoverInit C_SignRecoverInit;
CK_C_SignRecover C_SignRecover;
CK_C_VerifyInit C_VerifyInit;
CK_C_Verify C_Verify;
CK_C_VerifyUpdate C_VerifyUpdate;
CK_C_VerifyFinal C_VerifyFinal;
CK_C_VerifyRecoverInit C_VerifyRecoverInit;
CK_C_VerifyRecover C_VerifyRecover;
CK_C_DigestEncryptUpdate
C_DigestEncryptUpdate;
CK_C_DecryptDigestUpdate C_DecryptDigestUpdate;
CK_C_SignEncryptUpdate
C_SignEncryptUpdate;
CK_C_DecryptVerifyUpdate C_DecryptVerifyUpdate;
CK_C_GenerateKey C_GenerateKey;
CK_C_GenerateKeyPair C_GenerateKeyPair;
CK_C_WrapKey C_WrapKey;
CK_C_UnwrapKey C_UnwrapKey;
CK_C_DeriveKey C_DeriveKey;
CK_C_SeedRandom C_SeedRandom;
CK_C_GenerateRandom C_GenerateRandom;
CK_C_GetFunctionStatus C_GetFunctionStatus;
CK_C_CancelFunction C_CancelFunction;
CK_C_WaitForSlotEvent C_WaitForSlotEvent;
} CK_FUNCTION_LIST;
Each Cryptoki library has a static CK_FUNCTION_LIST
structure, and a pointer to it (or to a copy of it which is also owned by the
library) may be obtained by the C_GetFunctionList function (see Section 5.2). The value that this pointer points to can be used by an application to quickly
find out where the executable code for each function in the Cryptoki API is
located. Every function in the Cryptoki API MUST have an entry point defined
in the Cryptoki library’s CK_FUNCTION_LIST structure. If a particular
function in the Cryptoki API is not supported by a library, then the function
pointer for that function in the library’s CK_FUNCTION_LIST structure
should point to a function stub which simply returns
CKR_FUNCTION_NOT_SUPPORTED.
In this structure ‘version’ is the cryptoki specification
version number. The major and minor versions must be set to 0x02 and 0x28
indicating a version 2.40 compatible structure. The updated function list table
for this version of the specification may be returned via C_GetInterfaceList
or C_GetInterface.
An application may or may not be able to modify a Cryptoki
library’s static CK_FUNCTION_LIST structure. Whether or not it can, it
should never attempt to do so.
PKCS #11 modules must not add new functions at the end of
the CK_FUNCTION_LIST that are not contained within the defined
structure. If a PKCS#11 module needs to define additional functions, they
should be placed within a vendor defined interface returned via C_GetInterfaceList
or C_GetInterface.
CK_FUNCTION_LIST_PTR is a pointer to a CK_FUNCTION_LIST.
CK_FUNCTION_LIST_PTR_PTR is a pointer to a CK_FUNCTION_LIST_PTR.
¨ CK_FUNCTION_LIST_3_0;
CK_FUNCTION_LIST_3_0_PTR; CK_FUNCTION_LIST_3_0_PTR_PTR
CK_FUNCTION_LIST_3_0 is a structure which contains
the same function pointers as in CK_FUNCTION_LIST and additional
functions added to the end of the structure that were defined in Cryptoki
version 3.0. It is defined as follows:
typedef struct CK_FUNCTION_LIST_3_0 {
CK_VERSION version;
CK_C_Initialize C_Initialize;
CK_C_Finalize C_Finalize;
CK_C_GetInfo C_GetInfo;
CK_C_GetFunctionList C_GetFunctionList;
CK_C_GetSlotList
C_GetSlotList;
CK_C_GetSlotInfo
C_GetSlotInfo;
CK_C_GetTokenInfo C_GetTokenInfo;
CK_C_GetMechanismList
C_GetMechanismList;
CK_C_GetMechanismInfo C_GetMechanismInfo;
CK_C_InitToken C_InitToken;
CK_C_InitPIN C_InitPIN;
CK_C_SetPIN C_SetPIN;
CK_C_OpenSession
C_OpenSession;
CK_C_CloseSession C_CloseSession;
CK_C_CloseAllSessions C_CloseAllSessions;
CK_C_GetSessionInfo C_GetSessionInfo;
CK_C_GetOperationState C_GetOperationState;
CK_C_SetOperationState C_SetOperationState;
CK_C_Login C_Login;
CK_C_Logout C_Logout;
CK_C_CreateObject C_CreateObject;
CK_C_CopyObject C_CopyObject;
CK_C_DestroyObject C_DestroyObject;
CK_C_GetObjectSize C_GetObjectSize;
CK_C_GetAttributeValue C_GetAttributeValue;
CK_C_SetAttributeValue C_SetAttributeValue;
CK_C_FindObjectsInit C_FindObjectsInit;
CK_C_FindObjects C_FindObjects;
CK_C_FindObjectsFinal C_FindObjectsFinal;
CK_C_EncryptInit C_EncryptInit;
CK_C_Encrypt C_Encrypt;
CK_C_EncryptUpdate C_EncryptUpdate;
CK_C_EncryptFinal C_EncryptFinal;
CK_C_DecryptInit C_DecryptInit;
CK_C_Decrypt C_Decrypt;
CK_C_DecryptUpdate C_DecryptUpdate;
CK_C_DecryptFinal C_DecryptFinal;
CK_C_DigestInit
C_DigestInit;
CK_C_Digest C_Digest;
CK_C_DigestUpdate C_DigestUpdate;
CK_C_DigestKey C_DigestKey;
CK_C_DigestFinal C_DigestFinal;
CK_C_SignInit
C_SignInit;
CK_C_Sign C_Sign;
CK_C_SignUpdate C_SignUpdate;
CK_C_SignFinal C_SignFinal;
CK_C_SignRecoverInit C_SignRecoverInit;
CK_C_SignRecover C_SignRecover;
CK_C_VerifyInit C_VerifyInit;
CK_C_Verify C_Verify;
CK_C_VerifyUpdate C_VerifyUpdate;
CK_C_VerifyFinal C_VerifyFinal;
CK_C_VerifyRecoverInit C_VerifyRecoverInit;
CK_C_VerifyRecover C_VerifyRecover;
CK_C_DigestEncryptUpdate
C_DigestEncryptUpdate;
CK_C_DecryptDigestUpdate C_DecryptDigestUpdate;
CK_C_SignEncryptUpdate
C_SignEncryptUpdate;
CK_C_DecryptVerifyUpdate C_DecryptVerifyUpdate;
CK_C_GenerateKey C_GenerateKey;
CK_C_GenerateKeyPair C_GenerateKeyPair;
CK_C_WrapKey C_WrapKey;
CK_C_UnwrapKey C_UnwrapKey;
CK_C_DeriveKey C_DeriveKey;
CK_C_SeedRandom C_SeedRandom;
CK_C_GenerateRandom C_GenerateRandom;
CK_C_GetFunctionStatus C_GetFunctionStatus;
CK_C_CancelFunction C_CancelFunction;
CK_C_WaitForSlotEvent C_WaitForSlotEvent;
CK_C_GetInterfaceList
C_GetInterfaceList;
CK_C_GetInterface
C_GetInterface;
CK_C_LoginUser
C_LoginUser;
CK_C_SessionCancel
C_SessionCancel;
CK_C_MessageEncryptInit C_MessageEncryptInit;
CK_C_EncryptMessage C_EncryptMessage;
CK_C_EncryptMessageBegin C_EncryptMessageBegin;
CK_C_EncryptMessageNext C_EncryptMessageNext;
CK_C_MessageEncryptFinal C_MessageEncryptFinal;
CK_C_MessageDecryptInit C_MessageDecryptInit;
CK_C_DecryptMessage C_DecryptMessage;
CK_C_DecryptMessageBegin C_DecryptMessageBegin;
CK_C_DecryptMessageNext C_DecryptMessageNext;
CK_C_MessageDecryptFinal C_MessageDecryptFinal;
CK_C_MessageSignInit C_MessageSignInit;
CK_C_SignMessage C_SignMessage;
CK_C_SignMessageBegin C_SignMessageBegin;
CK_C_SignMessageNext C_SignMessageNext;
CK_C_MessageSignFinal C_MessageSignFinal;
CK_C_MessageVerifyInit C_MessageVerifyInit;
CK_C_VerifyMessage C_VerifyMessage;
CK_C_VerifyMessageBegin C_VerifyMessageBegin;
CK_C_VerifyMessageNext C_VerifyMessageNext;
CK_C_MessageVerifyFinal C_MessageVerifyFinal;
} CK_FUNCTION_LIST_3_0;
For a general description of CK_FUNCTION_LIST_3_0 see
CK_FUNCTION_LIST.
In this structure, version is the cryptoki
specification version number. It should match the value of cryptokiVersion
returned in the CK_INFO structure, but must be 3.0 at minimum.
This function list may be returned via C_GetInterfaceList
or C_GetInterface
CK_FUNCTION_LIST_3_0_PTR is a pointer to a CK_FUNCTION_LIST_3_0.
CK_FUNCTION_LIST_3_0_PTR_PTR is a pointer to a CK_FUNCTION_LIST_3_0_PTR.
¨
CK_INTERFACE;
CK_INTERFACE_PTR; CK_INTERFACE_PTR_PTR
CK_INTERFACE is a structure which contains an
interface name with a function list and flag.
It is defined as follows:
typedef struct CK_INTERFACE {
CK_UTF8CHAR_PTR pInterfaceName;
CK_VOID_PTR pFunctionList;
CK_FLAGS flags;
} CK_INTERFACE;
The fields of the structure have the following meanings:
pInterfaceName the
name of the interface
pFunctionList the
interface function list which must always begin with a CK_VERSION structure as
the first field
flags bit
flags specifying interface capabilities
The interface name “PKCS 11” is reserved for use by
interfaces defined within the cryptoki specification.
Interfaces starting with the string: “Vendor ” are reserved
for vendor use and will not oetherwise be defined as interfaces in the PKCS #11
specification. Vendors should supply new functions with interface names of
“Vendor {vendor name}”. For example “Vendor ACME Inc”.
The following table defines
the flags field:
Table 9, CK_INTERFACE Flags
|
Bit Flag
|
Mask
|
Meaning
|
|
CKF_INTERFACE_FORK_SAFE
|
0x00000001
|
The returned interface will have fork
tolerant semantics. When the application forks, each process will get its own
copy of all session objects, session states, login states, and encryption
states. Each process will also maintain access to token objects with their previously
supplied handles.
|
CK_INTERFACE_PTR is a pointer to a CK_INTERFACE.
CK_INTERFACE_PTR_PTR is a pointer to a CK_INTERFACE_PTR.
The types in this section are provided solely for
applications which need to access Cryptoki from multiple threads
simultaneously. Applications which will not do this need not use any of
these types.
¨
CK_CREATEMUTEX
CK_CREATEMUTEX is the type of a pointer to an
application-supplied function which creates a new mutex object and returns a
pointer to it. It is defined as follows:
typedef CK_CALLBACK_FUNCTION(CK_RV, CK_CREATEMUTEX)(
CK_VOID_PTR_PTR ppMutex
);
Calling a CK_CREATEMUTEX function returns the pointer to the
new mutex object in the location pointed to by ppMutex. Such a function should
return one of the following values:
CKR_OK, CKR_GENERAL_ERROR
CKR_HOST_MEMORY
¨
CK_DESTROYMUTEX
CK_DESTROYMUTEX is the type of a pointer to an
application-supplied function which destroys an existing mutex object. It is
defined as follows:
typedef CK_CALLBACK_FUNCTION(CK_RV, CK_DESTROYMUTEX)(
CK_VOID_PTR pMutex
);
The argument to a CK_DESTROYMUTEX function is a pointer to
the mutex object to be destroyed. Such a function should return one of the
following values:
CKR_OK, CKR_GENERAL_ERROR
CKR_HOST_MEMORY
CKR_MUTEX_BAD
¨
CK_LOCKMUTEX and CK_UNLOCKMUTEX
CK_LOCKMUTEX is the type of a pointer to an
application-supplied function which locks an existing mutex object. CK_UNLOCKMUTEX
is the type of a pointer to an application-supplied function which unlocks an
existing mutex object. The proper behavior for these types of functions is as
follows:
·
If a CK_LOCKMUTEX function is called on a mutex which is not
locked, the calling thread obtains a lock on that mutex and returns.
·
If a CK_LOCKMUTEX function is called on a mutex which is locked
by some thread other than the calling thread, the calling thread blocks and
waits for that mutex to be unlocked.
·
If a CK_LOCKMUTEX function is called on a mutex which is locked
by the calling thread, the behavior of the function call is undefined.
·
If a CK_UNLOCKMUTEX function is called on a mutex which is locked
by the calling thread, that mutex is unlocked and the function call returns.
Furthermore:
o
If exactly one thread was blocking on that particular mutex, then
that thread stops blocking, obtains a lock on that mutex, and its CK_LOCKMUTEX
call returns.
o
If more than one thread was blocking on that particular mutex,
then exactly one of the blocking threads is selected somehow. That lucky
thread stops blocking, obtains a lock on the mutex, and its CK_LOCKMUTEX call
returns. All other threads blocking on that particular mutex continue to
block.
·
If a CK_UNLOCKMUTEX function is called on a mutex which is not
locked, then the function call returns the error code CKR_MUTEX_NOT_LOCKED.
·
If a CK_UNLOCKMUTEX function is called on a mutex which is locked
by some thread other than the calling thread, the behavior of the function call
is undefined.
CK_LOCKMUTEX is defined as follows:
typedef CK_CALLBACK_FUNCTION(CK_RV, CK_LOCKMUTEX)(
CK_VOID_PTR pMutex
);
The argument to a CK_LOCKMUTEX function is a pointer to the
mutex object to be locked. Such a function should return one of the following
values:
CKR_OK, CKR_GENERAL_ERROR
CKR_HOST_MEMORY,
CKR_MUTEX_BAD
CK_UNLOCKMUTEX is defined as follows:
typedef CK_CALLBACK_FUNCTION(CK_RV, CK_UNLOCKMUTEX)(
CK_VOID_PTR pMutex
);
The argument to a CK_UNLOCKMUTEX function is a pointer to
the mutex object to be unlocked. Such a function should return one of the
following values:
CKR_OK, CKR_GENERAL_ERROR
CKR_HOST_MEMORY
CKR_MUTEX_BAD
CKR_MUTEX_NOT_LOCKED
¨
CK_C_INITIALIZE_ARGS; CK_C_INITIALIZE_ARGS_PTR
CK_C_INITIALIZE_ARGS is a structure containing the
optional arguments for the C_Initialize function. For this version of
Cryptoki, these optional arguments are all concerned with the way the library
deals with threads. CK_C_INITIALIZE_ARGS is defined as follows:
typedef struct CK_C_INITIALIZE_ARGS {
CK_CREATEMUTEX CreateMutex;
CK_DESTROYMUTEX DestroyMutex;
CK_LOCKMUTEX LockMutex;
CK_UNLOCKMUTEX UnlockMutex;
CK_FLAGS flags;
CK_VOID_PTR pReserved;
} CK_C_INITIALIZE_ARGS;
The fields of the structure have the following meanings:
CreateMutex pointer
to a function to use for creating mutex objects
DestroyMutex pointer
to a function to use for destroying mutex objects
LockMutex pointer
to a function to use for locking mutex objects
UnlockMutex pointer
to a function to use for unlocking mutex objects
flags bit
flags specifying options for C_Initialize; the flags are defined below
pReserved reserved
for future use. Should be NULL_PTR for this version of Cryptoki
The following table defines
the flags field:
Table 10, C_Initialize Parameter Flags
|
Bit Flag
|
Mask
|
Meaning
|
|
CKF_LIBRARY_CANT_CREATE_OS_THREADS
|
0x00000001
|
True if application threads which are
executing calls to the library may not use native operating system
calls to spawn new threads; false if they may
|
|
CKF_OS_LOCKING_OK
|
0x00000002
|
True if the library can use the native
operation system threading model for locking; false otherwise
|
CK_C_INITIALIZE_ARGS_PTR
is a pointer to a CK_C_INITIALIZE_ARGS.
Cryptoki recognizes a number of classes of objects, as
defined in the CK_OBJECT_CLASS data type. An object consists of a set
of attributes, each of which has a given value. Each attribute that an object
possesses has precisely one value. The following figure illustrates the
high-level hierarchy of the Cryptoki objects and some of the attributes they
support:
Figure 1, Object Attribute
Hierarchy
Cryptoki provides functions for creating, destroying, and
copying objects in general, and for obtaining and modifying the values of their
attributes. Some of the cryptographic functions (e.g., C_GenerateKey)
also create key objects to hold their results.
Objects are always “well-formed” in Cryptoki—that is, an
object always contains all required attributes, and the attributes are always
consistent with one another from the time the object is created. This
contrasts with some object-based paradigms where an object has no attributes
other than perhaps a class when it is created, and is uninitialized for some
time. In Cryptoki, objects are always initialized.
Tables throughout most of Section 4
define each Cryptoki attribute in terms of the data type of the attribute value
and the meaning of the attribute, which may include a default initial value.
Some of the data types are defined explicitly by Cryptoki (e.g., CK_OBJECT_CLASS).
Attribute values may also take the following types:
Byte array an
arbitrary string (array) of CK_BYTEs
Big integer a string
of CK_BYTEs representing an unsigned integer of arbitrary size,
most-significant byte first (e.g., the integer 32768 is represented as
the 2-byte string 0x80 0x00)
Local string an
unpadded string of CK_CHARs (see Table 3) with no null-termination
RFC2279 string an unpadded
string of CK_UTF8CHARs
with no null-termination
A token can hold several identical objects, i.e., it
is permissible for two or more objects to have exactly the same values for all
their attributes.
In most cases each type of object in the Cryptoki
specification possesses a completely well-defined set of Cryptoki attributes.
Some of these attributes possess default values, and need not be specified when
creating an object; some of these default values may even be the empty string
(“”). Nonetheless, the object possesses these attributes. A given object has
a single value for each attribute it possesses, even if the attribute is a
vendor-specific attribute whose meaning is outside the scope of Cryptoki.
In addition to possessing Cryptoki attributes, objects may
possess additional vendor-specific attributes whose meanings and values are not
specified by Cryptoki.
All Cryptoki functions that create, modify, or copy objects
take a template as one of their arguments, where the template specifies
attribute values. Cryptographic functions that create objects (see Section 5.18) may also contribute some additional attribute values themselves; which
attributes have values contributed by a cryptographic function call depends on
which cryptographic mechanism is being performed (see section 6
Mechanisms and [PKCS11-Hist] for specification of mechanisms for PKCS #11). In
any case, all the required attributes supported by an object class that do not
have default values MUST be specified when an object is created, either in the
template or by the function itself.
Objects may be created with the Cryptoki functions C_CreateObject
(see Section 5.7), C_GenerateKey, C_GenerateKeyPair, C_UnwrapKey,
and C_DeriveKey (see Section 5.18). In addition, copying an existing
object (with the function C_CopyObject) also creates a new object, but
we consider this type of object creation separately in Section 4.1.3.
Attempting to create an object with any of these functions
requires an appropriate template to be supplied.
1. If the
supplied template specifies a value for an invalid attribute, then the attempt
should fail with the error code CKR_ATTRIBUTE_TYPE_INVALID. An attribute is
valid if it is either one of the attributes described in the Cryptoki
specification or an additional vendor-specific attribute supported by the
library and token.
2. If the supplied
template specifies an invalid value for a valid attribute, then the attempt
should fail with the error code CKR_ATTRIBUTE_VALUE_INVALID. The valid values
for Cryptoki attributes are described in the Cryptoki specification.
3. If the
supplied template specifies a value for a read-only attribute, then the attempt
should fail with the error code CKR_ATTRIBUTE_READ_ONLY. Whether or not a
given Cryptoki attribute is read-only is explicitly stated in the Cryptoki
specification; however, a particular library and token may be even more
restrictive than Cryptoki specifies. In other words, an attribute which
Cryptoki says is not read-only may nonetheless be read-only under certain
circumstances (i.e., in conjunction with some combinations of other
attributes) for a particular library and token. Whether or not a given
non-Cryptoki attribute is read-only is obviously outside the scope of Cryptoki.
4. If the
attribute values in the supplied template, together with any default attribute
values and any attribute values contributed to the object by the
object-creation function itself, are insufficient to fully specify the object
to create, then the attempt should fail with the error code
CKR_TEMPLATE_INCOMPLETE.
5. If the
attribute values in the supplied template, together with any default attribute
values and any attribute values contributed to the object by the
object-creation function itself, are inconsistent, then the attempt should fail
with the error code CKR_TEMPLATE_INCONSISTENT. A set of attribute values is inconsistent
if not all of its members can be satisfied simultaneously by the token,
although each value individually is valid in Cryptoki. One example of an
inconsistent template would be using a template which specifies two different
values for the same attribute. Another example would be trying to create a
secret key object with an attribute which is appropriate for various types of
public keys or private keys, but not for secret keys. A final example would be
a template with an attribute that violates some token specific requirement.
Note that this final example of an inconsistent template is token-dependent—on
a different token, such a template might not be inconsistent.
6. If the
supplied template specifies the same value for a particular attribute more than
once (or the template specifies the same value for a particular attribute that
the object-creation function itself contributes to the object), then the
behavior of Cryptoki is not completely specified. The attempt to create an
object can either succeed—thereby creating the same object that would have been
created if the multiply-specified attribute had only appeared once—or it can
fail with error code CKR_TEMPLATE_INCONSISTENT. Library developers are
encouraged to make their libraries behave as though the attribute had only
appeared once in the template; application developers are strongly encouraged
never to put a particular attribute into a particular template more than once.
If more than one of the situations listed above applies to
an attempt to create an object, then the error code returned from the attempt
can be any of the error codes from above that applies.
Objects may be modified with the Cryptoki function C_SetAttributeValue
(see Section 5.7). The template supplied to C_SetAttributeValue can
contain new values for attributes which the object already possesses; values
for attributes which the object does not yet possess; or both.
Some attributes of an object may be modified after the
object has been created, and some may not. In addition, attributes which
Cryptoki specifies are modifiable may actually not be modifiable on some
tokens. That is, if a Cryptoki attribute is described as being modifiable,
that really means only that it is modifiable insofar as the Cryptoki
specification is concerned. A particular token might not actually support
modification of some such attributes. Furthermore, whether or not a particular
attribute of an object on a particular token is modifiable might depend on the
values of certain attributes of the object. For example, a secret key object’s
CKA_SENSITIVE attribute can be changed from CK_FALSE to CK_TRUE, but not
the other way around.
All the scenarios in Section 4.1.1—and
the error codes they return—apply to modifying objects with C_SetAttributeValue,
except for the possibility of a template being incomplete.
Unless an object's CKA_COPYABLE
(see Table 17) attribute is set to CK_FALSE,
it may be copied with the Cryptoki function C_CopyObject (see
Section 5.7). In the process of copying an object, C_CopyObject also
modifies the attributes of the newly-created copy according to an
application-supplied template.
The Cryptoki attributes which can be modified during the
course of a C_CopyObject operation are the same as the Cryptoki
attributes which are described as being modifiable, plus the four special
attributes CKA_TOKEN, CKA_PRIVATE, CKA_MODIFIABLE and
CKA_DESTROYABLE. To be more precise, these attributes are modifiable
during the course of a C_CopyObject operation insofar as the Cryptoki
specification is concerned. A particular token might not actually support
modification of some such attributes during the course of a C_CopyObject
operation. Furthermore, whether or not a particular attribute of an object on
a particular token is modifiable during the course of a C_CopyObject
operation might depend on the values of certain attributes of the object. For
example, a secret key object’s CKA_SENSITIVE attribute can be changed
from CK_FALSE to CK_TRUE during the course of a C_CopyObject operation,
but not the other way around.
If the CKA_COPYABLE attribute of the object to be copied is
set to CK_FALSE, C_CopyObject returns CKR_ACTION_PROHIBITED. Otherwise, the
scenarios described in 10.1.1 - and the error codes they return - apply to
copying objects with C_CopyObject, except for the possibility of a template
being incomplete.
4.2
Common attributes
Table 11, Common footnotes for object attribute tables
|
1 MUST be specified when object is created
with C_CreateObject.
2 MUST not be specified when object is
created with C_CreateObject.
3 MUST be specified when object is generated
with C_GenerateKey or C_GenerateKeyPair.
4 MUST not be specified when object is
generated with C_GenerateKey or C_GenerateKeyPair.
5 MUST be specified when object is unwrapped
with C_UnwrapKey.
6 MUST not be specified when object is
unwrapped with C_UnwrapKey.
7 Cannot be revealed if object has its CKA_SENSITIVE
attribute set to CK_TRUE or its CKA_EXTRACTABLE attribute set to
CK_FALSE.
8 May be modified after object is created with
a C_SetAttributeValue call, or in the process of copying object with a
C_CopyObject call. However, it is possible that a particular token
may not permit modification of the attribute during the course of a C_CopyObject
call.
9 Default value is token-specific, and may
depend on the values of other attributes.
10 Can only be set to CK_TRUE by the SO user.
11 Attribute cannot be changed once set to
CK_TRUE. It becomes a read only attribute.
12 Attribute cannot be changed once set to
CK_FALSE. It becomes a read only attribute.
|
Table 12, Common Object Attributes
|
Attribute
|
Data Type
|
Meaning
|
|
CKA_CLASS1
|
CK_OBJECT_CLASS
|
Object class (type)
|
Refer to Table 11 for footnotes
The above table defines the attributes common to all
objects.
This section defines the object class CKO_HW_FEATURE for
type CK_OBJECT_CLASS as used in the CKA_CLASS attribute of objects.
Hardware feature objects (CKO_HW_FEATURE) represent
features of the device. They provide an easily expandable method for
introducing new value-based features to the Cryptoki interface.
When searching for objects using C_FindObjectsInit
and C_FindObjects, hardware feature objects are not returned unless the CKA_CLASS
attribute in the template has the value CKO_HW_FEATURE. This protects
applications written to previous versions of Cryptoki from finding objects that
they do not understand.
Table 13, Hardware Feature
Common Attributes
|
Attribute
|
Data Type
|
Meaning
|
|
CKA_HW_FEATURE_TYPE1
|
CK_HW_FEATURE_TYPE
|
Hardware feature (type)
|
- Refer to Table 11 for footnotes
The CKA_HW_FEATURE_TYPE attribute takes the value CKH_CLOCK
of type CK_HW_FEATURE_TYPE.
Clock objects represent real-time clocks that exist on the
device. This represents the same clock source as the utcTime field in
the CK_TOKEN_INFO structure.
Table 14, Clock Object
Attributes
|
Attribute
|
Data Type
|
Meaning
|
|
CKA_VALUE
|
CK_CHAR[16]
|
Current time as a character-string of length
16, represented in the format YYYYMMDDhhmmssxx (4 characters for the year; 2
characters each for the month, the day, the hour, the minute, and the second;
and 2 additional reserved ‘0’ characters).
|
The CKA_VALUE attribute may be set using the C_SetAttributeValue
function if permitted by the device. The session used to set the time MUST be
logged in. The device may require the SO to be the user logged in to modify the
time value. C_SetAttributeValue will return the error
CKR_USER_NOT_LOGGED_IN to indicate that a different user type is required to
set the value.
The CKA_HW_FEATURE_TYPE attribute takes the value CKH_MONOTONIC_COUNTER
of type CK_HW_FEATURE_TYPE.
Monotonic counter objects represent hardware counters that
exist on the device. The counter is guaranteed to increase each time its value
is read, but not necessarily by one. This might be used by an application for
generating serial numbers to get some assurance of uniqueness per token.
Table 15, Monotonic Counter
Attributes
|
Attribute
|
Data Type
|
Meaning
|
|
CKA_RESET_ON_INIT1
|
CK_BBOOL
|
The value of the counter will reset to a
previously returned value if the token is initialized using C_InitToken.
|
|
CKA_HAS_RESET1
|
CK_BBOOL
|
The value of the counter has been reset at
least once at some point in time.
|
|
CKA_VALUE1
|
Byte Array
|
The current version of the monotonic counter.
The value is returned in big endian order.
|
1Read Only
The CKA_VALUE attribute may not be set by the client.
The CKA_HW_FEATURE_TYPE attribute takes the value
CKH_USER_INTERFACE of type CK_HW_FEATURE_TYPE.
User interface objects represent the presentation
capabilities of the device.
Table 16, User Interface Object
Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_PIXEL_X
|
CK_ULONG
|
Screen resolution (in pixels) in X-axis
(e.g. 1280)
|
|
CKA_PIXEL_Y
|
CK_ULONG
|
Screen resolution (in pixels) in Y-axis
(e.g. 1024)
|
|
CKA_RESOLUTION
|
CK_ULONG
|
DPI, pixels per inch
|
|
CKA_CHAR_ROWS
|
CK_ULONG
|
For character-oriented displays; number of
character rows (e.g. 24)
|
|
CKA_CHAR_COLUMNS
|
CK_ULONG
|
For character-oriented displays: number of
character columns (e.g. 80). If display is of proportional-font type, this
is the width of the display in “em”-s (letter “M”), see CC/PP Struct.
|
|
CKA_COLOR
|
CK_BBOOL
|
Color support
|
|
CKA_BITS_PER_PIXEL
|
CK_ULONG
|
The
number of bits of color or grayscale information per pixel.
|
|
CKA_CHAR_SETS
|
RFC 2279 string
|
String indicating supported character sets,
as defined by IANA MIBenum sets (www.iana.org). Supported
character sets are separated with “;”. E.g. a token supporting iso-8859-1
and US-ASCII would set the attribute value to “4;3”.
|
|
CKA_ENCODING_METHODS
|
RFC 2279 string
|
String indicating supported content transfer
encoding methods, as defined by IANA (www.iana.org). Supported methods
are separated with “;”. E.g. a token supporting 7bit, 8bit and base64 could
set the attribute value to “7bit;8bit;base64”.
|
|
CKA_MIME_TYPES
|
RFC 2279 string
|
String indicating supported (presentable)
MIME-types, as defined by IANA (www.iana.org). Supported types
are separated with “;”. E.g. a token supporting MIME types "a/b",
"a/c" and "a/d" would set the attribute value to
“a/b;a/c;a/d”.
|
The selection of attributes, and associated data types, has
been done in an attempt to stay as aligned with RFC 2534 and CC/PP Struct as
possible. The special value CK_UNAVAILABLE_INFORMATION may be used for
CK_ULONG-based attributes when information is not available or applicable.
None of the attribute values may be set by an application.
The value of the CKA_ENCODING_METHODS attribute may
be used when the application needs to send MIME objects with encoded content to
the token.
This is not an object class; hence no CKO_ definition is
required. It is a category of object classes with common attributes for the
object classes that follow.
Table 17, Common Storage Object
Attributes
|
Attribute
|
Data Type
|
Meaning
|
|
CKA_TOKEN
|
CK_BBOOL
|
CK_TRUE if object is a token object; CK_FALSE
if object is a session object. Default is CK_FALSE.
|
|
CKA_PRIVATE
|
CK_BBOOL
|
CK_TRUE if object is a private object;
CK_FALSE if object is a public object. Default value is token-specific, and
may depend on the values of other attributes of the object.
|
|
CKA_MODIFIABLE
|
CK_BBOOL
|
CK_TRUE if object can be modified Default is
CK_TRUE.
|
|
CKA_LABEL
|
RFC2279 string
|
Description of the object (default empty).
|
|
CKA_COPYABLE
|
CK_BBOOL
|
CK_TRUE if object can be copied
using C_CopyObject. Defaults to CK_TRUE. Can’t be set to TRUE once it is set to
FALSE.
|
|
CKA_DESTROYABLE
|
CK_BBOOL
|
CK_TRUE if the object can be
destroyed using C_DestroyObject. Default is CK_TRUE.
|
|
CKA_UNIQUE_ID246
|
RFC2279 string
|
The unique identifier assigned to
the object.
|
Only the CKA_LABEL attribute can be modified after
the object is created. (The CKA_TOKEN, CKA_PRIVATE, and CKA_MODIFIABLE
attributes can be changed in the process of copying an object, however.)
The CKA_TOKEN attribute identifies whether the object
is a token object or a session object.
When the CKA_PRIVATE attribute is CK_TRUE, a user may
not access the object until the user has been authenticated to the token.
The value of the CKA_MODIFIABLE attribute determines
whether or not an object is read-only.
The CKA_LABEL attribute is intended to assist users
in browsing.
The value of the CKA_COPYABLE
attribute determines whether or not an object can be copied. This attribute
can be used in conjunction with CKA_MODIFIABLE to prevent changes to the
permitted usages of keys and other objects.
The value of the CKA_DESTROYABLE
attribute determines whether the object can be destroyed using C_DestroyObject.
Any time a new object is created,
a value for CKA_UNIQUE_ID MUST be generated by the token and stored with the
object. The specific algorithm used to generate unique ID values for objects
is token-specific, but values generated MUST be unique across all objects
visible to any particular session, and SHOULD be unique across all objects created
by the token. Reinitializing the token, such as by calling C_InitToken, MAY
cause reuse of CKA_UNIQUE_ID values.
Any attempt to modify the
CKA_UNIQUE_ID attribute of an existing object or to specify the value of the
CKA_UNIQUE_ID attribute in the template for an operation that creates one or
more objects MUST fail. Operations failing for this reason return the error
code CKR_ATTRIBUTE_READ_ONLY.
This section defines the object class CKO_DATA for type CK_OBJECT_CLASS
as used in the CKA_CLASS attribute of objects.
Data objects (object class CKO_DATA) hold information
defined by an application. Other than providing access to it, Cryptoki does not
attach any special meaning to a data object. The following table lists the
attributes supported by data objects, in addition to the common attributes
defined for this object class:
Table
18, Data Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_APPLICATION
|
RFC2279 string
|
Description of the application that manages
the object (default empty)
|
|
CKA_OBJECT_ID
|
Byte Array
|
DER-encoding of the object identifier
indicating the data object type (default empty)
|
|
CKA_VALUE
|
Byte array
|
Value of the object (default empty)
|
The CKA_APPLICATION attribute provides a means for
applications to indicate ownership of the data objects they manage. Cryptoki
does not provide a means of ensuring that only a particular application has
access to a data object, however.
The CKA_OBJECT_ID attribute provides an application
independent and expandable way to indicate the type of the data object value.
Cryptoki does not provide a means of insuring that the data object identifier
matches the data value.
The following is a sample template containing attributes for
creating a data object:
CK_OBJECT_CLASS class = CKO_DATA;
CK_UTF8CHAR label[] = “A data object”;
CK_UTF8CHAR application[] = “An application”;
CK_BYTE data[] = “Sample data”;
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_APPLICATION, application, sizeof(application)-1},
{CKA_VALUE, data, sizeof(data)}
};
This section defines the object class CKO_CERTIFICATE for
type CK_OBJECT_CLASS as used in the CKA_CLASS attribute of objects.
Certificate objects (object class CKO_CERTIFICATE)
hold public-key or attribute certificates. Other than providing access to
certificate objects, Cryptoki does not attach any special meaning to
certificates. The following table defines the common certificate object
attributes, in addition to the common attributes defined for this object class:
Table 19,
Common Certificate Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_CERTIFICATE_TYPE1
|
CK_CERTIFICATE_TYPE
|
Type
of certificate
|
|
CKA_TRUSTED10
|
CK_BBOOL
|
The
certificate can be trusted for the application that it was created.
|
|
CKA_CERTIFICATE_CATEGORY
|
CKA_CERTIFICATE_CATEGORY
|
(default
CK_CERTIFICATE_CATEGORY_UNSPECIFIED)
|
|
CKA_CHECK_VALUE
|
Byte
array
|
Checksum
|
|
CKA_START_DATE
|
CK_DATE
|
Start
date for the certificate (default empty)
|
|
CKA_END_DATE
|
CK_DATE
|
End
date for the certificate (default empty)
|
|
CKA_PUBLIC_KEY_INFO
|
Byte
Array
|
DER-encoding
of the SubjectPublicKeyInfo for the public key contained in this certificate
(default empty)
|
- Refer to Table 11 for footnotes
Cryptoki does not enforce the relationship of the
CKA_PUBLIC_KEY_INFO to the public key in the certificate, but does recommend
that the key be extracted from the certificate to create this value.
The CKA_CERTIFICATE_TYPE attribute may not be
modified after an object is created. This version of Cryptoki supports the
following certificate types:
·
X.509
public key certificate
- WTLS
public key certificate
·
X.509
attribute certificate
The CKA_TRUSTED attribute cannot be set to CK_TRUE by
an application. It MUST be set by a token initialization application or by the
token’s SO. Trusted certificates cannot be modified.
The CKA_CERTIFICATE_CATEGORY
attribute is used to indicate if a stored certificate is a user certificate for
which the corresponding private key is available on the token (“token user”), a
CA certificate (“authority”), or another end-entity certificate (“other
entity”). This attribute may not be modified after an object is created.
The CKA_CERTIFICATE_CATEGORY and CKA_TRUSTED
attributes will together be used to map to the categorization of the
certificates.
CKA_CHECK_VALUE: The value of this attribute is
derived from the certificate by taking the first three bytes of the SHA-1 hash
of the certificate object’s CKA_VALUE attribute.
The CKA_START_DATE and CKA_END_DATE attributes
are for reference only; Cryptoki does not attach any special meaning to them.
When present, the application is responsible to set them to values that match
the certificate’s encoded “not before” and “not after” fields (if any).
X.509 certificate objects (certificate type CKC_X_509)
hold X.509 public key certificates. The following table defines the X.509
certificate object attributes, in addition to the common attributes defined for
this object class:
Table 20, X.509 Certificate
Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_SUBJECT1
|
Byte array
|
DER-encoding of the certificate subject name
|
|
CKA_ID
|
Byte array
|
Key identifier for public/private key pair
(default empty)
|
|
CKA_ISSUER
|
Byte array
|
DER-encoding of the certificate issuer name
(default empty)
|
|
CKA_SERIAL_NUMBER
|
Byte array
|
DER-encoding of the certificate serial number
(default empty)
|
|
CKA_VALUE2
|
Byte array
|
BER-encoding of the certificate
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1MUST be specified when the object is created.
2MUST be specified when the object is created. MUST be non-empty if
CKA_URL is empty.
3MUST be non-empty if CKA_VALUE is empty.
4Can only be empty if CKA_URL is empty.
Only the CKA_ID, CKA_ISSUER, and CKA_SERIAL_NUMBER
attributes may be modified after the object is created.
The CKA_ID attribute is intended as a means of
distinguishing multiple public-key/private-key pairs held by the same subject
(whether stored in the same token or not). (Since the keys are distinguished by
subject name as well as identifier, it is possible that keys for different
subjects may have the same CKA_ID value without introducing any
ambiguity.)
It is intended in the interests of interoperability that the
subject name and key identifier for a certificate will be the same as those for
the corresponding public and private keys (though it is not required that all
be stored in the same token). However, Cryptoki does not enforce this
association, or even the uniqueness of the key identifier for a given subject;
in particular, an application may leave the key identifier empty.
The CKA_ISSUER and CKA_SERIAL_NUMBER attributes
are for compatibility with PKCS #7 and Privacy Enhanced Mail (RFC1421). Note
that with the version 3 extensions to X.509 certificates, the key identifier
may be carried in the certificate. It is intended that the CKA_ID value
be identical to the key identifier in such a certificate extension, although
this will not be enforced by Cryptoki.
The CKA_URL attribute enables the support for storage
of the URL where the certificate can be found instead of the certificate
itself. Storage of a URL instead of the complete certificate is often used in
mobile environments.
The CKA_HASH_OF_SUBJECT_PUBLIC_KEY and CKA_HASH_OF_ISSUER_PUBLIC_KEY
attributes are used to store the hashes of the public keys of the subject and
the issuer. They are particularly important when only the URL is available to
be able to correlate a certificate with a private key and when searching for
the certificate of the issuer. The hash algorithm is defined by
CKA_NAME_HASH_ALGORITHM.
The CKA_JAVA_MIDP_SECURITY_DOMAIN attribute
associates a certificate with a Java MIDP security domain.
The following is a sample template for creating an X.509
certificate object:
CK_OBJECT_CLASS class = CKO_CERTIFICATE;
CK_CERTIFICATE_TYPE certType = CKC_X_509;
CK_UTF8CHAR label[] = “A certificate object”;
CK_BYTE subject[] = {...};
CK_BYTE id[] = {123};
CK_BYTE certificate[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_CERTIFICATE_TYPE, &certType, sizeof(certType)};
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_SUBJECT, subject, sizeof(subject)},
{CKA_ID, id, sizeof(id)},
{CKA_VALUE, certificate, sizeof(certificate)}
};
WTLS certificate objects (certificate type CKC_WTLS)
hold WTLS public key certificates. The following table defines the WTLS
certificate object attributes, in addition to the common attributes defined for
this object class.
Table 21: WTLS Certificate
Object Attributes
|
Attribute
|
Data
type
|
Meaning
|
|
CKA_SUBJECT1
|
Byte
array
|
WTLS-encoding
(Identifier type) of the certificate subject
|
|
CKA_ISSUER
|
Byte
array
|
WTLS-encoding
(Identifier type) of the certificate issuer (default empty)
|
|
CKA_VALUE2
|
Byte
array
|
WTLS-encoding
of the certificate
|
|
CKA_URL3
|
RFC2279
string
|
If
not empty this attribute gives the URL where the complete certificate can be
obtained
|
|
CKA_HASH_OF_SUBJECT_PUBLIC_KEY4
|
Byte
array
|
SHA-1
hash of the subject public key (default empty). Hash algorithm is defined by
CKA_NAME_HASH_ALGORITHM
|
|
CKA_HASH_OF_ISSUER_PUBLIC_KEY4
|
Byte
array
|
SHA-1
hash of the issuer public key (default empty). Hash algorithm is defined by
CKA_NAME_HASH_ALGORITHM
|
|
CKA_NAME_HASH_ALGORITHM
|
CK_MECHANISM_TYPE
|
Defines
the mechanism used to calculate CKA_HASH_OF_SUBJECT_PUBLIC_KEY and CKA_HASH_OF_ISSUER_PUBLIC_KEY.
If the attribute is not present then the type defaults to SHA-1.
|
1MUST be specified when the object is created.
Can only be empty if CKA_VALUE is empty.
2MUST be specified when the object is created.
MUST be non-empty if CKA_URL is empty.
3MUST be non-empty if CKA_VALUE is empty.
4Can only be empty if CKA_URL is empty.
Only the CKA_ISSUER attribute may be modified after
the object has been created.
The encoding for the CKA_SUBJECT, CKA_ISSUER,
and CKA_VALUE attributes can be found in [WTLS].
The CKA_URL attribute enables the support for storage
of the URL where the certificate can be found instead of the certificate
itself. Storage of a URL instead of the complete certificate is often used in
mobile environments.
The CKA_HASH_OF_SUBJECT_PUBLIC_KEY and CKA_HASH_OF_ISSUER_PUBLIC_KEY
attributes are used to store the hashes of the public keys of the subject and
the issuer. They are particularly important when only the URL is available to
be able to correlate a certificate with a private key and when searching for
the certificate of the issuer. The hash algorithm is defined by
CKA_NAME_HASH_ALGORITHM.
The following is a sample template for creating a WTLS
certificate object:
CK_OBJECT_CLASS class = CKO_CERTIFICATE;
CK_CERTIFICATE_TYPE certType = CKC_WTLS;
CK_UTF8CHAR label[] = “A certificate object”;
CK_BYTE subject[] = {...};
CK_BYTE certificate[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] =
{
{CKA_CLASS, &class, sizeof(class)},
{CKA_CERTIFICATE_TYPE, &certType, sizeof(certType)};
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_SUBJECT, subject, sizeof(subject)},
{CKA_VALUE, certificate, sizeof(certificate)}
};
X.509 attribute certificate objects (certificate type CKC_X_509_ATTR_CERT)
hold X.509 attribute certificates. The following table defines the X.509
attribute certificate object attributes, in addition to the common attributes
defined for this object class:
Table
22, X.509 Attribute Certificate Object Attributes
|
Attribute
|
Data Type
|
Meaning
|
|
CKA_OWNER1
|
Byte Array
|
DER-encoding of the
attribute certificate's subject field. This is distinct from the CKA_SUBJECT
attribute contained in CKC_X_509 certificates because the ASN.1 syntax and
encoding are different.
|
|
CKA_AC_ISSUER
|
Byte Array
|
DER-encoding of the
attribute certificate's issuer field. This is distinct from the CKA_ISSUER
attribute contained in CKC_X_509 certificates because the ASN.1 syntax and
encoding are different. (default empty)
|
|
CKA_SERIAL_NUMBER
|
Byte Array
|
DER-encoding of the
certificate serial number. (default empty)
|
|
CKA_ATTR_TYPES
|
Byte Array
|
BER-encoding of a
sequence of object identifier values corresponding to the attribute types
contained in the certificate. When
present, this field offers an opportunity for applications to search for a
particular attribute certificate without fetching and parsing the certificate
itself. (default empty)
|
|
CKA_VALUE1
|
Byte Array
|
BER-encoding of the
certificate.
|
1MUST be specified
when the object is created
Only the CKA_AC_ISSUER, CKA_SERIAL_NUMBER and CKA_ATTR_TYPES
attributes may be modified after the object is created.
The following is a sample template for creating an X.509
attribute certificate object:
CK_OBJECT_CLASS class = CKO_CERTIFICATE;
CK_CERTIFICATE_TYPE certType = CKC_X_509_ATTR_CERT;
CK_UTF8CHAR label[] = "An attribute certificate object";
CK_BYTE owner[] = {...};
CK_BYTE certificate[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_CERTIFICATE_TYPE, &certType, sizeof(certType)};
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_OWNER, owner, sizeof(owner)},
{CKA_VALUE, certificate, sizeof(certificate)}
};
There is no CKO_ definition for the base key object class,
only for the key types derived from it.
This section defines the object class CKO_PUBLIC_KEY,
CKO_PRIVATE_KEY and CKO_SECRET_KEY for type CK_OBJECT_CLASS as used in the
CKA_CLASS attribute of objects.
Key objects hold encryption or authentication keys, which
can be public keys, private keys, or secret keys. The following common
footnotes apply to all the tables describing attributes of keys:
The following table defines the attributes common to public
key, private key and secret key classes, in addition to the common attributes
defined for this object class:
Table 23, Common Key
Attributes
|
Attribute
|
Data Type
|
Meaning
|
|
CKA_KEY_TYPE1,5
|
CK_KEY_TYPE
|
Type
of key
|
|
CKA_ID8
|
Byte
array
|
Key
identifier for key (default empty)
|
|
CKA_START_DATE8
|
CK_DATE
|
Start
date for the key (default empty)
|
|
CKA_END_DATE8
|
CK_DATE
|
End
date for the key (default empty)
|
|
CKA_DERIVE8
|
CK_BBOOL
|
CK_TRUE
if key supports key derivation (i.e., if other keys can be derived
from this one (default CK_FALSE)
|
|
CKA_LOCAL2,4,6
|
CK_BBOOL
|
CK_TRUE
only if key was either
·
generated
locally (i.e., on the token) with a C_GenerateKey or C_GenerateKeyPair
call
·
created
with a C_CopyObject call as a copy of a key which had its CKA_LOCAL
attribute set to CK_TRUE
|
|
CKA_KEY_GEN_
MECHANISM2,4,6
|
CK_MECHANISM_TYPE
|
Identifier
of the mechanism used to generate the key material.
|
|
CKA_ALLOWED_MECHANISMS
|
CK_MECHANISM_TYPE _PTR, pointer to a
CK_MECHANISM_TYPE array
|
A list of mechanisms allowed to be used with
this key. The number of mechanisms in the array is the ulValueLen
component of the attribute divided by the size
of CK_MECHANISM_TYPE.
|
- Refer to Table 11 for footnotes
The CKA_ID field is intended to distinguish among
multiple keys. In the case of public and private keys, this field assists in
handling multiple keys held by the same subject; the key identifier for a
public key and its corresponding private key should be the same. The key
identifier should also be the same as for the corresponding certificate, if one
exists. Cryptoki does not enforce these associations, however. (See Section 4.6 for further commentary.)
In the case of secret keys, the meaning of the CKA_ID
attribute is up to the application.
Note that the CKA_START_DATE and CKA_END_DATE
attributes are for reference only; Cryptoki does not attach any special meaning
to them. In particular, it does not restrict usage of a key according to the
dates; doing this is up to the application.
The CKA_DERIVE attribute has the value CK_TRUE if and
only if it is possible to derive other keys from the key.
The CKA_LOCAL attribute has the value CK_TRUE if and
only if the value of the key was originally generated on the token by a C_GenerateKey
or C_GenerateKeyPair call.
The CKA_KEY_GEN_MECHANISM attribute identifies the
key generation mechanism used to generate the key material. It contains a valid
value only if the CKA_LOCAL attribute has the value CK_TRUE. If CKA_LOCAL
has the value CK_FALSE, the value of the attribute is
CK_UNAVAILABLE_INFORMATION.
Public key objects (object class CKO_PUBLIC_KEY) hold
public keys. The following table defines the attributes common to all public
keys, in addition to the common attributes defined for this object class:
Table 24, Common Public Key Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_SUBJECT8
|
Byte array
|
DER-encoding of the key subject name (default
empty)
|
|
CKA_ENCRYPT8
|
CK_BBOOL
|
CK_TRUE if key supports encryption9
|
|
CKA_VERIFY8
|
CK_BBOOL
|
CK_TRUE if key supports verification where
the signature is an appendix to the data9
|
|
CKA_VERIFY_RECOVER8
|
CK_BBOOL
|
CK_TRUE if key supports verification where
the data is recovered from the signature9
|
|
CKA_WRAP8
|
CK_BBOOL
|
CK_TRUE if key supports wrapping (i.e.,
can be used to wrap other keys)9
|
|
CKA_TRUSTED10
|
CK_BBOOL
|
The key can be trusted for the application
that it was created.
The wrapping key can be used to wrap keys
with CKA_WRAP_WITH_TRUSTED set to CK_TRUE.
|
|
CKA_WRAP_TEMPLATE
|
CK_ATTRIBUTE_PTR
|
For wrapping keys. The attribute template to
match against any keys wrapped using this wrapping key. Keys that do not
match cannot be wrapped. The number of attributes in the array is the ulValueLen
component of the attribute divided by the size of CK_ATTRIBUTE.
|
|
CKA_PUBLIC_KEY_INFO
|
Byte array
|
DER-encoding of the SubjectPublicKeyInfo for
this public key. (MAY be empty, DEFAULT derived from the underlying public
key data)
|
- Refer to Table 11 for footnotes
It is intended in the interests of interoperability that the
subject name and key identifier for a public key will be the same as those for
the corresponding certificate and private key. However, Cryptoki does not
enforce this, and it is not required that the certificate and private key also
be stored on the token.
To map between ISO/IEC 9594-8 (X.509) keyUsage flags
for public keys and the PKCS #11 attributes for public keys, use the following
table.
Table
25, Mapping of X.509 key usage flags to Cryptoki attributes for public keys
|
Key usage flags for public keys in X.509
public key certificates
|
Corresponding cryptoki attributes for public
keys.
|
|
dataEncipherment
|
CKA_ENCRYPT
|
|
digitalSignature, keyCertSign, cRLSign
|
CKA_VERIFY
|
|
digitalSignature, keyCertSign, cRLSign
|
CKA_VERIFY_RECOVER
|
|
keyAgreement
|
CKA_DERIVE
|
|
keyEncipherment
|
CKA_WRAP
|
|
nonRepudiation
|
CKA_VERIFY
|
|
nonRepudiation
|
CKA_VERIFY_RECOVER
|
The value of the
CKA_PUBLIC_KEY_INFO attribute is the DER encoded value of SubjectPublicKeyInfo:
SubjectPublicKeyInfo ::=
SEQUENCE {
algorithm AlgorithmIdentifier,
subjectPublicKey BIT_STRING
}
The encodings for the
subjectPublicKey field are specified in the description of the public key types
in the appropriate sections for the
key types defined within this specification.
Private key objects (object class CKO_PRIVATE_KEY)
hold private keys. The following table defines the attributes common to all
private keys, in addition to the common attributes defined for this object
class:
Table 26, Common Private Key Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_SUBJECT8
|
Byte array
|
DER-encoding of certificate subject name
(default empty)
|
|
CKA_SENSITIVE8,11
|
CK_BBOOL
|
CK_TRUE
if key is sensitive9
|
|
CKA_DECRYPT8
|
CK_BBOOL
|
CK_TRUE
if key supports decryption9
|
|
CKA_SIGN8
|
CK_BBOOL
|
CK_TRUE
if key supports signatures where the signature is an appendix to the data9
|
|
CKA_SIGN_RECOVER8
|
CK_BBOOL
|
CK_TRUE
if key supports signatures where the data can be recovered from the signature9
|
|
CKA_UNWRAP8
|
CK_BBOOL
|
CK_TRUE
if key supports unwrapping (i.e., can be used to unwrap other keys)9
|
|
CKA_EXTRACTABLE8,12
|
CK_BBOOL
|
CK_TRUE
if key is extractable and can be wrapped 9
|
|
CKA_ALWAYS_SENSITIVE2,4,6
|
CK_BBOOL
|
CK_TRUE
if key has always had the CKA_SENSITIVE attribute set to CK_TRUE
|
|
CKA_NEVER_EXTRACTABLE2,4,6
|
CK_BBOOL
|
CK_TRUE
if key has never had the CKA_EXTRACTABLE attribute set to CK_TRUE
|
|
CKA_WRAP_WITH_TRUSTED11
|
CK_BBOOL
|
CK_TRUE
if the key can only be wrapped with a wrapping key that has CKA_TRUSTED set
to CK_TRUE.
Default
is CK_FALSE.
|
|
CKA_UNWRAP_TEMPLATE
|
CK_ATTRIBUTE_PTR
|
For
wrapping keys. The attribute template to apply to any keys unwrapped using
this wrapping key. Any user supplied template is applied after this template
as if the object has already been created. The number of attributes in the
array is the ulValueLen component of the attribute divided by the size
of
CK_ATTRIBUTE.
|
|
CKA_ALWAYS_AUTHENTICATE
|
CK_BBOOL
|
If
CK_TRUE, the user has to supply the PIN for each use (sign or decrypt) with
the key. Default is CK_FALSE.
|
|
CKA_PUBLIC_KEY_INFO8
|
Byte
Array
|
DER-encoding
of the SubjectPublicKeyInfo for the associated public key (MAY be empty;
DEFAULT derived from the underlying private key data; MAY be manually set for
specific key types; if set; MUST be consistent with the underlying private
key data)
|
|
CKA_DERIVE_TEMPLATE
|
CK_ATTRIBUTE_PTR
|
For
deriving keys. The attribute template to match against any keys derived using
this derivation key. Any user supplied template is applied after this
template as if the object has already been created. The number of attributes
in the array is the ulValueLen component of the attribute divided by the size
of CK_ATTRIBUTE.
|
- Refer to Table 11 for footnotes
It is intended in the
interests of interoperability that the subject name and key identifier for a
private key will be the same as those for the corresponding certificate and public
key. However, this is not enforced by Cryptoki, and it is not required that
the certificate and public key also be stored on the token.
If the CKA_SENSITIVE attribute is CK_TRUE, or if the CKA_EXTRACTABLE
attribute is CK_FALSE, then certain attributes of the private key cannot be
revealed in plaintext outside the token. Which attributes these are is
specified for each type of private key in the attribute table in the section
describing that type of key.
The CKA_ALWAYS_AUTHENTICATE attribute can be used to
force re-authentication (i.e. force the user to provide a PIN) for each use of
a private key. “Use” in this case means a cryptographic operation such as sign
or decrypt. This attribute may only be set to CK_TRUE when CKA_PRIVATE
is also CK_TRUE.
Re-authentication occurs by calling C_Login with userType
set to CKU_CONTEXT_SPECIFIC immediately after a cryptographic operation
using the key has been initiated (e.g. after C_SignInit). In this call,
the actual user type is implicitly given by the usage requirements of the
active key. If C_Login returns CKR_OK the user was successfully
authenticated and this sets the active key in an authenticated state that lasts
until the cryptographic operation has successfully or unsuccessfully been
completed (e.g. by C_Sign, C_SignFinal,..). A return value
CKR_PIN_INCORRECT from C_Login means that the user was denied permission
to use the key and continuing the cryptographic operation will result in a
behavior as if C_Login had not been called. In both of these cases the
session state will remain the same, however repeated failed re-authentication
attempts may cause the PIN to be locked. C_Login returns in this case
CKR_PIN_LOCKED and this also logs the user out from the token. Failing or
omitting to re-authenticate when CKA_ALWAYS_AUTHENTICATE is set to CK_TRUE will
result in CKR_USER_NOT_LOGGED_IN to be returned from calls using the key. C_Login
will return CKR_OPERATION_NOT_INITIALIZED, but the active cryptographic
operation will not be affected, if an attempt is made to re-authenticate when
CKA_ALWAYS_AUTHENTICATE is set to CK_FALSE.
The CKA_PUBLIC_KEY_INFO attribute represents the
public key associated with this private key. The data it represents may either
be stored as part of the private key data, or regenerated as needed from the
private key.
If this attribute is supplied as part of a template for C_CreateObject,
C_CopyObject or C_SetAttributeValue for a private key, the token
MUST verify correspondence between the private key data and the public key data
as supplied in CKA_PUBLIC_KEY_INFO. This can be done either by deriving
a public key from the private key and comparing the values, or by doing a sign
and verify operation. If there is a mismatch, the command SHALL return CKR_ATTRIBUTE_VALUE_INVALID.
A token MAY choose not to support the CKA_PUBLIC_KEY_INFO attribute for
commands which create new private keys. If it does not support the attribute,
the command SHALL return CKR_ATTRIBUTE_TYPE_INVALID.
As a general guideline, private keys of any type SHOULD
store sufficient information to retrieve the public key information. In particular, the RSA private key description has been
modified in PKCS #11 V2.40 to add the CKA_PUBLIC_EXPONENT to the list of
attributes required for an RSA private key. All other private key types
described in this specification contain sufficient information to recover the
associated public key.
Secret key objects (object class CKO_SECRET_KEY) hold
secret keys. The following table defines the attributes common to all secret
keys, in addition to the common attributes defined for this object class:
Table 27,
Common Secret Key Attributes
|
Attribute
|
Data
type
|
Meaning
|
|
CKA_SENSITIVE8,11
|
CK_BBOOL
|
CK_TRUE
if object is sensitive (default CK_FALSE)
|
|
CKA_ENCRYPT8
|
CK_BBOOL
|
CK_TRUE
if key supports encryption9
|
|
CKA_DECRYPT8
|
CK_BBOOL
|
CK_TRUE
if key supports decryption9
|
|
CKA_SIGN8
|
CK_BBOOL
|
CK_TRUE
if key supports signatures (i.e., authentication codes) where the
signature is an appendix to the data9
|
|
CKA_VERIFY8
|
CK_BBOOL
|
CK_TRUE
if key supports verification (i.e., of authentication codes) where the
signature is an appendix to the data9
|
|
CKA_WRAP8
|
CK_BBOOL
|
CK_TRUE
if key supports wrapping (i.e., can be used to wrap other keys)9
|
|
CKA_UNWRAP8
|
CK_BBOOL
|
CK_TRUE
if key supports unwrapping (i.e., can be used to unwrap other keys)9
|
|
CKA_EXTRACTABLE8,12
|
CK_BBOOL
|
CK_TRUE
if key is extractable and can be wrapped 9
|
|
CKA_ALWAYS_SENSITIVE2,4,6
|
CK_BBOOL
|
CK_TRUE
if key has always had the CKA_SENSITIVE attribute set to CK_TRUE
|
|
CKA_NEVER_EXTRACTABLE2,4,6
|
CK_BBOOL
|
CK_TRUE
if key has never had the CKA_EXTRACTABLE attribute set to CK_TRUE
|
|
CKA_CHECK_VALUE
|
Byte
array
|
Key
checksum
|
|
CKA_WRAP_WITH_TRUSTED11
|
CK_BBOOL
|
CK_TRUE
if the key can only be wrapped with a wrapping key that has CKA_TRUSTED set
to CK_TRUE.
Default
is CK_FALSE.
|
|
CKA_TRUSTED10
|
CK_BBOOL
|
The
wrapping key can be used to wrap keys with CKA_WRAP_WITH_TRUSTED set to
CK_TRUE.
|
|
CKA_WRAP_TEMPLATE
|
CK_ATTRIBUTE_PTR
|
For
wrapping keys. The attribute template to match against any keys wrapped using
this wrapping key. Keys that do not match cannot be wrapped. The number of
attributes in the array is the
ulValueLen component of the
attribute divided by the size of
CK_ATTRIBUTE
|
|
CKA_UNWRAP_TEMPLATE
|
CK_ATTRIBUTE_PTR
|
For
wrapping keys. The attribute template to apply to any keys unwrapped using
this wrapping key. Any user supplied template is applied after this template
as if the object has already been created. The number of attributes in the
array is the ulValueLen component of the attribute divided by the size
of
CK_ATTRIBUTE.
|
|
A_DERIVE_TEMPLATE
|
CK_ATTRIBUTE_PTR
|
For
deriving keys. The attribute template to match against any keys derived using
this derivation key. Any user supplied template is applied after this
template as if the object has already been created. The number of attributes
in the array is the ulValueLen component of the attribute divided by the size
of CK_ATTRIBUTE.
|
- Refer to Table 11 for footnotes
If the CKA_SENSITIVE attribute is CK_TRUE, or if the CKA_EXTRACTABLE
attribute is CK_FALSE, then certain attributes of the secret key cannot be
revealed in plaintext outside the token. Which attributes these are is
specified for each type of secret key in the attribute table in the section
describing that type of key.
The
key check value (KCV) attribute for symmetric key objects to be called CKA_CHECK_VALUE,
of type byte array, length 3 bytes, operates like a fingerprint, or checksum of
the key. They are intended to be used to cross-check symmetric keys against
other systems where the same key is shared, and as a validity check after
manual key entry or restore from backup. Refer to object definitions of
specific key types for KCV algorithms.
Properties:
- For two keys that are cryptographically identical the value of
this attribute should be identical.
- CKA_CHECK_VALUE should not be usable to obtain any part of the
key value.
- Non-uniqueness. Two different keys can have the same
CKA_CHECK_VALUE. This is unlikely (the probability can easily be
calculated) but possible.
The attribute is optional, but if supported, regardless of
how the key object is created or derived, the value of the attribute is always
supplied. It SHALL be supplied even if the encryption operation for the key is
forbidden (i.e. when CKA_ENCRYPT is set to CK_FALSE).
If a value is supplied in the application template (allowed
but never necessary) then, if supported, it MUST match what the library
calculates it to be or the library returns a CKR_ATTRIBUTE_VALUE_INVALID. If
the library does not support the attribute then it should ignore it. Allowing
the attribute in the template this way does no harm and allows the attribute to
be treated like any other attribute for the purposes of key wrap and unwrap
where the attributes are preserved also.
The generation of the KCV may be prevented by the
application supplying the attribute in the template as a no-value (0 length)
entry. The application can query the value at any time like any other attribute
using C_GetAttributeValue. C_SetAttributeValue may be used to destroy the
attribute, by supplying no-value.
Unless otherwise specified for the object definition, the value
of this attribute is derived from the key object by taking the first three
bytes of an encryption of a single block of null (0x00) bytes, using the
default cipher and mode (e.g. ECB) associated with the key type of the secret
key object.
This section defines the object class CKO_DOMAIN_PARAMETERS
for type CK_OBJECT_CLASS as used in the CKA_CLASS attribute of objects.
This object class was created to support the storage of
certain algorithm's extended parameters. DSA and DH both use domain parameters
in the key-pair generation step. In particular, some libraries support the
generation of domain parameters (originally out of scope for PKCS11) so the
object class was added.
To use a domain parameter object you MUST extract the
attributes into a template and supply them (still in the template) to the
corresponding key-pair generation function.
Domain parameter objects (object class CKO_DOMAIN_PARAMETERS)
hold public domain parameters.
The following table defines the attributes common to domain
parameter objects in addition to the common attributes defined for this object
class:
Table 28, Common Domain
Parameter Attributes
|
Attribute
|
Data Type
|
Meaning
|
|
CKA_KEY_TYPE1
|
CK_KEY_TYPE
|
Type of key the domain parameters can be used
to generate.
|
|
CKA_LOCAL2,4
|
CK_BBOOL
|
CK_TRUE only if domain parameters were either
·
generated
locally (i.e., on the token) with a C_GenerateKey
·
created
with a C_CopyObject call as a copy of domain parameters which had its CKA_LOCAL
attribute set to CK_TRUE
|
- Refer to Table 11 for footnotes
The CKA_LOCAL attribute has the value CK_TRUE if and
only if the values of the domain parameters were originally generated on the
token by a C_GenerateKey call.
This section defines the object class CKO_MECHANISM for type
CK_OBJECT_CLASS as used in the CKA_CLASS attribute of objects.
Mechanism objects provide information about mechanisms
supported by a device beyond that given by the CK_MECHANISM_INFO
structure.
When searching for objects using C_FindObjectsInit
and C_FindObjects, mechanism objects are not returned unless the CKA_CLASS
attribute in the template has the value CKO_MECHANISM. This protects
applications written to previous versions of Cryptoki from finding objects that
they do not understand.
Table
29, Common Mechanism Attributes
|
Attribute
|
Data Type
|
Meaning
|
|
CKA_MECHANISM_TYPE
|
CK_MECHANISM_TYPE
|
The type of mechanism object
|
The CKA_MECHANISM_TYPE attribute may not be set.
This section defines the object class CKO_PROFILE for type
CK_OBJECT_CLASS as used in the CKA_CLASS attribute of objects.
4.13.2 Overview
Profile objects (object class CKO_PROFILE) describe which
PKCS #11 profiles the token implements. Profiles are defined in the OASIS PKCS
#11 Cryptographic Token Interface Profiles document. A given token can contain
more than one profile ID. The following table lists the attributes supported by
profile objects, in addition to the common attributes defined for this object
class:
Table 30, Profile Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_PROFILE_ID
|
CK_PROFILE_ID
|
ID of the supported profile.
|
The CKA_PROFILE_ID attribute identifies a profile
that the token supports.
Cryptoki's functions are organized into the following
categories:
·
general-purpose functions (4 functions)
·
slot and token management functions (9 functions)
·
session management functions (8 functions)
·
object management functions (9 functions)
·
encryption functions (4 functions)
·
message-based encryption functions (5 functions)
·
decryption functions (4 functions)
·
message digesting functions (5 functions)
·
signing and MACing functions (6 functions)
·
functions for verifying signatures and MACs (6 functions)
·
dual-purpose cryptographic functions (4 functions)
·
key management functions (5 functions)
·
random number generation functions (2 functions)
·
parallel function management functions (2 functions)
In addition to these functions, Cryptoki can use
application-supplied callback functions to notify an application of certain
events, and can also use application-supplied functions to handle mutex objects
for safe multi-threaded library access.
The Cryptoki API functions are presented in the following
table:
Table 31, Summary of Cryptoki Functions
|
Category
|
Function
|
Description
|
|
General
|
C_Initialize
|
initializes
Cryptoki
|
|
purpose
functions
|
C_Finalize
|
clean
up miscellaneous Cryptoki-associated resources
|
|
|
C_GetInfo
|
obtains
general information about Cryptoki
|
|
|
C_GetFunctionList
|
obtains
entry points of Cryptoki library functions
|
|
|
C_GetInterfaceList
|
obtains
list of interfaces supported by Cryptoki library
|
|
|
C_GetInterface
|
obtains
interface specific entry points to Cryptoki library functions
|
|
Slot
and token
|
C_GetSlotList
|
obtains
a list of slots in the system
|
|
management
|
C_GetSlotInfo
|
obtains
information about a particular slot
|
|
functions
|
C_GetTokenInfo
|
obtains
information about a particular token
|
|
|
C_WaitForSlotEvent
|
waits
for a slot event (token insertion, removal, etc.) to occur
|
|
|
C_GetMechanismList
|
obtains
a list of mechanisms supported by a token
|
|
|
C_GetMechanismInfo
|
obtains
information about a particular mechanism
|
|
|
C_InitToken
|
initializes
a token
|
|
|
C_InitPIN
|
initializes
the normal user’s PIN
|
|
|
C_SetPIN
|
modifies
the PIN of the current user
|
|
Session
management functions
|
C_OpenSession
|
opens
a connection between an application and a particular token or sets up an
application callback for token insertion
|
|
|
C_CloseSession
|
closes
a session
|
|
|
C_CloseAllSessions
|
closes
all sessions with a token
|
|
|
C_GetSessionInfo
|
obtains
information about the session
|
|
|
C_SessionCancel
|
terminates
active session based operations
|
|
|
C_GetOperationState
|
obtains
the cryptographic operations state of a session
|
|
|
C_SetOperationState
|
sets
the cryptographic operations state of a session
|
|
|
C_Login
|
logs
into a token
|
|
|
C_LoginUser
|
logs
into a token with explicit user name
|
|
|
C_Logout
|
logs
out from a token
|
|
Object
|
C_CreateObject
|
creates
an object
|
|
management
|
C_CopyObject
|
creates
a copy of an object
|
|
functions
|
C_DestroyObject
|
destroys
an object
|
|
|
C_GetObjectSize
|
obtains
the size of an object in bytes
|
|
|
C_GetAttributeValue
|
obtains
an attribute value of an object
|
|
|
C_SetAttributeValue
|
modifies
an attribute value of an object
|
|
|
C_FindObjectsInit
|
initializes
an object search operation
|
|
|
C_FindObjects
|
continues
an object search operation
|
|
|
C_FindObjectsFinal
|
finishes
an object search operation
|
|
Encryption
|
C_EncryptInit
|
initializes
an encryption operation
|
|
functions
|
C_Encrypt
|
encrypts
single-part data
|
|
|
C_EncryptUpdate
|
continues
a multiple-part encryption operation
|
|
|
C_EncryptFinal
|
finishes
a multiple-part encryption operation
|
|
Message-based
Encryption Functions
|
C_MessageEncryptInit
|
initializes a message-based encryption process
|
|
C_EncryptMessage
|
encrypts a single-part message
|
|
C_EncryptMessageBegin
|
begins a multiple-part message encryption operation
|
|
C_EncryptMessageNext
|
continues or finishes a multiple-part message encryption
operation
|
|
C_MessageEncryptFinal
|
finishes a message-based encryption process
|
|
Decryption
|
C_DecryptInit
|
initializes
a decryption operation
|
|
Functions
|
C_Decrypt
|
decrypts
single-part encrypted data
|
|
|
C_DecryptUpdate
|
continues
a multiple-part decryption operation
|
|
|
C_DecryptFinal
|
finishes
a multiple-part decryption operation
|
|
Message-based
Decryption
Functions
|
C_MessageDecryptInit
|
initializes
a message decryption operation
|
|
C_DecryptMessage
|
decrypts
single-part data
|
|
C_DecryptMessageBegin
|
starts
a multiple-part message decryption operation
|
|
C_DecryptMessageNext
|
Continues
and finishes a multiple-part message decryption operation
|
|
C_MessageDecryptFinal
|
finishes
a message decryption operation
|
|
Message
|
C_DigestInit
|
initializes
a message-digesting operation
|
|
Digesting
|
C_Digest
|
digests
single-part data
|
|
Functions
|
C_DigestUpdate
|
continues
a multiple-part digesting operation
|
|
|
C_DigestKey
|
digests
a key
|
|
|
C_DigestFinal
|
finishes
a multiple-part digesting operation
|
|
Signing
|
C_SignInit
|
initializes
a signature operation
|
|
and
MACing
|
C_Sign
|
signs
single-part data
|
|
functions
|
C_SignUpdate
|
continues
a multiple-part signature operation
|
|
|
C_SignFinal
|
finishes
a multiple-part signature operation
|
|
|
C_SignRecoverInit
|
initializes
a signature operation, where the data can be recovered from the signature
|
|
|
C_SignRecover
|
signs
single-part data, where the data can be recovered from the signature
|
|
Message-based
Signature functions
|
C_MessageSignInit
|
initializes
a message signature operation
|
|
C_SignMessage
|
signs
single-part data
|
|
C_SignMessageBegin
|
starts
a multiple-part message signature operation
|
|
C_SignMessageNext
|
continues
and finishes a multiple-part message signature operation
|
|
C_MessageSignFinal
|
finishes
a message signature operation
|
|
Functions
for verifying
|
C_VerifyInit
|
initializes
a verification operation
|
|
signatures
|
C_Verify
|
verifies
a signature on single-part data
|
|
and
MACs
|
C_VerifyUpdate
|
continues
a multiple-part verification operation
|
|
|
C_VerifyFinal
|
finishes
a multiple-part verification operation
|
|
|
C_VerifyRecoverInit
|
initializes
a verification operation where the data is recovered from the signature
|
|
|
C_VerifyRecover
|
verifies
a signature on single-part data, where the data is recovered from the
signature
|
|
Message-based
Functions for verifying signatures and MACs
|
C_MessageVerifyInit
|
initializes
a message verification operation
|
|
C_VerifyMessage
|
verifies
single-part data
|
|
C_VerifyMessageBegin
|
starts
a multiple-part message verification operation
|
|
C_VerifyMessageNext
|
continues
and finishes a multiple-part message verification operation
|
|
C_MessageVerifyFinal
|
finishes
a message verification operation
|
|
Dual-purpose
cryptographic
|
C_DigestEncryptUpdate
|
continues
simultaneous multiple-part digesting and encryption operations
|
|
functions
|
C_DecryptDigestUpdate
|
continues
simultaneous multiple-part decryption and digesting operations
|
|
|
C_SignEncryptUpdate
|
continues
simultaneous multiple-part signature and encryption operations
|
|
|
C_DecryptVerifyUpdate
|
continues
simultaneous multiple-part decryption and verification operations
|
|
Key
|
C_GenerateKey
|
generates
a secret key
|
|
management
|
C_GenerateKeyPair
|
generates
a public-key/private-key pair
|
|
functions
|
C_WrapKey
|
wraps
(encrypts) a key
|
|
|
C_UnwrapKey
|
unwraps
(decrypts) a key
|
|
|
C_DeriveKey
|
derives
a key from a base key
|
|
Random
number generation
|
C_SeedRandom
|
mixes
in additional seed material to the random number generator
|
|
functions
|
C_GenerateRandom
|
generates
random data
|
|
Parallel
function management
|
C_GetFunctionStatus
|
legacy
function which always returns CKR_FUNCTION_NOT_PARALLEL
|
|
functions
|
C_CancelFunction
|
legacy
function which always returns CKR_FUNCTION_NOT_PARALLEL
|
|
Callback
function
|
|
application-supplied
function to process notifications from Cryptoki
|
Execution of a Cryptoki function call is in general an
all-or-nothing affair, i.e., a function call accomplishes either its
entire goal, or nothing at all.
·
If a Cryptoki function executes successfully, it returns the
value CKR_OK.
·
If a Cryptoki function does not execute successfully, it returns
some value other than CKR_OK, and the token is in the same state as it was in
prior to the function call. If the function call was supposed to modify the
contents of certain memory addresses on the host computer, these memory
addresses may have been modified, despite the failure of the function.
·
In unusual (and extremely unpleasant!) circumstances, a function
can fail with the return value CKR_GENERAL_ERROR. When this happens, the token
and/or host computer may be in an inconsistent state, and the goals of the
function may have been partially achieved.
There are a small number of Cryptoki functions whose return
values do not behave precisely as described above; these exceptions are
documented individually with the description of the functions themselves.
A Cryptoki library need not support every function in the
Cryptoki API. However, even an unsupported function MUST have a “stub” in the
library which simply returns the value CKR_FUNCTION_NOT_SUPPORTED. The
function’s entry in the library’s CK_FUNCTION_LIST structure (as
obtained by C_GetFunctionList) should point to this stub function (see
Section 3.6).
The Cryptoki interface possesses a large number of functions
and return values. In Section 5.1, we enumerate the various possible return
values for Cryptoki functions; most of the remainder of Section 5.1 details the behavior of Cryptoki functions, including what values each of them
may return.
Because of the complexity of the Cryptoki specification, it
is recommended that Cryptoki applications attempt to give some leeway when
interpreting Cryptoki functions’ return values. We have attempted to specify
the behavior of Cryptoki functions as completely as was feasible; nevertheless,
there are presumably some gaps. For example, it is possible that a particular
error code which might apply to a particular Cryptoki function is unfortunately
not actually listed in the description of that function as a possible error
code. It is conceivable that the developer of a Cryptoki library might
nevertheless permit his/her implementation of that function to return that
error code. It would clearly be somewhat ungraceful if a Cryptoki application
using that library were to terminate by abruptly dumping core upon receiving
that error code for that function. It would be far preferable for the
application to examine the function’s return value, see that it indicates some
sort of error (even if the application doesn’t know precisely what kind
of error), and behave accordingly.
See Section 5.1.8 for some specific details on how a
developer might attempt to make an application that accommodates a range of
behaviors from Cryptoki libraries.
Any Cryptoki function can return any of the following
values:
·
CKR_GENERAL_ERROR: Some horrible, unrecoverable error has
occurred. In the worst case, it is possible that the function only partially
succeeded, and that the computer and/or token is in an inconsistent state.
·
CKR_HOST_MEMORY: The computer that the Cryptoki library is
running on has insufficient memory to perform the requested function.
·
CKR_FUNCTION_FAILED: The requested function could not be
performed, but detailed information about why not is not available in this
error return. If the failed function uses a session, it is possible that the CK_SESSION_INFO
structure that can be obtained by calling C_GetSessionInfo will hold
useful information about what happened in its ulDeviceError field. In
any event, although the function call failed, the situation is not necessarily
totally hopeless, as it is likely to be when CKR_GENERAL_ERROR is returned.
Depending on what the root cause of the error actually was, it is possible that
an attempt to make the exact same function call again would succeed.
·
CKR_OK: The function executed successfully. Technically, CKR_OK
is not quite a “universal” return value; in particular, the legacy
functions C_GetFunctionStatus and C_CancelFunction (see Section 5.20) cannot return CKR_OK.
The relative priorities of these errors are in the order
listed above, e.g., if either of CKR_GENERAL_ERROR or CKR_HOST_MEMORY
would be an appropriate error return, then CKR_GENERAL_ERROR should be
returned.
Any Cryptoki function that takes a session handle as one of
its arguments (i.e., any Cryptoki function except for C_Initialize, C_Finalize,
C_GetInfo, C_GetFunctionList, C_GetSlotList, C_GetSlotInfo, C_GetTokenInfo,
C_WaitForSlotEvent, C_GetMechanismList, C_GetMechanismInfo, C_InitToken,
C_OpenSession, and C_CloseAllSessions) can return the following values:
·
CKR_SESSION_HANDLE_INVALID: The specified session handle was
invalid at the time that the function was invoked. Note that this can
happen if the session’s token is removed before the function invocation, since
removing a token closes all sessions with it.
·
CKR_DEVICE_REMOVED: The token was removed from its slot during
the execution of the function.
·
CKR_SESSION_CLOSED: The session was closed during the
execution of the function. Note that, as stated in [PKCS11-UG], the behavior of Cryptoki is undefined
if multiple threads of an application attempt to access a common Cryptoki
session simultaneously. Therefore, there is actually no guarantee that a
function invocation could ever return the value CKR_SESSION_CLOSED. An example
of multiple threads accessing a common session simultaneously is where one
thread is using a session when another thread closes that same session.
The relative priorities of these errors are in the order
listed above, e.g., if either of CKR_SESSION_HANDLE_INVALID or
CKR_DEVICE_REMOVED would be an appropriate error return, then
CKR_SESSION_HANDLE_INVALID should be returned.
In practice, it is often not crucial (or possible) for a
Cryptoki library to be able to make a distinction between a token being removed
before a function invocation and a token being removed during a
function execution.
Any Cryptoki function that uses a particular token (i.e.,
any Cryptoki function except for C_Initialize, C_Finalize, C_GetInfo,
C_GetFunctionList, C_GetSlotList, C_GetSlotInfo, or
C_WaitForSlotEvent) can return any of the following values:
·
CKR_DEVICE_MEMORY: The token does not have sufficient memory to
perform the requested function.
·
CKR_DEVICE_ERROR: Some problem has occurred with the token and/or
slot. This error code can be returned by more than just the functions
mentioned above; in particular, it is possible for C_GetSlotInfo to
return CKR_DEVICE_ERROR.
·
CKR_TOKEN_NOT_PRESENT: The token was not present in its slot at
the time that the function was invoked.
·
CKR_DEVICE_REMOVED: The token was removed from its slot during
the execution of the function.
The relative priorities of these errors are in the order
listed above, e.g., if either of CKR_DEVICE_MEMORY or CKR_DEVICE_ERROR
would be an appropriate error return, then CKR_DEVICE_MEMORY should be
returned.
In practice, it is often not critical (or possible) for a
Cryptoki library to be able to make a distinction between a token being removed
before a function invocation and a token being removed during a
function execution.
There is a special-purpose return value which is not returned
by any function in the actual Cryptoki API, but which may be returned by an
application-supplied callback function. It is:
·
CKR_CANCEL: When a function executing in serial with an
application decides to give the application a chance to do some work, it calls
an application-supplied function with a CKN_SURRENDER callback (see Section 5.21). If the callback returns the value CKR_CANCEL, then the function aborts and
returns CKR_FUNCTION_CANCELED.
There are two other special-purpose return values which are
not returned by any actual Cryptoki functions. These values may be returned by
application-supplied mutex-handling functions, and they may safely be ignored
by application developers who are not using their own threading model. They
are:
·
CKR_MUTEX_BAD: This error code can be returned by mutex-handling
functions that are passed a bad mutex object as an argument. Unfortunately, it
is possible for such a function not to recognize a bad mutex object. There is
therefore no guarantee that such a function will successfully detect bad mutex
objects and return this value.
·
CKR_MUTEX_NOT_LOCKED: This error code can be returned by
mutex-unlocking functions. It indicates that the mutex supplied to the
mutex-unlocking function was not locked.
Descriptions of the other Cryptoki function return values
follow. Except as mentioned in the descriptions of particular error codes,
there are in general no particular priorities among the errors listed below, i.e.,
if more than one error code might apply to an execution of a function, then the
function may return any applicable error code.
·
CKR_ACTION_PROHIBITED: This value can only be returned by
C_CopyObject, C_SetAttributeValue and C_DestroyObject. It denotes that the
action may not be taken, either because of underlying policy restrictions on the
token, or because the object has the relevant CKA_COPYABLE, CKA_MODIFIABLE or
CKA_DESTROYABLE policy attribute set to CK_FALSE.
·
CKR_ARGUMENTS_BAD: This is a rather generic error code which
indicates that the arguments supplied to the Cryptoki function were in some way
not appropriate.
·
CKR_ATTRIBUTE_READ_ONLY: An attempt was made to set a value for
an attribute which may not be set by the application, or which may not be
modified by the application. See Section 4.1
for more information.
·
CKR_ATTRIBUTE_SENSITIVE: An attempt was made to obtain the value
of an attribute of an object which cannot be satisfied because the object is
either sensitive or un-extractable.
·
CKR_ATTRIBUTE_TYPE_INVALID: An invalid attribute type was
specified in a template. See Section 4.1 for more information.
·
CKR_ATTRIBUTE_VALUE_INVALID: An invalid value was specified for a
particular attribute in a template. See Section 4.1
for more information.
·
CKR_BUFFER_TOO_SMALL: The output of the function is too large to
fit in the supplied buffer.
·
CKR_CANT_LOCK: This value can only be returned by C_Initialize.
It means that the type of locking requested by the application for
thread-safety is not available in this library, and so the application cannot
make use of this library in the specified fashion.
·
CKR_CRYPTOKI_ALREADY_INITIALIZED: This value can only be returned
by C_Initialize. It means that the Cryptoki library has already been
initialized (by a previous call to C_Initialize which did not have a
matching C_Finalize call).
·
CKR_CRYPTOKI_NOT_INITIALIZED: This value can be returned by any
function other than C_Initialize, C_GetFunctionList,
C_GetInterfaceList and C_GetInterface. It indicates that the function
cannot be executed because the Cryptoki library has not yet been initialized by
a call to C_Initialize.
·
CKR_CURVE_NOT_SUPPORTED: This curve is not supported by this
token. Used with Elliptic Curve mechanisms.
·
CKR_DATA_INVALID: The plaintext input data to a cryptographic
operation is invalid. This return value has lower priority than
CKR_DATA_LEN_RANGE.
·
CKR_DATA_LEN_RANGE: The plaintext input data to a cryptographic
operation has a bad length. Depending on the operation’s mechanism, this could
mean that the plaintext data is too short, too long, or is not a multiple of
some particular block size. This return value has higher priority than
CKR_DATA_INVALID.
·
CKR_DOMAIN_PARAMS_INVALID: Invalid or unsupported domain
parameters were supplied to the function. Which representation methods of
domain parameters are supported by a given mechanism can vary from token to
token.
·
CKR_ENCRYPTED_DATA_INVALID: The encrypted input to a decryption
operation has been determined to be invalid ciphertext. This return value has
lower priority than CKR_ENCRYPTED_DATA_LEN_RANGE.
·
CKR_ENCRYPTED_DATA_LEN_RANGE: The ciphertext input to a
decryption operation has been determined to be invalid ciphertext solely on the
basis of its length. Depending on the operation’s mechanism, this could mean
that the ciphertext is too short, too long, or is not a multiple of some
particular block size. This return value has higher priority than
CKR_ENCRYPTED_DATA_INVALID.
·
CKR_EXCEEDED_MAX_ITERATIONS: An iterative algorithm (for key pair
generation, domain parameter generation etc.) failed because we have exceeded
the maximum number of iterations. This error code has precedence over
CKR_FUNCTION_FAILED. Examples of iterative algorithms include DSA signature
generation (retry if either r = 0 or s = 0) and generation of DSA primes p and
q specified in FIPS 186-4.
·
CKR_FIPS_SELF_TEST_FAILED: A FIPS 140-2 power-up self-test or
conditional self-test failed. The token entered an error state. Future calls
to cryptographic functions on the token will return CKR_GENERAL_ERROR.
CKR_FIPS_SELF_TEST_FAILED has a higher precedence over CKR_GENERAL_ERROR. This
error may be returned by C_Initialize, if a power-up self-test failed, by
C_GenerateRandom or C_SeedRandom, if the continuous random number generator
test failed, or by C_GenerateKeyPair, if the pair-wise consistency test failed.
·
CKR_FUNCTION_CANCELED: The function was canceled in
mid-execution. This happens to a cryptographic function if the function makes
a CKN_SURRENDER application callback which returns CKR_CANCEL (see
CKR_CANCEL). It also happens to a function that performs PIN entry through a
protected path. The method used to cancel a protected path PIN entry operation
is device dependent.
·
CKR_FUNCTION_NOT_PARALLEL: There is currently no function
executing in parallel in the specified session. This is a legacy error code
which is only returned by the legacy functions C_GetFunctionStatus and C_CancelFunction.
·
CKR_FUNCTION_NOT_SUPPORTED: The requested function is not
supported by this Cryptoki library. Even unsupported functions in the Cryptoki
API should have a “stub” in the library; this stub should simply return the
value CKR_FUNCTION_NOT_SUPPORTED.
·
CKR_FUNCTION_REJECTED: The signature request is rejected by the
user.
·
CKR_INFORMATION_SENSITIVE: The information requested could not be
obtained because the token considers it sensitive, and is not able or willing
to reveal it.
·
CKR_KEY_CHANGED: This value is only returned by C_SetOperationState.
It indicates that one of the keys specified is not the same key that was being
used in the original saved session.
·
CKR_KEY_FUNCTION_NOT_PERMITTED: An attempt has been made to use a
key for a cryptographic purpose that the key’s attributes are not set to allow
it to do. For example, to use a key for performing encryption, that key MUST
have its CKA_ENCRYPT attribute set to CK_TRUE (the fact that the key
MUST have a CKA_ENCRYPT attribute implies that the key cannot be a
private key). This return value has lower priority than CKR_KEY_TYPE_INCONSISTENT.
·
CKR_KEY_HANDLE_INVALID: The specified key handle is not valid.
It may be the case that the specified handle is a valid handle for an object
which is not a key. We reiterate here that 0 is never a valid key handle.
·
CKR_KEY_INDIGESTIBLE: This error code can only be returned by C_DigestKey.
It indicates that the value of the specified key cannot be digested for some
reason (perhaps the key isn’t a secret key, or perhaps the token simply can’t
digest this kind of key).
·
CKR_KEY_NEEDED: This value is only returned by C_SetOperationState.
It indicates that the session state cannot be restored because C_SetOperationState
needs to be supplied with one or more keys that were being used in the original
saved session.
·
CKR_KEY_NOT_NEEDED: An extraneous key was supplied to C_SetOperationState.
For example, an attempt was made to restore a session that had been performing
a message digesting operation, and an encryption key was supplied.
·
CKR_KEY_NOT_WRAPPABLE: Although the specified private or secret key
does not have its CKA_EXTRACTABLE attribute set to CK_FALSE, Cryptoki (or the
token) is unable to wrap the key as requested (possibly the token can only wrap
a given key with certain types of keys, and the wrapping key specified is not
one of these types). Compare with CKR_KEY_UNEXTRACTABLE.
·
CKR_KEY_SIZE_RANGE: Although the requested keyed
cryptographic operation could in principle be carried out, this Cryptoki
library (or the token) is unable to actually do it because the supplied key‘s
size is outside the range of key sizes that it can handle.
·
CKR_KEY_TYPE_INCONSISTENT: The specified key is not the correct
type of key to use with the specified mechanism. This return value has a
higher priority than CKR_KEY_FUNCTION_NOT_PERMITTED.
·
CKR_KEY_UNEXTRACTABLE: The specified private or secret key can’t
be wrapped because its CKA_EXTRACTABLE attribute is set to CK_FALSE. Compare
with CKR_KEY_NOT_WRAPPABLE.
·
CKR_LIBRARY_LOAD_FAILED: The Cryptoki library could not load a dependent shared library.
·
CKR_MECHANISM_INVALID: An invalid mechanism was specified to the
cryptographic operation. This error code is an appropriate return value if an
unknown mechanism was specified or if the mechanism specified cannot be used in
the selected token with the selected function.
·
CKR_MECHANISM_PARAM_INVALID: Invalid parameters were supplied to
the mechanism specified to the cryptographic operation. Which parameter values
are supported by a given mechanism can vary from token to token.
·
CKR_NEED_TO_CREATE_THREADS: This value can only be returned by C_Initialize.
It is returned when two conditions hold:
- The application called C_Initialize in a way which
tells the Cryptoki library that application threads executing calls to the
library cannot use native operating system methods to spawn new threads.
- The library cannot function properly without being able to
spawn new threads in the above fashion.
·
CKR_NO_EVENT: This value can only be returned by C_WaitForSlotEvent.
It is returned when C_WaitForSlotEvent is called in non-blocking mode and
there are no new slot events to return.
·
CKR_OBJECT_HANDLE_INVALID: The specified object handle is not
valid. We reiterate here that 0 is never a valid object handle.
·
CKR_OPERATION_ACTIVE: There is already an active operation (or
combination of active operations) which prevents Cryptoki from activating the
specified operation. For example, an active object-searching operation would
prevent Cryptoki from activating an encryption operation with C_EncryptInit.
Or, an active digesting operation and an active encryption operation would
prevent Cryptoki from activating a signature operation. Or, on a token which
doesn’t support simultaneous dual cryptographic operations in a session (see
the description of the CKF_DUAL_CRYPTO_OPERATIONS flag in the CK_TOKEN_INFO
structure), an active signature operation would prevent Cryptoki from
activating an encryption operation.
·
CKR_OPERATION_NOT_INITIALIZED: There is no active operation of an
appropriate type in the specified session. For example, an application cannot
call C_Encrypt in a session without having called C_EncryptInit
first to activate an encryption operation.
·
CKR_PIN_EXPIRED: The specified PIN has expired, and the requested
operation cannot be carried out unless C_SetPIN is called to change the PIN
value. Whether or not the normal user’s PIN on a token ever expires varies
from token to token.
·
CKR_PIN_INCORRECT: The specified PIN is incorrect, i.e.,
does not match the PIN stored on the token. More generally-- when
authentication to the token involves something other than a PIN-- the attempt
to authenticate the user has failed.
·
CKR_PIN_INVALID: The specified PIN has invalid characters in it.
This return code only applies to functions which attempt to set a PIN.
·
CKR_PIN_LEN_RANGE: The specified PIN is too long or too short.
This return code only applies to functions which attempt to set a PIN.
·
CKR_PIN_LOCKED: The specified PIN is “locked”, and cannot be
used. That is, because some particular number of failed authentication
attempts has been reached, the token is unwilling to permit further attempts at
authentication. Depending on the token, the specified PIN may or may not
remain locked indefinitely.
·
CKR_PIN_TOO_WEAK: The specified PIN is too weak so that it could
be easy to guess. If the PIN is too short, CKR_PIN_LEN_RANGE should be
returned instead. This return code only applies to functions which attempt to
set a PIN.
·
CKR_PUBLIC_KEY_INVALID: The public key fails a public key
validation. For example, an EC public key fails the public key validation
specified in Section 5.2.2 of ANSI X9.62. This error code may be returned by
C_CreateObject, when the public key is created, or by C_VerifyInit or
C_VerifyRecoverInit, when the public key is used. It may also be returned by
C_DeriveKey, in preference to CKR_MECHANISM_PARAM_INVALID, if the other
party's public key specified in the mechanism's parameters is invalid.
·
CKR_RANDOM_NO_RNG: This value can be returned by C_SeedRandom
and C_GenerateRandom. It indicates that the specified token doesn’t have
a random number generator. This return value has higher priority than
CKR_RANDOM_SEED_NOT_SUPPORTED.
·
CKR_RANDOM_SEED_NOT_SUPPORTED: This value can only be returned by
C_SeedRandom. It indicates that the token’s random number generator
does not accept seeding from an application. This return value has lower
priority than CKR_RANDOM_NO_RNG.
·
CKR_SAVED_STATE_INVALID: This value can only be returned by C_SetOperationState.
It indicates that the supplied saved cryptographic operations state is invalid,
and so it cannot be restored to the specified session.
·
CKR_SESSION_COUNT: This value can only be returned by C_OpenSession.
It indicates that the attempt to open a session failed, either because the
token has too many sessions already open, or because the token has too many
read/write sessions already open.
·
CKR_SESSION_EXISTS: This value can only be returned by C_InitToken.
It indicates that a session with the token is already open, and so the token
cannot be initialized.
·
CKR_SESSION_PARALLEL_NOT_SUPPORTED: The specified token does not
support parallel sessions. This is a legacy error code—in Cryptoki Version
2.01 and up, no token supports parallel sessions.
CKR_SESSION_PARALLEL_NOT_SUPPORTED can only be returned by C_OpenSession,
and it is only returned when C_OpenSession is called in a particular
[deprecated] way.
·
CKR_SESSION_READ_ONLY: The specified session was unable to
accomplish the desired action because it is a read-only session. This return
value has lower priority than CKR_TOKEN_WRITE_PROTECTED.
·
CKR_SESSION_READ_ONLY_EXISTS: A read-only session already exists,
and so the SO cannot be logged in.
·
CKR_SESSION_READ_WRITE_SO_EXISTS: A read/write SO session already
exists, and so a read-only session cannot be opened.
·
CKR_SIGNATURE_LEN_RANGE: The provided signature/MAC can be seen
to be invalid solely on the basis of its length. This return value has higher
priority than CKR_SIGNATURE_INVALID.
·
CKR_SIGNATURE_INVALID: The provided signature/MAC is invalid.
This return value has lower priority than CKR_SIGNATURE_LEN_RANGE.
·
CKR_SLOT_ID_INVALID: The specified slot ID is not valid.
·
CKR_STATE_UNSAVEABLE: The cryptographic operations state of the
specified session cannot be saved for some reason (possibly the token is simply
unable to save the current state). This return value has lower priority than
CKR_OPERATION_NOT_INITIALIZED.
·
CKR_TEMPLATE_INCOMPLETE: The template specified for creating an
object is incomplete, and lacks some necessary attributes. See Section 4.1 for more information.
·
CKR_TEMPLATE_INCONSISTENT: The template specified for creating an
object has conflicting attributes. See Section 4.1
for more information.
·
CKR_TOKEN_NOT_RECOGNIZED: The Cryptoki library and/or slot does
not recognize the token in the slot.
·
CKR_TOKEN_WRITE_PROTECTED: The requested action could not be
performed because the token is write-protected. This return value has higher
priority than CKR_SESSION_READ_ONLY.
·
CKR_UNWRAPPING_KEY_HANDLE_INVALID: This value can only be
returned by C_UnwrapKey. It indicates that the key handle specified to
be used to unwrap another key is not valid.
·
CKR_UNWRAPPING_KEY_SIZE_RANGE: This value can only be returned by
C_UnwrapKey. It indicates that although the requested unwrapping
operation could in principle be carried out, this Cryptoki library (or the
token) is unable to actually do it because the supplied key’s size is outside
the range of key sizes that it can handle.
·
CKR_UNWRAPPING_KEY_TYPE_INCONSISTENT: This value can only be
returned by C_UnwrapKey. It indicates that the type of the key
specified to unwrap another key is not consistent with the mechanism specified
for unwrapping.
·
CKR_USER_ALREADY_LOGGED_IN: This value can only be returned by C_Login.
It indicates that the specified user cannot be logged into the session, because
it is already logged into the session. For example, if an application has an
open SO session, and it attempts to log the SO into it, it will receive this
error code.
·
CKR_USER_ANOTHER_ALREADY_LOGGED_IN: This value can only be
returned by C_Login. It indicates that the specified user cannot be
logged into the session, because another user is already logged into the
session. For example, if an application has an open SO session, and it
attempts to log the normal user into it, it will receive this error code.
·
CKR_USER_NOT_LOGGED_IN: The desired action cannot be performed
because the appropriate user (or an appropriate user) is not logged in.
One example is that a session cannot be logged out unless it is logged in.
Another example is that a private object cannot be created on a token unless
the session attempting to create it is logged in as the normal user. A final
example is that cryptographic operations on certain tokens cannot be performed
unless the normal user is logged in.
·
CKR_USER_PIN_NOT_INITIALIZED: This value can only be returned by C_Login.
It indicates that the normal user’s PIN has not yet been initialized with C_InitPIN.
·
CKR_USER_TOO_MANY_TYPES: An attempt was made to have more
distinct users simultaneously logged into the token than the token and/or
library permits. For example, if some application has an open SO session, and
another application attempts to log the normal user into a session, the attempt
may return this error. It is not required to, however. Only if the
simultaneous distinct users cannot be supported does C_Login have to
return this value. Note that this error code generalizes to true multi-user
tokens.
·
CKR_USER_TYPE_INVALID: An invalid value was specified as a CK_USER_TYPE.
Valid types are CKU_SO, CKU_USER, and CKU_CONTEXT_SPECIFIC.
·
CKR_WRAPPED_KEY_INVALID: This value can only be returned by C_UnwrapKey.
It indicates that the provided wrapped key is not valid. If a call is made to C_UnwrapKey
to unwrap a particular type of key (i.e., some particular key type is
specified in the template provided to C_UnwrapKey), and the wrapped key
provided to C_UnwrapKey is recognizably not a wrapped key of the proper
type, then C_UnwrapKey should return CKR_WRAPPED_KEY_INVALID. This
return value has lower priority than CKR_WRAPPED_KEY_LEN_RANGE.
·
CKR_WRAPPED_KEY_LEN_RANGE: This value can only be returned by C_UnwrapKey.
It indicates that the provided wrapped key can be seen to be invalid solely on
the basis of its length. This return value has higher priority than
CKR_WRAPPED_KEY_INVALID.
·
CKR_WRAPPING_KEY_HANDLE_INVALID: This value can only be returned
by C_WrapKey. It indicates that the key handle specified to be used to
wrap another key is not valid.
·
CKR_WRAPPING_KEY_SIZE_RANGE: This value can only be returned by C_WrapKey.
It indicates that although the requested wrapping operation could in principle
be carried out, this Cryptoki library (or the token) is unable to actually do
it because the supplied wrapping key’s size is outside the range of key sizes
that it can handle.
·
CKR_WRAPPING_KEY_TYPE_INCONSISTENT: This value can only be
returned by C_WrapKey. It indicates that the type of the key specified
to wrap another key is not consistent with the mechanism specified for
wrapping.
·
CKR_OPERATION_CANCEL_FAILED: This value can only be returned by C_SessionCancel.
It means that one or more of the requested operations could not be cancelled
for implementation or vendor-specific reasons.
In general, when a Cryptoki call is made, error codes from
Section 5.1.1 (other than CKR_OK) take precedence over error codes from Section
5.1.2, which take precedence over error codes from Section 5.1.3,
which take precedence over error codes from Section 5.1.6.
One minor implication of this is that functions that use a session handle (i.e.,
most functions!) never return the error code CKR_TOKEN_NOT_PRESENT (they
return CKR_SESSION_HANDLE_INVALID instead). Other than these precedences, if
more than one error code applies to the result of a Cryptoki call, any of the
applicable error codes may be returned. Exceptions to this rule will be
explicitly mentioned in the descriptions of functions.
Here is a short list of a few particular things about return
values that Cryptoki developers might want to be aware of:
1. As
mentioned in Sections 5.1.2 and 5.1.3, a Cryptoki library may not be able to
make a distinction between a token being removed before a function
invocation and a token being removed during a function invocation.
2. As
mentioned in Section 5.1.2, an application should never count on getting a
CKR_SESSION_CLOSED error.
3. The
difference between CKR_DATA_INVALID and CKR_DATA_LEN_RANGE can be somewhat
subtle. Unless an application needs to be able to distinguish between
these return values, it is best to always treat them equivalently.
4. Similarly,
the difference between CKR_ENCRYPTED_DATA_INVALID and
CKR_ENCRYPTED_DATA_LEN_RANGE, and between CKR_WRAPPED_KEY_INVALID and
CKR_WRAPPED_KEY_LEN_RANGE, can be subtle, and it may be best to treat these
return values equivalently.
5. Even with
the guidance of Section 4.1, it can be difficult for a Cryptoki library
developer to know which of CKR_ATTRIBUTE_VALUE_INVALID,
CKR_TEMPLATE_INCOMPLETE, or CKR_TEMPLATE_INCONSISTENT to return. When
possible, it is recommended that application developers be generous in their
interpretations of these error codes.
A number of the functions defined in Cryptoki return output
produced by some cryptographic mechanism. The amount of output returned by
these functions is returned in a variable-length application-supplied buffer.
An example of a function of this sort is C_Encrypt, which takes some
plaintext as an argument, and outputs a buffer full of ciphertext.
These functions have some common calling conventions, which
we describe here. Two of the arguments to the function are a pointer to the
output buffer (say pBuf) and a pointer to a location which will hold the
length of the output produced (say pulBufLen). There are two ways for
an application to call such a function:
1. If pBuf
is NULL_PTR, then all that the function does is return (in *pulBufLen) a
number of bytes which would suffice to hold the cryptographic output produced
from the input to the function. This number may somewhat exceed the precise number
of bytes needed, but should not exceed it by a large amount. CKR_OK is
returned by the function.
2. If pBuf
is not NULL_PTR, then *pulBufLen MUST contain the size in bytes of the
buffer pointed to by pBuf. If that buffer is large enough to hold the
cryptographic output produced from the input to the function, then that
cryptographic output is placed there, and CKR_OK is returned by the function
and *pulBufLen is set to the exact number of bytes returned. If the buffer is
not large enough, then CKR_BUFFER_TOO_SMALL is returned and *pulBufLen
is set to at least the number of bytes needed to hold the cryptographic output
produced from the input to the function.
NOTE: This is a change from previous specs. The problem is
that in some decrypt cases, the token doesn’t know how big a buffer is needed
until the decrypt completes. The act of doing decrypt can mess up the internal
encryption state. Many tokens already implement this relaxed behavior, tokens
which implement the more precise behavior are still compliant. The one corner
case is applications using a token that knows exactly how big the decryption
is (through some out of band means), could get CKR_BUFFER_TOO_SMALL returned
when it supplied a buffer exactly big enough to hold the decrypted value when
it may previously have succeeded.
All functions which use the above convention will explicitly
say so.
Cryptographic functions which return output in a
variable-length buffer should always return as much output as can be computed
from what has been passed in to them thus far. As an example, consider a
session which is performing a multiple-part decryption operation with DES in
cipher-block chaining mode with PKCS padding. Suppose that, initially, 8 bytes
of ciphertext are passed to the C_DecryptUpdate function. The block
size of DES is 8 bytes, but the PKCS padding makes it unclear at this stage
whether the ciphertext was produced from encrypting a 0-byte string, or from
encrypting some string of length at least 8 bytes. Hence the call to C_DecryptUpdate
should return 0 bytes of plaintext. If a single additional byte of ciphertext
is supplied by a subsequent call to C_DecryptUpdate, then that call
should return 8 bytes of plaintext (one full DES block).
For the remainder of this section, we enumerate the various
functions defined in Cryptoki. Most functions will be shown in use in at least
one sample code snippet. For the sake of brevity, sample code will frequently
be somewhat incomplete. In particular, sample code will generally ignore
possible error returns from C library functions, and also will not deal with
Cryptoki error returns in a realistic fashion.
Cryptoki provides the following general-purpose functions:
CK_DECLARE_FUNCTION(CK_RV, C_Initialize) {
CK_VOID_PTR pInitArgs
);
C_Initialize initializes the Cryptoki library. pInitArgs
either has the value NULL_PTR or points to a CK_C_INITIALIZE_ARGS
structure containing information on how the library should deal with
multi-threaded access. If an application will not be accessing Cryptoki
through multiple threads simultaneously, it can generally supply the value
NULL_PTR to C_Initialize (the consequences of supplying this value will
be explained below).
If pInitArgs is non-NULL_PTR, C_Initialize
should cast it to a CK_C_INITIALIZE_ARGS_PTR and then dereference the
resulting pointer to obtain the CK_C_INITIALIZE_ARGS fields CreateMutex,
DestroyMutex, LockMutex, UnlockMutex, flags, and pReserved.
For this version of Cryptoki, the value of pReserved thereby obtained
MUST be NULL_PTR; if it’s not, then C_Initialize should return with the
value CKR_ARGUMENTS_BAD.
If the CKF_LIBRARY_CANT_CREATE_OS_THREADS flag in the
flags field is set, that indicates that application threads which are
executing calls to the Cryptoki library are not permitted to use the native
operation system calls to spawn off new threads. In other words, the library’s
code may not create its own threads. If the library is unable to function
properly under this restriction, C_Initialize should return with the
value CKR_NEED_TO_CREATE_THREADS.
A call to C_Initialize specifies one of four
different ways to support multi-threaded access via the value of the CKF_OS_LOCKING_OK
flag in the flags field and the values of the CreateMutex, DestroyMutex,
LockMutex, and UnlockMutex function pointer fields:
1. If the
flag isn’t set, and the function pointer fields aren’t supplied (i.e.,
they all have the value NULL_PTR), that means that the application won’t
be accessing the Cryptoki library from multiple threads simultaneously.
2. If the
flag is set, and the function pointer fields aren’t supplied (i.e.,
they all have the value NULL_PTR), that means that the application will
be performing multi-threaded Cryptoki access, and the library needs to use the
native operating system primitives to ensure safe multi-threaded access. If
the library is unable to do this, C_Initialize should return with the
value CKR_CANT_LOCK.
3. If the
flag isn’t set, and the function pointer fields are supplied (i.e.,
they all have non-NULL_PTR values), that means that the application will
be performing multi-threaded Cryptoki access, and the library needs to use the
supplied function pointers for mutex-handling to ensure safe multi-threaded
access. If the library is unable to do this, C_Initialize should return
with the value CKR_CANT_LOCK.
4. If the
flag is set, and the function pointer fields are supplied (i.e.,
they all have non-NULL_PTR values), that means that the application will
be performing multi-threaded Cryptoki access, and the library needs to use
either the native operating system primitives or the supplied function pointers
for mutex-handling to ensure safe multi-threaded access. If the library is
unable to do this, C_Initialize should return with the value
CKR_CANT_LOCK.
If some, but not all, of the supplied function pointers to C_Initialize
are non-NULL_PTR, then C_Initialize should return with the value
CKR_ARGUMENTS_BAD.
A call to C_Initialize with pInitArgs set to
NULL_PTR is treated like a call to C_Initialize with pInitArgs
pointing to a CK_C_INITIALIZE_ARGS which has the CreateMutex, DestroyMutex,
LockMutex, UnlockMutex, and pReserved fields set to
NULL_PTR, and has the flags field set to 0.
C_Initialize should be the first Cryptoki call made
by an application, except for calls to C_GetFunctionList,
C_GetInterfaceList, or C_GetInterface. What this function actually
does is implementation-dependent; typically, it might cause Cryptoki to
initialize its internal memory buffers, or any other resources it requires.
If several applications are using Cryptoki, each one should
call C_Initialize. Every call to C_Initialize should (eventually)
be succeeded by a single call to C_Finalize. See [PKCS11-UG] for further details.
Return values: CKR_ARGUMENTS_BAD, CKR_CANT_LOCK,
CKR_CRYPTOKI_ALREADY_INITIALIZED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR,
CKR_HOST_MEMORY, CKR_NEED_TO_CREATE_THREADS, CKR_OK.
Example: see C_GetInfo.
CK_DECLARE_FUNCTION(CK_RV, C_Finalize)(
CK_VOID_PTR pReserved
);
C_Finalize is called to indicate that an application
is finished with the Cryptoki library. It should be the last Cryptoki call
made by an application. The pReserved parameter is reserved for future
versions; for this version, it should be set to NULL_PTR (if C_Finalize
is called with a non-NULL_PTR value for pReserved, it should return the
value CKR_ARGUMENTS_BAD.
If several applications are using Cryptoki, each one should
call C_Finalize. Each application’s call to C_Finalize should be
preceded by a single call to C_Initialize; in between the two calls, an
application can make calls to other Cryptoki functions. See [PKCS11-UG] for further details.
Despite the fact that the parameters supplied to C_Initialize
can in general allow for safe multi-threaded access to a Cryptoki library, the
behavior of C_Finalize is nevertheless undefined if it is called by an
application while other threads of the application are making Cryptoki calls.
The exception to this exceptional behavior of C_Finalize occurs when a
thread calls C_Finalize while another of the application’s threads is
blocking on Cryptoki’s C_WaitForSlotEvent function. When this happens,
the blocked thread becomes unblocked and returns the value CKR_CRYPTOKI_NOT_INITIALIZED.
See C_WaitForSlotEvent for more information.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR,
CKR_HOST_MEMORY, CKR_OK.
Example: see C_GetInfo.
CK_DECLARE_FUNCTION(CK_RV, C_GetInfo)(
CK_INFO_PTR pInfo
);
C_GetInfo returns general information about
Cryptoki. pInfo points to the location that receives the information.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_OK.
Example:
CK_INFO info;
CK_RV rv;
CK_C_INITIALIZE_ARGS InitArgs;
InitArgs.CreateMutex = &MyCreateMutex;
InitArgs.DestroyMutex = &MyDestroyMutex;
InitArgs.LockMutex = &MyLockMutex;
InitArgs.UnlockMutex = &MyUnlockMutex;
InitArgs.flags = CKF_OS_LOCKING_OK;
InitArgs.pReserved = NULL_PTR;
rv = C_Initialize((CK_VOID_PTR)&InitArgs);
assert(rv == CKR_OK);
rv = C_GetInfo(&info);
assert(rv == CKR_OK);
if(info.cryptokiVersion.major == 2) {
/* Do lots of interesting cryptographic things with the
token */
.
.
}
rv = C_Finalize(NULL_PTR);
assert(rv == CKR_OK);
CK_DECLARE_FUNCTION(CK_RV, C_GetFunctionList)(
CK_FUNCTION_LIST_PTR_PTR ppFunctionList
);
C_GetFunctionList obtains a pointer to the Cryptoki
library’s list of function pointers. ppFunctionList points to a value
which will receive a pointer to the library’s CK_FUNCTION_LIST
structure, which in turn contains function pointers for all the Cryptoki API
routines in the library. The pointer thus obtained may point into memory
which is owned by the Cryptoki library, and which may or may not be writable.
Whether or not this is the case, no attempt should be made to write to this
memory.
C_GetFunctionList, C_GetInterfaceList, and
C_GetInterface are the only Cryptoki functions which an application may
call before calling C_Initialize. It is provided to make it easier and
faster for applications to use shared Cryptoki libraries and to use more than
one Cryptoki library simultaneously.
Return values: CKR_ARGUMENTS_BAD, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK.
Example:
CK_FUNCTION_LIST_PTR pFunctionList;
CK_C_Initialize pC_Initialize;
CK_RV rv;
/* It’s OK to call C_GetFunctionList before calling C_Initialize
*/
rv = C_GetFunctionList(&pFunctionList);
assert(rv == CKR_OK);
pC_Initialize = pFunctionList -> C_Initialize;
/* Call the C_Initialize function in the library */
rv = (*pC_Initialize)(NULL_PTR);
CK_DECLARE_FUNCTION(CK_RV, C_GetInterfaceList)(
CK_INTERFACE_PTR pInterfaceList,
CK_ULONG_PTR pulCount
);
C_GetInterfaceList is used to obtain a list of
interfaces supported by a Cryptoki library. pulCount points to the
location that receives the number of interfaces.
There are two ways for an application to call C_GetInterfaceList:
1. If pInterfaceList
is NULL_PTR, then all that C_GetInterfaceList does is return (in *pulCount)
the number of interfaces, without actually returning a list of interfaces. The
contents of *pulCount on entry to C_GetInterfaceList has no
meaning in this case, and the call returns the value CKR_OK.
2. If pIntrerfaceList
is not NULL_PTR, then *pulCount MUST contain the size (in terms of CK_INTERFACE
elements) of the buffer pointed to by pInterfaceList. If that buffer is
large enough to hold the list of interfaces, then the list is returned in it,
and CKR_OK is returned. If not, then the call to C_GetInterfaceList
returns the value CKR_BUFFER_TOO_SMALL. In either case, the value *pulCount
is set to hold the number of interfaces.
Because C_GetInterfaceList does not allocate any
space of its own, an application will often call C_GetInterfaceList
twice. However, this behavior is by no means required.
C_GetInterfaceList obtains (in *pFunctionList
of each interface) a pointer to the Cryptoki library’s list of function
pointers. The pointer thus obtained may point into memory which is owned by
the Cryptoki library, and which may or may not be writable. Whether or not
this is the case, no attempt should be made to write to this memory. The same
caveat applies to the interface names returned.
C_GetFunctionList, C_GetInterfaceList, and
C_GetInterface are the only Cryptoki functions which an application may
call before calling C_Initialize. It is provided to make it easier and
faster for applications to use shared Cryptoki libraries and to use more than
one Cryptoki library simultaneously.
Return values: CKR_BUFFER_TOO_SMALL, CKR_ARGUMENTS_BAD,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK.
Example:
CK_ULONG ulCount=0;
CK_INTERFACE_PTR interfaceList=NULL;
CK_RV rv;
int I;
/* get number of interfaces */
rv = C_GetInterfaceList(NULL,&ulCount);
if (rv == CKR_OK) {
/* get copy of interfaces */
interfaceList =
(CK_INTERFACE_PTR)malloc(ulCount*sizeof(CK_INTERFACE));
rv = C_GetInterfaceList(interfaceList,&ulCount);
for(i=0;i<ulCount;i++) {
printf("interface %s version %d.%d funcs %p flags
0x%lu\n",
interfaceList[i].pInterfaceName,
((CK_VERSION
*)interfaceList[i].pFunctionList)->major,
((CK_VERSION
*)interfaceList[i].pFunctionList)->minor,
interfaceList[i].pFunctionList,
interfaceList[i].flags);
}
}
CK_DECLARE_FUNCTION(CK_RV,C_GetInterface)(
CK_UTF8CHAR_PTR pInterfaceName,
CK_VERSION_PTR pVersion,
CK_INTERFACE_PTR_PTR ppInterface,
CK_FLAGS flags
);
C_GetInterface is used to obtain an interface
supported by a Cryptoki library. pInterfaceName specifies the name of
the interface, pVersion specifies the interface version, ppInterface
points to the location that receives the interface, flags specifies the
required interface flags.
There are multiple ways for an application to specify a
particular interface when calling C_GetInterface:
1. If pInterfaceName
is not NULL_PTR, the name of the interface returned must match. If pInterfaceName
is NULL_PTR, the cryptoki library can return a default interface of its choice
2. If pVersion
is not NULL_PTR, the version of the interface returned must match. If pVersion
is NULL_PTR, the cryptoki library can return an interface of any version
3. If flags
is non-zero, the interface returned must match all of the supplied flag values
(but may include additional flags not specified). If flags is 0, the
cryptoki library can return an interface with any flags
C_GetInterface obtains (in *pFunctionList of
each interface) a pointer to the Cryptoki library’s list of function pointers. The
pointer thus obtained may point into memory which is owned by the Cryptoki
library, and which may or may not be writable. Whether or not this is the
case, no attempt should be made to write to this memory. The same caveat
applies to the interface names returned.
C_GetFunctionList, C_GetInterfaceList, and
C_GetInterface are the only Cryptoki functions which an application may
call before calling C_Initialize. It is provided to make it easier and
faster for applications to use shared Cryptoki libraries and to use more than
one Cryptoki library simultaneously.
Return values: CKR_BUFFER_TOO_SMALL, CKR_ARGUMENTS_BAD,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK.
Example:
CK_INTERFACE_PTR interface;
CK_RV rv;
CK_VERSION version;
CK_FLAGS flags=CKF_ INTERFACE_FORK_SAFE;
/* get default interface */
rv = C_GetInterface(NULL,NULL,&interface,flags);
if (rv == CKR_OK) {
printf("interface %s version %d.%d funcs %p flags
0x%lu\n",
interface->pInterfaceName,
((CK_VERSION *)interface->pFunctionList)->major,
((CK_VERSION *)interface->pFunctionList)->minor,
interface->pFunctionList,
interface->flags);
}
/* get default standard interface */
rv = C_GetInterface((CK_UTF8CHAR_PTR)"PKCS
11",NULL,&interface,flags);
if (rv == CKR_OK) {
printf("interface %s version %d.%d funcs %p flags
0x%lu\n",
interface->pInterfaceName,
((CK_VERSION *)interface->pFunctionList)->major,
((CK_VERSION *)interface->pFunctionList)->minor,
interface->pFunctionList,
interface->flags);
}
/* get specific standard version interface */
version.major=3;
version.minor=0;
rv = C_GetInterface((CK_UTF8CHAR_PTR)"PKCS
11",&version,&interface,flags);
if (rv == CKR_OK) {
CK_FUNCTION_LIST_3_0_PTR
pkcs11=interface->pFunctionList;
/* ... use the new functions */
pkcs11->C_LoginUser(hSession,userType,pPin,ulPinLen,
pUsername,ulUsernameLen);
}
/* get specific vendor version interface */
version.major=1;
version.minor=0;
rv = C_GetInterface((CK_UTF8CHAR_PTR)
"Vendor
VendorName",&version,&interface,flags);
if (rv == CKR_OK) {
CK_FUNCTION_LIST_VENDOR_1_0_PTR
pkcs11=interface->pFunctionList;
/* ... use vendor specific functions */
pkcs11->C_VendorFunction1(param1,param2,param3);
}
Cryptoki provides the following functions for slot and token
management:
CK_DECLARE_FUNCTION(CK_RV, C_GetSlotList)(
CK_BBOOL tokenPresent,
CK_SLOT_ID_PTR pSlotList,
CK_ULONG_PTR pulCount
);
C_GetSlotList is used to obtain a list of slots in
the system. tokenPresent indicates whether the list obtained includes
only those slots with a token present (CK_TRUE), or all slots (CK_FALSE); pulCount
points to the location that receives the number of slots.
There are two ways for an application to call C_GetSlotList:
1. If pSlotList
is NULL_PTR, then all that C_GetSlotList does is return (in *pulCount)
the number of slots, without actually returning a list of slots. The contents
of the buffer pointed to by pulCount on entry to C_GetSlotList
has no meaning in this case, and the call returns the value CKR_OK.
2. If pSlotList
is not NULL_PTR, then *pulCount MUST contain the size (in terms of CK_SLOT_ID
elements) of the buffer pointed to by pSlotList. If that buffer is
large enough to hold the list of slots, then the list is returned in it, and
CKR_OK is returned. If not, then the call to C_GetSlotList returns the
value CKR_BUFFER_TOO_SMALL. In either case, the value *pulCount is set
to hold the number of slots.
Because C_GetSlotList does not allocate any space of
its own, an application will often call C_GetSlotList twice (or
sometimes even more times—if an application is trying to get a list of all
slots with a token present, then the number of such slots can (unfortunately)
change between when the application asks for how many such slots there are and
when the application asks for the slots themselves). However, multiple calls
to C_GetSlotList are by no means required.
All slots which C_GetSlotList
reports MUST be able to be queried as valid slots by C_GetSlotInfo.
Furthermore, the set of slots accessible through a Cryptoki library is checked
at the time that C_GetSlotList, for list length prediction (NULL
pSlotList argument) is called. If an application calls C_GetSlotList
with a non-NULL pSlotList, and then the user adds or removes a hardware
device, the changed slot list will only be visible and effective if C_GetSlotList
is called again with NULL. Even if C_ GetSlotList is successfully called
this way, it may or may not be the case that the changed slot list will be
successfully recognized depending on the library implementation. On some
platforms, or earlier PKCS11 compliant libraries, it may be necessary to
successfully call C_Initialize or to restart the entire system.
Return values: CKR_ARGUMENTS_BAD, CKR_BUFFER_TOO_SMALL,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR,
CKR_HOST_MEMORY, CKR_OK.
Example:
CK_ULONG ulSlotCount, ulSlotWithTokenCount;
CK_SLOT_ID_PTR pSlotList, pSlotWithTokenList;
CK_RV rv;
/* Get list of all slots */
rv = C_GetSlotList(CK_FALSE, NULL_PTR, &ulSlotCount);
if (rv == CKR_OK) {
pSlotList =
(CK_SLOT_ID_PTR) malloc(ulSlotCount*sizeof(CK_SLOT_ID));
rv = C_GetSlotList(CK_FALSE, pSlotList, &ulSlotCount);
if (rv == CKR_OK) {
/* Now use that list of all slots */
.
.
}
free(pSlotList);
}
/* Get list of all slots with a token present */
pSlotWithTokenList = (CK_SLOT_ID_PTR) malloc(0);
ulSlotWithTokenCount = 0;
while (1) {
rv = C_GetSlotList(
CK_TRUE, pSlotWithTokenList, &ulSlotWithTokenCount);
if (rv != CKR_BUFFER_TOO_SMALL)
break;
pSlotWithTokenList = realloc(
pSlotWithTokenList,
ulSlotWithTokenList*sizeof(CK_SLOT_ID));
}
if (rv == CKR_OK) {
/* Now use that list of all slots with a token present */
.
.
}
free(pSlotWithTokenList);
CK_DECLARE_FUNCTION(CK_RV, C_GetSlotInfo)(
CK_SLOT_ID slotID,
CK_SLOT_INFO_PTR pInfo
);
C_GetSlotInfo obtains information about a particular
slot in the system. slotID is the ID of the slot; pInfo points to
the location that receives the slot information.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_SLOT_ID_INVALID.
Example: see C_GetTokenInfo.
CK_DECLARE_FUNCTION(CK_RV, C_GetTokenInfo)(
CK_SLOT_ID slotID,
CK_TOKEN_INFO_PTR pInfo
);
C_GetTokenInfo obtains information about a particular
token in the system. slotID is the ID of the token’s slot; pInfo
points to the location that receives the token information.
Return values: CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_SLOT_ID_INVALID,
CKR_TOKEN_NOT_PRESENT, CKR_TOKEN_NOT_RECOGNIZED, CKR_ARGUMENTS_BAD.
Example:
CK_ULONG ulCount;
CK_SLOT_ID_PTR pSlotList;
CK_SLOT_INFO slotInfo;
CK_TOKEN_INFO tokenInfo;
CK_RV rv;
rv = C_GetSlotList(CK_FALSE, NULL_PTR, &ulCount);
if ((rv == CKR_OK) && (ulCount > 0)) {
pSlotList = (CK_SLOT_ID_PTR)
malloc(ulCount*sizeof(CK_SLOT_ID));
rv = C_GetSlotList(CK_FALSE, pSlotList, &ulCount);
assert(rv == CKR_OK);
/* Get slot information for first slot */
rv = C_GetSlotInfo(pSlotList[0], &slotInfo);
assert(rv == CKR_OK);
/* Get token information for first slot */
rv = C_GetTokenInfo(pSlotList[0], &tokenInfo);
if (rv == CKR_TOKEN_NOT_PRESENT) {
.
.
}
.
.
free(pSlotList);
}
CK_DECLARE_FUNCTION(CK_RV, C_WaitForSlotEvent)(
CK_FLAGS flags,
CK_SLOT_ID_PTR pSlot,
CK_VOID_PTR pReserved
);
C_WaitForSlotEvent waits for a slot event, such as
token insertion or token removal, to occur. flags determines whether or
not the C_WaitForSlotEvent call blocks (i.e., waits for a slot
event to occur); pSlot points to a location which will receive the ID of
the slot that the event occurred in. pReserved is reserved for future
versions; for this version of Cryptoki, it should be NULL_PTR.
At present, the only flag defined for use in the flags
argument is CKF_DONT_BLOCK:
Internally, each Cryptoki application has a flag for each
slot which is used to track whether or not any unrecognized events involving
that slot have occurred. When an application initially calls C_Initialize,
every slot’s event flag is cleared. Whenever a slot event occurs, the flag
corresponding to the slot in which the event occurred is set.
If C_WaitForSlotEvent is called with the CKF_DONT_BLOCK
flag set in the flags argument, and some slot’s event flag is set, then
that event flag is cleared, and the call returns with the ID of that slot in
the location pointed to by pSlot. If more than one slot’s event flag is
set at the time of the call, one such slot is chosen by the library to have its
event flag cleared and to have its slot ID returned.
If C_WaitForSlotEvent is called with the CKF_DONT_BLOCK
flag set in the flags argument, and no slot’s event flag is set, then
the call returns with the value CKR_NO_EVENT. In this case, the contents of
the location pointed to by pSlot when C_WaitForSlotEvent are
undefined.
If C_WaitForSlotEvent is called with the CKF_DONT_BLOCK
flag clear in the flags argument, then the call behaves as above, except
that it will block. That is, if no slot’s event flag is set at the time of the
call, C_WaitForSlotEvent will wait until some slot’s event flag becomes
set. If a thread of an application has a C_WaitForSlotEvent call
blocking when another thread of that application calls C_Finalize, the C_WaitForSlotEvent
call returns with the value CKR_CRYPTOKI_NOT_INITIALIZED.
Although the parameters supplied to C_Initialize
can in general allow for safe multi-threaded access to a Cryptoki library, C_WaitForSlotEvent
is exceptional in that the behavior of Cryptoki is undefined if multiple
threads of a single application make simultaneous calls to C_WaitForSlotEvent.
Return values: CKR_ARGUMENTS_BAD, CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_NO_EVENT, CKR_OK.
Example:
CK_FLAGS flags = 0;
CK_SLOT_ID slotID;
CK_SLOT_INFO slotInfo;
CK_RV rv;
.
.
/* Block and wait for a slot event */
rv = C_WaitForSlotEvent(flags, &slotID, NULL_PTR);
assert(rv == CKR_OK);
/* See what’s up with that slot */
rv = C_GetSlotInfo(slotID, &slotInfo);
assert(rv == CKR_OK);
CK_DECLARE_FUNCTION(CK_RV, C_GetMechanismList)(
CK_SLOT_ID slotID,
CK_MECHANISM_TYPE_PTR pMechanismList,
CK_ULONG_PTR pulCount
);
C_GetMechanismList is used to obtain a list of
mechanism types supported by a token. SlotID is the ID of the token’s
slot; pulCount points to the location that receives the number of
mechanisms.
There are two ways for an application to call C_GetMechanismList:
1. If pMechanismList
is NULL_PTR, then all that C_GetMechanismList does is return (in *pulCount)
the number of mechanisms, without actually returning a list of mechanisms. The
contents of *pulCount on entry to C_GetMechanismList has no
meaning in this case, and the call returns the value CKR_OK.
2. If pMechanismList
is not NULL_PTR, then *pulCount MUST contain the size (in terms of CK_MECHANISM_TYPE
elements) of the buffer pointed to by pMechanismList. If that buffer is
large enough to hold the list of mechanisms, then the list is returned in it,
and CKR_OK is returned. If not, then the call to C_GetMechanismList
returns the value CKR_BUFFER_TOO_SMALL. In either case, the value *pulCount
is set to hold the number of mechanisms.
Because C_GetMechanismList does not allocate any
space of its own, an application will often call C_GetMechanismList
twice. However, this behavior is by no means required.
Return values: CKR_BUFFER_TOO_SMALL,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_OK, CKR_SLOT_ID_INVALID, CKR_TOKEN_NOT_PRESENT, CKR_TOKEN_NOT_RECOGNIZED,
CKR_ARGUMENTS_BAD.
Example:
CK_SLOT_ID slotID;
CK_ULONG ulCount;
CK_MECHANISM_TYPE_PTR pMechanismList;
CK_RV rv;
.
.
rv = C_GetMechanismList(slotID, NULL_PTR, &ulCount);
if ((rv == CKR_OK) && (ulCount > 0)) {
pMechanismList =
(CK_MECHANISM_TYPE_PTR)
malloc(ulCount*sizeof(CK_MECHANISM_TYPE));
rv = C_GetMechanismList(slotID, pMechanismList,
&ulCount);
if (rv == CKR_OK) {
.
.
}
free(pMechanismList);
}
CK_DECLARE_FUNCTION(CK_RV, C_GetMechanismInfo)(
CK_SLOT_ID slotID,
CK_MECHANISM_TYPE type,
CK_MECHANISM_INFO_PTR pInfo
);
C_GetMechanismInfo obtains information about a
particular mechanism possibly supported by a token. slotID is the ID of
the token’s slot; type is the type of mechanism; pInfo points to
the location that receives the mechanism information.
Return values: CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_MECHANISM_INVALID, CKR_OK,
CKR_SLOT_ID_INVALID, CKR_TOKEN_NOT_PRESENT, CKR_TOKEN_NOT_RECOGNIZED,
CKR_ARGUMENTS_BAD.
Example:
CK_SLOT_ID slotID;
CK_MECHANISM_INFO info;
CK_RV rv;
.
.
/* Get information about the CKM_MD2 mechanism for this
token */
rv = C_GetMechanismInfo(slotID, CKM_MD2, &info);
if (rv == CKR_OK) {
if (info.flags & CKF_DIGEST) {
.
.
}
}
CK_DECLARE_FUNCTION(CK_RV, C_InitToken)(
CK_SLOT_ID slotID,
CK_UTF8CHAR_PTR pPin,
CK_ULONG ulPinLen,
CK_UTF8CHAR_PTR pLabel
);
C_InitToken initializes a token. slotID is the
ID of the token’s slot; pPin points to the SO’s initial PIN (which need not
be null-terminated); ulPinLen is the length in bytes of the PIN; pLabel
points to the 32-byte label of the token (which MUST be padded with blank
characters, and which MUST not be null-terminated). This standard allows
PIN values to contain any valid UTF8 character, but the token may impose subset
restrictions.
If the token has not been initialized (i.e. new from the
factory), then the pPin parameter becomes the initial value of the SO
PIN. If the token is being reinitialized, the pPin parameter is checked
against the existing SO PIN to authorize the initialization operation. In both
cases, the SO PIN is the value pPin after the function completes
successfully. If the SO PIN is lost, then the card MUST be reinitialized using
a mechanism outside the scope of this standard. The CKF_TOKEN_INITIALIZED flag
in the CK_TOKEN_INFO structure indicates the action that will result
from calling C_InitToken. If set, the token will be reinitialized, and
the client MUST supply the existing SO password in pPin.
When a token is initialized, all objects that can be
destroyed are destroyed (i.e., all except for “indestructible” objects
such as keys built into the token). Also, access by the normal user is
disabled until the SO sets the normal user’s PIN. Depending on the token, some
“default” objects may be created, and attributes of some objects may be set to
default values.
If the token has a “protected authentication path”, as
indicated by the CKF_PROTECTED_AUTHENTICATION_PATH flag in its CK_TOKEN_INFO
being set, then that means that there is some way for a user to be
authenticated to the token without having the application send a PIN through
the Cryptoki library. One such possibility is that the user enters a PIN on a
PINpad on the token itself, or on the slot device. To initialize a token with
such a protected authentication path, the pPin parameter to C_InitToken
should be NULL_PTR. During the execution of C_InitToken, the SO’s PIN
will be entered through the protected authentication path.
If the token has a protected authentication path other than
a PINpad, then it is token-dependent whether or not C_InitToken can be
used to initialize the token.
A token cannot be initialized if Cryptoki detects that any
application has an open session with it; when a call to C_InitToken is
made under such circumstances, the call fails with error CKR_SESSION_EXISTS.
Unfortunately, it may happen when C_InitToken is called that some other
application does have an open session with the token, but Cryptoki
cannot detect this, because it cannot detect anything about other applications
using the token. If this is the case, then the consequences of the C_InitToken
call are undefined.
The C_InitToken function may not be sufficient to
properly initialize complex tokens. In these situations, an initialization
mechanism outside the scope of Cryptoki MUST be employed. The definition of
“complex token” is product specific.
Return values: CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_PIN_INCORRECT,
CKR_PIN_LOCKED, CKR_SESSION_EXISTS, CKR_SLOT_ID_INVALID, CKR_TOKEN_NOT_PRESENT,
CKR_TOKEN_NOT_RECOGNIZED, CKR_TOKEN_WRITE_PROTECTED, CKR_ARGUMENTS_BAD.
Example:
CK_SLOT_ID slotID;
CK_UTF8CHAR pin[] = {“MyPIN”};
CK_UTF8CHAR label[32];
CK_RV rv;
.
.
memset(label, ‘ ’, sizeof(label));
memcpy(label, “My first token”, strlen(“My first token”));
rv = C_InitToken(slotID, pin, strlen(pin), label);
if (rv == CKR_OK) {
.
.
}
CK_DECLARE_FUNCTION(CK_RV, C_InitPIN)(
CK_SESSION_HANDLE hSession,
CK_UTF8CHAR_PTR pPin,
CK_ULONG ulPinLen
);
C_InitPIN initializes the normal user’s PIN. hSession
is the session’s handle; pPin points to the normal user’s PIN;
ulPinLen is the length in bytes of the PIN. This standard allows PIN values
to contain any valid UTF8 character, but the token may impose subset
restrictions.
C_InitPIN can only be called in the “R/W SO
Functions” state. An attempt to call it from a session in any other state
fails with error CKR_USER_NOT_LOGGED_IN.
If the token has a “protected authentication path”, as
indicated by the CKF_PROTECTED_AUTHENTICATION_PATH flag in its CK_TOKEN_INFO
being set, then that means that there is some way for a user to be authenticated
to the token without having to send a PIN through the Cryptoki library. One
such possibility is that the user enters a PIN on a PIN pad on the token
itself, or on the slot device. To initialize the normal user’s PIN on a token
with such a protected authentication path, the pPin parameter to C_InitPIN
should be NULL_PTR. During the execution of C_InitPIN, the SO will
enter the new PIN through the protected authentication path.
If the token has a protected authentication path other than
a PIN pad, then it is token-dependent whether or not C_InitPIN can be
used to initialize the normal user’s token access.
Return values: CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK,
CKR_PIN_INVALID, CKR_PIN_LEN_RANGE, CKR_SESSION_CLOSED, CKR_SESSION_READ_ONLY,
CKR_SESSION_HANDLE_INVALID, CKR_TOKEN_WRITE_PROTECTED, CKR_USER_NOT_LOGGED_IN,
CKR_ARGUMENTS_BAD.
Example:
CK_SESSION_HANDLE hSession;
CK_UTF8CHAR newPin[]= {“NewPIN”};
CK_RV rv;
rv = C_InitPIN(hSession, newPin, sizeof(newPin)-1);
if (rv == CKR_OK) {
.
.
}
CK_DECLARE_FUNCTION(CK_RV, C_SetPIN)(
CK_SESSION_HANDLE hSession,
CK_UTF8CHAR_PTR pOldPin,
CK_ULONG ulOldLen,
CK_UTF8CHAR_PTR pNewPin,
CK_ULONG ulNewLen
);
C_SetPIN modifies the PIN of the user that is
currently logged in, or the CKU_USER PIN if the session is not logged in. hSession
is the session’s handle; pOldPin points to the old PIN; ulOldLen
is the length in bytes of the old PIN; pNewPin points to the new PIN; ulNewLen
is the length in bytes of the new PIN. This standard allows PIN values to
contain any valid UTF8 character, but the token may impose subset restrictions.
C_SetPIN can only be called in the “R/W Public
Session” state, “R/W SO Functions” state, or “R/W User Functions” state. An
attempt to call it from a session in any other state fails with error
CKR_SESSION_READ_ONLY.
If the token has a “protected authentication path”, as
indicated by the CKF_PROTECTED_AUTHENTICATION_PATH flag in its CK_TOKEN_INFO
being set, then that means that there is some way for a user to be
authenticated to the token without having to send a PIN through the Cryptoki
library. One such possibility is that the user enters a PIN on a PIN pad on
the token itself, or on the slot device. To modify the current user’s PIN on a
token with such a protected authentication path, the pOldPin and pNewPin
parameters to C_SetPIN should be NULL_PTR. During the execution of C_SetPIN,
the current user will enter the old PIN and the new PIN through the protected
authentication path. It is not specified how the PIN pad should be used to
enter two PINs; this varies.
If the token has a protected authentication path other than
a PIN pad, then it is token-dependent whether or not C_SetPIN can be
used to modify the current user’s PIN.
Return values: CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK,
CKR_PIN_INCORRECT, CKR_PIN_INVALID, CKR_PIN_LEN_RANGE, CKR_PIN_LOCKED,
CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID, CKR_SESSION_READ_ONLY,
CKR_TOKEN_WRITE_PROTECTED, CKR_ARGUMENTS_BAD.
Example:
CK_SESSION_HANDLE hSession;
CK_UTF8CHAR oldPin[] = {“OldPIN”};
CK_UTF8CHAR newPin[] = {“NewPIN”};
CK_RV rv;
rv = C_SetPIN(
hSession, oldPin, sizeof(oldPin)-1, newPin,
sizeof(newPin)-1);
if (rv == CKR_OK) {
.
.
}
A typical application might perform the following series of
steps to make use of a token (note that there are other reasonable sequences of
events that an application might perform):
1. Select a
token.
2. Make one
or more calls to C_OpenSession to obtain one or more sessions with the
token.
3. Call C_Login
to log the user into the token. Since all sessions an application has with a
token have a shared login state, C_Login only needs to be called for one
of the sessions.
4. Perform
cryptographic operations using the sessions with the token.
5. Call C_CloseSession
once for each session that the application has with the token, or call C_CloseAllSessions
to close all the application’s sessions simultaneously.
As has been observed, an application may have concurrent
sessions with more than one token. It is also possible for a token to have
concurrent sessions with more than one application.
Cryptoki provides the following functions for session
management:
CK_DECLARE_FUNCTION(CK_RV, C_OpenSession)(
CK_SLOT_ID slotID,
CK_FLAGS flags,
CK_VOID_PTR pApplication,
CK_NOTIFY Notify,
CK_SESSION_HANDLE_PTR phSession
);
C_OpenSession opens a session between an application
and a token in a particular slot. slotID is the slot’s ID; flags
indicates the type of session; pApplication is an application-defined
pointer to be passed to the notification callback; Notify is the address
of the notification callback function (see Section 5.21);
phSession points to the location that receives the handle for the new
session.
When opening a session with C_OpenSession, the flags
parameter consists of the logical OR of zero or more bit flags defined in the CK_SESSION_INFO
data type. For legacy reasons, the CKF_SERIAL_SESSION bit MUST always
be set; if a call to C_OpenSession does not have this bit set, the call
should return unsuccessfully with the error code
CKR_SESSION_PARALLEL_NOT_SUPPORTED.
There may be a limit on the number of concurrent sessions an
application may have with the token, which may depend on whether the session is
“read-only” or “read/write”. An attempt to open a session which does not
succeed because there are too many existing sessions of some type should return
CKR_SESSION_COUNT.
If the token is write-protected (as indicated in the CK_TOKEN_INFO
structure), then only read-only sessions may be opened with it.
If the application calling C_OpenSession already has
a R/W SO session open with the token, then any attempt to open a R/O session
with the token fails with error code CKR_SESSION_READ_WRITE_SO_EXISTS (see [PKCS11-UG] for further details).
The Notify callback function is used by Cryptoki to
notify the application of certain events. If the application does not wish to
support callbacks, it should pass a value of NULL_PTR as the Notify
parameter. See Section 5.21 for more information about application callbacks.
Return values: CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_SESSION_COUNT, CKR_SESSION_PARALLEL_NOT_SUPPORTED,
CKR_SESSION_READ_WRITE_SO_EXISTS, CKR_SLOT_ID_INVALID, CKR_TOKEN_NOT_PRESENT,
CKR_TOKEN_NOT_RECOGNIZED, CKR_TOKEN_WRITE_PROTECTED, CKR_ARGUMENTS_BAD.
Example: see C_CloseSession.
CK_DECLARE_FUNCTION(CK_RV, C_CloseSession)(
CK_SESSION_HANDLE hSession
);
C_CloseSession closes a session between an
application and a token. hSession is the session’s handle.
When a session is closed, all session objects created by the
session are destroyed automatically, even if the application has other sessions
“using” the objects (see [PKCS11-UG] for
further details).
If this function is successful and it closes the last
session between the application and the token, the login state of the token for
the application returns to public sessions. Any new sessions to the token
opened by the application will be either R/O Public or R/W Public sessions.
Depending on the token, when the last open session any
application has with the token is closed, the token may be “ejected” from its
reader (if this capability exists).
Despite the fact this C_CloseSession is supposed to
close a session, the return value CKR_SESSION_CLOSED is an error
return. It actually indicates the (probably somewhat unlikely) event that
while this function call was executing, another call was made to C_CloseSession
to close this particular session, and that call finished executing first. Such
uses of sessions are a bad idea, and Cryptoki makes little promise of what will
occur in general if an application indulges in this sort of behavior.
Return values: CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID.
Example:
CK_SLOT_ID slotID;
CK_BYTE application;
CK_NOTIFY MyNotify;
CK_SESSION_HANDLE hSession;
CK_RV rv;
.
.
application = 17;
MyNotify = &EncryptionSessionCallback;
rv = C_OpenSession(
slotID, CKF_SERIAL_SESSION | CKF_RW_SESSION,
(CK_VOID_PTR) &application, MyNotify,
&hSession);
if (rv == CKR_OK) {
.
.
C_CloseSession(hSession);
}
CK_DECLARE_FUNCTION(CK_RV, C_CloseAllSessions)(
CK_SLOT_ID slotID
);
C_CloseAllSessions closes all sessions an application
has with a token. slotID specifies the token’s slot.
When a session is closed, all session objects created by the
session are destroyed automatically.
After successful execution of this function, the login state
of the token for the application returns to public sessions. Any new sessions
to the token opened by the application will be either R/O Public or R/W Public
sessions.
Depending on the token, when the last open session any
application has with the token is closed, the token may be “ejected” from its
reader (if this capability exists).
Return values: CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_SLOT_ID_INVALID,
CKR_TOKEN_NOT_PRESENT.
Example:
CK_SLOT_ID slotID;
CK_RV rv;
.
.
rv = C_CloseAllSessions(slotID);
CK_DECLARE_FUNCTION(CK_RV, C_GetSessionInfo)(
CK_SESSION_HANDLE hSession,
CK_SESSION_INFO_PTR pInfo
);
C_GetSessionInfo obtains information about a
session. hSession is the session’s handle; pInfo points to the
location that receives the session information.
Return values: CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_ARGUMENTS_BAD.
Example:
CK_SESSION_HANDLE hSession;
CK_SESSION_INFO info;
CK_RV rv;
.
.
rv = C_GetSessionInfo(hSession, &info);
if (rv == CKR_OK) {
if (info.state == CKS_RW_USER_FUNCTIONS) {
.
.
}
.
.
}
CK_DECLARE_FUNCTION(CK_RV, C_SessionCancel)(
CK_SESSION_HANDLE hSession
CK_FLAGS flags
);
C_SessionCancel terminates active session based
operations. hSession is the session’s handle; flags indicates
the operations to cancel.
To identify which operation(s) should be terminated, the flags
parameter should be assigned the logical bitwise OR of one or more of the bit
flags defined in the CK_MECHANISM_INFO structure.
If no flags are set, the session state will not be modified
and CKR_OK will be returned.
If a flag is set for an operation that has not been
initialized in the session, the operation flag will be ignored and C_SessionCancel
will behave as if the operation flag was not set.
If any of the operations indicated by the flags
parameter cannot be cancelled, CKR_OPERATION_CANCEL_FAILED must be returned.
If multiple operation flags were set and CKR_OPERATION_CANCEL_FAILED is
returned, this function does not provide any information about which
operation(s) could not be cancelled. If an application desires to know if any
single operation could not be cancelled, the application should not call C_SessionCancel
with multiple flags set.
If C_SessionCancel is called from an application
callback (see Section 5.21), no action will be taken by the library and
CKR_FUNCTION_FAILED must be returned.
If C_SessionCancel is used to cancel one half of a
dual-function operation, the remaining operation should still be left in an
active state. However, it is expected that some Cryptoki implementations may
not support this and return CKR_OPERATION_CANCEL_FAILED unless flags for both
operations are provided.
Return values: CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_OPERATION_CANCEL_FAILED,
CKR_TOKEN_NOT_PRESENT.
Example:
CK_SESSION_HANDLE hSession;
CK_RV rv;
rv = C_EncryptInit(hSession, &mechanism, hKey);
if (rv != CKR_OK)
{
.
.
}
rv = C_SessionCancel (hSession, CKF_ENCRYPT);
if (rv != CKR_OK)
{
.
.
}
rv = C_EncryptInit(hSession, &mechanism, hKey);
if (rv != CKR_OK)
{
.
.
}
Below are modifications to existing API
descriptions to allow an alternate method of cancelling individual operations.
The additional text is highlighted.
CK_DECLARE_FUNCTION(CK_RV, C_GetOperationState)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pOperationState,
CK_ULONG_PTR pulOperationStateLen
);
C_GetOperationState obtains a copy of the
cryptographic operations state of a session, encoded as a string of bytes. hSession
is the session’s handle; pOperationState points to the location that
receives the state; pulOperationStateLen points to the location that
receives the length in bytes of the state.
Although the saved state output by C_GetOperationState
is not really produced by a “cryptographic mechanism”, C_GetOperationState
nonetheless uses the convention described in Section 5.2
on producing output.
Precisely what the “cryptographic operations state” this
function saves is varies from token to token; however, this state is what is
provided as input to C_SetOperationState to restore the cryptographic
activities of a session.
Consider a session which is performing a message digest
operation using SHA-1 (i.e., the session is using the CKM_SHA_1
mechanism). Suppose that the message digest operation was initialized
properly, and that precisely 80 bytes of data have been supplied so far as
input to SHA-1. The application now wants to “save the state” of this digest
operation, so that it can continue it later. In this particular case, since
SHA-1 processes 512 bits (64 bytes) of input at a time, the cryptographic
operations state of the session most likely consists of three distinct parts:
the state of SHA-1’s 160-bit internal chaining variable; the 16 bytes of
unprocessed input data; and some administrative data indicating that this saved
state comes from a session which was performing SHA-1 hashing. Taken together,
these three pieces of information suffice to continue the current hashing
operation at a later time.
Consider next a session which is performing an encryption
operation with DES (a block cipher with a block size of 64 bits) in CBC
(cipher-block chaining) mode (i.e., the session is using the CKM_DES_CBC
mechanism). Suppose that precisely 22 bytes of data (in addition to an IV for
the CBC mode) have been supplied so far as input to DES, which means that the
first two 8-byte blocks of ciphertext have already been produced and output.
In this case, the cryptographic operations state of the session most likely
consists of three or four distinct parts: the second 8-byte block of ciphertext
(this will be used for cipher-block chaining to produce the next block of
ciphertext); the 6 bytes of data still awaiting encryption; some administrative
data indicating that this saved state comes from a session which was performing
DES encryption in CBC mode; and possibly the DES key being used for encryption
(see C_SetOperationState for more information on whether or not the key
is present in the saved state).
If a session is performing two cryptographic operations
simultaneously (see Section 5.14), then the cryptographic operations state of
the session will contain all the necessary information to restore both
operations.
An attempt to save the cryptographic operations state of a
session which does not currently have some active savable cryptographic
operation(s) (encryption, decryption, digesting, signing without message
recovery, verification without message recovery, or some legal combination of
two of these) should fail with the error CKR_OPERATION_NOT_INITIALIZED.
An attempt to save the cryptographic operations state of a
session which is performing an appropriate cryptographic operation (or two),
but which cannot be satisfied for any of various reasons (certain necessary
state information and/or key information can’t leave the token, for example)
should fail with the error CKR_STATE_UNSAVEABLE.
Return values: CKR_BUFFER_TOO_SMALL,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_OK, CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_STATE_UNSAVEABLE, CKR_ARGUMENTS_BAD.
Example: see C_SetOperationState.
CK_DECLARE_FUNCTION(CK_RV, C_SetOperationState)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pOperationState,
CK_ULONG ulOperationStateLen,
CK_OBJECT_HANDLE hEncryptionKey,
CK_OBJECT_HANDLE hAuthenticationKey
);
C_SetOperationState
restores the cryptographic operations state of a session from a string of bytes
obtained with C_GetOperationState. hSession is the session’s
handle; pOperationState points to the location holding the saved state; ulOperationStateLen
holds the length of the saved state; hEncryptionKey holds a handle to
the key which will be used for an ongoing encryption or decryption operation in
the restored session (or 0 if no encryption or decryption key is needed, either
because no such operation is ongoing in the stored session or because all the
necessary key information is present in the saved state); hAuthenticationKey
holds a handle to the key which will be used for an ongoing signature, MACing,
or verification operation in the restored session (or 0 if no such key is
needed, either because no such operation is ongoing in the stored session or
because all the necessary key information is present in the saved state).
The state need not have been obtained from the same session
(the “source session”) as it is being restored to (the “destination session”).
However, the source session and destination session should have a common
session state (e.g., CKS_RW_USER_FUNCTIONS), and should be with a common
token. There is also no guarantee that cryptographic operations state may be
carried across logins, or across different Cryptoki implementations.
If C_SetOperationState is supplied with alleged saved
cryptographic operations state which it can determine is not valid saved state
(or is cryptographic operations state from a session with a different session
state, or is cryptographic operations state from a different token), it fails
with the error CKR_SAVED_STATE_INVALID.
Saved state obtained from calls to C_GetOperationState
may or may not contain information about keys in use for ongoing cryptographic
operations. If a saved cryptographic operations state has an ongoing
encryption or decryption operation, and the key in use for the operation is not
saved in the state, then it MUST be supplied to C_SetOperationState in
the hEncryptionKey argument. If it is not, then C_SetOperationState
will fail and return the error CKR_KEY_NEEDED. If the key in use for the
operation is saved in the state, then it can be supplied in the hEncryptionKey
argument, but this is not required.
Similarly, if a saved cryptographic operations state has an
ongoing signature, MACing, or verification operation, and the key in use for
the operation is not saved in the state, then it MUST be supplied to C_SetOperationState
in the hAuthenticationKey argument. If it is not, then C_SetOperationState
will fail with the error CKR_KEY_NEEDED. If the key in use for the operation is
saved in the state, then it can be supplied in the hAuthenticationKey
argument, but this is not required.
If an irrelevant key is supplied to C_SetOperationState
call (e.g., a nonzero key handle is submitted in the hEncryptionKey
argument, but the saved cryptographic operations state supplied does not have
an ongoing encryption or decryption operation, then C_SetOperationState
fails with the error CKR_KEY_NOT_NEEDED.
If a key is supplied as an argument to C_SetOperationState,
and C_SetOperationState can somehow detect that this key was not the key
being used in the source session for the supplied cryptographic operations
state (it may be able to detect this if the key or a hash of the key is present
in the saved state, for example), then C_SetOperationState fails with
the error CKR_KEY_CHANGED.
An application can look at the CKF_RESTORE_KEY_NOT_NEEDED
flag in the flags field of the CK_TOKEN_INFO field for a token to
determine whether or not it needs to supply key handles to C_SetOperationState
calls. If this flag is true, then a call to C_SetOperationState never
needs a key handle to be supplied to it. If this flag is false, then at least
some of the time, C_SetOperationState requires a key handle, and so the
application should probably always pass in any relevant key handles when
restoring cryptographic operations state to a session.
C_SetOperationState can successfully restore
cryptographic operations state to a session even if that session has active
cryptographic or object search operations when C_SetOperationState is
called (the ongoing operations are abruptly cancelled).
Return values: CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_KEY_CHANGED, CKR_KEY_NEEDED,
CKR_KEY_NOT_NEEDED, CKR_OK, CKR_SAVED_STATE_INVALID, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_ARGUMENTS_BAD.
Example:
CK_SESSION_HANDLE hSession;
CK_MECHANISM digestMechanism;
CK_BYTE_PTR pState;
CK_ULONG ulStateLen;
CK_BYTE data1[] = {0x01, 0x03, 0x05, 0x07};
CK_BYTE data2[] = {0x02, 0x04, 0x08};
CK_BYTE data3[] = {0x10, 0x0F, 0x0E, 0x0D, 0x0C};
CK_BYTE pDigest[20];
CK_ULONG ulDigestLen;
CK_RV rv;
.
.
/* Initialize hash operation */
rv = C_DigestInit(hSession,
&digestMechanism);
assert(rv == CKR_OK);
/* Start hashing */
rv = C_DigestUpdate(hSession, data1, sizeof(data1));
assert(rv == CKR_OK);
/* Find out how big the state might be */
rv = C_GetOperationState(hSession, NULL_PTR,
&ulStateLen);
assert(rv == CKR_OK);
/* Allocate some memory and then get the state */
pState = (CK_BYTE_PTR) malloc(ulStateLen);
rv = C_GetOperationState(hSession, pState, &ulStateLen);
/* Continue hashing */
rv = C_DigestUpdate(hSession, data2, sizeof(data2));
assert(rv == CKR_OK);
/* Restore state. No key handles needed */
rv = C_SetOperationState(hSession, pState, ulStateLen, 0,
0);
assert(rv == CKR_OK);
/* Continue hashing from where we saved state */
rv = C_DigestUpdate(hSession, data3, sizeof(data3));
assert(rv == CKR_OK);
/* Conclude hashing operation */
ulDigestLen = sizeof(pDigest);
rv = C_DigestFinal(hSession, pDigest,
&ulDigestLen);
if (rv == CKR_OK) {
/* pDigest[] now contains the hash of 0x01030507100F0E0D0C
*/
.
.
}
CK_DECLARE_FUNCTION(CK_RV, C_Login)(
CK_SESSION_HANDLE hSession,
CK_USER_TYPE userType,
CK_UTF8CHAR_PTR pPin,
CK_ULONG ulPinLen
);
C_Login logs a user into a token. hSession is
a session handle; userType is the user type; pPin points to the
user’s PIN; ulPinLen is the length of the PIN. This standard allows PIN
values to contain any valid UTF8 character, but the token may impose subset
restrictions.
When the user type is either CKU_SO or CKU_USER, if the call
succeeds, each of the application's sessions will enter either the "R/W SO
Functions" state, the "R/W User Functions" state, or the
"R/O User Functions" state. If the user type is CKU_CONTEXT_SPECIFIC,
the behavior of C_Login depends on the context in which it is called. Improper
use of this user type will result in a return value
CKR_OPERATION_NOT_INITIALIZED..
If the token has a “protected authentication path”, as
indicated by the CKF_PROTECTED_AUTHENTICATION_PATH flag in its CK_TOKEN_INFO
being set, then that means that there is some way for a user to be
authenticated to the token without having to send a PIN through the Cryptoki
library. One such possibility is that the user enters a PIN on a PIN pad on
the token itself, or on the slot device. Or the user might not even use a
PIN—authentication could be achieved by some fingerprint-reading device, for
example. To log into a token with a protected authentication path, the pPin
parameter to C_Login should be NULL_PTR. When C_Login returns,
whatever authentication method supported by the token will have been performed;
a return value of CKR_OK means that the user was successfully authenticated,
and a return value of CKR_PIN_INCORRECT means that the user was denied access.
If there are any active cryptographic or object finding
operations in an application’s session, and then C_Login is successfully
executed by that application, it may or may not be the case that those
operations are still active. Therefore, before logging in, any active
operations should be finished.
If the application calling C_Login has a R/O session
open with the token, then it will be unable to log the SO into a session (see [PKCS11-UG] for further details). An attempt
to do this will result in the error code CKR_SESSION_READ_ONLY_EXISTS.
C_Login may be called repeatedly, without intervening C_Logout
calls, if (and only if) a key with the CKA_ALWAYS_AUTHENTICATE attribute set to
CK_TRUE exists, and the user needs to do cryptographic operation on this key.
See further Section 4.9.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED,
CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_OK, CKR_OPERATION_NOT_INITIALIZED, CKR_PIN_INCORRECT, CKR_PIN_LOCKED,
CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID, CKR_SESSION_READ_ONLY_EXISTS,
CKR_USER_ALREADY_LOGGED_IN, CKR_USER_ANOTHER_ALREADY_LOGGED_IN,
CKR_USER_PIN_NOT_INITIALIZED, CKR_USER_TOO_MANY_TYPES, CKR_USER_TYPE_INVALID.
Example: see C_Logout.
CK_DECLARE_FUNCTION(CK_RV, C_LoginUser)(
CK_SESSION_HANDLE hSession,
CK_USER_TYPE userType,
CK_UTF8CHAR_PTR pPin,
CK_ULONG ulPinLen,
CK_UTF8CHAR_PTR pUsername,
CK_ULONG ulUsernameLen
);
C_LoginUser logs a user into a token. hSession
is a session handle; userType is the user type; pPin points to
the user’s PIN; ulPinLen is the length of the PIN, pUsername
points to the user name, ulUsernameLen is the length of the user name.
This standard allows PIN and user name values to contain any valid UTF8
character, but the token may impose subset restrictions.
When the user type is either CKU_SO or CKU_USER, if the call
succeeds, each of the application's sessions will enter either the "R/W SO
Functions" state, the "R/W User Functions" state, or the
"R/O User Functions" state. If the user type is
CKU_CONTEXT_SPECIFIC, the behavior of C_LoginUser depends on the context
in which it is called. Improper use of this user type will result in a return
value CKR_OPERATION_NOT_INITIALIZED.
If the token has a “protected authentication path”, as
indicated by the CKF_PROTECTED_AUTHENTICATION_PATH flag in its CK_TOKEN_INFO
being set, then that means that there is some way for a user to be
authenticated to the token without having to send a PIN through the Cryptoki
library. One such possibility is that the user enters a PIN on a PIN pad on
the token itself, or on the slot device. The user might not even use a
PIN—authentication could be achieved by some fingerprint-reading device, for
example. To log into a token with a protected authentication path, the pPin
parameter to C_LoginUser should be NULL_PTR. When C_LoginUser
returns, whatever authentication method supported by the token will have been
performed; a return value of CKR_OK means that the user was successfully
authenticated, and a return value of CKR_PIN_INCORRECT means that the user was
denied access.
If there are any active cryptographic or object finding
operations in an application’s session, and then C_LoginUser is
successfully executed by that application, it may or may not be the case that
those operations are still active. Therefore, before logging in, any active
operations should be finished.
If the application calling C_LoginUser has a R/O
session open with the token, then it will be unable to log the SO into a
session (see [PKCS11-UG] for further details). An attempt to do this will
result in the error code CKR_SESSION_READ_ONLY_EXISTS.
C_LoginUser may be called repeatedly, without
intervening C_Logout calls, if (and only if) a key with the
CKA_ALWAYS_AUTHENTICATE attribute set to CK_TRUE exists, and the user needs to
do cryptographic operation on this key. See further Section 4.9.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_OPERATION_NOT_INITIALIZED,
CKR_PIN_INCORRECT, CKR_PIN_LOCKED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_SESSION_READ_ONLY_EXISTS,
CKR_USER_ALREADY_LOGGED_IN, CKR_USER_ANOTHER_ALREADY_LOGGED_IN,
CKR_USER_PIN_NOT_INITIALIZED, CKR_USER_TOO_MANY_TYPES, CKR_USER_TYPE_INVALID.
Example:
CK_SESSION_HANDLE hSession;
CK_UTF8CHAR userPin[] = {“MyPIN”};
CK_UTF8CHAR userName[] = {“MyUserName”};
CK_RV rv;
rv = C_LoginUser(hSession, CKU_USER, userPin,
sizeof(userPin)-1, userName,
sizeof(userName)-1);
if (rv == CKR_OK) {
.
.
rv = C_Logout(hSession);
if (rv == CKR_OK) {
.
.
}
}
CK_DECLARE_FUNCTION(CK_RV, C_Logout)(
CK_SESSION_HANDLE hSession
);
C_Logout logs a user out from a token. hSession
is the session’s handle.
Depending on the current user type, if the call succeeds,
each of the application’s sessions will enter either the “R/W Public Session”
state or the “R/O Public Session” state.
When C_Logout successfully executes, any of the
application’s handles to private objects become invalid (even if a user is
later logged back into the token, those handles remain invalid). In addition,
all private session objects from sessions belonging to the application are
destroyed.
If there are any active cryptographic or object-finding
operations in an application’s session, and then C_Logout is
successfully executed by that application, it may or may not be the case that
those operations are still active. Therefore, before logging out, any active
operations should be finished.
Return values: CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN.
Example:
CK_SESSION_HANDLE hSession;
CK_UTF8CHAR userPin[] = {“MyPIN”};
CK_RV rv;
rv = C_Login(hSession, CKU_USER, userPin,
sizeof(userPin)-1);
if (rv == CKR_OK) {
.
.
rv = C_Logout(hSession);
if (rv == CKR_OK) {
.
.
}
}
Cryptoki provides the following functions for managing
objects. Additional functions provided specifically for managing key objects
are described in Section 5.18.
CK_DECLARE_FUNCTION(CK_RV, C_CreateObject)(
CK_SESSION_HANDLE hSession,
CK_ATTRIBUTE_PTR pTemplate,
CK_ULONG ulCount,
CK_OBJECT_HANDLE_PTR phObject
);
C_CreateObject creates a new object. hSession
is the session’s handle; pTemplate points to the object’s template; ulCount
is the number of attributes in the template; phObject points to the
location that receives the new object’s handle.
If a call to C_CreateObject cannot support the
precise template supplied to it, it will fail and return without creating any
object.
If C_CreateObject is used to create a key object, the
key object will have its CKA_LOCAL attribute set to CK_FALSE. If that
key object is a secret or private key then the new key will have the CKA_ALWAYS_SENSITIVE
attribute set to CK_FALSE, and the CKA_NEVER_EXTRACTABLE attribute set
to CK_FALSE.
Only session objects can be created during a read-only
session. Only public objects can be created unless the normal user is logged
in.
Whenever an object is created, a
value for CKA_UNIQUE_ID is generated and assigned to the new object (See
Section 4.4.1).
Return values: CKR_ARGUMENTS_BAD, CKR_ATTRIBUTE_READ_ONLY,
CKR_ATTRIBUTE_TYPE_INVALID, CKR_ATTRIBUTE_VALUE_INVALID,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_CURVE_NOT_SUPPORTED, CKR_DEVICE_ERROR,
CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_DOMAIN_PARAMS_INVALID,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK,
CKR_PIN_EXPIRED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID,
CKR_SESSION_READ_ONLY, CKR_TEMPLATE_INCOMPLETE, CKR_TEMPLATE_INCONSISTENT,
CKR_TOKEN_WRITE_PROTECTED, CKR_USER_NOT_LOGGED_IN.
Example:
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE
hData,
hCertificate,
hKey;
CK_OBJECT_CLASS
dataClass = CKO_DATA,
certificateClass = CKO_CERTIFICATE,
keyClass = CKO_PUBLIC_KEY;
CK_KEY_TYPE keyType = CKK_RSA;
CK_UTF8CHAR application[] = {“My Application”};
CK_BYTE dataValue[] = {...};
CK_BYTE subject[] = {...};
CK_BYTE id[] = {...};
CK_BYTE certificateValue[] = {...};
CK_BYTE modulus[] = {...};
CK_BYTE exponent[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE dataTemplate[] = {
{CKA_CLASS, &dataClass, sizeof(dataClass)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_APPLICATION, application, sizeof(application)-1},
{CKA_VALUE, dataValue, sizeof(dataValue)}
};
CK_ATTRIBUTE certificateTemplate[] = {
{CKA_CLASS, &certificateClass,
sizeof(certificateClass)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_SUBJECT, subject, sizeof(subject)},
{CKA_ID, id, sizeof(id)},
{CKA_VALUE, certificateValue, sizeof(certificateValue)}
};
CK_ATTRIBUTE keyTemplate[] = {
{CKA_CLASS, &keyClass, sizeof(keyClass)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_WRAP, &true, sizeof(true)},
{CKA_MODULUS, modulus, sizeof(modulus)},
{CKA_PUBLIC_EXPONENT, exponent, sizeof(exponent)}
};
CK_RV rv;
.
.
/* Create a data object */
rv = C_CreateObject(hSession, dataTemplate, 4, &hData);
if (rv == CKR_OK) {
.
.
}
/* Create a certificate object */
rv = C_CreateObject(
hSession, certificateTemplate, 5, &hCertificate);
if (rv == CKR_OK) {
.
.
}
/* Create an RSA public key object */
rv = C_CreateObject(hSession, keyTemplate, 5, &hKey);
if (rv == CKR_OK) {
.
.
}
CK_DECLARE_FUNCTION(CK_RV, C_CopyObject)(
CK_SESSION_HANDLE hSession,
CK_OBJECT_HANDLE hObject,
CK_ATTRIBUTE_PTR pTemplate,
CK_ULONG ulCount,
CK_OBJECT_HANDLE_PTR phNewObject
);
C_CopyObject copies an object, creating a new object
for the copy. hSession is the session’s handle; hObject is the
object’s handle; pTemplate points to the template for the new object; ulCount
is the number of attributes in the template; phNewObject points to the
location that receives the handle for the copy of the object.
The template may specify new values for any attributes of
the object that can ordinarily be modified (e.g., in the course of
copying a secret key, a key’s CKA_EXTRACTABLE attribute may be changed
from CK_TRUE to CK_FALSE, but not the other way around. If this change is
made, the new key’s CKA_NEVER_EXTRACTABLE attribute will have the value
CK_FALSE. Similarly, the template may specify that the new key’s CKA_SENSITIVE
attribute be CK_TRUE; the new key will have the same value for its CKA_ALWAYS_SENSITIVE
attribute as the original key). It may also specify new values of the CKA_TOKEN
and CKA_PRIVATE attributes (e.g., to copy a session object to a
token object). If the template specifies a value of an attribute which is
incompatible with other existing attributes of the object, the call fails with
the return code CKR_TEMPLATE_INCONSISTENT.
If a call to C_CopyObject cannot support the precise
template supplied to it, it will fail and return without creating any object.
If the object indicated by hObject has its CKA_COPYABLE attribute set to
CK_FALSE, C_CopyObject will return CKR_ACTION_PROHIBITED.
Whenever an object is copied, a
new value for CKA_UNIQUE_ID is generated and assigned to the new object (See
Section 4.4.1).
Only session objects can be created during a read-only
session. Only public objects can be created unless the normal user is logged
in.
Return values: , CKR_ACTION_PROHIBITED, CKR_ARGUMENTS_BAD,
CKR_ATTRIBUTE_READ_ONLY, CKR_ATTRIBUTE_TYPE_INVALID, CKR_ATTRIBUTE_VALUE_INVALID,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_OBJECT_HANDLE_INVALID, CKR_OK, CKR_PIN_EXPIRED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_SESSION_READ_ONLY, CKR_TEMPLATE_INCONSISTENT,
CKR_TOKEN_WRITE_PROTECTED, CKR_USER_NOT_LOGGED_IN.
Example:
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hKey, hNewKey;
CK_OBJECT_CLASS keyClass = CKO_SECRET_KEY;
CK_KEY_TYPE keyType = CKK_DES;
CK_BYTE id[] = {...};
CK_BYTE keyValue[] = {...};
CK_BBOOL false = CK_FALSE;
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE keyTemplate[] = {
{CKA_CLASS, &keyClass, sizeof(keyClass)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &false, sizeof(false)},
{CKA_ID, id, sizeof(id)},
{CKA_VALUE, keyValue, sizeof(keyValue)}
};
CK_ATTRIBUTE copyTemplate[] = {
{CKA_TOKEN, &true, sizeof(true)}
};
CK_RV rv;
.
.
/* Create a DES secret key session object */
rv = C_CreateObject(hSession, keyTemplate, 5, &hKey);
if (rv == CKR_OK) {
/* Create a copy which is a token object */
rv = C_CopyObject(hSession, hKey, copyTemplate, 1,
&hNewKey);
.
.
}
CK_DECLARE_FUNCTION(CK_RV, C_DestroyObject)(
CK_SESSION_HANDLE hSession,
CK_OBJECT_HANDLE hObject
);
C_DestroyObject destroys an object. hSession
is the session’s handle; and hObject is the object’s handle.
Only session objects can be destroyed during a read-only
session. Only public objects can be destroyed unless the normal user is logged
in.
Certain objects may not be destroyed. Calling
C_DestroyObject on such objects will result in the CKR_ACTION_PROHIBITED error
code. An application can consult the object's CKA_DESTROYABLE attribute to determine
if an object may be destroyed or not.
Return values: CKR_ACTION_PROHIBITED,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_OBJECT_HANDLE_INVALID, CKR_OK, CKR_PIN_EXPIRED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_SESSION_READ_ONLY, CKR_TOKEN_WRITE_PROTECTED.
Example: see C_GetObjectSize.
CK_DECLARE_FUNCTION(CK_RV, C_GetObjectSize)(
CK_SESSION_HANDLE hSession,
CK_OBJECT_HANDLE hObject,
CK_ULONG_PTR pulSize
);
C_GetObjectSize gets the size of an object in bytes.
hSession is the session’s handle; hObject is the object’s handle;
pulSize points to the location that receives the size in bytes of the
object.
Cryptoki does not specify what the precise meaning of an
object’s size is. Intuitively, it is some measure of how much token memory the
object takes up. If an application deletes (say) a private object of size S,
it might be reasonable to assume that the ulFreePrivateMemory field of
the token’s CK_TOKEN_INFO structure increases by approximately S.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_INFORMATION_SENSITIVE, CKR_OBJECT_HANDLE_INVALID, CKR_OK,
CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID.
Example:
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hObject;
CK_OBJECT_CLASS dataClass = CKO_DATA;
CK_UTF8CHAR application[] = {“My Application”};
CK_BYTE value[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &dataClass, sizeof(dataClass)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_APPLICATION, application, sizeof(application)-1},
{CKA_VALUE, value, sizeof(value)}
};
CK_ULONG ulSize;
CK_RV rv;
.
.
rv = C_CreateObject(hSession, template, 4, &hObject);
if (rv == CKR_OK) {
rv = C_GetObjectSize(hSession, hObject, &ulSize);
if (rv != CKR_INFORMATION_SENSITIVE) {
.
.
}
rv = C_DestroyObject(hSession, hObject);
.
.
}
CK_DECLARE_FUNCTION(CK_RV, C_GetAttributeValue)(
CK_SESSION_HANDLE hSession,
CK_OBJECT_HANDLE hObject,
CK_ATTRIBUTE_PTR pTemplate,
CK_ULONG ulCount
);
C_GetAttributeValue obtains the value of one or more
attributes of an object. hSession is the session’s handle; hObject
is the object’s handle; pTemplate points to a template that specifies
which attribute values are to be obtained, and receives the attribute values; ulCount
is the number of attributes in the template.
For each (type, pValue, ulValueLen)
triple in the template, C_GetAttributeValue performs the following
algorithm:
1. If the
specified attribute (i.e., the attribute specified by the type field) for the
object cannot be revealed because the object is sensitive or unextractable,
then the ulValueLen field in that triple is modified to hold the value
CK_UNAVAILABLE_INFORMATION.
2. Otherwise,
if the specified value for the object is invalid (the object does not possess
such an attribute), then the ulValueLen field in that triple is modified to
hold the value CK_UNAVAILABLE_INFORMATION.
3. Otherwise,
if the pValue field has the value NULL_PTR, then the ulValueLen
field is modified to hold the exact length of the specified attribute for the
object.
4. Otherwise,
if the length specified in ulValueLen is large enough to hold the value
of the specified attribute for the object, then that attribute is copied into
the buffer located at pValue, and the ulValueLen field is
modified to hold the exact length of the attribute.
5. Otherwise,
the ulValueLen field is modified to hold the value CK_UNAVAILABLE_INFORMATION.
If case 1 applies to any of the requested attributes, then
the call should return the value CKR_ATTRIBUTE_SENSITIVE. If case 2 applies to
any of the requested attributes, then the call should return the value
CKR_ATTRIBUTE_TYPE_INVALID. If case 5 applies to any of the requested
attributes, then the call should return the value CKR_BUFFER_TOO_SMALL. As
usual, if more than one of these error codes is applicable, Cryptoki may return
any of them. Only if none of them applies to any of the requested attributes
will CKR_OK be returned.
In the special case of an attribute whose value is an array
of attributes, for example CKA_WRAP_TEMPLATE, where it is passed in with pValue
not NULL, the length specified in ulValueLen MUST be large enough to hold all
attributes in the array. If the pValue of elements within the array is
NULL_PTR then the ulValueLen of elements within the array will be set to the
required length. If the pValue of elements within the array is not NULL_PTR,
then the ulValueLen element of attributes within the array MUST reflect the
space that the corresponding pValue points to, and pValue is filled in if there
is sufficient room. Therefore it is important to initialize the contents of a
buffer before calling C_GetAttributeValue to get such an array value. Note that
the type element of attributes within the array MUST be ignored on input and
MUST be set on output. If any ulValueLen within the array isn't large enough,
it will be set to CK_UNAVAILABLE_INFORMATION and the function will return
CKR_BUFFER_TOO_SMALL, as it does if an attribute in the pTemplate argument has
ulValueLen too small. Note that any attribute whose value is an array of
attributes is identifiable by virtue of the attribute type having the
CKF_ARRAY_ATTRIBUTE bit set.
Note that the error codes CKR_ATTRIBUTE_SENSITIVE,
CKR_ATTRIBUTE_TYPE_INVALID, and CKR_BUFFER_TOO_SMALL do not denote true errors
for C_GetAttributeValue. If a call to C_GetAttributeValue
returns any of these three values, then the call MUST nonetheless have
processed every attribute in the template supplied to C_GetAttributeValue.
Each attribute in the template whose value can be returned by the call
to C_GetAttributeValue will be returned by the call to C_GetAttributeValue.
Return values: CKR_ARGUMENTS_BAD, CKR_ATTRIBUTE_SENSITIVE,
CKR_ATTRIBUTE_TYPE_INVALID, CKR_BUFFER_TOO_SMALL, CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OBJECT_HANDLE_INVALID, CKR_OK,
CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID.
Example:
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hObject;
CK_BYTE_PTR pModulus, pExponent;
CK_ATTRIBUTE template[] = {
{CKA_MODULUS, NULL_PTR, 0},
{CKA_PUBLIC_EXPONENT, NULL_PTR, 0}
};
CK_RV rv;
.
.
rv = C_GetAttributeValue(hSession, hObject, template, 2);
if (rv == CKR_OK) {
pModulus = (CK_BYTE_PTR) malloc(template[0].ulValueLen);
template[0].pValue = pModulus;
/* template[0].ulValueLen was set by C_GetAttributeValue
*/
pExponent = (CK_BYTE_PTR) malloc(template[1].ulValueLen);
template[1].pValue = pExponent;
/* template[1].ulValueLen was set by C_GetAttributeValue
*/
rv = C_GetAttributeValue(hSession, hObject, template, 2);
if (rv == CKR_OK) {
.
.
}
free(pModulus);
free(pExponent);
}
CK_DECLARE_FUNCTION(CK_RV, C_SetAttributeValue)(
CK_SESSION_HANDLE hSession,
CK_OBJECT_HANDLE hObject,
CK_ATTRIBUTE_PTR pTemplate,
CK_ULONG ulCount
);
C_SetAttributeValue modifies the value of one or more
attributes of an object. hSession is the session’s handle; hObject
is the object’s handle; pTemplate points to a template that specifies
which attribute values are to be modified and their new values; ulCount
is the number of attributes in the template.
Certain objects may not be modified. Calling
C_SetAttributeValue on such objects will result in the CKR_ACTION_PROHIBITED
error code. An application can consult the object's CKA_MODIFIABLE attribute to
determine if an object may be modified or not.
Only session objects can be modified during a read-only
session.
The template may specify new values for any attributes of
the object that can be modified. If the template specifies a value of an
attribute which is incompatible with other existing attributes of the object,
the call fails with the return code CKR_TEMPLATE_INCONSISTENT.
Not all attributes can be modified; see Section 4.1.2 for more details.
Return values: CKR_ACTION_PROHIBITED, CKR_ARGUMENTS_BAD,
CKR_ATTRIBUTE_READ_ONLY, CKR_ATTRIBUTE_TYPE_INVALID,
CKR_ATTRIBUTE_VALUE_INVALID, CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR,
CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR,
CKR_HOST_MEMORY, CKR_OBJECT_HANDLE_INVALID, CKR_OK, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_SESSION_READ_ONLY, CKR_TEMPLATE_INCONSISTENT,
CKR_TOKEN_WRITE_PROTECTED, CKR_USER_NOT_LOGGED_IN.
Example:
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hObject;
CK_UTF8CHAR label[] = {“New label”};
CK_ATTRIBUTE template[] = {
{CKA_LABEL, label, sizeof(label)-1}
};
CK_RV rv;
.
.
rv = C_SetAttributeValue(hSession, hObject, template, 1);
if (rv == CKR_OK) {
.
.
}
CK_DECLARE_FUNCTION(CK_RV, C_FindObjectsInit)(
CK_SESSION_HANDLE hSession,
CK_ATTRIBUTE_PTR pTemplate,
CK_ULONG ulCount
);
C_FindObjectsInit initializes a search for token and
session objects that match a template. hSession is the session’s handle;
pTemplate points to a search template that specifies the attribute
values to match; ulCount is the number of attributes in the search
template. The matching criterion is an exact byte-for-byte match with all
attributes in the template. To find all objects, set ulCount to 0.
After calling C_FindObjectsInit, the application may
call C_FindObjects one or more times to obtain handles for objects
matching the template, and then eventually call C_FindObjectsFinal to
finish the active search operation. At most one search operation may be active
at a given time in a given session.
The object search operation will only find objects that the
session can view. For example, an object search in an “R/W Public Session”
will not find any private objects (even if one of the attributes in the search
template specifies that the search is for private objects).
If a search operation is active, and objects are created or
destroyed which fit the search template for the active search operation, then
those objects may or may not be found by the search operation. Note that this
means that, under these circumstances, the search operation may return invalid
object handles.
Even though C_FindObjectsInit can return the values
CKR_ATTRIBUTE_TYPE_INVALID and CKR_ATTRIBUTE_VALUE_INVALID, it is not required
to. For example, if it is given a search template with nonexistent attributes
in it, it can return CKR_ATTRIBUTE_TYPE_INVALID, or it can initialize a search
operation which will match no objects and return CKR_OK.
If the CKA_UNIQUE_ID attribute is
present in the search template, either zero or one objects will be found, since
at most one object can have any particular CKA_UNIQUE_ID value.
Return values: CKR_ARGUMENTS_BAD,
CKR_ATTRIBUTE_TYPE_INVALID, CKR_ATTRIBUTE_VALUE_INVALID,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_OK, CKR_OPERATION_ACTIVE, CKR_PIN_EXPIRED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID.
Example: see C_FindObjectsFinal.
CK_DECLARE_FUNCTION(CK_RV, C_FindObjects)(
CK_SESSION_HANDLE hSession,
CK_OBJECT_HANDLE_PTR phObject,
CK_ULONG ulMaxObjectCount,
CK_ULONG_PTR pulObjectCount
);
C_FindObjects continues a search for token and session
objects that match a template, obtaining additional object handles. hSession
is the session’s handle; phObject points to the location that receives
the list (array) of additional object handles; ulMaxObjectCount is the
maximum number of object handles to be returned; pulObjectCount points
to the location that receives the actual number of object handles returned.
If there are no more objects matching the template, then the
location that pulObjectCount points to receives the value 0.
The search MUST have been initialized with C_FindObjectsInit.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_OK, CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID.
Example: see C_FindObjectsFinal.
CK_DECLARE_FUNCTION(CK_RV, C_FindObjectsFinal)(
CK_SESSION_HANDLE hSession
);
C_FindObjectsFinal terminates a search for token and
session objects. hSession is the session’s handle.
Return values: CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_OPERATION_NOT_INITIALIZED,
CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID.
Example:
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hObject;
CK_ULONG ulObjectCount;
CK_RV rv;
.
.
rv = C_FindObjectsInit(hSession, NULL_PTR, 0);
assert(rv == CKR_OK);
while (1) {
rv = C_FindObjects(hSession, &hObject, 1,
&ulObjectCount);
if (rv != CKR_OK || ulObjectCount == 0)
break;
.
.
}
rv = C_FindObjectsFinal(hSession);
assert(rv == CKR_OK);
Cryptoki provides the following functions for encrypting
data:
CK_DECLARE_FUNCTION(CK_RV, C_EncryptInit)(
CK_SESSION_HANDLE hSession,
CK_MECHANISM_PTR pMechanism,
CK_OBJECT_HANDLE hKey
);
C_EncryptInit initializes an encryption operation. hSession
is the session’s handle; pMechanism points to the encryption mechanism; hKey
is the handle of the encryption key.
The CKA_ENCRYPT attribute of the encryption key,
which indicates whether the key supports encryption, MUST be CK_TRUE.
After calling C_EncryptInit, the application can
either call C_Encrypt to encrypt data in a single part; or call C_EncryptUpdate
zero or more times, followed by C_EncryptFinal, to encrypt data in
multiple parts. The encryption operation is active until the application uses
a call to C_Encrypt or C_EncryptFinal to actually obtain
the final piece of ciphertext. To process additional data (in single or
multiple parts), the application MUST call C_EncryptInit again.
C_EncryptInit can be called with pMechanism
set to NULL_PTR to terminate an active encryption operation. If an active
operation operations cannot be cancelled, CKR_OPERATION_CANCEL_FAILED must be
returned.
Return values: CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_KEY_FUNCTION_NOT_PERMITTED, CKR_KEY_HANDLE_INVALID, CKR_KEY_SIZE_RANGE,
CKR_KEY_TYPE_INCONSISTENT, CKR_MECHANISM_INVALID, CKR_MECHANISM_PARAM_INVALID,
CKR_OK, CKR_OPERATION_ACTIVE, CKR_PIN_EXPIRED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN,
CKR_OPERATION_CANCEL_FAILED.
Example: see C_EncryptFinal.
CK_DECLARE_FUNCTION(CK_RV, C_Encrypt)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pData,
CK_ULONG ulDataLen,
CK_BYTE_PTR pEncryptedData,
CK_ULONG_PTR pulEncryptedDataLen
);
C_Encrypt encrypts single-part data. hSession
is the session’s handle; pData points to the data; ulDataLen is
the length in bytes of the data; pEncryptedData points to the location
that receives the encrypted data; pulEncryptedDataLen points to the
location that holds the length in bytes of the encrypted data.
C_Encrypt uses the convention described in Section 5.2 on producing output.
The encryption operation MUST have been initialized with C_EncryptInit.
A call to C_Encrypt always terminates the active encryption operation
unless it returns CKR_BUFFER_TOO_SMALL or is a successful call (i.e.,
one which returns CKR_OK) to determine the length of the buffer needed to hold
the ciphertext.
C_Encrypt cannot be used to terminate a
multi-part operation, and MUST be called after C_EncryptInit without
intervening C_EncryptUpdate calls.
For some encryption mechanisms, the input plaintext data has
certain length constraints (either because the mechanism can only encrypt
relatively short pieces of plaintext, or because the mechanism’s input data
MUST consist of an integral number of blocks). If these constraints are not
satisfied, then C_Encrypt will fail with return code CKR_DATA_LEN_RANGE.
The plaintext and ciphertext can be in the same place, i.e.,
it is OK if pData and pEncryptedData point to the same location.
For most mechanisms, C_Encrypt is equivalent to a
sequence of C_EncryptUpdate operations followed by C_EncryptFinal.
Return values: CKR_ARGUMENTS_BAD, CKR_BUFFER_TOO_SMALL,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DATA_INVALID, CKR_DATA_LEN_RANGE,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK,
CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID.
Example: see C_EncryptFinal for an example of similar
functions.
CK_DECLARE_FUNCTION(CK_RV, C_EncryptUpdate)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pPart,
CK_ULONG ulPartLen,
CK_BYTE_PTR pEncryptedPart,
CK_ULONG_PTR pulEncryptedPartLen
);
C_EncryptUpdate continues a multiple-part encryption
operation, processing another data part. hSession is the session’s
handle; pPart points to the data part; ulPartLen is the length of
the data part; pEncryptedPart points to the location that receives the
encrypted data part; pulEncryptedPartLen points to the location that
holds the length in bytes of the encrypted data part.
C_EncryptUpdate uses the convention described in
Section 5.2 on producing output.
The encryption operation MUST have been initialized with C_EncryptInit.
This function may be called any number of times in succession. A call to C_EncryptUpdate
which results in an error other than CKR_BUFFER_TOO_SMALL terminates the
current encryption operation.
The plaintext and ciphertext can be in the same place, i.e.,
it is OK if pPart and pEncryptedPart point to the same location.
Return values: CKR_ARGUMENTS_BAD, CKR_BUFFER_TOO_SMALL,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DATA_LEN_RANGE, CKR_DEVICE_ERROR,
CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK,
CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID.
Example: see C_EncryptFinal.
CK_DECLARE_FUNCTION(CK_RV, C_EncryptFinal)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pLastEncryptedPart,
CK_ULONG_PTR pulLastEncryptedPartLen
);
C_EncryptFinal finishes a multiple-part encryption
operation. hSession is the session’s handle; pLastEncryptedPart
points to the location that receives the last encrypted data part, if any; pulLastEncryptedPartLen
points to the location that holds the length of the last encrypted data part.
C_EncryptFinal uses the convention described in
Section 5.2 on producing output.
The encryption operation MUST have been initialized with C_EncryptInit.
A call to C_EncryptFinal always terminates the active encryption
operation unless it returns CKR_BUFFER_TOO_SMALL or is a successful call (i.e.,
one which returns CKR_OK) to determine the length of the buffer needed to hold
the ciphertext.
For some multi-part encryption mechanisms, the input
plaintext data has certain length constraints, because the mechanism’s input
data MUST consist of an integral number of blocks. If these constraints are
not satisfied, then C_EncryptFinal will fail with return code
CKR_DATA_LEN_RANGE.
Return values: CKR_ARGUMENTS_BAD, CKR_BUFFER_TOO_SMALL,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DATA_LEN_RANGE, CKR_DEVICE_ERROR,
CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK,
CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID.
Example:
#define PLAINTEXT_BUF_SZ 200
#define CIPHERTEXT_BUF_SZ 256
CK_ULONG firstPieceLen, secondPieceLen;
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hKey;
CK_BYTE iv[8];
CK_MECHANISM mechanism = {
CKM_DES_CBC_PAD, iv, sizeof(iv)
};
CK_BYTE data[PLAINTEXT_BUF_SZ];
CK_BYTE encryptedData[CIPHERTEXT_BUF_SZ];
CK_ULONG ulEncryptedData1Len;
CK_ULONG ulEncryptedData2Len;
CK_ULONG ulEncryptedData3Len;
CK_RV rv;
.
.
firstPieceLen = 90;
secondPieceLen = PLAINTEXT_BUF_SZ-firstPieceLen;
rv = C_EncryptInit(hSession, &mechanism, hKey);
if (rv == CKR_OK) {
/* Encrypt first piece */
ulEncryptedData1Len = sizeof(encryptedData);
rv = C_EncryptUpdate(
hSession,
&data[0], firstPieceLen,
&encryptedData[0], &ulEncryptedData1Len);
if (rv != CKR_OK) {
.
.
}
/* Encrypt second piece */
ulEncryptedData2Len = sizeof(encryptedData)-ulEncryptedData1Len;
rv = C_EncryptUpdate(
hSession,
&data[firstPieceLen], secondPieceLen,
&encryptedData[ulEncryptedData1Len],
&ulEncryptedData2Len);
if (rv != CKR_OK) {
.
.
}
/* Get last little encrypted bit */
ulEncryptedData3Len =
sizeof(encryptedData)-ulEncryptedData1Len-ulEncryptedData2Len;
rv = C_EncryptFinal(
hSession,
&encryptedData[ulEncryptedData1Len+ulEncryptedData2Len],
&ulEncryptedData3Len);
if (rv != CKR_OK) {
.
.
}
}
Message-based encryption refers to
the process of encrypting multiple messages using the same encryption mechanism
and encryption key. The encryption mechanism can be either an authenticated
encryption with associated data (AEAD) algorithm or a pure encryption
algorithm.
Cryptoki provides the following
functions for message-based encryption:
CK_DECLARE_FUNCTION(CK_RV, C_MessageEncryptInit)(
CK_SESSION_HANDLE hSession,
CK_MECHANISM_PTR pMechanism,
CK_OBJECT_HANDLE hKey
);
C_MessageEncryptInit prepares a session for one or more encryption operations
that use the same encryption mechanism and encryption key. hSession is the
session’s handle; pMechanism points to the encryption mechanism; hKey is the
handle of the encryption key.
The CKA_ENCRYPT attribute of the
encryption key, which indicates whether the key supports encryption, MUST be
CK_TRUE.
After calling C_MessageEncryptInit,
the application can either call C_EncryptMessage to encrypt a message in
a single part, or call C_EncryptMessageBegin, followed by C_EncryptMessageNext
one or more times, to encrypt a message in multiple parts. This may be repeated
several times. The message-based encryption process is active until the
application calls C_MessageEncryptFinal to finish the message-based encryption
process.
C_MessageEncryptInit can be called
with pMechanism set to NULL_PTR to terminate a message-based encryption
process. If a multi-part message encryption operation is active, it will also
be terminated. If an active operation has been initialized and it cannot be
cancelled, CKR_OPERATION_CANCEL_FAILED must be returned.
Return values:
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_KEY_FUNCTION_NOT_PERMITTED,
CKR_KEY_HANDLE_INVALID, CKR_KEY_SIZE_RANGE, CKR_KEY_TYPE_INCONSISTENT,
CKR_MECHANISM_INVALID, CKR_MECHANISM_PARAM_INVALID, CKR_OK,
CKR_OPERATION_ACTIVE, CKR_PIN_EXPIRED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN,
CKR_OPERATION_CANCEL_FAILED.
CK_DECLARE_FUNCTION(CK_RV,
C_EncryptMessage)(
CK_SESSION_HANDLE hSession,
CK_VOID_PTR pParameter,
CK_ULONG ulParameterLen,
CK_BYTE_PTR pAssociatedData,
CK_ULONG ulAssociatedDataLen,
CK_BYTE_PTR pPlaintext,
CK_ULONG ulPlaintextLen,
CK_BYTE_PTR pCiphertext,
CK_ULONG_PTR pulCiphertextLen
);
C_EncryptMessage encrypts a message in a single part. hSession is
the session’s handle; pParameter and ulParameterLen specify any
mechanism-specific parameters for the message encryption operation; pAssociatedData
and ulAssociatedDataLen specify the associated data for an AEAD
mechanism; pPlaintext points to the plaintext data; ulPlaintextLen
is the length in bytes of the plaintext data; pCiphertext points to the
location that receives the encrypted data; pulCiphertextLen points to
the location that holds the length in bytes of the encrypted data.
Typically, pParameter is an
initialization vector (IV) or nonce. Depending on the mechanism parameter
passed to C_MessageEncryptInit, pParameter may be either an input
or an output parameter. For example, if the mechanism parameter specifies an
IV generator mechanism, the IV generated by the IV generator will be output to
the pParameter buffer.
If the encryption mechanism is not
AEAD, pAssociatedData and ulAssociatedDataLen are not used and
should be set to (NULL, 0).
C_EncryptMessage uses the convention described in Section 5.2 on producing output.
The message-based encryption
process MUST have been initialized with C_MessageEncryptInit. A call to C_EncryptMessage
begins and terminates a message encryption operation.
C_EncryptMessage cannot be called in the middle of a multi-part message
encryption operation.
For some encryption mechanisms,
the input plaintext data has certain length constraints (either because the
mechanism can only encrypt relatively short pieces of plaintext, or because the
mechanism’s input data MUST consist of an integral number of blocks). If these
constraints are not satisfied, then C_EncryptMessage will fail with
return code CKR_DATA_LEN_RANGE. The plaintext and ciphertext can be in the same
place, i.e., it is OK if pPlaintext and pCiphertext point to the
same location.
For most mechanisms, C_EncryptMessage
is equivalent to C_EncryptMessageBegin followed by a sequence of C_EncryptMessageNext
operations.
Return values: CKR_ARGUMENTS_BAD,
CKR_BUFFER_TOO_SMALL, CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DATA_INVALID,
CKR_DATA_LEN_RANGE, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED,
CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_OK, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID.
CK_DECLARE_FUNCTION(CK_RV,
C_EncryptMessageBegin)(
CK_SESSION_HANDLE
hSession,
CK_VOID_PTR pParameter,
CK_ULONG
ulParameterLen,
CK_BYTE_PTR
pAssociatedData,
CK_ULONG
ulAssociatedDataLen
);
C_EncryptMessageBegin begins a multiple-part message encryption operation. hSession
is the session’s handle; pParameter and ulParameterLen specify
any mechanism-specific parameters for the message encryption operation;
pAssociatedData and ulAssociatedDataLen specify the associated data
for an AEAD mechanism.
Typically, pParameter is an
initialization vector (IV) or nonce. Depending on the mechanism parameter
passed to C_MessageEncryptInit, pParameter may be either an input
or an output parameter. For example, if the mechanism parameter specifies an
IV generator mechanism, the IV generated by the IV generator will be output to
the pParameter buffer.
If the mechanism is not AEAD, pAssociatedData
and ulAssociatedDataLen are not used and should be set to (NULL, 0).
After calling C_EncryptMessageBegin,
the application should call C_EncryptMessageNext one or more times to
encrypt the message in multiple parts. The message encryption operation is
active until the application uses a call to C_EncryptMessageNext with
flags=CKF_END_OF_MESSAGE to actually obtain the final piece of ciphertext. To
process additional messages (in single or multiple parts), the application MUST
call C_EncryptMessage or C_EncryptMessageBegin again.
Return values:
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_OPERATION_ACTIVE,
CKR_PIN_EXPIRED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID,
CKR_USER_NOT_LOGGED_IN.
CK_DECLARE_FUNCTION(CK_RV,
C_EncryptMessageNext)(
CK_SESSION_HANDLE hSession,
CK_VOID_PTR pParameter,
CK_ULONG ulParameterLen,
CK_BYTE_PTR pPlaintextPart,
CK_ULONG ulPlaintextPartLen,
CK_BYTE_PTR pCiphertextPart,
CK_ULONG_PTR pulCiphertextPartLen,
CK_FLAGS flags
);
C_EncryptMessageNext continues a multiple-part message encryption operation,
processing another message part. hSession is the session’s handle; pParameter
and ulParameterLen specify any mechanism-specific parameters for the
message encryption operation; pPlaintextPart points to the plaintext
message part; ulPlaintextPartLen is the length of the plaintext message
part; pCiphertextPart points to the location that receives the encrypted
message part; pulCiphertextPartLen points to the location that holds the
length in bytes of the encrypted message part; flags is set to 0 if
there is more plaintext data to follow, or set to CKF_END_OF_MESSAGE if this is
the last plaintext part.
Typically, pParameter is an
initialization vector (IV) or nonce. Depending on the mechanism parameter
passed to C_EncryptMessageNext, pParameter may be either an input
or an output parameter. For example, if the mechanism parameter specifies an
IV generator mechanism, the IV generated by the IV generator will be output to
the pParameter buffer.
C_EncryptMessageNext uses the convention described in Section 5.2 on producing output.
The message encryption operation
MUST have been started with C_EncryptMessageBegin. This function may be
called any number of times in succession. A call to C_EncryptMessageNext with
flags=0 which results in an error other than CKR_BUFFER_TOO_SMALL terminates
the current message encryption operation. A call to C_EncryptMessageNext
with flags=CKF_END_OF_MESSAGE always terminates the active message encryption
operation unless it returns CKR_BUFFER_TOO_SMALL or is a successful call (i.e.,
one which returns CKR_OK) to determine the length of the buffer needed
to hold the ciphertext.
Although the last C_EncryptMessageNext
call ends the encryption of a message, it does not finish the message-based
encryption process. Additional C_EncryptMessage or C_EncryptMessageBegin
and C_EncryptMessageNext calls may be made on the session.
The plaintext and ciphertext can
be in the same place, i.e., it is OK if pPlaintextPart and pCiphertextPart
point to the same location.
For some multi-part encryption
mechanisms, the input plaintext data has certain length constraints, because
the mechanism’s input data MUST consist of an integral number of blocks. If
these constraints are not satisfied when the final message part is supplied
(i.e., with flags=CKF_END_OF_MESSAGE), then C_EncryptMessageNext will
fail with return code CKR_DATA_LEN_RANGE.
Return values: CKR_ARGUMENTS_BAD,
CKR_BUFFER_TOO_SMALL, CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DATA_LEN_RANGE,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK,
CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID.
CK_DECLARE_FUNCTION(CK_RV, C_MessageEncryptFinal)(
CK_SESSION_HANDLE hSession
);
C_MessageEncryptFinal finishes a message-based encryption process. hSession is
the session’s handle.
The message-based encryption
process MUST have been initialized with C_MessageEncryptInit.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR,
CKR_HOST_MEMORY, CKR_OK,
CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID.
Example:
#define PLAINTEXT_BUF_SZ 200
#define AUTH_BUF_SZ 100
#define CIPHERTEXT_BUF_SZ 256
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hKey;
CK_BYTE iv[] = { 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 };
CK_BYTE tag[16];
CK_GCM_MESSAGE_PARAMS gcmParams = {
iv,
sizeof(iv) * 8,
0,
CKG_NO_GENERATE,
tag,
sizeof(tag) * 8
};
CK_MECHANISM mechanism = {
CKM_AES_GCM, &gcmParams, sizeof(gcmParams)
};
CK_BYTE data[2][PLAINTEXT_BUF_SZ];
CK_BYTE auth[2][AUTH_BUF_SZ];
CK_BYTE encryptedData[2][CIPHERTEXT_BUF_SZ];
CK_ULONG ulEncryptedDataLen, ulFirstEncryptedDataLen;
CK_ULONG firstPieceLen = PLAINTEXT_BUF_SZ / 2;
/* error handling is omitted for better readability */
.
.
C_MessageEncryptInit(hSession, &mechanism, hKey);
/* encrypt message en bloc with given IV */
ulEncryptedDataLen = sizeof(encryptedData[0]);
C_EncryptMessage(hSession,
&gcmParams, sizeof(gcmParams),
&auth[0][0], sizeof(auth[0]),
&data[0][0], sizeof(data[0]),
&encryptedData[0][0], &ulEncryptedDataLen);
/* iv and tag are set now for message */
/* encrypt message in two steps with generated IV */
gcmParams.ivGenerator = CKG_GENERATE;
C_EncryptMessageBegin(hSession,
&gcmParams, sizeof(gcmParams),
&auth[1][0], sizeof(auth[1])
);
/* encrypt first piece */
ulFirstEncryptedDataLen = sizeof(encryptedData[1]);
C_EncryptMessageNext(hSession,
&gcmParams, sizeof(gcmParams),
&data[1][0], firstPieceLen,
&encryptedData[1][0], &ulFirstEncryptedDataLen,
0
);
/* encrypt second piece */
ulEncryptedDataLen = sizeof(encryptedData[1]) -
ulFirstEncryptedDataLen;
C_EncryptMessageNext(hSession,
&gcmParams, sizeof(gcmParams),
&data[1][firstPieceLen],
sizeof(data[1])-firstPieceLen,
&encryptedData[1][ulFirstEncryptedDataLen],
&ulEncryptedDataLen,
CKF_END_OF_MESSAGE
);
/* tag is set now for message */
/* finalize */
C_MessageEncryptFinal(hSession);
Cryptoki
provides the following functions for decrypting data:
CK_DECLARE_FUNCTION(CK_RV, C_DecryptInit)(
CK_SESSION_HANDLE hSession,
CK_MECHANISM_PTR pMechanism,
CK_OBJECT_HANDLE hKey
);
C_DecryptInit initializes a decryption operation. hSession
is the session’s handle; pMechanism points to the decryption mechanism; hKey
is the handle of the decryption key.
The CKA_DECRYPT attribute of the decryption key,
which indicates whether the key supports decryption, MUST be CK_TRUE.
After calling C_DecryptInit, the application can
either call C_Decrypt to decrypt data in a single part; or call C_DecryptUpdate
zero or more times, followed by C_DecryptFinal, to decrypt data in
multiple parts. The decryption operation is active until the application uses
a call to C_Decrypt or C_DecryptFinal to actually obtain
the final piece of plaintext. To process additional data (in single or
multiple parts), the application MUST call C_DecryptInit again.
C_DecryptInit can be called with pMechanism
set to NULL_PTR to terminate an active decryption operation. If an active
operation cannot be cancelled, CKR_OPERATION_CANCEL_FAILED must be returned.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_KEY_FUNCTION_NOT_PERMITTED,
CKR_KEY_HANDLE_INVALID, CKR_KEY_SIZE_RANGE, CKR_KEY_TYPE_INCONSISTENT,
CKR_MECHANISM_INVALID, CKR_MECHANISM_PARAM_INVALID, CKR_OK,
CKR_OPERATION_ACTIVE, CKR_PIN_EXPIRED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN,
CKR_OPERATION_CANCEL_FAILED.
Example: see C_DecryptFinal.
CK_DECLARE_FUNCTION(CK_RV, C_Decrypt)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pEncryptedData,
CK_ULONG ulEncryptedDataLen,
CK_BYTE_PTR pData,
CK_ULONG_PTR pulDataLen
);
C_Decrypt decrypts encrypted data in a single part. hSession
is the session’s handle; pEncryptedData points to the encrypted data; ulEncryptedDataLen
is the length of the encrypted data; pData points to the location that
receives the recovered data; pulDataLen points to the location that
holds the length of the recovered data.
C_Decrypt uses the convention described in Section 5.2 on producing output.
The decryption operation MUST have been initialized with C_DecryptInit.
A call to C_Decrypt always terminates the active decryption operation
unless it returns CKR_BUFFER_TOO_SMALL or is a successful call (i.e.,
one which returns CKR_OK) to determine the length of the buffer needed to hold
the plaintext.
C_Decrypt cannot be used to terminate a
multi-part operation, and MUST be called after C_DecryptInit without
intervening C_DecryptUpdate calls.
The ciphertext and plaintext can be in the same place, i.e.,
it is OK if pEncryptedData and pData point to the same location.
If the input ciphertext data cannot be decrypted because it
has an inappropriate length, then either CKR_ENCRYPTED_DATA_INVALID or CKR_ENCRYPTED_DATA_LEN_RANGE
may be returned.
For most mechanisms, C_Decrypt is equivalent to a
sequence of C_DecryptUpdate operations followed by C_DecryptFinal.
Return values: CKR_ARGUMENTS_BAD, CKR_BUFFER_TOO_SMALL, CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED,
CKR_ENCRYPTED_DATA_INVALID, CKR_ENCRYPTED_DATA_LEN_RANGE,
CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_OK, CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN.
Example: see C_DecryptFinal
for an example of similar functions.
CK_DECLARE_FUNCTION(CK_RV, C_DecryptUpdate)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pEncryptedPart,
CK_ULONG ulEncryptedPartLen,
CK_BYTE_PTR pPart,
CK_ULONG_PTR pulPartLen
);
C_DecryptUpdate continues a multiple-part decryption
operation, processing another encrypted data part. hSession is the
session’s handle; pEncryptedPart points to the encrypted data part; ulEncryptedPartLen
is the length of the encrypted data part; pPart points to the location
that receives the recovered data part; pulPartLen points to the location
that holds the length of the recovered data part.
C_DecryptUpdate uses the
convention described in Section 5.2 on producing output.
The decryption operation MUST have been initialized with C_DecryptInit.
This function may be called any number of times in succession. A call to C_DecryptUpdate
which results in an error other than CKR_BUFFER_TOO_SMALL terminates the
current decryption operation.
The ciphertext and plaintext can be in the same place, i.e.,
it is OK if pEncryptedPart and pPart point to the same location.
Return values: CKR_ARGUMENTS_BAD, CKR_BUFFER_TOO_SMALL, CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED,
CKR_ENCRYPTED_DATA_INVALID, CKR_ENCRYPTED_DATA_LEN_RANGE,
CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_OK, CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN.
Example: See C_DecryptFinal.
CK_DECLARE_FUNCTION(CK_RV, C_DecryptFinal)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pLastPart,
CK_ULONG_PTR pulLastPartLen
);
C_DecryptFinal finishes a multiple-part decryption
operation. hSession is the session’s handle; pLastPart points to
the location that receives the last recovered data part, if any; pulLastPartLen
points to the location that holds the length of the last recovered data part.
C_DecryptFinal uses the convention described in
Section 5.2 on producing output.
The decryption operation MUST have been initialized with C_DecryptInit.
A call to C_DecryptFinal always terminates the active decryption
operation unless it returns CKR_BUFFER_TOO_SMALL or is a successful call (i.e.,
one which returns CKR_OK) to determine the length of the buffer needed to hold
the plaintext.
If the input ciphertext data cannot be decrypted because it
has an inappropriate length, then either CKR_ENCRYPTED_DATA_INVALID or
CKR_ENCRYPTED_DATA_LEN_RANGE may be returned.
Return values: CKR_ARGUMENTS_BAD, CKR_BUFFER_TOO_SMALL,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_ENCRYPTED_DATA_INVALID, CKR_ENCRYPTED_DATA_LEN_RANGE,
CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_OK, CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN.
Example:
#define CIPHERTEXT_BUF_SZ 256
#define PLAINTEXT_BUF_SZ 256
CK_ULONG firstEncryptedPieceLen, secondEncryptedPieceLen;
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hKey;
CK_BYTE iv[8];
CK_MECHANISM mechanism = {
CKM_DES_CBC_PAD, iv, sizeof(iv)
};
CK_BYTE data[PLAINTEXT_BUF_SZ];
CK_BYTE encryptedData[CIPHERTEXT_BUF_SZ];
CK_ULONG ulData1Len, ulData2Len, ulData3Len;
CK_RV rv;
.
.
firstEncryptedPieceLen = 90;
secondEncryptedPieceLen =
CIPHERTEXT_BUF_SZ-firstEncryptedPieceLen;
rv = C_DecryptInit(hSession, &mechanism, hKey);
if (rv == CKR_OK) {
/* Decrypt first piece */
ulData1Len = sizeof(data);
rv = C_DecryptUpdate(
hSession,
&encryptedData[0], firstEncryptedPieceLen,
&data[0], &ulData1Len);
if (rv != CKR_OK) {
.
.
}
/* Decrypt second piece */
ulData2Len = sizeof(data)-ulData1Len;
rv = C_DecryptUpdate(
hSession,
&encryptedData[firstEncryptedPieceLen],
secondEncryptedPieceLen,
&data[ulData1Len], &ulData2Len);
if (rv != CKR_OK) {
.
.
}
/* Get last little decrypted bit */
ulData3Len = sizeof(data)-ulData1Len-ulData2Len;
rv = C_DecryptFinal(
hSession,
&data[ulData1Len+ulData2Len], &ulData3Len);
if (rv != CKR_OK) {
.
.
}
}
Message-based decryption refers to
the process of decrypting multiple encrypted messages using the same decryption
mechanism and decryption key. The decryption mechanism can be either an
authenticated encryption with associated data (AEAD) algorithm or a pure
encryption algorithm.
Cryptoki provides the following
functions for message-based decryption.
CK_DECLARE_FUNCTION(CK_RV,
C_MessageDecryptInit)(
CK_SESSION_HANDLE hSession,
CK_MECHANISM_PTR pMechanism,
CK_OBJECT_HANDLE hKey
);
C_MessageDecryptInit initializes a message-based decryption process, preparing
a session for one or more decryption operations that use the same decryption
mechanism and decryption key. hSession is the session’s handle; pMechanism
points to the decryption mechanism; hKey is the handle of the decryption
key.
The CKA_DECRYPT attribute of the
decryption key, which indicates whether the key supports decryption, MUST be
CK_TRUE.
After calling C_MessageDecryptInit,
the application can either call C_DecryptMessage to decrypt an encrypted
message in a single part; or call C_DecryptMessageBegin, followed by C_DecryptMessageNext
one or more times, to decrypt an encrypted message in multiple parts. This may
be repeated several times. The message-based decryption process is active
until the application uses a call to C_MessageDecryptFinal to finish the
message-based decryption process.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED,
CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_KEY_FUNCTION_NOT_PERMITTED, CKR_KEY_HANDLE_INVALID, CKR_KEY_SIZE_RANGE,
CKR_KEY_TYPE_INCONSISTENT, CKR_MECHANISM_INVALID, CKR_MECHANISM_PARAM_INVALID,
CKR_OK, CKR_OPERATION_ACTIVE, CKR_PIN_EXPIRED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN,
CKR_OPERATION_CANCEL_FAILED.
CK_DECLARE_FUNCTION(CK_RV,
C_DecryptMessage)(
CK_SESSION_HANDLE hSession,
CK_VOID_PTR pParameter,
CK_ULONG ulParameterLen,
CK_BYTE_PTR pAssociatedData,
CK_ULONG ulAssociatedDataLen,
CK_BYTE_PTR pCiphertext,
CK_ULONG ulCiphertextLen,
CK_BYTE_PTR pPlaintext,
CK_ULONG_PTR pulPlaintextLen
);
C_DecryptMessage decrypts an encrypted message in a single part. hSession
is the session’s handle; pParameter and ulParameterLen
specify any mechanism-specific parameters for the message decryption operation;
pAssociatedData and ulAssociatedDataLen specify the associated
data for an AEAD mechanism; pCiphertext points to the encrypted message;
ulCiphertextLen is the length of the encrypted message; pPlaintext
points to the location that receives the recovered message; pulPlaintextLen
points to the location that holds the length of the recovered message.
Typically, pParameter is an
initialization vector (IV) or nonce. Unlike the pParameter parameter of C_EncryptMessage,
pParameter is always an input parameter.
If the decryption mechanism is not
AEAD, pAssociatedData and ulAssociatedDataLen are not used and
should be set to (NULL, 0).
C_DecryptMessage uses the convention described in Section 5.2 on producing output.
The message-based decryption
process MUST have been initialized with C_MessageDecryptInit. A call to C_DecryptMessage
begins and terminates a message decryption operation.
C_DecryptMessage cannot be called in the middle of a multi-part message
decryption operation.
The ciphertext and plaintext can
be in the same place, i.e., it is OK if pCiphertext and pPlaintext
point to the same location.
If the input ciphertext data
cannot be decrypted because it has an inappropriate length, then either
CKR_ENCRYPTED_DATA_INVALID or CKR_ENCRYPTED_DATA_LEN_RANGE may be returned.
If the decryption mechanism is an
AEAD algorithm and the authenticity of the associated data or ciphertext cannot
be verified, then CKR_AEAD_DECRYPT_FAILED is returned.
For most mechanisms, C_DecryptMessage
is equivalent to C_DecryptMessageBegin followed by a sequence of C_DecryptMessageNext
operations.
Return values: CKR_ARGUMENTS_BAD,
CKR_BUFFER_TOO_SMALL, CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR,
CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_ENCRYPTED_DATA_INVALID,
CKR_ENCRYPTED_DATA_LEN_RANGE, CKR_AEAD_DECRYPT_FAILED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK,
CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID,
CKR_USER_NOT_LOGGED_IN, CKR_OPERATION_CANCEL_FAILED.
CK_DECLARE_FUNCTION(CK_RV,
C_DecryptMessageBegin)(
CK_SESSION_HANDLE hSession,
CK_VOID_PTR pParameter,
CK_ULONG ulParameterLen,
CK_BYTE_PTR pAssociatedData,
CK_ULONG ulAssociatedDataLen
);
C_DecryptMessageBegin begins a multiple-part message decryption operation. hSession
is the session’s handle; pParameter and ulParameterLen specify
any mechanism-specific parameters for the message decryption operation; pAssociatedData
and ulAssociatedDataLen specify the associated data for an AEAD
mechanism.
Typically, pParameter is an
initialization vector (IV) or nonce. Unlike the pParameter parameter of
C_EncryptMessageBegin, pParameter is always an input parameter.
If the decryption mechanism is not
AEAD, pAssociatedData and ulAssociatedDataLen are not used and
should be set to (NULL, 0).
After calling C_DecryptMessageBegin,
the application should call C_DecryptMessageNext one or more times to
decrypt the encrypted message in multiple parts. The message decryption
operation is active until the application uses a call to C_DecryptMessageNext
with flags=CKF_END_OF_MESSAGE to actually obtain the final piece of plaintext.
To process additional encrypted messages (in single or multiple parts), the
application MUST call C_DecryptMessage or C_DecryptMessageBegin
again.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_OPERATION_ACTIVE,
CKR_PIN_EXPIRED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID,
CKR_USER_NOT_LOGGED_IN.
CK_DECLARE_FUNCTION(CK_RV,
C_DecryptMessageNext)(
CK_SESSION_HANDLE hSession,
CK_VOID_PTR pParameter,
CK_ULONG ulParameterLen,
CK_BYTE_PTR pCiphertextPart,
CK_ULONG ulCiphertextPartLen,
CK_BYTE_PTR pPlaintextPart,
CK_ULONG_PTR pulPlaintextPartLen,
CK_FLAGS flags
);
C_DecryptMessageNext continues a multiple-part message decryption operation,
processing another encrypted message part. hSession is the session’s
handle; pParameter and ulParameterLen specify any
mechanism-specific parameters for the message decryption operation; pCiphertextPart
points to the encrypted message part; ulCiphertextPartLen is the length
of the encrypted message part; pPlaintextPart points to the location
that receives the recovered message part; pulPlaintextPartLen points to
the location that holds the length of the recovered message part; flags
is set to 0 if there is more ciphertext data to follow, or set to
CKF_END_OF_MESSAGE if this is the last ciphertext part.
Typically, pParameter is an
initialization vector (IV) or nonce. Unlike the pParameter parameter of C_EncryptMessageNext,
pParameter is always an input parameter.
C_DecryptMessageNext uses the convention described in Section 5.2 on producing output.
The message decryption operation
MUST have been started with C_DecryptMessageBegin. This function may be
called any number of times in succession. A call to C_DecryptMessageNext
with flags=0 which results in an error other than CKR_BUFFER_TOO_SMALL
terminates the current message decryption operation. A call to C_DecryptMessageNext
with flags=CKF_END_OF_MESSAGE always terminates the active message
decryption operation unless it returns CKR_BUFFER_TOO_SMALL or is a successful
call (i.e., one which returns CKR_OK) to determine the length of the buffer
needed to hold the plaintext.
The ciphertext and plaintext can
be in the same place, i.e., it is OK if pCiphertextPart and pPlaintextPart
point to the same location.
Although the last C_DecryptMessageNext
call ends the decryption of a message, it does not finish the message-based
decryption process. Additional C_DecryptMessage or
C_DecryptMessageBegin and C_DecryptMessageNext calls may be made on
the session.
If the input ciphertext data cannot
be decrypted because it has an inappropriate length, then either
CKR_ENCRYPTED_DATA_INVALID or CKR_ENCRYPTED_DATA_LEN_RANGE may be returned by
the last C_DecryptMessageNext call.
If the decryption mechanism is an
AEAD algorithm and the authenticity of the associated data or ciphertext cannot
be verified, then CKR_AEAD_DECRYPT_FAILED is returned by the last C_DecryptMessageNext
call.
Return values: CKR_ARGUMENTS_BAD,
CKR_BUFFER_TOO_SMALL, CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR,
CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_ENCRYPTED_DATA_INVALID,
CKR_ENCRYPTED_DATA_LEN_RANGE, CKR_AEAD_DECRYPT_FAILED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK,
CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID,
CKR_USER_NOT_LOGGED_IN.
CK_DECLARE_FUNCTION(CK_RV, C_MessageDecryptFinal)(
CK_SESSION_HANDLE hSession
);
C_MessageDecryptFinal finishes a message-based decryption process. hSession
is the session’s handle.
The message-based decryption
process MUST have been initialized with C_MessageDecryptInit.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR,
CKR_HOST_MEMORY, CKR_OK, CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN.
Cryptoki provides the following functions for digesting
data:
CK_DECLARE_FUNCTION(CK_RV, C_DigestInit)(
CK_SESSION_HANDLE hSession,
CK_MECHANISM_PTR pMechanism
);
C_DigestInit initializes a message-digesting
operation. hSession is the session’s handle; pMechanism points to
the digesting mechanism.
After calling C_DigestInit, the application can
either call C_Digest to digest data in a single part; or call C_DigestUpdate
zero or more times, followed by C_DigestFinal, to digest data in
multiple parts. The message-digesting operation is active until the
application uses a call to C_Digest or C_DigestFinal to
actually obtain the message digest. To process additional data (in single
or multiple parts), the application MUST call C_DigestInit again.
C_DigestInit can be called with pMechanism set
to NULL_PTR to terminate an active message-digesting operation. If an
operation has been initialized and it cannot be cancelled,
CKR_OPERATION_CANCEL_FAILED must be returned.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_MECHANISM_INVALID,
CKR_MECHANISM_PARAM_INVALID, CKR_OK, CKR_OPERATION_ACTIVE, CKR_PIN_EXPIRED,
CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN,
CKR_OPERATION_CANCEL_FAILED.
Example: see C_DigestFinal.
CK_DECLARE_FUNCTION(CK_RV, C_Digest)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pData,
CK_ULONG ulDataLen,
CK_BYTE_PTR pDigest,
CK_ULONG_PTR pulDigestLen
);
C_Digest digests data in a single part. hSession
is the session’s handle, pData points to the data; ulDataLen is
the length of the data; pDigest points to the location that receives the
message digest; pulDigestLen points to the location that holds the
length of the message digest.
C_Digest uses the convention described in Section 5.2 on producing output.
The digest operation MUST have been initialized with C_DigestInit.
A call to C_Digest always terminates the active digest operation unless
it returns CKR_BUFFER_TOO_SMALL or is a successful call (i.e., one which
returns CKR_OK) to determine the length of the buffer needed to hold the
message digest.
C_Digest cannot be used to terminate a
multi-part operation, and MUST be called after C_DigestInit without
intervening C_DigestUpdate calls.
The input data and digest output can be in the same place, i.e.,
it is OK if pData and pDigest point to the same location.
C_Digest is equivalent to a sequence of C_DigestUpdate
operations followed by C_DigestFinal.
Return values: CKR_ARGUMENTS_BAD, CKR_BUFFER_TOO_SMALL,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_OPERATION_NOT_INITIALIZED,
CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID.
Example: see C_DigestFinal for an example of similar
functions.
CK_DECLARE_FUNCTION(CK_RV, C_DigestUpdate)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pPart,
CK_ULONG ulPartLen
);
C_DigestUpdate continues a multiple-part
message-digesting operation, processing another data part. hSession is
the session’s handle, pPart points to the data part; ulPartLen is
the length of the data part.
The message-digesting operation MUST
have been initialized with C_DigestInit. Calls to this function and C_DigestKey
may be interspersed any number of times in any order. A call to C_DigestUpdate
which results in an error terminates the current digest operation.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_OPERATION_NOT_INITIALIZED,
CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID.
Example: see C_DigestFinal.
CK_DECLARE_FUNCTION(CK_RV, C_DigestKey)(
CK_SESSION_HANDLE hSession,
CK_OBJECT_HANDLE hKey
);
C_DigestKey continues a multiple-part
message-digesting operation by digesting the value of a secret key. hSession
is the session’s handle; hKey is the handle of the secret key to be
digested.
The message-digesting operation MUST have been initialized
with C_DigestInit. Calls to this function and C_DigestUpdate may
be interspersed any number of times in any order.
If the value of the supplied key cannot be digested purely
for some reason related to its length, C_DigestKey should return the
error code CKR_KEY_SIZE_RANGE.
Return values: CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_KEY_HANDLE_INVALID, CKR_KEY_INDIGESTIBLE, CKR_KEY_SIZE_RANGE, CKR_OK,
CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID.
Example: see C_DigestFinal.
CK_DECLARE_FUNCTION(CK_RV, C_DigestFinal)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pDigest,
CK_ULONG_PTR pulDigestLen
);
C_DigestFinal finishes a multiple-part message-digesting
operation, returning the message digest. hSession is the session’s
handle; pDigest points to the location that receives the message digest;
pulDigestLen points to the location that holds the length of the message
digest.
C_DigestFinal uses the convention described in
Section 5.2 on producing output.
The digest operation MUST have been initialized with C_DigestInit.
A call to C_DigestFinal always terminates the active digest operation
unless it returns CKR_BUFFER_TOO_SMALL or is a successful call (i.e.,
one which returns CKR_OK) to determine the length of the buffer needed to hold
the message digest.
Return values: CKR_ARGUMENTS_BAD, CKR_BUFFER_TOO_SMALL,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_OPERATION_NOT_INITIALIZED,
CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID.
Example:
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hKey;
CK_MECHANISM mechanism = {
CKM_MD5, NULL_PTR, 0
};
CK_BYTE data[] = {...};
CK_BYTE digest[16];
CK_ULONG ulDigestLen;
CK_RV rv;
.
.
rv = C_DigestInit(hSession, &mechanism);
if (rv != CKR_OK) {
.
.
}
rv = C_DigestUpdate(hSession, data, sizeof(data));
if (rv != CKR_OK) {
.
.
}
rv = C_DigestKey(hSession, hKey);
if (rv != CKR_OK) {
.
.
}
ulDigestLen = sizeof(digest);
rv = C_DigestFinal(hSession, digest,
&ulDigestLen);
.
.
5.13 Signing and MACing functions
Cryptoki provides the following functions for signing data
(for the purposes of Cryptoki, these operations also encompass message
authentication codes).
CK_DECLARE_FUNCTION(CK_RV, C_SignInit)(
CK_SESSION_HANDLE hSession,
CK_MECHANISM_PTR pMechanism,
CK_OBJECT_HANDLE hKey
);
C_SignInit initializes a signature operation, where
the signature is an appendix to the data. hSession is the session’s
handle; pMechanism points to the signature mechanism; hKey is the
handle of the signature key.
The CKA_SIGN attribute of the signature key, which
indicates whether the key supports signatures with appendix, MUST be CK_TRUE.
After calling C_SignInit, the application can either
call C_Sign to sign in a single part; or call C_SignUpdate one or
more times, followed by C_SignFinal, to sign data in multiple parts.
The signature operation is active until the application uses a call to C_Sign
or C_SignFinal to actually obtain the signature. To process
additional data (in single or multiple parts), the application MUST call C_SignInit
again.
C_SignInit can be called with pMechanism
set to NULL_PTR to terminate an active signature operation. If an operation
has been initialized and it cannot be cancelled, CKR_OPERATION_CANCEL_FAILED
must be returned.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR,
CKR_HOST_MEMORY, CKR_KEY_FUNCTION_NOT_PERMITTED,CKR_KEY_HANDLE_INVALID,
CKR_KEY_SIZE_RANGE, CKR_KEY_TYPE_INCONSISTENT, CKR_MECHANISM_INVALID,
CKR_MECHANISM_PARAM_INVALID, CKR_OK, CKR_OPERATION_ACTIVE, CKR_PIN_EXPIRED,
CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN,
CKR_OPERATION_CANCEL_FAILED.
Example: see C_SignFinal.
CK_DECLARE_FUNCTION(CK_RV, C_Sign)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pData,
CK_ULONG ulDataLen,
CK_BYTE_PTR pSignature,
CK_ULONG_PTR pulSignatureLen
);
C_Sign signs data in a single part, where the
signature is an appendix to the data. hSession is the session’s handle; pData
points to the data; ulDataLen is the length of the data; pSignature
points to the location that receives the signature; pulSignatureLen
points to the location that holds the length of the signature.
C_Sign uses the convention described in Section 5.2 on producing output.
The signing operation MUST have been initialized with C_SignInit.
A call to C_Sign always terminates the active signing operation unless
it returns CKR_BUFFER_TOO_SMALL or is a successful call (i.e., one which
returns CKR_OK) to determine the length of the buffer needed to hold the
signature.
C_Sign cannot be used to terminate a multi-part
operation, and MUST be called after C_SignInit without intervening C_SignUpdate
calls.
For most mechanisms, C_Sign is equivalent to a
sequence of C_SignUpdate operations followed by C_SignFinal.
Return values: CKR_ARGUMENTS_BAD, CKR_BUFFER_TOO_SMALL, CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DATA_INVALID, CKR_DATA_LEN_RANGE, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_OPERATION_NOT_INITIALIZED,
CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN,
CKR_FUNCTION_REJECTED, CKR_TOKEN_RESOURCE_EXCEEDED.
Example: see C_SignFinal for an example of similar
functions.
CK_DECLARE_FUNCTION(CK_RV, C_SignUpdate)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pPart,
CK_ULONG ulPartLen
);
C_SignUpdate continues a multiple-part signature
operation, processing another data part. hSession is the session’s
handle, pPart points to the data part; ulPartLen is the length of
the data part.
The signature operation MUST have been initialized with C_SignInit.
This function may be called any number of times in succession. A call to C_SignUpdate
which results in an error terminates the current signature operation.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DATA_LEN_RANGE, CKR_DEVICE_ERROR,
CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK,
CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID,
CKR_USER_NOT_LOGGED_IN, CKR_TOKEN_RESOURCE_EXCEEDED.
Example: see C_SignFinal.
CK_DECLARE_FUNCTION(CK_RV, C_SignFinal)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pSignature,
CK_ULONG_PTR pulSignatureLen
);
C_SignFinal finishes a multiple-part signature
operation, returning the signature. hSession is the session’s handle; pSignature
points to the location that receives the signature; pulSignatureLen
points to the location that holds the length of the signature.
C_SignFinal uses the convention described in Section 5.2 on producing output.
The signing operation MUST have been initialized with C_SignInit.
A call to C_SignFinal always terminates the active signing operation
unless it returns CKR_BUFFER_TOO_SMALL or is a successful call (i.e.,
one which returns CKR_OK) to determine the length of the buffer needed to hold
the signature.
Return values: CKR_ARGUMENTS_BAD, CKR_BUFFER_TOO_SMALL,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DATA_LEN_RANGE, CKR_DEVICE_ERROR,
CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_OPERATION_NOT_INITIALIZED,
CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN,
CKR_FUNCTION_REJECTED, CKR_TOKEN_RESOURCE_EXCEEDED.
Example:
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hKey;
CK_MECHANISM mechanism = {
CKM_DES_MAC, NULL_PTR, 0
};
CK_BYTE data[] = {...};
CK_BYTE mac[4];
CK_ULONG ulMacLen;
CK_RV rv;
.
.
rv = C_SignInit(hSession, &mechanism, hKey);
if (rv == CKR_OK) {
rv = C_SignUpdate(hSession, data, sizeof(data));
.
.
ulMacLen = sizeof(mac);
rv = C_SignFinal(hSession, mac, &ulMacLen);
.
.
}
CK_DECLARE_FUNCTION(CK_RV, C_SignRecoverInit)(
CK_SESSION_HANDLE hSession,
CK_MECHANISM_PTR pMechanism,
CK_OBJECT_HANDLE hKey
);
C_SignRecoverInit initializes a signature operation,
where the data can be recovered from the signature. hSession is the
session’s handle; pMechanism points to the structure that specifies the
signature mechanism; hKey is the handle of the signature key.
The CKA_SIGN_RECOVER attribute of the signature key,
which indicates whether the key supports signatures where the data can be
recovered from the signature, MUST be CK_TRUE.
After calling C_SignRecoverInit, the application may
call C_SignRecover to sign in a single part. The signature operation is
active until the application uses a call to C_SignRecover to actually
obtain the signature. To process additional data in a single part, the
application MUST call C_SignRecoverInit again.
C_SignRecoverInit can be called with pMechanism
set to NULL_PTR to terminate an active signature with data recovery operation.
If an active operation has been initialized and it cannot be cancelled,
CKR_OPERATION_CANCEL_FAILED must be returned.
Return values: CKR_ARGUMENTS_BAD, CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_KEY_FUNCTION_NOT_PERMITTED, CKR_KEY_HANDLE_INVALID, CKR_KEY_SIZE_RANGE,
CKR_KEY_TYPE_INCONSISTENT, CKR_MECHANISM_INVALID, CKR_MECHANISM_PARAM_INVALID,
CKR_OK, CKR_OPERATION_ACTIVE, CKR_PIN_EXPIRED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN,
CKR_OPERATION_CANCEL_FAILED.
Example: see C_SignRecover.
CK_DECLARE_FUNCTION(CK_RV, C_SignRecover)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pData,
CK_ULONG ulDataLen,
CK_BYTE_PTR pSignature,
CK_ULONG_PTR pulSignatureLen
);
C_SignRecover signs data in a single operation, where
the data can be recovered from the signature. hSession is the session’s
handle; pData points to the data; uLDataLen is the length of the
data; pSignature points to the location that receives the signature; pulSignatureLen
points to the location that holds the length of the signature.
C_SignRecover uses the convention described in
Section 5.2 on producing output.
The signing operation MUST have been initialized with C_SignRecoverInit.
A call to C_SignRecover always terminates the active signing operation
unless it returns CKR_BUFFER_TOO_SMALL or is a successful call (i.e.,
one which returns CKR_OK) to determine the length of the buffer needed to hold
the signature.
Return values: CKR_ARGUMENTS_BAD, CKR_BUFFER_TOO_SMALL,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DATA_INVALID, CKR_DATA_LEN_RANGE,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK,
CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID,
CKR_USER_NOT_LOGGED_IN, CKR_TOKEN_RESOURCE_EXCEEDED.
Example:
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hKey;
CK_MECHANISM mechanism = {
CKM_RSA_9796, NULL_PTR, 0
};
CK_BYTE data[] = {...};
CK_BYTE signature[128];
CK_ULONG ulSignatureLen;
CK_RV rv;
.
.
rv = C_SignRecoverInit(hSession, &mechanism, hKey);
if (rv == CKR_OK) {
ulSignatureLen = sizeof(signature);
rv = C_SignRecover(
hSession, data, sizeof(data), signature,
&ulSignatureLen);
if (rv == CKR_OK) {
.
.
}
}
Message-based signature refers to the process of signing
multiple messages using the same signature mechanism and signature key.
Cryptoki provides the following functions for for signing
messages (for the purposes of Cryptoki, these operations also encompass message
authentication codes).
CK_DECLARE_FUNCTION(CK_RV, C_MessageSignInit)(
CK_SESSION_HANDLE hSession,
CK_MECHANISM_PTR pMechanism,
CK_OBJECT_HANDLE hKey
);
C_MessageSignInit initializes a message-based
signature process, preparing a session for one or more signature operations
(where the signature is an appendix to the data) that use the same signature
mechanism and signature key. hSession is the session’s handle; pMechanism
points to the signature mechanism; hKey is the handle of the signature
key.
The CKA_SIGN attribute of the signature key, which
indicates whether the key supports signatures with appendix, MUST be CK_TRUE.
After calling C_MessageSignInit, the application can
either call C_SignMessage to sign a message in a single part; or call C_SignMessageBegin,
followed by C_SignMessageNext one or more times, to sign a message in
multiple parts. This may be repeated several times. The message-based signature
process is active until the application calls C_MessageSignFinal to
finish the message-based signature process.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_KEY_FUNCTION_NOT_PERMITTED,CKR_KEY_HANDLE_INVALID,
CKR_KEY_SIZE_RANGE, CKR_KEY_TYPE_INCONSISTENT, CKR_MECHANISM_INVALID,
CKR_MECHANISM_PARAM_INVALID, CKR_OK, CKR_OPERATION_ACTIVE, CKR_PIN_EXPIRED,
CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN.
CK_DECLARE_FUNCTION(CK_RV, C_SignMessage)(
CK_SESSION_HANDLE hSession,
CK_VOID_PTR pParameter,
CK_ULONG ulParameterLen,
CK_BYTE_PTR pData,
CK_ULONG ulDataLen,
CK_BYTE_PTR pSignature,
CK_ULONG_PTR pulSignatureLen
);
C_SignMessage signs a message in a single part, where
the signature is an appendix to the message. C_MessageSignInit must
previously been called on the session. hSession is the session’s handle;
pParameter and ulParameterLen specify any mechanism-specific
parameters for the message signature operation; pData points to the
data; ulDataLen is the length of the data; pSignature points to
the location that receives the signature; pulSignatureLen points to the
location that holds the length of the signature.
Depending on the mechanism parameter passed to C_MessageSignInit,
pParameter may be either an input or an output parameter.
C_SignMessage uses the convention described in
Section 5.2 on producing output.
The message-based signing process MUST have been initialized
with C_MessageSignInit. A call to C_SignMessage begins and
terminates a message signing operation unless it returns CKR_BUFFER_TOO_SMALL
to determine the length of the buffer needed to hold the signature, or is a
successful call (i.e., one which returns CKR_OK).
C_SignMessage cannot be called in the middle of a
multi-part message signing operation.
C_SignMessage does not finish the message-based
signing process. Additional C_SignMessage or C_SignMessageBegin
and C_SignMessageNext calls may be made on the session.
For most mechanisms, C_SignMessage is equivalent to C_SignMessageBegin
followed by a sequence of C_SignMessageNext operations.
Return values: CKR_ARGUMENTS_BAD, CKR_BUFFER_TOO_SMALL,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DATA_INVALID, CKR_DATA_LEN_RANGE,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_OPERATION_NOT_INITIALIZED,
CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN,
CKR_FUNCTION_REJECTED, CKR_TOKEN_RESOURCE_EXCEEDED.
CK_DECLARE_FUNCTION(CK_RV, C_SignMessageBegin)(
CK_SESSION_HANDLE hSession,
CK_VOID_PTR pParameter,
CK_ULONG ulParameterLen
);
C_SignMessageBegin begins a multiple-part message
signature operation, where the signature is an appendix to the message. C_MessageSignInit
must previously been called on the session. hSession is the session’s
handle; pParameter and ulParameterLen specify any
mechanism-specific parameters for the message signature operation.
Depending on the mechanism parameter passed to C_MessageSignInit,
pParameter may be either an input or an output parameter.
After calling C_SignMessageBegin, the application
should call C_SignMessageNext one or more times to sign the message in
multiple parts. The message signature operation is active until the application
uses a call to C_SignMessageNext with a non-NULL pulSignatureLen
to actually obtain the signature. To process additional messages (in single or
multiple parts), the application MUST call C_SignMessage or C_SignMessageBegin
again.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_OPERATION_ACTIVE,
CKR_PIN_EXPIRED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID,
CKR_USER_NOT_LOGGED_IN, CKR_TOKEN_RESOURCE_EXCEEDED.
CK_DECLARE_FUNCTION(CK_RV, C_SignMessageNext)(
CK_SESSION_HANDLE hSession,
CK_VOID_PTR pParameter,
CK_ULONG ulParameterLen,
CK_BYTE_PTR pDataPart,
CK_ULONG ulDataPartLen,
CK_BYTE_PTR pSignature,
CK_ULONG_PTR pulSignatureLen
);
C_SignMessageNext continues a multiple-part message
signature operation, processing another data part, or finishes a multiple-part
message signature operation, returning the signature. hSession is the
session’s handle, pDataPart points to the data part; pParameter
and ulParameterLen specify any mechanism-specific parameters for the
message signature operation; ulDataPartLen is the length of the data
part; pSignature points to the location that receives the signature; pulSignatureLen
points to the location that holds the length of the signature.
The pulSignatureLen argument is set to NULL if there
is more data part to follow, or set to a non-NULL value (to receive the
signature length) if this is the last data part.
C_SignMessageNext uses the convention described in
Section 5.2 on producing output.
The message signing operation MUST have been started with C_SignMessageBegin.
This function may be called any number of times in succession. A call to C_SignMessageNext
with a NULL pulSignatureLen which results in an error terminates the
current message signature operation. A call to C_SignMessageNext with a
non-NULL pulSignatureLen always terminates the active message signing
operation unless it returns CKR_BUFFER_TOO_SMALL to determine the length of the
buffer needed to hold the signature, or is a successful call (i.e., one which
returns CKR_OK).
Although the last C_SignMessageNext call ends the
signing of a message, it does not finish the message-based signing process.
Additional C_SignMessage or C_SignMessageBegin and C_SignMessageNext
calls may be made on the session.
Return values: CKR_ARGUMENTS_BAD, CKR_BUFFER_TOO_SMALL,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DATA_LEN_RANGE, CKR_DEVICE_ERROR,
CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK,
CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID,
CKR_USER_NOT_LOGGED_IN, CKR_FUNCTION_REJECTED, CKR_TOKEN_RESOURCE_EXCEEDED.
CK_DECLARE_FUNCTION(CK_RV, C_MessageSignFinal)(
CK_SESSION_HANDLE hSession
);
C_MessageSignFinal finishes a message-based signing
process. hSession is the session’s handle.
The message-based signing process MUST have been initialized
with C_MessageSignInit.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_OPERATION_NOT_INITIALIZED,
CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN,
CKR_FUNCTION_REJECTED, CKR_TOKEN_RESOURCE_EXCEEDED.
Cryptoki provides the following functions for verifying
signatures on data (for the purposes of Cryptoki, these operations also
encompass message authentication codes):
CK_DECLARE_FUNCTION(CK_RV, C_VerifyInit)(
CK_SESSION_HANDLE hSession,
CK_MECHANISM_PTR pMechanism,
CK_OBJECT_HANDLE hKey
);
C_VerifyInit initializes a verification operation,
where the signature is an appendix to the data. hSession is the
session’s handle; pMechanism points to the structure that specifies the
verification mechanism; hKey is the handle of the verification key.
The CKA_VERIFY attribute of the verification key,
which indicates whether the key supports verification where the signature is an
appendix to the data, MUST be CK_TRUE.
After calling C_VerifyInit, the application can
either call C_Verify to verify a signature on data in a single part; or
call C_VerifyUpdate one or more times, followed by C_VerifyFinal,
to verify a signature on data in multiple parts. The verification operation is
active until the application calls C_Verify or C_VerifyFinal. To
process additional data (in single or multiple parts), the application MUST
call C_VerifyInit again.
C_VerifyInit can be called with pMechanism
set to NULL_PTR to terminate an active verification operation. If an active
operation has been initialized and it cannot be cancelled,
CKR_OPERATION_CANCEL_FAILED must be returned.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_KEY_FUNCTION_NOT_PERMITTED,
CKR_KEY_HANDLE_INVALID, CKR_KEY_SIZE_RANGE, CKR_KEY_TYPE_INCONSISTENT,
CKR_MECHANISM_INVALID, CKR_MECHANISM_PARAM_INVALID, CKR_OK,
CKR_OPERATION_ACTIVE, CKR_PIN_EXPIRED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN, CKR_OPERATION_CANCEL_FAILED.
Example: see C_VerifyFinal.
CK_DECLARE_FUNCTION(CK_RV, C_Verify)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pData,
CK_ULONG ulDataLen,
CK_BYTE_PTR pSignature,
CK_ULONG ulSignatureLen
);
C_Verify verifies a signature in a single-part operation,
where the signature is an appendix to the data. hSession is the
session’s handle; pData points to the data; ulDataLen is the
length of the data; pSignature points to the signature; ulSignatureLen
is the length of the signature.
The verification operation MUST have been initialized with C_VerifyInit.
A call to C_Verify always terminates the active verification operation.
A successful call to C_Verify should return either
the value CKR_OK (indicating that the supplied signature is valid) or CKR_SIGNATURE_INVALID
(indicating that the supplied signature is invalid). If the signature can be
seen to be invalid purely on the basis of its length, then
CKR_SIGNATURE_LEN_RANGE should be returned. In any of these cases, the active
signing operation is terminated.
C_Verify cannot be used to terminate a
multi-part operation, and MUST be called after C_VerifyInit without
intervening C_VerifyUpdate calls.
For most mechanisms, C_Verify is equivalent to a
sequence of C_VerifyUpdate operations followed by C_VerifyFinal.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DATA_INVALID, CKR_DATA_LEN_RANGE,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK,
CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID,
CKR_SIGNATURE_INVALID, CKR_SIGNATURE_LEN_RANGE, CKR_TOKEN_RESOURCE_EXCEEDED.
Example: see C_VerifyFinal for an example of similar
functions.
CK_DECLARE_FUNCTION(CK_RV, C_VerifyUpdate)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pPart,
CK_ULONG ulPartLen
);
C_VerifyUpdate continues a multiple-part verification
operation, processing another data part. hSession is the session’s
handle, pPart points to the data part; ulPartLen is the length of
the data part.
The verification operation MUST have been initialized with C_VerifyInit.
This function may be called any number of times in succession. A call to C_VerifyUpdate
which results in an error terminates the current verification operation.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DATA_LEN_RANGE, CKR_DEVICE_ERROR,
CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK,
CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID,
CKR_TOKEN_RESOURCE_EXCEEDED.
Example: see C_VerifyFinal.
CK_DECLARE_FUNCTION(CK_RV, C_VerifyFinal)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pSignature,
CK_ULONG ulSignatureLen
);
C_VerifyFinal finishes a multiple-part verification
operation, checking the signature. hSession is the session’s handle; pSignature
points to the signature; ulSignatureLen is the length of the signature.
The verification operation MUST have been initialized with C_VerifyInit.
A call to C_VerifyFinal always terminates the active verification
operation.
A successful call to C_VerifyFinal should return
either the value CKR_OK (indicating that the supplied signature is valid) or
CKR_SIGNATURE_INVALID (indicating that the supplied signature is invalid). If
the signature can be seen to be invalid purely on the basis of its length, then
CKR_SIGNATURE_LEN_RANGE should be returned. In any of these cases, the active
verifying operation is terminated.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DATA_LEN_RANGE, CKR_DEVICE_ERROR,
CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK,
CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID,
CKR_SIGNATURE_INVALID, CKR_SIGNATURE_LEN_RANGE, CKR_TOKEN_RESOURCE_EXCEEDED.
Example:
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hKey;
CK_MECHANISM mechanism = {
CKM_DES_MAC, NULL_PTR, 0
};
CK_BYTE data[] = {...};
CK_BYTE mac[4];
CK_RV rv;
.
.
rv = C_VerifyInit(hSession, &mechanism, hKey);
if (rv == CKR_OK) {
rv = C_VerifyUpdate(hSession, data, sizeof(data));
.
.
rv = C_VerifyFinal(hSession, mac, sizeof(mac));
.
.
}
CK_DECLARE_FUNCTION(CK_RV, C_VerifyRecoverInit)(
CK_SESSION_HANDLE hSession,
CK_MECHANISM_PTR pMechanism,
CK_OBJECT_HANDLE hKey
);
C_VerifyRecoverInit initializes a signature
verification operation, where the data is recovered from the signature. hSession
is the session’s handle; pMechanism points to the structure that
specifies the verification mechanism; hKey is the handle of the
verification key.
The CKA_VERIFY_RECOVER attribute of the verification
key, which indicates whether the key supports verification where the data is
recovered from the signature, MUST be CK_TRUE.
After calling C_VerifyRecoverInit, the application
may call C_VerifyRecover to verify a signature on data in a single
part. The verification operation is active until the application uses a call
to C_VerifyRecover to actually obtain the recovered message.
C_VerifyRecoverInit can be called
with pMechanism set to NULL_PTR to terminate an active verification with
data recovery operation. If an active operations has been initialized and it
cannot be cancelled, CKR_OPERATION_CANCEL_FAILED must be returned.
Return values: CKR_ARGUMENTS_BAD, CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_KEY_FUNCTION_NOT_PERMITTED, CKR_KEY_HANDLE_INVALID, CKR_KEY_SIZE_RANGE,
CKR_KEY_TYPE_INCONSISTENT, CKR_MECHANISM_INVALID, CKR_MECHANISM_PARAM_INVALID,
CKR_OK, CKR_OPERATION_ACTIVE, CKR_PIN_EXPIRED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN,
CKR_OPERATION_CANCEL_FAILED.
Example: see C_VerifyRecover.
CK_DECLARE_FUNCTION(CK_RV, C_VerifyRecover)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pSignature,
CK_ULONG ulSignatureLen,
CK_BYTE_PTR pData,
CK_ULONG_PTR pulDataLen
);
C_VerifyRecover verifies a signature in a single-part
operation, where the data is recovered from the signature. hSession is
the session’s handle; pSignature points to the signature; ulSignatureLen
is the length of the signature; pData points to the location that
receives the recovered data; and pulDataLen points to the location that
holds the length of the recovered data.
C_VerifyRecover uses the
convention described in Section 5.2 on producing output.
The verification operation MUST have been initialized with C_VerifyRecoverInit.
A call to C_VerifyRecover always terminates the active verification
operation unless it returns CKR_BUFFER_TOO_SMALL or is a successful call (i.e.,
one which returns CKR_OK) to determine the length of the buffer needed to hold
the recovered data.
A successful call to C_VerifyRecover should return
either the value CKR_OK (indicating that the supplied signature is valid) or
CKR_SIGNATURE_INVALID (indicating that the supplied signature is invalid). If
the signature can be seen to be invalid purely on the basis of its length, then
CKR_SIGNATURE_LEN_RANGE should be returned. The return codes
CKR_SIGNATURE_INVALID and CKR_SIGNATURE_LEN_RANGE have a higher priority than
the return code CKR_BUFFER_TOO_SMALL, i.e., if C_VerifyRecover is
supplied with an invalid signature, it will never return CKR_BUFFER_TOO_SMALL.
Return values: CKR_ARGUMENTS_BAD, CKR_BUFFER_TOO_SMALL,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DATA_INVALID, CKR_DATA_LEN_RANGE,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK,
CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID,
CKR_SIGNATURE_LEN_RANGE, CKR_SIGNATURE_INVALID, CKR_TOKEN_RESOURCE_EXCEEDED.
Example:
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hKey;
CK_MECHANISM mechanism = {
CKM_RSA_9796, NULL_PTR, 0
};
CK_BYTE data[] = {...};
CK_ULONG ulDataLen;
CK_BYTE signature[128];
CK_RV rv;
.
.
rv = C_VerifyRecoverInit(hSession, &mechanism, hKey);
if (rv == CKR_OK) {
ulDataLen = sizeof(data);
rv = C_VerifyRecover(
hSession, signature, sizeof(signature), data,
&ulDataLen);
.
.
}
Message-based verification refers to the process of
verifying signatures on multiple messages using the same verification mechanism
and verification key.
Cryptoki provides the following functions for verifying
signatures on messages (for the purposes of Cryptoki, these operations also
encompass message authentication codes).
CK_DECLARE_FUNCTION(CK_RV, C_MessageVerifyInit)(
CK_SESSION_HANDLE hSession,
CK_MECHANISM_PTR pMechanism,
CK_OBJECT_HANDLE hKey
);
C_MessageVerifyInit initializes a message-based
verification process, preparing a session for one or more verification
operations (where the signature is an appendix to the data) that use the same
verification mechanism and verification key. hSession is the session’s
handle; pMechanism points to the structure that specifies the
verification mechanism; hKey is the handle of the verification key.
The CKA_VERIFY attribute of the verification key,
which indicates whether the key supports verification where the signature is an
appendix to the data, MUST be CK_TRUE.
After calling C_MessageVerifyInit, the application
can either call C_VerifyMessage to verify a signature on a message in a
single part; or call C_VerifyMessageBegin, followed by C_VerifyMessageNext
one or more times, to verify a signature on a message in multiple parts. This
may be repeated several times. The message-based verification process is active
until the application calls C_MessageVerifyFinal to finish the
message-based verification process.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_KEY_FUNCTION_NOT_PERMITTED,
CKR_KEY_HANDLE_INVALID, CKR_KEY_SIZE_RANGE, CKR_KEY_TYPE_INCONSISTENT,
CKR_MECHANISM_INVALID, CKR_MECHANISM_PARAM_INVALID, CKR_OK,
CKR_OPERATION_ACTIVE, CKR_PIN_EXPIRED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN.
CK_DECLARE_FUNCTION(CK_RV, C_VerifyMessage)(
CK_SESSION_HANDLE hSession,
CK_VOID_PTR pParameter,
CK_ULONG ulParameterLen,
CK_BYTE_PTR pData,
CK_ULONG ulDataLen,
CK_BYTE_PTR pSignature,
CK_ULONG ulSignatureLen
);
C_VerifyMessage verifies a signature on a message in
a single part operation, where the signature is an appendix to the data. C_MessageVerifyInit
must previously been called on the session. hSession is the session’s
handle; pParameter and ulParameterLen specify any
mechanism-specific parameters for the message verification operation; pData
points to the data; ulDataLen is the length of the data; pSignature
points to the signature; ulSignatureLen is the length of the signature.
Unlike the pParameter parameter of C_SignMessage,
pParameter is always an input parameter.
The message-based verification process MUST have been
initialized with C_MessageVerifyInit. A call to C_VerifyMessage
starts and terminates a message verification operation.
A successful call to C_VerifyMessage should return
either the value CKR_OK (indicating that the supplied signature is valid) or
CKR_SIGNATURE_INVALID (indicating that the supplied signature is invalid). If
the signature can be seen to be invalid purely on the basis of its length, then
CKR_SIGNATURE_LEN_RANGE should be returned.
C_VerifyMessage does not finish the message-based
verification process. Additional C_VerifyMessage or C_VerifyMessageBegin
and C_VerifyMessageNext calls may be made on the session.
For most mechanisms, C_VerifyMessage is equivalent to
C_VerifyMessageBegin followed by a sequence of C_VerifyMessageNext
operations.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DATA_INVALID, CKR_DATA_LEN_RANGE,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK,
CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID,
CKR_SIGNATURE_INVALID, CKR_SIGNATURE_LEN_RANGE, CKR_TOKEN_RESOURCE_EXCEEDED.
CK_DECLARE_FUNCTION(CK_RV, C_VerifyMessageBegin)(
CK_SESSION_HANDLE hSession,
CK_VOID_PTR pParameter,
CK_ULONG ulParameterLen
);
C_VerifyMessageBegin begins a multiple-part message
verification operation, where the signature is an appendix to the message. C_MessageVerifyInit
must previously been called on the session. hSession is the session’s
handle; pParameter and ulParameterLen specify any mechanism-specific
parameters for the message verification operation.
Unlike the pParameter parameter of C_SignMessageBegin,
pParameter is always an input parameter.
After calling C_VerifyMessageBegin, the application
should call C_VerifyMessageNext one or more times to verify a signature
on a message in multiple parts. The message verification operation is active
until the application calls C_VerifyMessageNext with a non-NULL pSignature.
To process additional messages (in single or multiple parts), the application MUST
call C_VerifyMessage or C_VerifyMessageBegin again.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_OPERATION_ACTIVE,
CKR_PIN_EXPIRED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID,
CKR_USER_NOT_LOGGED_IN.
CK_DECLARE_FUNCTION(CK_RV, C_VerifyMessageNext)(
CK_SESSION_HANDLE hSession,
CK_VOID_PTR pParameter,
CK_ULONG ulParameterLen,
CK_BYTE_PTR pDataPart,
CK_ULONG ulDataPartLen,
CK_BYTE_PTR pSignature,
CK_ULONG ulSignatureLen
);
C_VerifyMessageNext continues a multiple-part message
verification operation, processing another data part, or finishes a multiple-part
message verification operation, checking the signature. hSession is the
session’s handle, pParameter and ulParameterLen specify any
mechanism-specific parameters for the message verification operation, pPart
points to the data part; ulPartLen is the length of the data part; pSignature
points to the signature; ulSignatureLen is the length of the signature.
The pSignature argument is set to NULL if there is
more data part to follow, or set to a non-NULL value (pointing to the signature
to verify) if this is the last data part.
The message verification operation MUST have been started
with C_VerifyMessageBegin. This function may be called any number of
times in succession. A call to C_VerifyMessageNext with a NULL pSignature
which results in an error terminates the current message verification
operation. A call to C_VerifyMessageNext with a non-NULL pSignature
always terminates the active message verification operation.
A successful call to C_VerifyMessageNext with a
non-NULL pSignature should return either the value CKR_OK (indicating
that the supplied signature is valid) or CKR_SIGNATURE_INVALID (indicating that
the supplied signature is invalid). If the signature can be seen to be invalid
purely on the basis of its length, then CKR_SIGNATURE_LEN_RANGE should be
returned. In any of these cases, the active message verifying operation is
terminated.
Although the last C_VerifyMessageNext call ends the
verification of a message, it does not finish the message-based verification
process. Additional C_VerifyMessage or C_VerifyMessageBegin and C_VerifyMessageNext
calls may be made on the session.
Return values: CKR_ARGUMENTS_BAD, CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DATA_LEN_RANGE, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED,
CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_OK, CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID,
CKR_SIGNATURE_INVALID, CKR_SIGNATURE_LEN_RANGE, CKR_TOKEN_RESOURCE_EXCEEDED.
CK_DECLARE_FUNCTION(CK_RV,C_MessageVerifyFinal)(
CK_SESSION_HANDLE hSession
);
C_MessageVerifyFinal finishes a message-based
verification process. hSession is the session’s handle.
The message-based verification process MUST have been
initialized with C_MessageVerifyInit.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DATA_LEN_RANGE, CKR_DEVICE_ERROR,
CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK,
CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID,
CKR_TOKEN_RESOURCE_EXCEEDED.
Cryptoki provides the following functions to perform two
cryptographic operations “simultaneously” within a session. These functions
are provided so as to avoid unnecessarily passing data back and forth to and
from a token.
CK_DECLARE_FUNCTION(CK_RV, C_DigestEncryptUpdate)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pPart,
CK_ULONG ulPartLen,
CK_BYTE_PTR pEncryptedPart,
CK_ULONG_PTR pulEncryptedPartLen
);
C_DigestEncryptUpdate continues multiple-part digest
and encryption operations, processing another data part. hSession is the
session’s handle; pPart points to the data part; ulPartLen is the
length of the data part; pEncryptedPart points to the location that
receives the digested and encrypted data part; pulEncryptedPartLen
points to the location that holds the length of the encrypted data part.
C_DigestEncryptUpdate uses the convention described
in Section 5.2 on producing output. If a C_DigestEncryptUpdate call
does not produce encrypted output (because an error occurs, or because pEncryptedPart
has the value NULL_PTR, or because pulEncryptedPartLen is too small to
hold the entire encrypted part output), then no plaintext is passed to the
active digest operation.
Digest and encryption operations MUST both be active (they
MUST have been initialized with C_DigestInit and C_EncryptInit,
respectively). This function may be called any number of times in succession,
and may be interspersed with C_DigestUpdate, C_DigestKey, and C_EncryptUpdate
calls (it would be somewhat unusual to intersperse calls to C_DigestEncryptUpdate
with calls to C_DigestKey, however).
Return values: CKR_ARGUMENTS_BAD, CKR_BUFFER_TOO_SMALL,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DATA_LEN_RANGE, CKR_DEVICE_ERROR,
CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK,
CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID.
Example:
#define BUF_SZ 512
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hKey;
CK_BYTE iv[8];
CK_MECHANISM digestMechanism = {
CKM_MD5, NULL_PTR, 0
};
CK_MECHANISM encryptionMechanism = {
CKM_DES_ECB, iv, sizeof(iv)
};
CK_BYTE encryptedData[BUF_SZ];
CK_ULONG ulEncryptedDataLen;
CK_BYTE digest[16];
CK_ULONG ulDigestLen;
CK_BYTE data[(2*BUF_SZ)+8];
CK_RV rv;
int i;
.
.
memset(iv, 0, sizeof(iv));
memset(data, ‘A’, ((2*BUF_SZ)+5));
rv = C_EncryptInit(hSession, &encryptionMechanism,
hKey);
if (rv != CKR_OK) {
.
.
}
rv = C_DigestInit(hSession, &digestMechanism);
if (rv != CKR_OK) {
.
.
}
ulEncryptedDataLen = sizeof(encryptedData);
rv = C_DigestEncryptUpdate(
hSession,
&data[0], BUF_SZ,
encryptedData, &ulEncryptedDataLen);
.
.
ulEncryptedDataLen = sizeof(encryptedData);
rv = C_DigestEncryptUpdate(
hSession,
&data[BUF_SZ], BUF_SZ,
encryptedData, &ulEncryptedDataLen);
.
.
/*
* The last portion of the buffer needs to be
* handled with separate calls to deal with
* padding issues in ECB mode
*/
/* First, complete the digest on the buffer */
rv = C_DigestUpdate(hSession, &data[BUF_SZ*2],
5);
.
.
ulDigestLen = sizeof(digest);
rv = C_DigestFinal(hSession, digest,
&ulDigestLen);
.
.
/* Then, pad last part with 3 0x00 bytes, and complete
encryption */
for(i=0;i<3;i++)
data[((BUF_SZ*2)+5)+i] = 0x00;
/* Now, get second-to-last piece of ciphertext */
ulEncryptedDataLen = sizeof(encryptedData);
rv = C_EncryptUpdate(
hSession,
&data[BUF_SZ*2], 8,
encryptedData, &ulEncryptedDataLen);
.
.
/* Get last piece of ciphertext (should have length 0, here)
*/
ulEncryptedDataLen = sizeof(encryptedData);
rv = C_EncryptFinal(hSession, encryptedData,
&ulEncryptedDataLen);
.
.
CK_DECLARE_FUNCTION(CK_RV, C_DecryptDigestUpdate)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pEncryptedPart,
CK_ULONG ulEncryptedPartLen,
CK_BYTE_PTR pPart,
CK_ULONG_PTR pulPartLen
);
C_DecryptDigestUpdate continues a multiple-part
combined decryption and digest operation, processing another data part. hSession
is the session’s handle; pEncryptedPart points to the encrypted data
part; ulEncryptedPartLen is the length of the encrypted data part; pPart
points to the location that receives the recovered data part; pulPartLen
points to the location that holds the length of the recovered data part.
C_DecryptDigestUpdate uses the convention described
in Section 5.2 on producing output. If a C_DecryptDigestUpdate call does
not produce decrypted output (because an error occurs, or because pPart
has the value NULL_PTR, or because pulPartLen is too small to hold the
entire decrypted part output), then no plaintext is passed to the active digest
operation.
Decryption and digesting operations MUST both be active
(they MUST have been initialized with C_DecryptInit and C_DigestInit,
respectively). This function may be called any number of times in succession,
and may be interspersed with C_DecryptUpdate, C_DigestUpdate, and
C_DigestKey calls (it would be somewhat unusual to intersperse calls to C_DigestEncryptUpdate
with calls to C_DigestKey, however).
Use of C_DecryptDigestUpdate involves a pipelining
issue that does not arise when using C_DigestEncryptUpdate, the “inverse
function” of C_DecryptDigestUpdate. This is because when C_DigestEncryptUpdate
is called, precisely the same input is passed to both the active digesting
operation and the active encryption operation; however, when C_DecryptDigestUpdate
is called, the input passed to the active digesting operation is the output
of the active decryption operation. This issue comes up only when the
mechanism used for decryption performs padding.
In particular, envision a 24-byte ciphertext which was
obtained by encrypting an 18-byte plaintext with DES in CBC mode with PKCS padding.
Consider an application which will simultaneously decrypt this ciphertext and
digest the original plaintext thereby obtained.
After initializing decryption and digesting operations, the
application passes the 24-byte ciphertext (3 DES blocks) into C_DecryptDigestUpdate.
C_DecryptDigestUpdate returns exactly 16 bytes of plaintext, since at
this point, Cryptoki doesn’t know if there’s more ciphertext coming, or if the
last block of ciphertext held any padding. These 16 bytes of plaintext are
passed into the active digesting operation.
Since there is no more ciphertext, the application calls C_DecryptFinal.
This tells Cryptoki that there’s no more ciphertext coming, and the call
returns the last 2 bytes of plaintext. However, since the active decryption
and digesting operations are linked only through the C_DecryptDigestUpdate
call, these 2 bytes of plaintext are not passed on to be digested.
A call to C_DigestFinal, therefore, would compute the
message digest of the first 16 bytes of the plaintext, not the message
digest of the entire plaintext. It is crucial that, before C_DigestFinal
is called, the last 2 bytes of plaintext get passed into the active digesting
operation via a C_DigestUpdate call.
Because of this, it is critical that when an application
uses a padded decryption mechanism with C_DecryptDigestUpdate, it knows
exactly how much plaintext has been passed into the active digesting
operation. Extreme caution is warranted when using a padded decryption
mechanism with C_DecryptDigestUpdate.
Return values: CKR_ARGUMENTS_BAD, CKR_BUFFER_TOO_SMALL,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_ENCRYPTED_DATA_INVALID, CKR_ENCRYPTED_DATA_LEN_RANGE,
CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_OK, CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID.
Example:
#define BUF_SZ 512
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hKey;
CK_BYTE iv[8];
CK_MECHANISM decryptionMechanism = {
CKM_DES_ECB, iv, sizeof(iv)
};
CK_MECHANISM digestMechanism = {
CKM_MD5, NULL_PTR, 0
};
CK_BYTE encryptedData[(2*BUF_SZ)+8];
CK_BYTE digest[16];
CK_ULONG ulDigestLen;
CK_BYTE data[BUF_SZ];
CK_ULONG ulDataLen, ulLastUpdateSize;
CK_RV rv;
.
.
memset(iv, 0, sizeof(iv));
memset(encryptedData, ‘A’, ((2*BUF_SZ)+8));
rv = C_DecryptInit(hSession, &decryptionMechanism,
hKey);
if (rv != CKR_OK) {
.
.
}
rv = C_DigestInit(hSession, &digestMechanism);
if (rv != CKR_OK){
.
.
}
ulDataLen = sizeof(data);
rv = C_DecryptDigestUpdate(
hSession,
&encryptedData[0], BUF_SZ,
data, &ulDataLen);
.
.
ulDataLen = sizeof(data);
rv = C_DecryptDigestUpdate(
hSession,
&encryptedData[BUF_SZ], BUF_SZ,
data, &ulDataLen);
.
.
/*
* The last portion of the buffer needs to be handled with
* separate calls to deal with padding issues in ECB mode
*/
/* First, complete the decryption of the buffer */
ulLastUpdateSize = sizeof(data);
rv = C_DecryptUpdate(
hSession,
&encryptedData[BUF_SZ*2], 8,
data, &ulLastUpdateSize);
.
.
/* Get last piece of plaintext (should have length 0, here)
*/
ulDataLen = sizeof(data)-ulLastUpdateSize;
rv = C_DecryptFinal(hSession, &data[ulLastUpdateSize],
&ulDataLen);
if (rv != CKR_OK) {
.
.
}
/* Digest last bit of plaintext */
rv = C_DigestUpdate(hSession, data, 5);
if (rv != CKR_OK) {
.
.
}
ulDigestLen = sizeof(digest);
rv = C_DigestFinal(hSession, digest,
&ulDigestLen);
if (rv != CKR_OK) {
.
.
}
CK_DECLARE_FUNCTION(CK_RV, C_SignEncryptUpdate)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pPart,
CK_ULONG ulPartLen,
CK_BYTE_PTR pEncryptedPart,
CK_ULONG_PTR pulEncryptedPartLen
);
C_SignEncryptUpdate continues a multiple-part
combined signature and encryption operation, processing another data part. hSession
is the session’s handle; pPart points to the data part; ulPartLen
is the length of the data part; pEncryptedPart points to the location
that receives the digested and encrypted data part; and pulEncryptedPartLen
points to the location that holds the length of the encrypted data part.
C_SignEncryptUpdate uses the convention described in
Section 5.2 on producing output. If a C_SignEncryptUpdate call does not
produce encrypted output (because an error occurs, or because pEncryptedPart
has the value NULL_PTR, or because pulEncryptedPartLen is too small to
hold the entire encrypted part output), then no plaintext is passed to the
active signing operation.
Signature and encryption operations MUST both be active
(they MUST have been initialized with C_SignInit and C_EncryptInit,
respectively). This function may be called any number of times in succession,
and may be interspersed with C_SignUpdate and C_EncryptUpdate
calls.
Return values: CKR_ARGUMENTS_BAD, CKR_BUFFER_TOO_SMALL,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DATA_LEN_RANGE, CKR_DEVICE_ERROR,
CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK,
CKR_OPERATION_NOT_INITIALIZED, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID,
CKR_USER_NOT_LOGGED_IN.
Example:
#define BUF_SZ 512
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hEncryptionKey, hMacKey;
CK_BYTE iv[8];
CK_MECHANISM signMechanism = {
CKM_DES_MAC, NULL_PTR, 0
};
CK_MECHANISM encryptionMechanism = {
CKM_DES_ECB, iv, sizeof(iv)
};
CK_BYTE encryptedData[BUF_SZ];
CK_ULONG ulEncryptedDataLen;
CK_BYTE MAC[4];
CK_ULONG ulMacLen;
CK_BYTE data[(2*BUF_SZ)+8];
CK_RV rv;
int i;
.
.
memset(iv, 0, sizeof(iv));
memset(data, ‘A’, ((2*BUF_SZ)+5));
rv = C_EncryptInit(hSession, &encryptionMechanism,
hEncryptionKey);
if (rv != CKR_OK) {
.
.
}
rv = C_SignInit(hSession, &signMechanism, hMacKey);
if (rv != CKR_OK) {
.
.
}
ulEncryptedDataLen = sizeof(encryptedData);
rv = C_SignEncryptUpdate(
hSession,
&data[0], BUF_SZ,
encryptedData, &ulEncryptedDataLen);
.
.
ulEncryptedDataLen = sizeof(encryptedData);
rv = C_SignEncryptUpdate(
hSession,
&data[BUF_SZ], BUF_SZ,
encryptedData, &ulEncryptedDataLen);
.
.
/*
* The last portion of the buffer needs to be handled with
* separate calls to deal with padding issues in ECB mode
*/
/* First, complete the signature on the buffer */
rv = C_SignUpdate(hSession, &data[BUF_SZ*2], 5);
.
.
ulMacLen = sizeof(MAC);
rv = C_SignFinal(hSession, MAC, &ulMacLen);
.
.
/* Then pad last part with 3 0x00 bytes, and complete
encryption */
for(i=0;i<3;i++)
data[((BUF_SZ*2)+5)+i] = 0x00;
/* Now, get second-to-last piece of ciphertext */
ulEncryptedDataLen = sizeof(encryptedData);
rv = C_EncryptUpdate(
hSession,
&data[BUF_SZ*2], 8,
encryptedData, &ulEncryptedDataLen);
.
.
/* Get last piece of ciphertext (should have length 0, here)
*/
ulEncryptedDataLen = sizeof(encryptedData);
rv = C_EncryptFinal(hSession, encryptedData,
&ulEncryptedDataLen);
.
.
CK_DECLARE_FUNCTION(CK_RV, C_DecryptVerifyUpdate)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pEncryptedPart,
CK_ULONG ulEncryptedPartLen,
CK_BYTE_PTR pPart,
CK_ULONG_PTR pulPartLen
);
C_DecryptVerifyUpdate
continues a multiple-part combined decryption and verification operation,
processing another data part. hSession is the session’s handle; pEncryptedPart
points to the encrypted data; ulEncryptedPartLen is the length of the
encrypted data; pPart points to the location that receives the recovered
data; and pulPartLen points to the location that holds the length of the
recovered data.
C_DecryptVerifyUpdate uses the convention described
in Section 5.2 on producing output. If a C_DecryptVerifyUpdate call
does not produce decrypted output (because an error occurs, or because pPart
has the value NULL_PTR, or because pulPartLen is too small to hold the
entire encrypted part output), then no plaintext is passed to the active
verification operation.
Decryption and signature operations MUST both be active
(they MUST have been initialized with C_DecryptInit and C_VerifyInit,
respectively). This function may be called any number of times in succession,
and may be interspersed with C_DecryptUpdate and C_VerifyUpdate
calls.
Use of C_DecryptVerifyUpdate involves a pipelining
issue that does not arise when using C_SignEncryptUpdate, the “inverse
function” of C_DecryptVerifyUpdate. This is because when C_SignEncryptUpdate
is called, precisely the same input is passed to both the active signing
operation and the active encryption operation; however, when C_DecryptVerifyUpdate
is called, the input passed to the active verifying operation is the output
of the active decryption operation. This issue comes up only when the
mechanism used for decryption performs padding.
In particular, envision a 24-byte ciphertext which was
obtained by encrypting an 18-byte plaintext with DES in CBC mode with PKCS
padding. Consider an application which will simultaneously decrypt this
ciphertext and verify a signature on the original plaintext thereby obtained.
After initializing decryption and verification operations,
the application passes the 24-byte ciphertext (3 DES blocks) into C_DecryptVerifyUpdate.
C_DecryptVerifyUpdate returns exactly 16 bytes of plaintext, since at
this point, Cryptoki doesn’t know if there’s more ciphertext coming, or if the
last block of ciphertext held any padding. These 16 bytes of plaintext are
passed into the active verification operation.
Since there is no more ciphertext, the application calls C_DecryptFinal.
This tells Cryptoki that there’s no more ciphertext coming, and the call
returns the last 2 bytes of plaintext. However, since the active decryption
and verification operations are linked only through the C_DecryptVerifyUpdate
call, these 2 bytes of plaintext are not passed on to the verification
mechanism.
A call to C_VerifyFinal, therefore, would verify
whether or not the signature supplied is a valid signature on the first 16
bytes of the plaintext, not on the entire plaintext. It is crucial that,
before C_VerifyFinal is called, the last 2 bytes of plaintext get passed
into the active verification operation via a C_VerifyUpdate call.
Because of this, it is critical that when an application
uses a padded decryption mechanism with C_DecryptVerifyUpdate, it knows
exactly how much plaintext has been passed into the active verification
operation. Extreme caution is warranted when using a padded decryption
mechanism with C_DecryptVerifyUpdate.
Return values: CKR_ARGUMENTS_BAD, CKR_BUFFER_TOO_SMALL,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DATA_LEN_RANGE, CKR_DEVICE_ERROR,
CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_ENCRYPTED_DATA_INVALID,
CKR_ENCRYPTED_DATA_LEN_RANGE, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_OPERATION_NOT_INITIALIZED,
CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID.
Example:
#define BUF_SZ 512
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hDecryptionKey, hMacKey;
CK_BYTE iv[8];
CK_MECHANISM decryptionMechanism = {
CKM_DES_ECB, iv, sizeof(iv)
};
CK_MECHANISM verifyMechanism = {
CKM_DES_MAC, NULL_PTR, 0
};
CK_BYTE encryptedData[(2*BUF_SZ)+8];
CK_BYTE MAC[4];
CK_ULONG ulMacLen;
CK_BYTE data[BUF_SZ];
CK_ULONG ulDataLen, ulLastUpdateSize;
CK_RV rv;
.
.
memset(iv, 0, sizeof(iv));
memset(encryptedData, ‘A’, ((2*BUF_SZ)+8));
rv = C_DecryptInit(hSession, &decryptionMechanism,
hDecryptionKey);
if (rv != CKR_OK) {
.
.
}
rv = C_VerifyInit(hSession, &verifyMechanism, hMacKey);
if (rv != CKR_OK){
.
.
}
ulDataLen = sizeof(data);
rv = C_DecryptVerifyUpdate(
hSession,
&encryptedData[0], BUF_SZ,
data, &ulDataLen);
.
.
ulDataLen = sizeof(data);
rv = C_DecryptVerifyUpdate(
hSession,
&encryptedData[BUF_SZ], BUF_SZ,
data, &ulDataLen);
.
.
/*
* The last portion of the buffer needs to be handled with
* separate calls to deal with padding issues in ECB mode
*/
/* First, complete the decryption of the buffer */
ulLastUpdateSize = sizeof(data);
rv = C_DecryptUpdate(
hSession,
&encryptedData[BUF_SZ*2], 8,
data, &ulLastUpdateSize);
.
.
/* Get last little piece of plaintext. Should have length 0
*/
ulDataLen = sizeof(data)-ulLastUpdateSize;
rv = C_DecryptFinal(hSession, &data[ulLastUpdateSize],
&ulDataLen);
if (rv != CKR_OK) {
.
.
}
/* Send last bit of plaintext to verification operation */
rv = C_VerifyUpdate(hSession, data, 5);
if (rv != CKR_OK) {
.
.
}
rv = C_VerifyFinal(hSession, MAC, ulMacLen);
if (rv == CKR_SIGNATURE_INVALID) {
.
.
}
Cryptoki provides the following functions for key
management:
CK_DECLARE_FUNCTION(CK_RV, C_GenerateKey)(
CK_SESSION_HANDLE hSession
CK_MECHANISM_PTR pMechanism,
CK_ATTRIBUTE_PTR pTemplate,
CK_ULONG ulCount,
CK_OBJECT_HANDLE_PTR phKey
);
C_GenerateKey generates a secret key or set of domain
parameters, creating a new object. hSession is the session’s handle; pMechanism
points to the generation mechanism; pTemplate points to the template for
the new key or set of domain parameters; ulCount is the number of
attributes in the template; phKey points to the location that receives
the handle of the new key or set of domain parameters.
If the generation mechanism is for domain parameter
generation, the CKA_CLASS attribute will have the value
CKO_DOMAIN_PARAMETERS; otherwise, it will have the value CKO_SECRET_KEY.
Since the type of key or domain parameters to be generated
is implicit in the generation mechanism, the template does not need to supply a
key type. If it does supply a key type which is inconsistent with the
generation mechanism, C_GenerateKey fails and returns the error code
CKR_TEMPLATE_INCONSISTENT. The CKA_CLASS attribute is treated similarly.
If a call to C_GenerateKey cannot support the precise
template supplied to it, it will fail and return without creating an object.
The object created by a successful call to C_GenerateKey
will have its CKA_LOCAL attribute set to CK_TRUE.
In addition, the object created will have a value for CKA_UNIQUE_ID generated
and assigned (See Section 4.4.1).
Return values: CKR_ARGUMENTS_BAD, CKR_ATTRIBUTE_READ_ONLY, CKR_ATTRIBUTE_TYPE_INVALID,
CKR_ATTRIBUTE_VALUE_INVALID, CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_CURVE_NOT_SUPPORTED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_MECHANISM_INVALID,
CKR_MECHANISM_PARAM_INVALID, CKR_OK, CKR_OPERATION_ACTIVE, CKR_PIN_EXPIRED,
CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID, CKR_SESSION_READ_ONLY,
CKR_TEMPLATE_INCOMPLETE, CKR_TEMPLATE_INCONSISTENT, CKR_TOKEN_WRITE_PROTECTED,
CKR_USER_NOT_LOGGED_IN.
Example:
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hKey;
CK_MECHANISM mechanism = {
CKM_DES_KEY_GEN, NULL_PTR, 0
};
CK_RV rv;
.
.
rv = C_GenerateKey(hSession, &mechanism, NULL_PTR, 0,
&hKey);
if (rv == CKR_OK) {
.
.
}
CK_DECLARE_FUNCTION(CK_RV, C_GenerateKeyPair)(
CK_SESSION_HANDLE hSession,
CK_MECHANISM_PTR pMechanism,
CK_ATTRIBUTE_PTR pPublicKeyTemplate,
CK_ULONG ulPublicKeyAttributeCount,
CK_ATTRIBUTE_PTR pPrivateKeyTemplate,
CK_ULONG ulPrivateKeyAttributeCount,
CK_OBJECT_HANDLE_PTR phPublicKey,
CK_OBJECT_HANDLE_PTR phPrivateKey
);
C_GenerateKeyPair generates a public/private key
pair, creating new key objects. hSession is the session’s handle; pMechanism
points to the key generation mechanism; pPublicKeyTemplate points to the
template for the public key; ulPublicKeyAttributeCount is the number of
attributes in the public-key template; pPrivateKeyTemplate points to the
template for the private key; ulPrivateKeyAttributeCount is the number
of attributes in the private-key template; phPublicKey points to the
location that receives the handle of the new public key; phPrivateKey
points to the location that receives the handle of the new private key.
Since the types of keys to be generated are implicit in the
key pair generation mechanism, the templates do not need to supply key types.
If one of the templates does supply a key type which is inconsistent with the
key generation mechanism, C_GenerateKeyPair fails and returns the error
code CKR_TEMPLATE_INCONSISTENT. The CKA_CLASS attribute is treated similarly.
If a call to C_GenerateKeyPair cannot support the
precise templates supplied to it, it will fail and return without creating any
key objects.
A call to C_GenerateKeyPair will never create just
one key and return. A call can fail, and create no keys; or it can succeed,
and create a matching public/private key pair.
The key objects created by a successful call to C_GenerateKeyPair
will have their CKA_LOCAL attributes set to CK_TRUE. In addition, the key objects created will both have
values for CKA_UNIQUE_ID generated and assigned (See Section 4.4.1).
Note carefully the order of the arguments to C_GenerateKeyPair.
The last two arguments do not have the same order as they did in the original
Cryptoki Version 1.0 document. The order of these two arguments has caused
some unfortunate confusion.
Return values: CKR_ARGUMENTS_BAD, CKR_ATTRIBUTE_READ_ONLY,
CKR_ATTRIBUTE_TYPE_INVALID, CKR_ATTRIBUTE_VALUE_INVALID,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_CURVE_NOT_SUPPORTED, CKR_DEVICE_ERROR,
CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_DOMAIN_PARAMS_INVALID,
CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_MECHANISM_INVALID, CKR_MECHANISM_PARAM_INVALID, CKR_OK,
CKR_OPERATION_ACTIVE, CKR_PIN_EXPIRED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_SESSION_READ_ONLY, CKR_TEMPLATE_INCOMPLETE,
CKR_TEMPLATE_INCONSISTENT, CKR_TOKEN_WRITE_PROTECTED, CKR_USER_NOT_LOGGED_IN.
Example:
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hPublicKey, hPrivateKey;
CK_MECHANISM mechanism = {
CKM_RSA_PKCS_KEY_PAIR_GEN, NULL_PTR, 0
};
CK_ULONG modulusBits = 3072;
CK_BYTE publicExponent[] = { 3 };
CK_BYTE subject[] = {...};
CK_BYTE id[] = {123};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE publicKeyTemplate[] = {
{CKA_ENCRYPT, &true, sizeof(true)},
{CKA_VERIFY, &true, sizeof(true)},
{CKA_WRAP, &true, sizeof(true)},
{CKA_MODULUS_BITS, &modulusBits, sizeof(modulusBits)},
{CKA_PUBLIC_EXPONENT, publicExponent, sizeof
(publicExponent)}
};
CK_ATTRIBUTE privateKeyTemplate[] = {
{CKA_TOKEN, &true, sizeof(true)},
{CKA_PRIVATE, &true, sizeof(true)},
{CKA_SUBJECT, subject, sizeof(subject)},
{CKA_ID, id, sizeof(id)},
{CKA_SENSITIVE, &true, sizeof(true)},
{CKA_DECRYPT, &true, sizeof(true)},
{CKA_SIGN, &true, sizeof(true)},
{CKA_UNWRAP, &true, sizeof(true)}
};
CK_RV rv;
rv = C_GenerateKeyPair(
hSession, &mechanism,
publicKeyTemplate, 5,
privateKeyTemplate, 8,
&hPublicKey, &hPrivateKey);
if (rv == CKR_OK) {
.
.
}
CK_DECLARE_FUNCTION(CK_RV, C_WrapKey)(
CK_SESSION_HANDLE hSession,
CK_MECHANISM_PTR pMechanism,
CK_OBJECT_HANDLE hWrappingKey,
CK_OBJECT_HANDLE hKey,
CK_BYTE_PTR pWrappedKey,
CK_ULONG_PTR pulWrappedKeyLen
);
C_WrapKey wraps (i.e., encrypts) a private or
secret key. hSession is the session’s handle; pMechanism points
to the wrapping mechanism; hWrappingKey is the handle of the wrapping
key; hKey is the handle of the key to be wrapped; pWrappedKey
points to the location that receives the wrapped key; and pulWrappedKeyLen
points to the location that receives the length of the wrapped key.
C_WrapKey uses the convention described in Section 5.2 on producing output.
The CKA_WRAP attribute of the wrapping key, which
indicates whether the key supports wrapping, MUST be CK_TRUE. The CKA_EXTRACTABLE
attribute of the key to be wrapped MUST also be CK_TRUE.
If the key to be wrapped cannot be wrapped for some
token-specific reason, despite its having its CKA_EXTRACTABLE attribute
set to CK_TRUE, then C_WrapKey fails with error code
CKR_KEY_NOT_WRAPPABLE. If it cannot be wrapped with the specified wrapping key
and mechanism solely because of its length, then C_WrapKey fails with
error code CKR_KEY_SIZE_RANGE.
C_WrapKey can be used in the following situations:
·
To wrap any secret key with a public key that supports encryption
and decryption.
·
To wrap any secret key with any other secret key. Consideration
MUST be given to key size and mechanism strength or the token may not allow the
operation.
·
To wrap a private key with any secret key.
Of course, tokens vary in which types of keys can actually
be wrapped with which mechanisms.
To partition the wrapping keys so they can only wrap a
subset of extractable keys the attribute CKA_WRAP_TEMPLATE can be used on the
wrapping key to specify an attribute set that will be compared against the
attributes of the key to be wrapped. If all attributes match according to the
C_FindObject rules of attribute matching then the wrap will proceed. The value
of this attribute is an attribute template and the size is the number of items
in the template times the size of CK_ATTRIBUTE. If this attribute is not
supplied then any template is acceptable. If an attribute is not present, it
will not be checked. If any attribute mismatch occurs on an attempt to wrap a
key then the function SHALL return CKR_KEY_HANDLE_INVALID.
Return Values: CKR_ARGUMENTS_BAD, CKR_BUFFER_TOO_SMALL,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_KEY_HANDLE_INVALID,
CKR_KEY_NOT_WRAPPABLE, CKR_KEY_SIZE_RANGE, CKR_KEY_UNEXTRACTABLE,
CKR_MECHANISM_INVALID, CKR_MECHANISM_PARAM_INVALID, CKR_OK,
CKR_OPERATION_ACTIVE, CKR_PIN_EXPIRED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN,
CKR_WRAPPING_KEY_HANDLE_INVALID, CKR_WRAPPING_KEY_SIZE_RANGE,
CKR_WRAPPING_KEY_TYPE_INCONSISTENT.
Example:
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hWrappingKey, hKey;
CK_MECHANISM mechanism = {
CKM_DES3_ECB, NULL_PTR, 0
};
CK_BYTE wrappedKey[8];
CK_ULONG ulWrappedKeyLen;
CK_RV rv;
.
.
ulWrappedKeyLen = sizeof(wrappedKey);
rv = C_WrapKey(
hSession, &mechanism,
hWrappingKey, hKey,
wrappedKey, &ulWrappedKeyLen);
if (rv == CKR_OK) {
.
.
}
CK_DECLARE_FUNCTION(CK_RV, C_UnwrapKey)(
CK_SESSION_HANDLE hSession,
CK_MECHANISM_PTR pMechanism,
CK_OBJECT_HANDLE hUnwrappingKey,
CK_BYTE_PTR pWrappedKey,
CK_ULONG ulWrappedKeyLen,
CK_ATTRIBUTE_PTR pTemplate,
CK_ULONG ulAttributeCount,
CK_OBJECT_HANDLE_PTR phKey
);
C_UnwrapKey unwraps (i.e. decrypts) a wrapped
key, creating a new private key or secret key object. hSession is the
session’s handle; pMechanism points to the unwrapping mechanism; hUnwrappingKey
is the handle of the unwrapping key; pWrappedKey points to the wrapped
key; ulWrappedKeyLen is the length of the wrapped key; pTemplate
points to the template for the new key; ulAttributeCount is the number
of attributes in the template; phKey points to the location that
receives the handle of the recovered key.
The CKA_UNWRAP attribute of the unwrapping key, which
indicates whether the key supports unwrapping, MUST be CK_TRUE.
The new key will have the CKA_ALWAYS_SENSITIVE
attribute set to CK_FALSE, and the CKA_NEVER_EXTRACTABLE attribute set
to CK_FALSE. The CKA_EXTRACTABLE attribute is by default set to CK_TRUE.
Some mechanisms may modify, or attempt to modify. the contents
of the pMechanism structure at the same time that the key is unwrapped.
If a call to C_UnwrapKey cannot support the precise
template supplied to it, it will fail and return without creating any key
object.
The key object created by a successful call to C_UnwrapKey
will have its CKA_LOCAL attribute set to CK_FALSE. In addition, the object created will have a value for
CKA_UNIQUE_ID generated and assigned (See Section 4.4.1).
To partition the unwrapping keys so they can only unwrap a
subset of keys the attribute CKA_UNWRAP_TEMPLATE can be used on the unwrapping
key to specify an attribute set that will be added to attributes of the key to
be unwrapped. If the attributes do not conflict with the user supplied
attribute template, in ‘pTemplate’, then the unwrap will proceed. The value of
this attribute is an attribute template and the size is the number of items in
the template times the size of CK_ATTRIBUTE. If this attribute is not present
on the unwrapping key then no additional attributes will be added. If any
attribute conflict occurs on an attempt to unwrap a key then the function SHALL
return CKR_TEMPLATE_INCONSISTENT.
Return values: CKR_ARGUMENTS_BAD, CKR_ATTRIBUTE_READ_ONLY,
CKR_ATTRIBUTE_TYPE_INVALID, CKR_ATTRIBUTE_VALUE_INVALID, CKR_BUFFER_TOO_SMALL,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_CURVE_NOT_SUPPORTED, CKR_DEVICE_ERROR,
CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_DOMAIN_PARAMS_INVALID,
CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_MECHANISM_INVALID, CKR_MECHANISM_PARAM_INVALID, CKR_OK,
CKR_OPERATION_ACTIVE, CKR_PIN_EXPIRED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_SESSION_READ_ONLY, CKR_TEMPLATE_INCOMPLETE,
CKR_TEMPLATE_INCONSISTENT, CKR_TOKEN_WRITE_PROTECTED,
CKR_UNWRAPPING_KEY_HANDLE_INVALID, CKR_UNWRAPPING_KEY_SIZE_RANGE,
CKR_UNWRAPPING_KEY_TYPE_INCONSISTENT, CKR_USER_NOT_LOGGED_IN,
CKR_WRAPPED_KEY_INVALID, CKR_WRAPPED_KEY_LEN_RANGE.
Example:
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hUnwrappingKey, hKey;
CK_MECHANISM mechanism = {
CKM_DES3_ECB, NULL_PTR, 0
};
CK_BYTE wrappedKey[8] = {...};
CK_OBJECT_CLASS keyClass = CKO_SECRET_KEY;
CK_KEY_TYPE keyType = CKK_DES;
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &keyClass, sizeof(keyClass)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_ENCRYPT, &true, sizeof(true)},
{CKA_DECRYPT, &true, sizeof(true)}
};
CK_RV rv;
.
.
rv = C_UnwrapKey(
hSession, &mechanism, hUnwrappingKey,
wrappedKey, sizeof(wrappedKey), template, 4, &hKey);
if (rv == CKR_OK) {
.
.
}
CK_DECLARE_FUNCTION(CK_RV, C_DeriveKey)(
CK_SESSION_HANDLE hSession,
CK_MECHANISM_PTR pMechanism,
CK_OBJECT_HANDLE hBaseKey,
CK_ATTRIBUTE_PTR pTemplate,
CK_ULONG ulAttributeCount,
CK_OBJECT_HANDLE_PTR phKey
);
C_DeriveKey derives a key from a base key, creating a
new key object. hSession is the session’s handle; pMechanism
points to a structure that specifies the key derivation mechanism; hBaseKey
is the handle of the base key; pTemplate points to the template for the
new key; ulAttributeCount is the number of attributes in the template;
and phKey points to the location that receives the handle of the derived
key.
The values of the CKA_SENSITIVE, CKA_ALWAYS_SENSITIVE,
CKA_EXTRACTABLE, and CKA_NEVER_EXTRACTABLE attributes for the
base key affect the values that these attributes can hold for the newly-derived
key. See the description of each particular key-derivation mechanism in
Section 5.21.2 for any constraints of this type.
If a call to C_DeriveKey cannot support the precise
template supplied to it, it will fail and return without creating any key
object.
The key object created by a successful call to C_DeriveKey
will have its CKA_LOCAL attribute set to CK_FALSE. In addition, the object created will have a value for
CKA_UNIQUE_ID generated and assigned (See Section 4.4.1).
To partition the derivation keys so they can only derive a
subset of keys the attribute CKA_DERIVE_TEMPLATE can be used on the derivation
keys to specify an attribute set that will be added to attributes of the key to
be derived. If the attributes do not conflict with the user supplied attribute
template, in ‘pTemplate’, then the derivation will proceed. The value of this
attribute is an attribute template and the size is the number of items in the
template times the size of CK_ATTRIBUTE. If this attribute is not present on
the base derivation keys then no additional attributes will be added. If any
attribute conflict occurs on an attempt to derive a key then the function SHALL
return CKR_TEMPLATE_INCONSISTENT.
Return values: CKR_ARGUMENTS_BAD, CKR_ATTRIBUTE_READ_ONLY,
CKR_ATTRIBUTE_TYPE_INVALID, CKR_ATTRIBUTE_VALUE_INVALID,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_CURVE_NOT_SUPPORTED, CKR_DEVICE_ERROR,
CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_DOMAIN_PARAMS_INVALID,
CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY,
CKR_KEY_HANDLE_INVALID, CKR_KEY_SIZE_RANGE, CKR_KEY_TYPE_INCONSISTENT,
CKR_MECHANISM_INVALID, CKR_MECHANISM_PARAM_INVALID, CKR_OK,
CKR_OPERATION_ACTIVE, CKR_PIN_EXPIRED, CKR_SESSION_CLOSED,
CKR_SESSION_HANDLE_INVALID, CKR_SESSION_READ_ONLY, CKR_TEMPLATE_INCOMPLETE,
CKR_TEMPLATE_INCONSISTENT, CKR_TOKEN_WRITE_PROTECTED, CKR_USER_NOT_LOGGED_IN.
Example:
CK_SESSION_HANDLE hSession;
CK_OBJECT_HANDLE hPublicKey, hPrivateKey, hKey;
CK_MECHANISM keyPairMechanism = {
CKM_DH_PKCS_KEY_PAIR_GEN, NULL_PTR, 0
};
CK_BYTE prime[] = {...};
CK_BYTE base[] = {...};
CK_BYTE publicValue[128];
CK_BYTE otherPublicValue[128];
CK_MECHANISM mechanism = {
CKM_DH_PKCS_DERIVE, otherPublicValue,
sizeof(otherPublicValue)
};
CK_ATTRIBUTE template[] = {
{CKA_VALUE, &publicValue, sizeof(publicValue)}
};
CK_OBJECT_CLASS keyClass = CKO_SECRET_KEY;
CK_KEY_TYPE keyType = CKK_DES;
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE publicKeyTemplate[] = {
{CKA_PRIME, prime, sizeof(prime)},
{CKA_BASE, base, sizeof(base)}
};
CK_ATTRIBUTE privateKeyTemplate[] = {
{CKA_DERIVE, &true, sizeof(true)}
};
CK_ATTRIBUTE derivedKeyTemplate[] = {
{CKA_CLASS, &keyClass, sizeof(keyClass)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_ENCRYPT, &true, sizeof(true)},
{CKA_DECRYPT, &true, sizeof(true)}
};
CK_RV rv;
.
.
rv = C_GenerateKeyPair(
hSession, &keyPairMechanism,
publicKeyTemplate, 2,
privateKeyTemplate, 1,
&hPublicKey, &hPrivateKey);
if (rv == CKR_OK) {
rv = C_GetAttributeValue(hSession, hPublicKey, template,
1);
if (rv == CKR_OK) {
/* Put other guy’s public value in otherPublicValue */
.
.
rv = C_DeriveKey(
hSession, &mechanism,
hPrivateKey, derivedKeyTemplate, 4, &hKey);
if (rv == CKR_OK) {
.
.
}
}
}
Cryptoki provides the following
functions for generating random numbers:
CK_DECLARE_FUNCTION(CK_RV, C_SeedRandom)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pSeed,
CK_ULONG ulSeedLen
);
C_SeedRandom mixes additional seed material into the
token’s random number generator. hSession is the session’s handle; pSeed
points to the seed material; and ulSeedLen is the length in bytes of the
seed material.
Return values: CKR_ARGUMENTS_BAD, CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY, CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED,
CKR_FUNCTION_FAILED, CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK,
CKR_OPERATION_ACTIVE, CKR_RANDOM_SEED_NOT_SUPPORTED, CKR_RANDOM_NO_RNG,
CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID, CKR_USER_NOT_LOGGED_IN.
Example: see C_GenerateRandom.
CK_DECLARE_FUNCTION(CK_RV, C_GenerateRandom)(
CK_SESSION_HANDLE hSession,
CK_BYTE_PTR pRandomData,
CK_ULONG ulRandomLen
);
C_GenerateRandom generates random or pseudo-random
data. hSession is the session’s handle; pRandomData points to the
location that receives the random data; and ulRandomLen is the length in
bytes of the random or pseudo-random data to be generated.
Return values: CKR_ARGUMENTS_BAD,
CKR_CRYPTOKI_NOT_INITIALIZED, CKR_DEVICE_ERROR, CKR_DEVICE_MEMORY,
CKR_DEVICE_REMOVED, CKR_FUNCTION_CANCELED, CKR_FUNCTION_FAILED,
CKR_GENERAL_ERROR, CKR_HOST_MEMORY, CKR_OK, CKR_OPERATION_ACTIVE,
CKR_RANDOM_NO_RNG, CKR_SESSION_CLOSED, CKR_SESSION_HANDLE_INVALID,
CKR_USER_NOT_LOGGED_IN.
Example:
CK_SESSION_HANDLE hSession;
CK_BYTE seed[] = {...};
CK_BYTE randomData[] = {...};
CK_RV rv;
.
.
rv = C_SeedRandom(hSession, seed, sizeof(seed));
if (rv != CKR_OK) {
.
.
}
rv = C_GenerateRandom(hSession, randomData,
sizeof(randomData));
if (rv == CKR_OK) {
.
.
}
Cryptoki provides the following functions for managing
parallel execution of cryptographic functions. These functions exist only for backwards
compatibility.
CK_DECLARE_FUNCTION(CK_RV, C_GetFunctionStatus)(
CK_SESSION_HANDLE hSession
);
In previous versions of Cryptoki, C_GetFunctionStatus
obtained the status of a function running in parallel with an application. Now,
however, C_GetFunctionStatus is a legacy function which should simply
return the value CKR_FUNCTION_NOT_PARALLEL.
Return values: CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_FUNCTION_FAILED, CKR_FUNCTION_NOT_PARALLEL, CKR_GENERAL_ERROR,
CKR_HOST_MEMORY, CKR_SESSION_HANDLE_INVALID, CKR_SESSION_CLOSED.
CK_DECLARE_FUNCTION(CK_RV, C_CancelFunction)(
CK_SESSION_HANDLE hSession
);
In previous versions of Cryptoki, C_CancelFunction
cancelled a function running in parallel with an application. Now, however, C_CancelFunction
is a legacy function which should simply return the value
CKR_FUNCTION_NOT_PARALLEL.
Return values: CKR_CRYPTOKI_NOT_INITIALIZED,
CKR_FUNCTION_FAILED, CKR_FUNCTION_NOT_PARALLEL, CKR_GENERAL_ERROR,
CKR_HOST_MEMORY, CKR_SESSION_HANDLE_INVALID, CKR_SESSION_CLOSED.
Cryptoki sessions can use function pointers of type CK_NOTIFY
to notify the application of certain events.
Cryptographic functions (i.e., any functions falling
under one of these categories: encryption functions; decryption functions;
message digesting functions; signing and MACing functions; functions for
verifying signatures and MACs; dual-purpose cryptographic functions; key
management functions; random number generation functions) executing in Cryptoki
sessions can periodically surrender control to the application who called them
if the session they are executing in had a notification callback function
associated with it when it was opened. They do this by calling the session’s callback
with the arguments (hSession, CKN_SURRENDER, pApplication), where hSession is
the session’s handle and pApplication was supplied to C_OpenSession when
the session was opened. Surrender callbacks should return either the value
CKR_OK (to indicate that Cryptoki should continue executing the function) or
the value CKR_CANCEL (to indicate that Cryptoki should abort execution of the
function). Of course, before returning one of these values, the callback
function can perform some computation, if desired.
A typical use of a surrender callback might be to give an
application user feedback during a lengthy key pair generation operation. Each
time the application receives a callback, it could display an additional “.” to
the user. It might also examine the keyboard’s activity since the last
surrender callback, and abort the key pair generation operation (probably by
returning the value CKR_CANCEL) if the user hit <ESCAPE>.
A Cryptoki library is not required to make any
surrender callbacks.
Library vendors can also define additional types of
callbacks. Because of this extension capability, application-supplied
notification callback routines should examine each callback they receive, and
if they are unfamiliar with the type of that callback, they should immediately
give control back to the library by returning with the value CKR_OK.
Table 32,
Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_RSA_PKCS_KEY_PAIR_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_RSA_X9_31_KEY_PAIR_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_RSA_PKCS
|
ü2
|
ü2
|
ü
|
|
|
ü
|
|
|
CKM_RSA_PKCS_OAEP
|
ü2
|
|
|
|
|
ü
|
|
|
CKM_RSA_PKCS_PSS
|
|
ü2
|
|
|
|
|
|
|
CKM_RSA_9796
|
|
ü2
|
ü
|
|
|
|
|
|
CKM_RSA_X_509
|
ü2
|
ü2
|
ü
|
|
|
ü
|
|
|
CKM_RSA_X9_31
|
|
ü2
|
|
|
|
|
|
|
CKM_SHA1_RSA_PKCS
|
|
ü
|
|
|
|
|
|
|
CKM_SHA224_RSA_PKCS
|
|
ü
|
|
|
|
|
|
|
CKM_SHA256_RSA_PKCS
|
|
ü
|
|
|
|
|
|
|
CKM_SHA384_RSA_PKCS
|
|
ü
|
|
|
|
|
|
|
CKM_SHA512_RSA_PKCS
|
|
ü
|
|
|
|
|
|
|
CKM_SHA1_RSA_PKCS_PSS
|
|
ü
|
|
|
|
|
|
|
CKM_SHA224_RSA_PKCS_PSS
|
|
ü
|
|
|
|
|
|
|
CKM_SHA256_RSA_PKCS_PSS
|
|
ü
|
|
|
|
|
|
|
CKM_SHA384_RSA_PKCS_PSS
|
|
ü
|
|
|
|
|
|
|
CKM_SHA512_RSA_PKCS_PSS
|
|
ü
|
|
|
|
|
|
|
CKM_SHA1_RSA_X9_31
|
|
ü
|
|
|
|
|
|
|
CKM_RSA_PKCS_TPM_1_1
|
ü2
|
|
|
|
|
ü
|
|
|
CKM_RSA_PKCS_OAEP_TPM_1_1
|
ü2
|
|
|
|
|
ü
|
|
|
CKM_SHA3_224_RSA_PKCS
|
|
ü
|
|
|
|
|
|
|
CKM_SHA3_256_RSA_PKCS
|
|
ü
|
|
|
|
|
|
|
CKM_SHA3_384_RSA_PKCS
|
|
ü
|
|
|
|
|
|
|
CKM_SHA3_512_RSA_PKCS
|
|
ü
|
|
|
|
|
|
|
CKM_SHA3_224_RSA_PKCS_PSS
|
|
ü
|
|
|
|
|
|
|
CKM_SHA3_256_RSA_PKCS_PSS
|
|
ü
|
|
|
|
|
|
|
CKM_SHA3_384_RSA_PKCS_PSS
|
|
ü
|
|
|
|
|
|
|
CKM_SHA3_512_RSA_PKCS_PSS
|
|
ü
|
|
|
|
|
|
This section defines the RSA key type “CKK_RSA” for type
CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of RSA key objects.
Mechanisms:
CKM_RSA_PKCS_KEY_PAIR_GEN
CKM_RSA_PKCS
CKM_RSA_9796
CKM_RSA_X_509
CKM_MD2_RSA_PKCS
CKM_MD5_RSA_PKCS
CKM_SHA1_RSA_PKCS
CKM_SHA224_RSA_PKCS
CKM_SHA256_RSA_PKCS
CKM_SHA384_RSA_PKCS
CKM_SHA512_RSA_PKCS
CKM_RIPEMD128_RSA_PKCS
CKM_RIPEMD160_RSA_PKCS
CKM_RSA_PKCS_OAEP
CKM_RSA_X9_31_KEY_PAIR_GEN
CKM_RSA_X9_31
CKM_SHA1_RSA_X9_31
CKM_RSA_PKCS_PSS
CKM_SHA1_RSA_PKCS_PSS
CKM_SHA224_RSA_PKCS_PSS
CKM_SHA256_RSA_PKCS_PSS
CKM_SHA512_RSA_PKCS_PSS
CKM_SHA384_RSA_PKCS_PSS
CKM_RSA_PKCS_TPM_1_1
CKM_RSA_PKCS_OAEP_TPM_1_1
CKM_RSA_AES_KEY_WRAP
CKM_SHA3_224_RSA_PKCS
CKM_SHA3_256_RSA_PKCS
CKM_SHA3_384_RSA_PKCS
CKM_SHA3_512_RSA_PKCS
CKM_SHA3_224_RSA_PKCS_PSS
CKM_SHA3_256_RSA_PKCS_PSS
CKM_SHA3_384_RSA_PKCS_PSS
CKM_SHA3_512_RSA_PKCS_PSS
6.1.2 RSA public key objects
RSA public key objects (object class CKO_PUBLIC_KEY, key
type CKK_RSA) hold RSA public keys. The following table defines the RSA
public key object attributes, in addition to the common attributes defined for
this object class:
Table 33, RSA Public Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_MODULUS1,4
|
Big integer
|
Modulus n
|
|
CKA_MODULUS_BITS2,3
|
CK_ULONG
|
Length in bits of modulus n
|
|
CKA_PUBLIC_EXPONENT1
|
Big integer
|
Public exponent e
|
- Refer to Table 11 for footnotes
Depending on the token, there may be limits on the length of
key components. See PKCS #1 for more information on RSA keys.
The following is a sample template for creating an RSA
public key object:
CK_OBJECT_CLASS class = CKO_PUBLIC_KEY;
CK_KEY_TYPE keyType = CKK_RSA;
CK_UTF8CHAR label[] = “An RSA public key object”;
CK_BYTE modulus[] = {...};
CK_BYTE exponent[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_WRAP, &true, sizeof(true)},
{CKA_ENCRYPT, &true, sizeof(true)},
{CKA_MODULUS, modulus, sizeof(modulus)},
{CKA_PUBLIC_EXPONENT, exponent, sizeof(exponent)}
};
RSA private key objects (object class CKO_PRIVATE_KEY, key
type CKK_RSA) hold RSA private keys. The following table defines the
RSA private key object attributes, in addition to the common attributes defined
for this object class:
Table 34, RSA Private Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_MODULUS1,4,6
|
Big integer
|
Modulus n
|
|
CKA_PUBLIC_EXPONENT1,4,6
|
Big integer
|
Public exponent e
|
|
CKA_PRIVATE_EXPONENT1,4,6,7
|
Big integer
|
Private exponent d
|
|
CKA_PRIME_14,6,7
|
Big integer
|
Prime p
|
|
CKA_PRIME_24,6,7
|
Big integer
|
Prime q
|
|
CKA_EXPONENT_14,6,7
|
Big integer
|
Private exponent d modulo p-1
|
|
CKA_EXPONENT_24,6,7
|
Big integer
|
Private exponent d modulo q-1
|
|
CKA_COEFFICIENT4,6,7
|
Big integer
|
CRT coefficient q-1 mod p
|
- Refer to Table 11 for footnotes
Depending on the token, there may be limits on the length of
the key components. See PKCS #1 for more information on RSA keys.
Tokens vary in what they actually store for RSA private
keys. Some tokens store all of the above attributes, which can assist in
performing rapid RSA computations. Other tokens might store only the CKA_MODULUS
and CKA_PRIVATE_EXPONENT values. Effective with
version 2.40, tokens MUST also store CKA_PUBLIC_EXPONENT. This
permits the retrieval of sufficient data to reconstitute the associated public
key.
Because of this, Cryptoki is flexible in dealing with RSA
private key objects. When a token generates an RSA private key, it stores
whichever of the fields in Table 34 it keeps track of. Later, if an
application asks for the values of the key’s various attributes, Cryptoki
supplies values only for attributes whose values it can obtain (i.e., if
Cryptoki is asked for the value of an attribute it cannot obtain, the request
fails). Note that a Cryptoki implementation may or may not be able and/or
willing to supply various attributes of RSA private keys which are not actually
stored on the token. E.g., if a particular token stores values only for
the CKA_PRIVATE_EXPONENT, CKA_PRIME_1, and CKA_PRIME_2
attributes, then Cryptoki is certainly able to report values for all the
attributes above (since they can all be computed efficiently from these three
values). However, a Cryptoki implementation may or may not actually do this
extra computation. The only attributes from Table 34 for which a Cryptoki implementation is required to be able to return
values are CKA_MODULUS, CKA_PUBLIC_EXPONENT and CKA_PRIVATE_EXPONENT.
A token SHOULD also be able to return CKA_PUBLIC_KEY_INFO for an RSA
private key.
If an RSA private key object is created on a token, and more
attributes from Table 34 are supplied to the object creation call than are
supported by the token, the extra attributes are likely to be thrown away. If
an attempt is made to create an RSA private key object on a token with
insufficient attributes for that particular token, then the object creation call
fails and returns CKR_TEMPLATE_INCOMPLETE.
Note that when generating an RSA private key, there is no CKA_MODULUS_BITS
attribute specified. This is because RSA private keys are only generated as
part of an RSA key pair, and the CKA_MODULUS_BITS attribute for
the pair is specified in the template for the RSA public key.
The following is a sample template for creating an RSA
private key object:
CK_OBJECT_CLASS class = CKO_PRIVATE_KEY;
CK_KEY_TYPE keyType = CKK_RSA;
CK_UTF8CHAR label[] = “An RSA private key object”;
CK_BYTE subject[] = {...};
CK_BYTE id[] = {123};
CK_BYTE modulus[] = {...};
CK_BYTE publicExponent[] = {...};
CK_BYTE privateExponent[] = {...};
CK_BYTE prime1[] = {...};
CK_BYTE prime2[] = {...};
CK_BYTE exponent1[] = {...};
CK_BYTE exponent2[] = {...};
CK_BYTE coefficient[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_SUBJECT, subject, sizeof(subject)},
{CKA_ID, id, sizeof(id)},
{CKA_SENSITIVE, &true, sizeof(true)},
{CKA_DECRYPT, &true, sizeof(true)},
{CKA_SIGN, &true, sizeof(true)},
{CKA_MODULUS, modulus, sizeof(modulus)},
{CKA_PUBLIC_EXPONENT, publicExponent, sizeof(publicExponent)},
{CKA_PRIVATE_EXPONENT, privateExponent,
sizeof(privateExponent)},
{CKA_PRIME_1, prime1, sizeof(prime1)},
{CKA_PRIME_2, prime2, sizeof(prime2)},
{CKA_EXPONENT_1, exponent1, sizeof(exponent1)},
{CKA_EXPONENT_2, exponent2, sizeof(exponent2)},
{CKA_COEFFICIENT, coefficient, sizeof(coefficient)}
};
The PKCS #1 RSA key pair generation mechanism, denoted CKM_RSA_PKCS_KEY_PAIR_GEN,
is a key pair generation mechanism based on the RSA public-key cryptosystem, as
defined in PKCS #1.
It does not have a parameter.
The mechanism generates RSA public/private key pairs with a
particular modulus length in bits and public exponent, as specified in the CKA_MODULUS_BITS
and CKA_PUBLIC_EXPONENT attributes of the template for the public key.
The CKA_PUBLIC_EXPONENT may be omitted in which case the mechanism
shall supply the public exponent attribute using the default value of
0x10001 (65537). Specific implementations may use a random value or an
alternative default if 0x10001 cannot be used by the token.
Note: Implementations strictly compliant with version
2.11 or prior versions may generate an error if this attribute is
omitted from the template. Experience has shown that many implementations of
2.11 and prior did allow the CKA_PUBLIC_EXPONENT attribute to be
omitted from the template, and behaved as described above. The mechanism
contributes the CKA_CLASS, CKA_KEY_TYPE, CKA_MODULUS, and CKA_PUBLIC_EXPONENT
attributes to the new public key. CKA_PUBLIC_EXPONENT will be
copied from the template if supplied. CKR_TEMPLATE_INCONSISTENT shall
be returned if the implementation cannot use the supplied exponent value. It
contributes the CKA_CLASS and CKA_KEY_TYPE attributes to the new
private key; it may also contribute some of the following attributes to the new
private key: CKA_MODULUS, CKA_PUBLIC_EXPONENT, CKA_PRIVATE_EXPONENT,
CKA_PRIME_1, CKA_PRIME_2, CKA_EXPONENT_1, CKA_EXPONENT_2,
CKA_COEFFICIENT. Other attributes supported by the RSA public and
private key types (specifically, the flags indicating which functions the keys
support) may also be specified in the templates for the keys, or else are
assigned default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
RSA modulus sizes, in bits.
The X9.31 RSA key pair generation mechanism, denoted CKM_RSA_X9_31_KEY_PAIR_GEN,
is a key pair generation mechanism based on the RSA public-key cryptosystem, as
defined in X9.31.
It does not have a parameter.
The mechanism generates RSA public/private key pairs with a
particular modulus length in bits and public exponent, as specified in the CKA_MODULUS_BITS
and CKA_PUBLIC_EXPONENT attributes of the template for the public key.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
CKA_MODULUS, and CKA_PUBLIC_EXPONENT attributes to the new public
key. It contributes the CKA_CLASS and CKA_KEY_TYPE attributes to
the new private key; it may also contribute some of the following attributes to
the new private key: CKA_MODULUS, CKA_PUBLIC_EXPONENT, CKA_PRIVATE_EXPONENT,
CKA_PRIME_1, CKA_PRIME_2, CKA_EXPONENT_1, CKA_EXPONENT_2,
CKA_COEFFICIENT. Other attributes supported by the RSA public and
private key types (specifically, the flags indicating which functions the keys
support) may also be specified in the templates for the keys, or else are
assigned default initial values. Unlike the CKM_RSA_PKCS_KEY_PAIR_GEN
mechanism, this mechanism is guaranteed to generate p and q
values, CKA_PRIME_1 and CKA_PRIME_2 respectively, that meet the
strong primes requirement of X9.31.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
RSA modulus sizes, in bits.
The PKCS #1 v1.5 RSA mechanism, denoted CKM_RSA_PKCS,
is a multi-purpose mechanism based on the RSA public-key cryptosystem and the
block formats initially defined in PKCS #1 v1.5. It supports single-part
encryption and decryption; single-part signatures and verification with and
without message recovery; key wrapping; and key unwrapping. This mechanism
corresponds only to the part of PKCS #1 v1.5 that involves RSA; it does not
compute a message digest or a DigestInfo encoding as specified for the
md2withRSAEncryption and md5withRSAEncryption algorithms in PKCS #1 v1.5 .
This mechanism does not have a parameter.
This mechanism can wrap and unwrap any secret key of appropriate
length. Of course, a particular token may not be able to wrap/unwrap every
appropriate-length secret key that it supports. For wrapping, the “input” to
the encryption operation is the value of the CKA_VALUE attribute of the
key that is wrapped; similarly for unwrapping. The mechanism does not wrap the
key type or any other information about the key, except the key length; the
application must convey these separately. In particular, the mechanism
contributes only the CKA_CLASS and CKA_VALUE (and CKA_VALUE_LEN,
if the key has it) attributes to the recovered key during unwrapping; other
attributes must be specified in the template.
Constraints on key types and the length of the data are
summarized in the following table. For encryption, decryption, signatures and
signature verification, the input and output data may begin at the same
location in memory. In the table, k is the length in bytes of the RSA
modulus.
Table
35, PKCS #1 v1.5 RSA: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
Comments
|
|
C_Encrypt1
|
RSA public key
|
≤ k-11
|
k
|
block type 02
|
|
C_Decrypt1
|
RSA private key
|
k
|
≤ k-11
|
block type 02
|
|
C_Sign1
|
RSA private key
|
≤ k-11
|
k
|
block type 01
|
|
C_SignRecover
|
RSA private key
|
≤ k-11
|
k
|
block type 01
|
|
C_Verify1
|
RSA public key
|
≤ k-11, k2
|
N/A
|
block type 01
|
|
C_VerifyRecover
|
RSA public key
|
k
|
≤ k-11
|
block type 01
|
|
C_WrapKey
|
RSA public key
|
≤ k-11
|
k
|
block type 02
|
|
C_UnwrapKey
|
RSA private key
|
k
|
≤ k-11
|
block type 02
|
1
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
RSA modulus sizes, in bits.
¨
CK_RSA_PKCS_MGF_TYPE;
CK_RSA_PKCS_MGF_TYPE_PTR
CK_RSA_PKCS_MGF_TYPE is used to indicate the Mask
Generation Function (MGF) applied to a message block when formatting a message
block for the PKCS #1 OAEP encryption scheme or the PKCS #1 PSS signature
scheme. It is defined as follows:
typedef CK_ULONG CK_RSA_PKCS_MGF_TYPE;
The following MGFs are defined in PKCS #1. The following
table lists the defined functions.
Table 36, PKCS #1 Mask Generation Functions
|
Source Identifier
|
Value
|
|
CKG_MGF1_SHA1
|
0x00000001UL
|
|
CKG_MGF1_SHA224
|
0x00000005UL
|
|
CKG_MGF1_SHA256
|
0x00000002UL
|
|
CKG_MGF1_SHA384
|
0x00000003UL
|
|
CKG_MGF1_SHA512
|
0x00000004UL
|
|
CKG_MGF1_SHA3_224
|
0x00000006UL
|
|
CKG_MGF1_SHA3_256
|
0x00000007UL
|
|
CKG_MGF1_SHA3_384
|
0x00000008UL
|
|
CKG_MGF1_SHA3_512
|
0x00000009UL
|
CK_RSA_PKCS_MGF_TYPE_PTR is a pointer to a CK_RSA_PKCS_MGF_TYPE.
¨
CK_RSA_PKCS_OAEP_SOURCE_TYPE;
CK_RSA_PKCS_OAEP_SOURCE_TYPE_PTR
CK_RSA_PKCS_OAEP_SOURCE_TYPE is used to indicate the
source of the encoding parameter when formatting a message block for the PKCS
#1 OAEP encryption scheme. It is defined as follows:
typedef CK_ULONG CK_RSA_PKCS_OAEP_SOURCE_TYPE;
The following encoding parameter sources are defined in PKCS
#1. The following table lists the defined sources along with the corresponding
data type for the pSourceData field in the CK_RSA_PKCS_OAEP_PARAMS
structure defined below.
Table 37, PKCS #1 RSA OAEP: Encoding parameter sources
|
Source Identifier
|
Value
|
Data Type
|
|
CKZ_DATA_SPECIFIED
|
0x00000001UL
|
Array
of CK_BYTE containing the value of the encoding parameter. If the parameter
is empty, pSourceData must be NULL and ulSourceDataLen must be
zero.
|
CK_RSA_PKCS_OAEP_SOURCE_TYPE_PTR is a pointer to a CK_RSA_PKCS_OAEP_SOURCE_TYPE.
¨
CK_RSA_PKCS_OAEP_PARAMS;
CK_RSA_PKCS_OAEP_PARAMS_PTR
CK_RSA_PKCS_OAEP_PARAMS is a structure that provides
the parameters to the CKM_RSA_PKCS_OAEP mechanism. The structure is
defined as follows:
typedef struct CK_RSA_PKCS_OAEP_PARAMS {
CK_MECHANISM_TYPE hashAlg;
CK_RSA_PKCS_MGF_TYPE mgf;
CK_RSA_PKCS_OAEP_SOURCE_TYPE source;
CK_VOID_PTR pSourceData;
CK_ULONG ulSourceDataLen;
} CK_RSA_PKCS_OAEP_PARAMS;
The fields of the structure have the following meanings:
hashAlg mechanism
ID of the message digest algorithm used to calculate the digest of the encoding
parameter
mgf mask
generation function to use on the encoded block
source source
of the encoding parameter
pSourceData data used
as the input for the encoding parameter source
ulSourceDataLen length of the
encoding parameter source input
CK_RSA_PKCS_OAEP_PARAMS_PTR is a pointer to a CK_RSA_PKCS_OAEP_PARAMS.
The PKCS #1 RSA OAEP mechanism, denoted CKM_RSA_PKCS_OAEP,
is a multi-purpose mechanism based on the RSA public-key cryptosystem and the
OAEP block format defined in PKCS #1. It supports single-part encryption and
decryption; key wrapping; and key unwrapping.
It has a parameter, a CK_RSA_PKCS_OAEP_PARAMS
structure.
This mechanism can wrap and unwrap any secret key of
appropriate length. Of course, a particular token may not be able to
wrap/unwrap every appropriate-length secret key that it supports. For
wrapping, the “input” to the encryption operation is the value of the CKA_VALUE
attribute of the key that is wrapped; similarly for unwrapping. The mechanism
does not wrap the key type or any other information about the key, except the
key length; the application must convey these separately. In particular, the
mechanism contributes only the CKA_CLASS and CKA_VALUE (and CKA_VALUE_LEN,
if the key has it) attributes to the recovered key during unwrapping; other
attributes must be specified in the template.
Constraints on key types and the length of the data are
summarized in the following table. For encryption and decryption, the input
and output data may begin at the same location in memory. In the table, k
is the length in bytes of the RSA modulus, and hLen is the output length
of the message digest algorithm specified by the hashAlg field of the CK_RSA_PKCS_OAEP_PARAMS
structure.
Table 38, PKCS #1 RSA OAEP: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Encrypt1
|
RSA public key
|
≤ k-2-2hLen
|
k
|
|
C_Decrypt1
|
RSA private key
|
k
|
≤ k-2-2hLen
|
|
C_WrapKey
|
RSA public key
|
≤ k-2-2hLen
|
k
|
|
C_UnwrapKey
|
RSA private key
|
k
|
≤ k-2-2hLen
|
1
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
RSA modulus sizes, in bits.
¨
CK_RSA_PKCS_PSS_PARAMS;
CK_RSA_PKCS_PSS_PARAMS_PTR
CK_RSA_PKCS_PSS_PARAMS is a structure that provides
the parameters to the CKM_RSA_PKCS_PSS mechanism. The structure is
defined as follows:
typedef struct CK_RSA_PKCS_PSS_PARAMS {
CK_MECHANISM_TYPE hashAlg;
CK_RSA_PKCS_MGF_TYPE mgf;
CK_ULONG sLen;
} CK_RSA_PKCS_PSS_PARAMS;
The fields of the structure have the following meanings:
hashAlg hash algorithm
used in the PSS encoding; if the signature mechanism does not include message
hashing, then this value must be the mechanism used by the application to
generate the message hash; if the signature mechanism includes hashing, then
this value must match the hash algorithm indicated by the signature mechanism
mgf mask
generation function to use on the encoded block
sLen length,
in bytes, of the salt value used in the PSS encoding; typical values are the
length of the message hash and zero
CK_RSA_PKCS_PSS_PARAMS_PTR is a pointer to a CK_RSA_PKCS_PSS_PARAMS.
The PKCS #1 RSA PSS mechanism, denoted CKM_RSA_PKCS_PSS,
is a mechanism based on the RSA public-key cryptosystem and the PSS block
format defined in PKCS #1. It supports single-part signature generation and
verification without message recovery. This mechanism corresponds only to the
part of PKCS #1 that involves block formatting and RSA, given a hash value; it
does not compute a hash value on the message to be signed.
It has a parameter, a CK_RSA_PKCS_PSS_PARAMS
structure. The sLen field must be less than or equal to k*-2-hLen
and hLen is the length of the input to the C_Sign or C_Verify
function. k* is the length in bytes of the RSA modulus, except if the
length in bits of the RSA modulus is one more than a multiple of 8, in which
case k* is one less than the length in bytes of the RSA modulus.
Constraints on key types and the length of the data are
summarized in the following table. In the table, k is the length in
bytes of the RSA.
Table 39, PKCS #1 RSA PSS: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Sign1
|
RSA private key
|
hLen
|
k
|
|
C_Verify1
|
RSA public key
|
hLen, k
|
N/A
|
1
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
RSA modulus sizes, in bits.
The ISO/IEC 9796 RSA mechanism, denoted CKM_RSA_9796,
is a mechanism for single-part signatures and verification with and without
message recovery based on the RSA public-key cryptosystem and the block formats
defined in ISO/IEC 9796 and its annex A.
This mechanism processes only byte strings, whereas ISO/IEC
9796 operates on bit strings. Accordingly, the following transformations are
performed:
·
Data is converted between byte and bit string formats by
interpreting the most-significant bit of the leading byte of the byte string as
the leftmost bit of the bit string, and the least-significant bit of the
trailing byte of the byte string as the rightmost bit of the bit string (this
assumes the length in bits of the data is a multiple of 8).
·
A signature is converted from a bit string to a byte string by
padding the bit string on the left with 0 to 7 zero bits so that the resulting
length in bits is a multiple of 8, and converting the resulting bit string as
above; it is converted from a byte string to a bit string by converting the
byte string as above, and removing bits from the left so that the resulting
length in bits is the same as that of the RSA modulus.
This mechanism does not have a parameter.
Constraints on key types and the length of input and output
data are summarized in the following table. In the table, k is the
length in bytes of the RSA modulus.
Table
40, ISO/IEC 9796 RSA: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Sign1
|
RSA private key
|
≤ ⌊k/2⌋
|
k
|
|
C_SignRecover
|
RSA private key
|
≤ ⌊k/2⌋
|
k
|
|
C_Verify1
|
RSA public key
|
≤ ⌊k/2⌋, k2
|
N/A
|
|
C_VerifyRecover
|
RSA public key
|
k
|
≤ ⌊k/2⌋
|
1
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
RSA modulus sizes, in bits.
The X.509 (raw) RSA mechanism, denoted CKM_RSA_X_509,
is a multi-purpose mechanism based on the RSA public-key cryptosystem. It
supports single-part encryption and decryption; single-part signatures and
verification with and without message recovery; key wrapping; and key
unwrapping. All these operations are based on so-called “raw” RSA, as assumed
in X.509.
“Raw” RSA as defined here encrypts a byte string by
converting it to an integer, most-significant byte first, applying “raw” RSA
exponentiation, and converting the result to a byte string, most-significant
byte first. The input string, considered as an integer, must be less than the
modulus; the output string is also less than the modulus.
This mechanism does not have a parameter.
This mechanism can wrap and unwrap any secret key of
appropriate length. Of course, a particular token may not be able to
wrap/unwrap every appropriate-length secret key that it supports. For
wrapping, the “input” to the encryption operation is the value of the CKA_VALUE
attribute of the key that is wrapped; similarly for unwrapping. The
mechanism does not wrap the key type, key length, or any other information
about the key; the application must convey these separately, and supply them
when unwrapping the key.
Unfortunately, X.509 does not specify how to perform padding
for RSA encryption. For this mechanism, padding should be performed by
prepending plaintext data with 0-valued bytes. In effect, to encrypt the
sequence of plaintext bytes b1 b2 … bn (n ≤
k), Cryptoki forms P=2n-1b1+2n-2b2+…+bn.
This number must be less than the RSA modulus. The k-byte ciphertext (k
is the length in bytes of the RSA modulus) is produced by raising P to the RSA
public exponent modulo the RSA modulus. Decryption of a k-byte
ciphertext C is accomplished by raising C to the RSA private exponent modulo
the RSA modulus, and returning the resulting value as a sequence of exactly k
bytes. If the resulting plaintext is to be used to produce an unwrapped key,
then however many bytes are specified in the template for the length of the key
are taken from the end of this sequence of bytes.
Technically, the above procedures may differ very slightly
from certain details of what is specified in X.509.
Executing cryptographic operations using this mechanism can
result in the error returns CKR_DATA_INVALID (if plaintext is supplied which
has the same length as the RSA modulus and is numerically at least as large as
the modulus) and CKR_ENCRYPTED_DATA_INVALID (if ciphertext is supplied which
has the same length as the RSA modulus and is numerically at least as large as
the modulus).
Constraints on key types and the length of input and output
data are summarized in the following table. In the table, k is the
length in bytes of the RSA modulus.
Table
41, X.509 (Raw) RSA: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Encrypt1
|
RSA public key
|
≤ k
|
k
|
|
C_Decrypt1
|
RSA private key
|
k
|
k
|
|
C_Sign1
|
RSA private key
|
≤ k
|
k
|
|
C_SignRecover
|
RSA private key
|
≤ k
|
k
|
|
C_Verify1
|
RSA public key
|
≤ k, k2
|
N/A
|
|
C_VerifyRecover
|
RSA public key
|
k
|
k
|
|
C_WrapKey
|
RSA public key
|
≤ k
|
k
|
|
C_UnwrapKey
|
RSA private key
|
k
|
≤ k
(specified in template)
|
1
For this
mechanism, the ulMinKeySize and ulMaxKeySize fields of the CK_MECHANISM_INFO
structure specify the supported range of RSA modulus sizes, in bits.
This mechanism is intended for compatibility with
applications that do not follow the PKCS #1 or ISO/IEC 9796 block formats.
The ANSI X9.31 RSA mechanism, denoted CKM_RSA_X9_31,
is a mechanism for single-part signatures and verification without message
recovery based on the RSA public-key cryptosystem and the block formats defined
in ANSI X9.31.
This mechanism applies the header and padding fields of the
hash encapsulation. The trailer field must be applied by the application.
This mechanism processes only byte strings, whereas ANSI
X9.31 operates on bit strings. Accordingly, the following transformations are
performed:
·
Data is converted between byte and bit string formats by
interpreting the most-significant bit of the leading byte of the byte string as
the leftmost bit of the bit string, and the least-significant bit of the
trailing byte of the byte string as the rightmost bit of the bit string (this
assumes the length in bits of the data is a multiple of 8).
·
A signature is converted from a bit string to a byte string by padding
the bit string on the left with 0 to 7 zero bits so that the resulting length
in bits is a multiple of 8, and converting the resulting bit string as above;
it is converted from a byte string to a bit string by converting the byte
string as above, and removing bits from the left so that the resulting length
in bits is the same as that of the RSA modulus.
This mechanism does not have a parameter.
Constraints on key types and the length of input and output
data are summarized in the following table. In the table, k is the
length in bytes of the RSA modulus. For all operations, the k value must
be at least 128 and a multiple of 32 as specified in ANSI X9.31.
Table 42, ANSI X9.31 RSA: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Sign1
|
RSA private key
|
≤ k-2
|
k
|
|
C_Verify1
|
RSA public key
|
≤ k-2, k2
|
N/A
|
1
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
RSA modulus sizes, in bits.
The PKCS #1 v1.5 RSA signature with MD2 mechanism, denoted CKM_MD2_RSA_PKCS,
performs single- and multiple-part digital signatures and verification
operations without message recovery. The operations performed are as described
initially in PKCS #1 v1.5 with the object identifier md2WithRSAEncryption, and
as in the scheme RSASSA-PKCS1-v1_5 in the current version of PKCS #1, where the
underlying hash function is MD2.
Similarly, the PKCS #1 v1.5 RSA signature with MD5
mechanism, denoted CKM_MD5_RSA_PKCS, performs the same operations
described in PKCS #1 with the object identifier md5WithRSAEncryption. The PKCS
#1 v1.5 RSA signature with SHA-1 mechanism, denoted CKM_SHA1_RSA_PKCS,
performs the same operations, except that it uses the hash function SHA-1 with
object identifier sha1WithRSAEncryption.
Likewise, the PKCS #1 v1.5 RSA signature with SHA-256,
SHA-384, and SHA-512 mechanisms, denoted CKM_SHA256_RSA_PKCS, CKM_SHA384_RSA_PKCS,
and CKM_SHA512_RSA_PKCS respectively, perform the same operations using
the SHA-256, SHA-384 and SHA-512 hash functions with the object identifiers
sha256WithRSAEncryption, sha384WithRSAEncryption and sha512WithRSAEncryption
respectively.
The PKCS #1 v1.5 RSA signature with RIPEMD-128 or
RIPEMD-160, denoted CKM_RIPEMD128_RSA_PKCS and CKM_RIPEMD160_RSA_PKCS
respectively, perform the same operations using the RIPE-MD 128 and RIPE-MD 160
hash functions.
None of these mechanisms has a parameter.
Constraints on key types and the length of the data for
these mechanisms are summarized in the following table. In the table, k
is the length in bytes of the RSA modulus. For the PKCS #1 v1.5 RSA signature
with MD2 and PKCS #1 v1.5 RSA signature with MD5 mechanisms, k must be
at least 27; for the PKCS #1 v1.5 RSA signature with SHA-1 mechanism, k
must be at least 31, and so on for other underlying hash functions, where the
minimum is always 11 bytes more than the length of the hash value.
Table 43, PKCS #1 v1.5 RSA Signatures with Various Hash
Functions: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
Comments
|
|
C_Sign
|
RSA private key
|
any
|
k
|
block type 01
|
|
C_Verify
|
RSA public key
|
any, k2
|
N/A
|
block type 01
|
2
For these mechanisms, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
RSA modulus sizes, in bits.
The PKCS #1 v1.5 RSA signature with SHA-224 mechanism,
denoted CKM_SHA224_RSA_PKCS, performs similarly as the other CKM_SHAX_RSA_PKCS
mechanisms but uses the SHA-224 hash function.
The PKCS #1 RSA PSS signature with SHA-224 mechanism,
denoted CKM_SHA224_RSA_PKCS_PSS, performs similarly as the other CKM_SHAX_RSA_
PKCS_PSS mechanisms but uses the SHA-224 hash function.
The PKCS #1 RSA PSS signature with SHA-1 mechanism, denoted CKM_SHA1_RSA_PKCS_PSS,
performs single- and multiple-part digital signatures and verification
operations without message recovery. The operations performed are as described
in PKCS #1 with the object identifier id-RSASSA-PSS, i.e., as in the scheme
RSASSA-PSS in PKCS #1 where the underlying hash function is SHA-1.
The PKCS #1 RSA PSS signature with SHA-256, SHA-384, and
SHA-512 mechanisms, denoted CKM_SHA256_RSA_PKCS_PSS, CKM_SHA384_RSA_PKCS_PSS,
and CKM_SHA512_RSA_PKCS_PSS respectively, perform the same operations
using the SHA-256, SHA-384 and SHA-512 hash functions.
The mechanisms have a parameter, a CK_RSA_PKCS_PSS_PARAMS
structure. The sLen field must be less than or equal to k*-2-hLen
where hLen is the length in bytes of the hash value. k* is the
length in bytes of the RSA modulus, except if the length in bits of the RSA
modulus is one more than a multiple of 8, in which case k* is one less
than the length in bytes of the RSA modulus.
Constraints on key types and the length of the data are
summarized in the following table. In the table, k is the length in
bytes of the RSA modulus.
Table 44, PKCS #1 RSA PSS Signatures with Various Hash Functions: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Sign
|
RSA private key
|
any
|
k
|
|
C_Verify
|
RSA public key
|
any, k2
|
N/A
|
2
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
RSA modulus sizes, in bits.
The PKCS #1 v1.5 RSA signature with SHA3-224, SHA3-256,
SHA3-384, SHA3-512 mechanisms, denoted CKM_SHA3_224_RSA_PKCS,
CKM_SHA3_256_RSA_PKCS, CKM_SHA3_384_RSA_PKCS, and
CKM_SHA3_512_RSA_PKCS respectively, performs similarly as the other CKM_SHAX_RSA_PKCS
mechanisms but uses the corresponding SHA3 hash functions.
The PKCS #1 RSA PSS signature with SHA3-224, SHA3-256, SHA3-384,
SHA3-512 mechanisms, denoted CKM_SHA3_224_RSA_PKCS_PSS,
CKM_SHA3_256_RSA_PKCS_PSS, CKM_SHA3_384_RSA_PKCS_PSS, and
CKM_SHA3_512_RSA_PKCS_PSS respectively, performs similarly as the other CKM_SHAX_RSA_PKCS_PSS
mechanisms but uses the corresponding SHA-3 hash functions.
The ANSI X9.31 RSA signature with SHA-1 mechanism, denoted CKM_SHA1_RSA_X9_31,
performs single- and multiple-part digital signatures and verification
operations without message recovery. The operations performed are as described
in ANSI X9.31.
This mechanism does not have a parameter.
Constraints on key types and the length of the data for
these mechanisms are summarized in the following table. In the table, k
is the length in bytes of the RSA modulus. For all operations, the k
value must be at least 128 and a multiple of 32 as specified in ANSI X9.31.
Table 45, ANSI X9.31 RSA Signatures with SHA-1: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Sign
|
RSA private key
|
any
|
k
|
|
C_Verify
|
RSA public key
|
any, k2
|
N/A
|
2
For these mechanisms, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
RSA modulus sizes, in bits.
The TPM 1.1b and TPM 1.2 PKCS #1 v1.5 RSA mechanism, denoted
CKM_RSA_PKCS_TPM_1_1, is a multi-use mechanism based on the RSA
public-key cryptosystem and the block formats initially defined in PKCS #1
v1.5, with additional formatting rules defined in TCPA TPM Specification
Version 1.1b. Additional formatting rules remained the same in TCG TPM
Specification 1.2 The mechanism supports single-part encryption and
decryption; key wrapping; and key unwrapping.
This mechanism does not have a parameter. It differs from
the standard PKCS#1 v1.5 RSA encryption mechanism in that the plaintext is
wrapped in a TCPA_BOUND_DATA (TPM_BOUND_DATA for TPM 1.2) structure before
being submitted to the PKCS#1 v1.5 encryption process. On encryption, the
version field of the TCPA_BOUND_DATA (TPM_BOUND_DATA for TPM 1.2) structure
must contain 0x01, 0x01, 0x00, 0x00. On decryption, any structure of the form
0x01, 0x01, 0xXX, 0xYY may be accepted.
This mechanism can wrap and unwrap any secret key of
appropriate length. Of course, a particular token may not be able to
wrap/unwrap every appropriate-length secret key that it supports. For
wrapping, the “input” to the encryption operation is the value of the CKA_VALUE
attribute of the key that is wrapped; similarly for unwrapping. The mechanism
does not wrap the key type or any other information about the key, except the
key length; the application must convey these separately. In particular, the
mechanism contributes only the CKA_CLASS and CKA_VALUE (and CKA_VALUE_LEN,
if the key has it) attributes to the recovered key during unwrapping; other
attributes must be specified in the template.
Constraints on key types and the length of the data are
summarized in the following table. For encryption and decryption, the input
and output data may begin at the same location in memory. In the table, k
is the length in bytes of the RSA modulus.
Table 46, TPM 1.1b and TPM 1.2 PKCS #1 v1.5 RSA: Key
And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Encrypt1
|
RSA public key
|
≤ k-11-5
|
k
|
|
C_Decrypt1
|
RSA private key
|
k
|
≤ k-11-5
|
|
C_WrapKey
|
RSA public key
|
≤ k-11-5
|
k
|
|
C_UnwrapKey
|
RSA private key
|
k
|
≤ k-11-5
|
1
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
RSA modulus sizes, in bits.
The TPM 1.1b and TPM 1.2 PKCS #1 RSA OAEP mechanism, denoted
CKM_RSA_PKCS_OAEP_TPM_1_1, is a multi-purpose mechanism based on the RSA
public-key cryptosystem and the OAEP block format defined in PKCS #1, with
additional formatting defined in TCPA TPM Specification Version 1.1b.
Additional formatting rules remained the same in TCG TPM Specification 1.2.
The mechanism supports single-part encryption and decryption; key wrapping; and
key unwrapping.
This mechanism does not have a parameter. It differs from
the standard PKCS#1 OAEP RSA encryption mechanism in that the plaintext is
wrapped in a TCPA_BOUND_DATA (TPM_BOUND_DATA for TPM 1.2) structure before
being submitted to the encryption process and that all of the values of the
parameters that are passed to a standard CKM_RSA_PKCS_OAEP operation are fixed.
On encryption, the version field of the TCPA_BOUND_DATA (TPM_BOUND_DATA for TPM
1.2) structure must contain 0x01, 0x01, 0x00, 0x00. On decryption, any
structure of the form 0x01, 0x01, 0xXX, 0xYY may be accepted.
This mechanism can wrap and unwrap any secret key of
appropriate length. Of course, a particular token may not be able to
wrap/unwrap every appropriate-length secret key that it supports. For
wrapping, the “input” to the encryption operation is the value of the CKA_VALUE
attribute of the key that is wrapped; similarly for unwrapping. The mechanism
does not wrap the key type or any other information about the key, except the
key length; the application must convey these separately. In particular, the
mechanism contributes only the CKA_CLASS and CKA_VALUE (and CKA_VALUE_LEN,
if the key has it) attributes to the recovered key during unwrapping; other
attributes must be specified in the template.
Constraints on key types and the length of the data are
summarized in the following table. For encryption and decryption, the input
and output data may begin at the same location in memory. In the table, k
is the length in bytes of the RSA modulus.
Table 47, TPM 1.1b and TPM 1.2 PKCS #1 RSA OAEP: Key
And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Encrypt1
|
RSA public key
|
≤ k-2-40-5
|
k
|
|
C_Decrypt1
|
RSA private key
|
k
|
≤ k-2-40-5
|
|
C_WrapKey
|
RSA public key
|
≤ k-2-40-5
|
k
|
|
C_UnwrapKey
|
RSA private key
|
k
|
≤ k-2-40-5
|
1
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
RSA modulus sizes, in bits.
The RSA AES key wrap mechanism, denoted CKM_RSA_AES_KEY_WRAP,
is a mechanism based on the RSA public-key cryptosystem and the AES key wrap
mechanism. It supports single-part key wrapping; and key unwrapping.
It has a parameter, a CK_RSA_AES_KEY_WRAP_PARAMS
structure.
The
mechanism can wrap and unwrap a target asymmetric key of any length and type
using an RSA key.
-
A temporary AES key
is used for wrapping the target key using CKM_AES_KEY_WRAP_KWP mechanism.
-
The temporary AES
key is wrapped with the wrapping RSA key using CKM_RSA_PKCS_OAEP mechanism.
For
wrapping, the mechanism -
- Generates
a temporary random AES key of ulAESKeyBits length. This key is not
accessible to the user - no handle is returned.
- Wraps
the AES key with the wrapping RSA key using CKM_RSA_PKCS_OAEP with
parameters of OAEPParams.
- Wraps
the target key with the temporary AES key using CKM_AES_KEY_WRAP_KWP.
- Zeroizes
the temporary AES key
- Concatenates
two wrapped keys and outputs the concatenated blob. The first is the
wrapped AES key, and the second is the wrapped target key.
The private
target key will be encoded as defined in section 6.7.
The use of
Attributes in the PrivateKeyInfo structure is OPTIONAL. In case of conflicts
between the object attribute template, and Attributes in the PrivateKeyInfo
structure, an error should be thrown
For
unwrapping, the mechanism -
- Splits
the input into two parts. The first is the wrapped AES key, and the second
is the wrapped target key. The length of the first part is equal to the
length of the unwrapping RSA key.
- Un-wraps
the temporary AES key from the first part with the private RSA key using CKM_RSA_PKCS_OAEP
with parameters of OAEPParams.
- Un-wraps
the target key from the second part with the temporary AES key using CKM_AES_KEY_WRAP_KWP.
- Zeroizes
the temporary AES key.
- Returns
the handle to the newly unwrapped target key.
Table
48, CKM_RSA_AES_KEY_WRAP Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_RSA_AES_KEY_WRAP
|
|
|
|
|
|
ü
|
|
|
1SR =
SignRecover, VR = VerifyRecover
|
¨ CK_RSA_AES_KEY_WRAP_PARAMS;
CK_RSA_AES_KEY_WRAP_PARAMS_PTR
CK_RSA_AES_KEY_WRAP_PARAMS is a structure that
provides the parameters to the CKM_RSA_AES_KEY_WRAP mechanism. It
is defined as follows:
typedef struct CK_RSA_AES_KEY_WRAP_PARAMS {
CK_ULONG ulAESKeyBits;
CK_RSA_PKCS_OAEP_PARAMS_PTR pOAEPParams;
} CK_RSA_AES_KEY_WRAP_PARAMS;
The fields of the structure have
the following meanings:
ulAESKeyBits length of
the temporary AES key in bits. Can be only 128, 192 or 256.
pOAEPParams pointer to the
parameters of the temporary AES key wrapping. See also the description of PKCS
#1 RSA OAEP mechanism parameters.
CK_RSA_AES_KEY_WRAP_PARAMS_PTR is a pointer to a CK_RSA_AES_KEY_WRAP_PARAMS.
When CKM_RSA_PKCS is operated in FIPS mode, the length of
the modulus SHALL only be 1024, 2048, or 3072 bits.
Table
49, DSA Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_DSA_KEY_PAIR_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_DSA_PARAMETER_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_DSA_PROBABILISTIC_PARAMETER_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_DSA_SHAWE_TAYLOR_PARAMETER_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_DSA_FIPS_G_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_DSA
|
|
ü2
|
|
|
|
|
|
|
CKM_DSA_SHA1
|
|
ü
|
|
|
|
|
|
|
CKM_DSA_SHA224
|
|
ü
|
|
|
|
|
|
|
CKM_DSA_SHA256
|
|
ü
|
|
|
|
|
|
|
CKM_DSA_SHA384
|
|
ü
|
|
|
|
|
|
|
CKM_DSA_SHA512
|
|
ü
|
|
|
|
|
|
|
CKM_DSA_SHA3_224
|
|
ü
|
|
|
|
|
|
|
CKM_DSA_SHA3_256
|
|
ü
|
|
|
|
|
|
|
CKM_DSA_SHA3_384
|
|
ü
|
|
|
|
|
|
|
CKM_DSA_SHA3_512
|
|
ü
|
|
|
|
|
|
This section defines the key type “CKK_DSA” for type
CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of DSA key objects.
Mechanisms:
CKM_DSA_KEY_PAIR_GEN
CKM_DSA
CKM_DSA_SHA1
CKM_DSA_SHA224
CKM_DSA_SHA256
CKM_DSA_SHA384
CKM_DSA_SHA512
CKM_DSA_SHA3_224
CKM_DSA_SHA3_256
CKM_DSA_SHA3_384
CKM_DSA_SHA3_512
CKM_DSA_PARAMETER_GEN
CKM_DSA_PROBABILISTIC_PARAMETER_GEN
CKM_DSA_SHAWE_TAYLOR_PARAMETER_GEN
CKM_DSA_FIPS_G_GEN
¨
CK_DSA_PARAMETER_GEN_PARAM
CK_DSA_PARAMETER_GEN_PARAM is a structure which provides and
returns parameters for the NIST FIPS 186-4 parameter generating algorithms.
CK_DSA_PARAMETER_GEN_PARAM_PTR is a pointer to a
CK_DSA_PARAMETER_GEN_PARAM.
typedef struct CK_DSA_PARAMETER_GEN_PARAM {
CK_MECHANISM_TYPE hash;
CK_BYTE_PTR pSeed;
CK_ULONG ulSeedLen;
CK_ULONG ulIndex;
} CK_DSA_PARAMETER_GEN_PARAM;
The fields of the structure have
the following meanings:
hash Mechanism
value for the base hash used in PQG generation, Valid values are CKM_SHA_1,
CKM_SHA224, CKM_SHA256, CKM_SHA384, CKM_SHA512.
pSeed Seed
value used to generate PQ and G. This value is returned by
CKM_DSA_PROBABILISTIC_PARAMETER_GEN, CKM_DSA_SHAWE_TAYLOR_PARAMETER_GEN, and
passed into CKM_DSA_FIPS_G_GEN.
ulSeedLen Length of
seed value.
ulIndex Index
value for generating G. Input for CKM_DSA_FIPS_G_GEN. Ignored by
CKM_DSA_PROBABILISTIC_PARAMETER_GEN and CKM_DSA_SHAWE_TAYLOR_PARAMETER_GEN.
DSA public key objects (object class CKO_PUBLIC_KEY, key
type CKK_DSA) hold DSA public keys. The following table defines the DSA
public key object attributes, in addition to the common attributes defined for
this object class:
Table 50, DSA Public Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_PRIME1,3
|
Big integer
|
Prime p (512 to 3072 bits, in steps of
64 bits)
|
|
CKA_SUBPRIME1,3
|
Big integer
|
Subprime q (160, 224 bits, or 256
bits)
|
|
CKA_BASE1,3
|
Big integer
|
Base g
|
|
CKA_VALUE1,4
|
Big integer
|
Public value y
|
- Refer to Table 11 for footnotes
The CKA_PRIME, CKA_SUBPRIME and CKA_BASE
attribute values are collectively the “DSA domain parameters”. See FIPS PUB
186-4 for more information on DSA keys.
The following is a sample template for creating a DSA public
key object:
CK_OBJECT_CLASS class = CKO_PUBLIC_KEY;
CK_KEY_TYPE keyType = CKK_DSA;
CK_UTF8CHAR label[] = “A DSA public key object”;
CK_BYTE prime[] = {...};
CK_BYTE subprime[] = {...};
CK_BYTE base[] = {...};
CK_BYTE value[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_PRIME, prime, sizeof(prime)},
{CKA_SUBPRIME, subprime, sizeof(subprime)},
{CKA_BASE, base, sizeof(base)},
{CKA_VALUE, value, sizeof(value)}
};
FIPS PUB 186-4 specifies permitted combinations of prime and
sub-prime lengths. They are:
- Prime: 1024 bits, Subprime: 160
- Prime: 2048 bits, Subprime: 224
- Prime: 2048 bits, Subprime: 256
- Prime: 3072 bits, Subprime: 256
Earlier versions of FIPS 186 permitted smaller prime
lengths, and those are included here for backwards compatibility. An
implementation that is compliant to FIPS 186-4 does not permit the use of
primes of any length less than 1024 bits.
DSA private key objects (object class CKO_PRIVATE_KEY, key
type CKK_DSA) hold DSA private keys. The following table defines the
DSA private key object attributes, in addition to the common attributes defined
for this object class:
Table 51, DSA Private Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_PRIME1,4,6
|
Big integer
|
Prime p (512 to 1024 bits, in steps of
64 bits)
|
|
CKA_SUBPRIME1,4,6
|
Big integer
|
Subprime q (160 bits, 224 bits, or 256
bits)
|
|
CKA_BASE1,4,6
|
Big integer
|
Base g
|
|
CKA_VALUE1,4,6,7
|
Big integer
|
Private value x
|
- Refer to Table 11 for footnotes
The CKA_PRIME, CKA_SUBPRIME and CKA_BASE
attribute values are collectively the “DSA domain parameters”. See FIPS PUB
186-4 for more information on DSA keys.
Note that when generating a DSA private key, the DSA domain
parameters are not specified in the key’s template. This is because DSA
private keys are only generated as part of a DSA key pair, and the DSA
domain parameters for the pair are specified in the template for the DSA public
key.
The following is a sample
template for creating a DSA private key object:
CK_OBJECT_CLASS class = CKO_PRIVATE_KEY;
CK_KEY_TYPE keyType = CKK_DSA;
CK_UTF8CHAR label[] = “A DSA private key object”;
CK_BYTE subject[] = {...};
CK_BYTE id[] = {123};
CK_BYTE prime[] = {...};
CK_BYTE subprime[] = {...};
CK_BYTE base[] = {...};
CK_BYTE value[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_SUBJECT, subject, sizeof(subject)},
{CKA_ID, id, sizeof(id)},
{CKA_SENSITIVE, &true, sizeof(true)},
{CKA_SIGN, &true, sizeof(true)},
{CKA_PRIME, prime, sizeof(prime)},
{CKA_SUBPRIME, subprime, sizeof(subprime)},
{CKA_BASE, base, sizeof(base)},
{CKA_VALUE, value, sizeof(value)}
};
DSA domain parameter objects (object class CKO_DOMAIN_PARAMETERS,
key type CKK_DSA) hold DSA domain parameters. The following table
defines the DSA domain parameter object attributes, in addition to the common
attributes defined for this object class:
Table 52, DSA Domain Parameter Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_PRIME1,4
|
Big integer
|
Prime p (512 to 1024 bits, in steps of
64 bits)
|
|
CKA_SUBPRIME1,4
|
Big integer
|
Subprime q (160 bits, 224 bits, or 256
bits)
|
|
CKA_BASE1,4
|
Big integer
|
Base g
|
|
CKA_PRIME_BITS2,3
|
CK_ULONG
|
Length of the prime value.
|
- Refer to Table 11 for footnotes
The CKA_PRIME, CKA_SUBPRIME and CKA_BASE
attribute values are collectively the “DSA domain parameters”. See FIPS PUB
186-4 for more information on DSA domain parameters.
To ensure backwards compatibility, if CKA_SUBPRIME_BITS
is not specified for a call to C_GenerateKey, it takes on a default
based on the value of CKA_PRIME_BITS as follows:
- If CKA_PRIME_BITS is less than or equal to 1024
then CKA_SUBPRIME_BITS shall be 160 bits
- If CKA_PRIME_BITS equals 2048 then
CKA_SUBPRIME_BITS shall be 224 bits
- If CKA_PRIME_BITS equals 3072 then
CKA_SUBPRIME_BITS shall be 256 bits
The following is a sample template for creating a DSA domain
parameter object:
CK_OBJECT_CLASS class = CKO_DOMAIN_PARAMETERS;
CK_KEY_TYPE keyType = CKK_DSA;
CK_UTF8CHAR label[] = “A DSA domain parameter object”;
CK_BYTE prime[] = {...};
CK_BYTE subprime[] = {...};
CK_BYTE base[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_PRIME, prime, sizeof(prime)},
{CKA_SUBPRIME, subprime, sizeof(subprime)},
{CKA_BASE, base, sizeof(base)},
};
The DSA key pair generation mechanism, denoted CKM_DSA_KEY_PAIR_GEN,
is a key pair generation mechanism based on the Digital Signature Algorithm
defined in FIPS PUB 186-2.
This mechanism does not have a parameter.
The mechanism generates DSA public/private key pairs with a
particular prime, subprime and base, as specified in the CKA_PRIME, CKA_SUBPRIME,
and CKA_BASE attributes of the template for the public key.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new public key and the CKA_CLASS,
CKA_KEY_TYPE, CKA_PRIME, CKA_SUBPRIME, CKA_BASE,
and CKA_VALUE attributes to the new private key. Other attributes
supported by the DSA public and private key types (specifically, the flags
indicating which functions the keys support) may also be specified in the
templates for the keys, or else are assigned default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
DSA prime sizes, in bits.
The DSA domain parameter generation mechanism, denoted CKM_DSA_PARAMETER_GEN,
is a domain parameter generation mechanism based on the Digital Signature
Algorithm defined in FIPS PUB 186-2.
This mechanism does not have a parameter.
The mechanism generates DSA domain parameters with a
particular prime length in bits, as specified in the CKA_PRIME_BITS
attribute of the template.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
CKA_PRIME, CKA_SUBPRIME, CKA_BASE and CKA_PRIME_BITS
attributes to the new object. Other attributes supported by the DSA domain
parameter types may also be specified in the template, or else are assigned
default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
DSA prime sizes, in bits.
The DSA probabilistic domain parameter generation mechanism,
denoted CKM_DSA_PROBABILISTIC_PARAMETER_GEN, is a domain parameter
generation mechanism based on the Digital Signature Algorithm defined in FIPS
PUB 186-4, section Appendix A.1.1 Generation and Validation of Probable
Primes..
This mechanism takes a CK_DSA_PARAMETER_GEN_PARAM
which supplies the base hash and returns the seed (pSeed) and the length
(ulSeedLen).
The mechanism generates DSA the prime and subprime domain
parameters with a particular prime length in bits, as specified in the CKA_PRIME_BITS
attribute of the template and the subprime length as specified in the CKA_SUBPRIME_BITS
attribute of the template.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
CKA_PRIME, CKA_SUBPRIME, CKA_PRIME_BITS, and CKA_SUBPRIME_BITS
attributes to the new object. CKA_BASE is not set by this call. Other
attributes supported by the DSA domain parameter types may also be specified in
the template, or else are assigned default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
DSA prime sizes, in bits.
The DSA Shawe-Taylor domain parameter generation mechanism,
denoted CKM_DSA_SHAWE_TAYLOR_PARAMETER_GEN, is a domain parameter
generation mechanism based on the Digital Signature Algorithm defined in FIPS
PUB 186-4, section Appendix A.1.2 Construction and Validation of Provable
Primes p and q.
This mechanism takes a CK_DSA_PARAMETER_GEN_PARAM
which supplies the base hash and returns the seed (pSeed) and the length
(ulSeedLen).
The mechanism generates DSA the prime and subprime domain
parameters with a particular prime length in bits, as specified in the
CKA_PRIME_BITS attribute of the template and the subprime length as specified
in the CKA_SUBPRIME_BITS attribute of the template.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
CKA_PRIME, CKA_SUBPRIME, CKA_PRIME_BITS, and CKA_SUBPRIME_BITS
attributes to the new object. CKA_BASE is not set by this call. Other
attributes supported by the DSA domain parameter types may also be specified in
the template, or else are assigned default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
DSA prime sizes, in bits.
The DSA base domain parameter generation mechanism, denoted CKM_DSA_FIPS_G_GEN,
is a base parameter generation mechanism based on the Digital Signature
Algorithm defined in FIPS PUB 186-4, section Appendix A.2 Generation of
Generator G.
This mechanism takes a CK_DSA_PARAMETER_GEN_PARAM
which supplies the base hash the seed (pSeed) and the length (ulSeedLen) and
the index value.
The mechanism generates the DSA base with the domain
parameter specified in the CKA_PRIME and CKA_SUBPRIME attributes
of the template.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_BASE attributes to the new object. Other attributes
supported by the DSA domain parameter types may also be specified in the
template, or else are assigned default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
DSA prime sizes, in bits.
The DSA without hashing mechanism, denoted CKM_DSA,
is a mechanism for single-part signatures and verification based on the Digital
Signature Algorithm defined in FIPS PUB 186-2. (This mechanism corresponds only
to the part of DSA that processes the 20-byte hash value; it does not compute
the hash value.)
For the purposes of this mechanism, a DSA signature is a
40-byte string, corresponding to the concatenation of the DSA values r
and s, each represented most-significant byte first.
It does not have a parameter.
Constraints on key types and the length of data are
summarized in the following table:
Table
53, DSA: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Sign1
|
DSA private key
|
20, 28, 32, 48, or 64
bytes
|
2*length of subprime
|
|
C_Verify1
|
DSA public key
|
(20, 28, 32, 48, or
64 bytes), (2*length of subprime)2
|
N/A
|
1
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
DSA prime sizes, in bits.
The DSA with SHA-1 mechanism, denoted CKM_DSA_SHA1,
is a mechanism for single- and multiple-part signatures and verification based
on the Digital Signature Algorithm defined in FIPS PUB 186-2. This mechanism
computes the entire DSA specification, including the hashing with SHA-1.
For the purposes of this mechanism, a DSA signature is a
40-byte string, corresponding to the concatenation of the DSA values r
and s, each represented most-significant byte first.
This mechanism does not have a parameter.
Constraints on key types and the length of data are
summarized in the following table:
Table 54, DSA with SHA-1: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Sign
|
DSA private key
|
any
|
2*subprime length
|
|
C_Verify
|
DSA public key
|
any, 2*subprime
length2
|
N/A
|
2
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
DSA prime sizes, in bits.
When CKM_DSA is operated in FIPS mode, only the following
bit lengths of p and q, represented by L and N, SHALL be used:
L = 1024, N = 160
L = 2048, N = 224
L = 2048, N = 256
L = 3072, N = 256
The DSA with SHA-224 mechanism, denoted CKM_DSA_SHA224,
is a mechanism for single- and multiple-part signatures and verification based
on the Digital Signature Algorithm defined in FIPS PUB 186-4. This mechanism
computes the entire DSA specification, including the hashing with SHA-224.
For the purposes of this mechanism, a DSA signature is a
string of length 2*subprime, corresponding to the concatenation of the DSA
values r and s, each represented most-significant byte first.
This mechanism does not have a parameter.
Constraints on key types and the length of data are
summarized in the following table:
Table 55, DSA with SHA-244: Key
And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Sign
|
DSA private key
|
any
|
2*subprime length
|
|
C_Verify
|
DSA public key
|
any, 2*subprime
length2
|
N/A
|
2 Data length, signature length.
For this mechanism, the ulMinKeySize
and ulMaxKeySize fields of the CK_MECHANISM_INFO structure
specify the supported range of DSA prime sizes, in bits.
The DSA with SHA-256 mechanism, denoted CKM_DSA_SHA256,
is a mechanism for single- and multiple-part signatures and verification based
on the Digital Signature Algorithm defined in FIPS PUB 186-4. This mechanism
computes the entire DSA specification, including the hashing with SHA-256.
For the purposes of this mechanism, a DSA signature is a
string of length 2*subprime, corresponding to the concatenation of the DSA
values r and s, each represented most-significant byte first.
This mechanism does not have a parameter.
Constraints on key types and the length of data are
summarized in the following table:
Table 56, DSA with SHA-256: Key
And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Sign
|
DSA private key
|
any
|
2*subprime length
|
|
C_Verify
|
DSA public key
|
any, 2*subprime
length2
|
N/A
|
2 Data length, signature length.
The DSA with SHA-384 mechanism, denoted CKM_DSA_SHA384,
is a mechanism for single- and multiple-part signatures and verification based
on the Digital Signature Algorithm defined in FIPS PUB 186-4. This mechanism
computes the entire DSA specification, including the hashing with SHA-384.
For the purposes of this mechanism, a DSA signature is a
string of length 2*subprime, corresponding to the concatenation of the DSA
values r and s, each represented most-significant byte first.
This mechanism does not have a parameter.
Constraints on key types and the length of data are
summarized in the following table:
Table 57, DSA with SHA-384: Key
And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Sign
|
DSA private key
|
any
|
2*subprime length
|
|
C_Verify
|
DSA public key
|
any, 2*subprime
length2
|
N/A
|
2 Data length, signature length.
The DSA with SHA-512 mechanism, denoted CKM_DSA_SHA512,
is a mechanism for single- and multiple-part signatures and verification based
on the Digital Signature Algorithm defined in FIPS PUB 186-4. This mechanism
computes the entire DSA specification, including the hashing with SHA-512.
For the purposes of this mechanism, a DSA signature is a
string of length 2*subprime, corresponding to the concatenation of the DSA
values r and s, each represented most-significant byte first.
This mechanism does not have a parameter.
Constraints on key types and the length of data are
summarized in the following table:
Table 58, DSA with SHA-512: Key
And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Sign
|
DSA private key
|
any
|
2*subprime length
|
|
C_Verify
|
DSA public key
|
any, 2*subprime
length2
|
N/A
|
2 Data length, signature length.
6.2.18
DSA with SHA3-224
The DSA with SHA3-224 mechanism, denoted CKM_DSA_SHA3_224,
is a mechanism for single- and multiple-part signatures and verification based
on the Digital Signature Algorithm defined in FIPS PUB 186-4. This mechanism
computes the entire DSA specification, including the hashing with SHA3-224.
For the purposes of this mechanism, a DSA signature is a
string of length 2*subprime, corresponding to the concatenation of the DSA
values r and s, each represented most-significant byte first.
This mechanism does not have a parameter.
Constraints on key types and the length of data are
summarized in the following table:
Table 59,
DSA with SHA3-224: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Sign
|
DSA private key
|
any
|
2*subprime length
|
|
C_Verify
|
DSA public key
|
any, 2*subprime
length2
|
N/A
|
2 Data length, signature length.
For this mechanism, the ulMinKeySize
and ulMaxKeySize fields of the CK_MECHANISM_INFO structure
specify the supported range of DSA prime sizes, in bits.
The DSA with SHA3-256 mechanism, denoted CKM_DSA_SHA3_256,
is a mechanism for single- and multiple-part signatures and verification based
on the Digital Signature Algorithm defined in FIPS PUB 186-4. This mechanism
computes the entire DSA specification, including the hashing with SHA3-256.
For the purposes of this mechanism, a DSA signature is a
string of length 2*subprime, corresponding to the concatenation of the DSA values
r and s, each represented most-significant byte first.
This mechanism does not have a parameter.
Constraints on key types and the length of data are
summarized in the following table:
Table 60,
DSA with SHA3-256: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Sign
|
DSA private key
|
any
|
2*subprime length
|
|
C_Verify
|
DSA public key
|
any, 2*subprime
length2
|
N/A
|
2 Data length, signature length.
The DSA with SHA3-384 mechanism, denoted CKM_DSA_SHA3_384,
is a mechanism for single- and multiple-part signatures and verification based
on the Digital Signature Algorithm defined in FIPS PUB 186-4. This mechanism
computes the entire DSA specification, including the hashing with SHA3-384.
For the purposes of this mechanism, a DSA signature is a
string of length 2*subprime, corresponding to the concatenation of the DSA
values r and s, each represented most-significant byte first.
This mechanism does not have a parameter.
Constraints on key types and the length of data are
summarized in the following table:
Table 61,
DSA with SHA3-384: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Sign
|
DSA private key
|
any
|
2*subprime length
|
|
C_Verify
|
DSA public key
|
any, 2*subprime
length2
|
N/A
|
2 Data length, signature length.
The DSA with SHA3-512 mechanism, denoted CKM_DSA_SHA3_512,
is a mechanism for single- and multiple-part signatures and verification based
on the Digital Signature Algorithm defined in FIPS PUB 186-4. This mechanism
computes the entire DSA specification, including the hashing with SH3A-512.
For the purposes of this mechanism, a DSA signature is a
string of length 2*subprime, corresponding to the concatenation of the DSA
values r and s, each represented most-significant byte first.
This mechanism does not have a parameter.
Constraints on key types and the length of data are
summarized in the following table:
Table 62,
DSA with SHA3-512: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Sign
|
DSA private key
|
any
|
2*subprime length
|
|
C_Verify
|
DSA public key
|
any, 2*subprime
length2
|
N/A
|
2 Data length, signature length.
The Elliptic Curve (EC) cryptosystem in this document was
originally based on the one described in the ANSI X9.62 and X9.63 standards
developed by the ANSI X9F1 working group.
The EC cryptosystem developed by the ANSI X9F1 working group
was created at a time when EC curves were always represented in their
Weierstrass form. Since that time, new curves represented in Edwards form (RFC
8032) and Montgomery form (RFC 7748) have become more common. To support these
new curves, the EC cryptosystem in this document has been extended from the
original. Additional key generation mechanisms have been added as well as an
additional signature generation mechanism.
Table
63, Elliptic Curve Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_EC_KEY_PAIR_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_EC_KEY_PAIR_GEN_W_EXTRA_BITS
|
|
|
|
|
ü
|
|
|
|
CKM_EC_EDWARDS_KEY_PAIR_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_EC_MONTGOMERY_KEY_PAIR_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_ECDSA
|
|
ü2
|
|
|
|
|
|
|
CKM_ECDSA_SHA1
|
|
ü
|
|
|
|
|
|
|
CKM_ECDSA_SHA224
|
|
ü
|
|
|
|
|
|
|
CKM_ECDSA_SHA256
|
|
ü
|
|
|
|
|
|
|
CKM_ECDSA_SHA384
|
|
ü
|
|
|
|
|
|
|
CKM_ECDSA_SHA512
|
|
ü
|
|
|
|
|
|
|
CKM_ECDSA_SHA3_224
|
|
ü
|
|
|
|
|
|
|
CKM_ECDSA_SHA3_256
|
|
ü
|
|
|
|
|
|
|
CKM_ECDSA_SHA3_384
|
|
ü
|
|
|
|
|
|
|
CKM_ECDSA_SHA3_512
|
|
ü
|
|
|
|
|
|
|
CKM_EDDSA
|
|
ü
|
|
|
|
|
|
|
CKM_XEDDSA
|
|
ü
|
|
|
|
|
|
|
CKM_ECDH1_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_ECDH1_COFACTOR_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_ECMQV_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_ECDH_AES_KEY_WRAP
|
|
|
|
|
|
ü
|
|
Table 64, Mechanism Information Flags
|
CKF_EC_F_P
|
0x00100000UL
|
True
if the mechanism can be used with EC domain parameters over Fp
|
|
CKF_EC_F_2M
|
0x00200000UL
|
True
if the mechanism can be used with EC domain parameters over F2m
|
|
CKF_EC_ECPARAMETERS
|
0x00400000UL
|
True
if the mechanism can be used with EC domain parameters of the choice
ecParameters
|
|
CKF_EC_OID
|
0x00800000UL
|
True
if the mechanism can be used with EC domain parameters of the choice oId
|
|
CKF_EC_UNCOMPRESS
|
0x01000000UL
|
True
if the mechanism can be used with Elliptic Curve point uncompressed
|
|
CKF_EC_COMPRESS
|
0x02000000UL
|
True
if the mechanism can be used with Elliptic Curve point compressed
|
|
CKF_EC_CURVENAME
|
0x04000000UL
|
True
of the mechanism can be used with EC domain parameters of the choice curveName
|
Note: CKF_EC_NAMEDCURVE is deprecated with PKCS#11 3.00. It
is replaced by CKF_EC_OID.
In these standards, there are two different varieties of EC
defined:
1. EC using a
field with an odd prime number of elements (i.e. the finite field Fp).
2. EC using a
field of characteristic two (i.e. the finite field F2m).
An EC key in Cryptoki contains information about which
variety of EC it is suited for. It is preferable that a Cryptoki library,
which can perform EC mechanisms, be capable of performing operations with the
two varieties of EC, however this is not required. The CK_MECHANISM_INFO
structure CKF_EC_F_P flag identifies a Cryptoki library supporting EC
keys over Fp whereas the CKF_EC_F_2M flag identifies a
Cryptoki library supporting EC keys over F2m. A Cryptoki library
that can perform EC mechanisms must set either or both of these flags for each
EC mechanism.
In these specifications there are also four representation
methods to define the domain parameters for an EC key. Only the ecParameters,
the oId and the curveName choices are supported in Cryptoki.
The CK_MECHANISM_INFO structure CKF_EC_ECPARAMETERS flag
identifies a Cryptoki library supporting the ecParameters choice whereas
the CKF_EC_OID flag identifies a Cryptoki library supporting the oId
choice, and the CKF_EC_CURVENAME flag identifies a Cryptoki library supporting
the curveName choice. A Cryptoki library that can perform EC mechanisms
must set the appropriate flag(s) for each EC mechanism.
In these specifications, an EC public key (i.e. EC point Q)
or the base point G when the ecParameters choice is used can be
represented as an octet string of the uncompressed form or the compressed
form. The CK_MECHANISM_INFO structure CKF_EC_UNCOMPRESS flag
identifies a Cryptoki library supporting the uncompressed form whereas the
CKF_EC_COMPRESS flag identifies a Cryptoki library supporting the
compressed form. A Cryptoki library that can
perform EC mechanisms must set either or both of these flags for each EC
mechanism.
Note that an
implementation of a Cryptoki library supporting EC with only one variety, one
representation of domain parameters or one form may encounter difficulties
achieving interoperability with other implementations.
If an attempt to create, generate, derive or unwrap an EC
key of an unsupported curve is made, the attempt should fail with the error
code CKR_CURVE_NOT_SUPPORTED. If an attempt to create, generate, derive, or
unwrap an EC key with invalid or of an unsupported representation of domain
parameters is made, that attempt should fail with the error code CKR_DOMAIN_PARAMS_INVALID.
If an attempt to create, generate, derive, or unwrap an EC key of an
unsupported form is made, that attempt should fail with the error code
CKR_TEMPLATE_INCONSISTENT.
For the purposes of these mechanisms, an ECDSA signature is
an octet string of even length which is at most two times nLen octets,
where nLen is the length in octets of the base point order n. The
signature octets correspond to the concatenation of the ECDSA values r
and s, both represented as an octet string of equal length of at most nLen
with the most significant byte first. If r and s have different
octet length, the shorter of both must be padded with leading zero octets such
that both have the same octet length. Loosely spoken, the first half of the
signature is r and the second half is s. For signatures created
by a token, the resulting signature is always of length 2nLen. For
signatures passed to a token for verification, the signature may have a shorter
length but must be composed as specified before.
If the length of the hash value is larger than the bit
length of n, only the leftmost bits of the hash up to the length of n
will be used. Any truncation is done by the token.
Note: For applications, it is recommended to encode the
signature as an octet string of length two times nLen if possible. This
ensures that the application works with PKCS#11 modules which have been
implemented based on an older version of this document. Older versions required
all signatures to have length two times nLen. It may be impossible to
encode the signature with the maximum length of two times nLen if the
application just gets the integer values of r and s (i.e. without
leading zeros), but does not know the base point order n, because r
and s can have any value between zero and the base point order n.
An EdDSA signature is an octet string of even length which
is two times nLen octets, where nLen is calculated as EdDSA parameter b divided
by 8. The signature octets correspond to the concatenation of the EdDSA values
R and S as defined in [RFC 8032], both represented as an octet string of equal
length of nLen bytes in little endian order.
This section defines the key types “CKK_EC”, “CKK_EC_EDWARDS”
and “CKK_EC_MONTGOMERY” for type CK_KEY_TYPE as used in the CKA_KEY_TYPE
attribute of key objects.
Note: CKK_ECDSA is deprecated. It is replaced by CKK_EC.
Mechanisms:
CKM_EC_KEY_PAIR_GEN
CKM_EC_EDWARDS_KEY_PAIR_GEN
CKM_EC_MONTGOMERY_KEY_PAIR_GEN
CKM_ECDSA
CKM_ECDSA_SHA1
CKM_ECDSA_SHA224
CKM_ECDSA_SHA256
CKM_ECDSA_SHA384
CKM_ECDSA_SHA512
CKM_ECDSA_SHA3_224
CKM_ECDSA_SHA3_256
CKM_ECDSA_SHA3_384
CKM_ECDSA_SHA3_512
CKM_EDDSA
CKM_XEDDSA
CKM_ECDH1_DERIVE
CKM_ECDH1_COFACTOR_DERIVE
CKM_ECMQV_DERIVE
CKM_ECDH_AES_KEY_WRAP
CKD_NULL
CKD_SHA1_KDF
CKD_SHA224_KDF
CKD_SHA256_KDF
CKD_SHA384_KDF
CKD_SHA512_KDF
CKD_SHA3_224_KDF
CKD_SHA3_256_KDF
CKD_SHA3_384_KDF
CKD_SHA3_512_KDF
CKD_SHA1_KDF_SP800
CKD_SHA224_KDF_SP800
CKD_SHA256_KDF_SP800
CKD_SHA384_KDF_SP800
CKD_SHA512_KDF_SP800
CKD_SHA3_224_KDF_SP800
CKD_SHA3_256_KDF_SP800
CKD_SHA3_384_KDF_SP800
CKD_SHA3_512_KDF_SP800
CKD_BLAKE2B_160_KDF
CKD_BLAKE2B_256_KDF
CKD_BLAKE2B_384_KDF
CKD_BLAKE2B_512_KDF
Short Weierstrass EC public key objects (object class CKO_PUBLIC_KEY,
key type CKK_EC) hold EC public keys. The following table defines
the EC public key object attributes, in addition to the common attributes
defined for this object class:
Table 65, Elliptic Curve Public Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_EC_PARAMS1,3
|
Byte array
|
DER-encoding of an ANSI X9.62 Parameters
value
|
|
CKA_EC_POINT1,4
|
Byte array
|
DER-encoding of ANSI X9.62 ECPoint value Q
|
- Refer to Table 11 for footnotes
Note: CKA_ECDSA_PARAMS is deprecated. It is replaced by
CKA_EC_PARAMS.
The CKA_EC_PARAMS attribute value is known as the “EC
domain parameters” and is defined in ANSI X9.62 as a choice of three parameter
representation methods with the following syntax:
Parameters ::= CHOICE {
ecParameters ECParameters,
oId CURVES.&id({CurveNames}),
implicitlyCA NULL,
curveName PrintableString
}
This allows detailed specification of all required values
using choice ecParameters, the use of oId as an object identifier
substitute for a particular set of Elliptic Curve domain parameters, or implicitlyCA
to indicate that the domain parameters are explicitly defined elsewhere, or curveName
to specify a curve name as e.g. define in [ANSI X9.62], [BRAINPOOL], [SEC 2],
[LEGIFRANCE]. The use of oId or curveName is recommended over
the choice ecParameters. The choice implicitlyCA must not be
used in Cryptoki.
The following is a sample template for creating an short
Weierstrass EC public key object:
CK_OBJECT_CLASS class = CKO_PUBLIC_KEY;
CK_KEY_TYPE keyType = CKK_EC;
CK_UTF8CHAR label[] = “An EC public key object”;
CK_BYTE ecParams[] = {...};
CK_BYTE ecPoint[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_EC_PARAMS, ecParams, sizeof(ecParams)},
{CKA_EC_POINT, ecPoint, sizeof(ecPoint)}
};
Short Weierstrass EC private key objects (object class CKO_PRIVATE_KEY,
key type CKK_EC) hold EC private keys. See Section 6.3 for more information about EC. The following table defines the EC private key
object attributes, in addition to the common attributes defined for this object
class:
Table 66, Elliptic Curve Private Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_EC_PARAMS1,4,6
|
Byte array
|
DER-encoding of an ANSI X9.62 Parameters
value
|
|
CKA_VALUE1,4,6,7
|
Big integer
|
ANSI X9.62 private value d
|
- Refer to Table 11 for footnotes
The CKA_EC_PARAMS attribute value is known as the “EC
domain parameters” and is defined in ANSI X9.62 as
a choice of three parameter representation methods with the following syntax:
Parameters ::= CHOICE {
ecParameters ECParameters,
oId CURVES.&id({CurveNames}),
implicitlyCA NULL,
curveName PrintableString
}
This allows detailed specification of all required values
using choice ecParameters, the use of oId as an object identifier
substitute for a particular set of Elliptic Curve domain parameters, or implicitlyCA
to indicate that the domain parameters are explicitly defined elsewhere, or curveName
to specify a curve name as e.g. define in [ANSI X9.62], [BRAINPOOL], [SEC 2],
[LEGIFRANCE]. The use of oId or curveName
is recommended over the choice ecParameters. The choice implicitlyCA
must not be used in Cryptoki.Note that when generating an EC private key, the
EC domain parameters are not specified in the key’s template. This is
because EC private keys are only generated as part of an EC key pair,
and the EC domain parameters for the pair are specified in the template for the
EC public key.
The following is a sample template for creating an short
Weierstrass EC private key object:
CK_OBJECT_CLASS class = CKO_PRIVATE_KEY;
CK_KEY_TYPE keyType = CKK_EC;
CK_UTF8CHAR label[] = “An EC private key object”;
CK_BYTE subject[] = {...};
CK_BYTE id[] = {123};
CK_BYTE ecParams[] = {...};
CK_BYTE value[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_SUBJECT, subject, sizeof(subject)},
{CKA_ID, id, sizeof(id)},
{CKA_SENSITIVE, &true, sizeof(true)},
{CKA_DERIVE, &true, sizeof(true)},
{CKA_EC_PARAMS, ecParams, sizeof(ecParams)},
{CKA_VALUE, value, sizeof(value)}
};
Edwards EC public key objects (object class CKO_PUBLIC_KEY,
key type CKK_EC_EDWARDS) hold Edwards EC public keys. The following
table defines the Edwards EC public key object attributes, in addition to the
common attributes defined for this object class:
Table 67, Edwards Elliptic Curve
Public Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_EC_PARAMS1,3
|
Byte array
|
DER-encoding of a Parameters value as defined
above
|
|
CKA_EC_POINT1,4
|
Byte array
|
Public key bytes in little endian order as
defined in RFC 8032
|
- Refer to Table 11 for footnotes
The CKA_EC_PARAMS attribute value is known as the “EC
domain parameters” and is defined in ANSI X9.62 as a choice of three parameter
representation methods. A 4th choice is added to support Edwards and
Montgomery Elliptic Curves. The CKA_EC_PARAMS attribute has the following
syntax:
Parameters
::= CHOICE {
ecParameters ECParameters,
oId CURVES.&id({CurveNames}),
implicitlyCA NULL,
curveName PrintableString
}
Edwards EC public keys only support the use of the curveName
selection to specify a curve name as defined in [RFC 8032] and the use of the oID
selection to specify a curve through an EdDSA algorithm as defined in [RFC
8410]. Note that keys defined by RFC 8032 and RFC 8410 are incompatible.
The following is a sample template for creating an Edwards
EC public key object with Edwards25519 being specified as curveName:
CK_OBJECT_CLASS class = CKO_PUBLIC_KEY;
CK_KEY_TYPE keyType = CKK_EC_EDWARDS;
CK_UTF8CHAR label[] = “An Edwards EC public key object”;
CK_BYTE ecParams[] = {0x13, 0x0c, 0x65, 0x64, 0x77, 0x61, 0x72,
0x64, 0x73, 0x32, 0x35, 0x35, 0x31, 0x39};
CK_BYTE ecPoint[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_EC_PARAMS, ecParams, sizeof(ecParams)},
{CKA_EC_POINT, ecPoint, sizeof(ecPoint)}
};
Edwards EC private key objects (object class CKO_PRIVATE_KEY,
key type CKK_EC_EDWARDS) hold Edwards EC private keys. See Section 6.3 for more information about EC. The following table defines the Edwards EC
private key object attributes, in addition to the common attributes defined for
this object class:
Table 68, Edwards Elliptic Curve
Private Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_EC_PARAMS1,4,6
|
Byte array
|
DER-encoding of a Parameters value as defined
above
|
|
CKA_VALUE1,4,6,7
|
Big integer
|
Private key bytes in little endian order as
defined in RFC 8032
|
- Refer to Table 11 for footnotes
The CKA_EC_PARAMS attribute value is known as the “EC
domain parameters” and is defined in ANSI X9.62 as a choice of three parameter
representation methods. A 4th choice is added to support Edwards and
Montgomery Elliptic Curves. The CKA_EC_PARAMS attribute has the following
syntax:
Parameters
::= CHOICE {
ecParameters ECParameters,
oId CURVES.&id({CurveNames}),
implicitlyCA NULL,
curveName PrintableString
}
Edwards EC private keys only support the use of the curveName
selection to specify a curve name as defined in [RFC 8032] and the use of the oID
selection to specify a curve through an EdDSA algorithm as defined in [RFC
8410]. Note that keys defined by RFC 8032 and RFC 8410 are incompatible.
Note that when generating an Edwards EC private key, the EC
domain parameters are not specified in the key’s template. This is
because Edwards EC private keys are only generated as part of an Edwards EC key
pair, and the EC domain parameters for the pair are specified in the
template for the Edwards EC public key.
The following is a sample template for creating an Edwards
EC private key object:
CK_OBJECT_CLASS class = CKO_PRIVATE_KEY;
CK_KEY_TYPE keyType = CKK_EC_EDWARDS;
CK_UTF8CHAR label[] = “An Edwards EC private key object”;
CK_BYTE subject[] = {...};
CK_BYTE id[] = {123};
CK_BYTE ecParams[] = {...};
CK_BYTE value[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_SUBJECT, subject, sizeof(subject)},
{CKA_ID, id, sizeof(id)},
{CKA_SENSITIVE, &true, sizeof(true)},
{CKA_DERIVE, &true, sizeof(true)},
{CKA_VALUE, value, sizeof(value)}
};
Montgomery EC public key objects (object class CKO_PUBLIC_KEY,
key type CKK_EC_MONTGOMERY) hold Montgomery EC public keys. The
following table defines the Montgomery EC public key object attributes, in
addition to the common attributes defined for this object class:
Table 69, Montgomery Elliptic
Curve Public Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_EC_PARAMS1,3
|
Byte array
|
DER-encoding of a Parameters value as defined
above
|
|
CKA_EC_POINT1,4
|
Byte array
|
Public key bytes in little endian order as
defined in RFC 7748
|
- Refer to Table 11 for footnotes
The CKA_EC_PARAMS attribute value is known as the “EC
domain parameters” and is defined in ANSI X9.62 as a choice of three parameter
representation methods. A 4th choice is added to support Edwards and
Montgomery Elliptic Curves. The CKA_EC_PARAMS attribute has the following
syntax:
Parameters
::= CHOICE {
ecParameters ECParameters,
oId CURVES.&id({CurveNames}),
implicitlyCA NULL,
curveName PrintableString
}
Montgomery EC public keys only support the use of the curveName
selection to specify a curve name as defined in [RFC7748] and the use of the oID
selection to specify a curve through an ECDH algorithm as defined in [RFC
8410]. Note that keys defined by RFC 7748 and RFC 8410 are incompatible.
The following is a sample template for creating a Montgomery
EC public key object:
CK_OBJECT_CLASS class = CKO_PUBLIC_KEY;
CK_KEY_TYPE keyType = CKK_EC_MONTGOMERY;
CK_UTF8CHAR label[] = “A Montgomery EC public key object”;
CK_BYTE ecParams[] = {...};
CK_BYTE ecPoint[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_EC_PARAMS, ecParams, sizeof(ecParams)},
{CKA_EC_POINT, ecPoint, sizeof(ecPoint)}
};
Montgomery EC private key objects (object class CKO_PRIVATE_KEY,
key type CKK_EC_MONTGOMERY) hold Montgomery EC private keys. See
Section 6.3 for more information about EC. The following table defines the
Montgomery EC private key object attributes, in addition to the common
attributes defined for this object class:
Table 70, Montgomery Elliptic
Curve Private Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_EC_PARAMS1,4,6
|
Byte array
|
DER-encoding of a Parameters value as defined
above
|
|
CKA_VALUE1,4,6,7
|
Big integer
|
Private key bytes in little endian order as
defined in RFC 7748
|
- Refer to Table 11 for footnotes
The CKA_EC_PARAMS attribute value is known as the “EC
domain parameters” and is defined in ANSI X9.62 as a choice of three parameter
representation methods. A 4th choice is added to support Edwards and
Montgomery Elliptic Curves. The CKA_EC_PARAMS attribute has the following
syntax:
Parameters
::= CHOICE {
ecParameters ECParameters,
oId CURVES.&id({CurveNames}),
implicitlyCA NULL,
curveName PrintableString
}
Montgomery EC private keys only support the use of the curveName
selection to specify a curve name as defined in [RFC7748] and the use of the oID
selection to specify a curve through an ECDH algorithm as defined in [RFC
8410]. Note that keys defined by RFC 7748 and RFC 8410 are incompatible.
Note that when generating a Montgomery EC private key, the
EC domain parameters are not specified in the key’s template. This is
because Montgomery EC private keys are only generated as part of a Montgomery EC
key pair, and the EC domain parameters for the pair are specified in the
template for the Montgomery EC public key.
The following is a sample template for creating a Montgomery
EC private key object:
CK_OBJECT_CLASS class = CKO_PRIVATE_KEY;
CK_KEY_TYPE keyType = CKK_EC_MONTGOMERY;
CK_UTF8CHAR label[] = “A Montgomery EC private key object”;
CK_BYTE subject[] = {...};
CK_BYTE id[] = {123};
CK_BYTE ecParams[] = {...};
CK_BYTE value[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_SUBJECT, subject, sizeof(subject)},
{CKA_ID, id, sizeof(id)},
{CKA_SENSITIVE, &true, sizeof(true)},
{CKA_DERIVE, &true, sizeof(true)},
{CKA_VALUE, value, sizeof(value)}
};
The short Weierstrass ECkey pair generation mechanism,
denoted CKM_EC_KEY_PAIR_GEN, is a key pair generation mechanism that uses the
method defined by the ANSI X9.62 and X9.63 standards.
The short Weierstrass EC key pair generation mechanism,
denoted CKM_EC_KEY_PAIR_GEN_W_EXTRA_BITS, is a key pair generation mechanism
that uses the method defined by FIPS 186-4 Appendix B.4.1.
These mechanisms do not have a parameter.
These mechanisms generate EC public/private key pairs with
particular EC domain parameters, as specified in the CKA_EC_PARAMS
attribute of the template for the public key. Note that this version of
Cryptoki does not include a mechanism for generating these EC domain
parameters.
These mechanism contribute the CKA_CLASS, CKA_KEY_TYPE,
and CKA_EC_POINT attributes to the new public key and the CKA_CLASS,
CKA_KEY_TYPE, CKA_EC_PARAMS and CKA_VALUE attributes to
the new private key. Other attributes supported by the EC public and private
key types (specifically, the flags indicating which functions the keys support)
may also be specified in the templates for the keys, or else are assigned
default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the minimum and
maximum supported number of bits in the field sizes, respectively. For
example, if a Cryptoki library supports only ECDSA using a field of
characteristic 2 which has between 2200 and 2300
elements, then ulMinKeySize = 201 and ulMaxKeySize = 301 (when
written in binary notation, the number 2200 consists of a 1 bit
followed by 200 0 bits. It is therefore a 201-bit number. Similarly, 2300
is a 301-bit number).
The Edwards EC key pair generation mechanism, denoted CKM_EC_EDWARDS_KEY_PAIR_GEN,
is a key pair generation mechanism for EC keys over curves represented in
Edwards form.
This mechanism does not have a parameter.
The mechanism can only generate EC public/private key pairs
over the curves edwards25519 and edwards448 as defined in RFC 8032 or the
curves id-Ed25519 and id-Ed448 as defined in RFC 8410. These curves can only
be specified in the CKA_EC_PARAMS attribute of the template for the
public key using the curveName or the oID methods. Attempts to generate
keys over these curves using any other EC key pair generation mechanism will
fail with CKR_CURVE_NOT_SUPPORTED.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_EC_POINT attributes to the new public key and the CKA_CLASS,
CKA_KEY_TYPE, CKA_EC_PARAMS and CKA_VALUE attributes to
the new private key. Other attributes supported by the Edwards EC public and
private key types (specifically, the flags indicating which functions the keys
support) may also be specified in the templates for the keys, or else are
assigned default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the minimum and
maximum supported number of bits in the field sizes, respectively. For this
mechanism, the only allowed values are 255 and 448 as RFC 8032 only defines
curves of these two sizes. A Cryptoki implementation may support one or both
of these curves and should set the ulMinKeySize and ulMaxKeySize
fields accordingly.
The Montgomery EC key pair generation mechanism, denoted CKM_EC_MONTGOMERY_KEY_PAIR_GEN,
is a key pair generation mechanism for EC keys over curves represented in
Montgomery form.
This mechanism does not have a parameter.
The mechanism can only generate Montgomery EC public/private
key pairs over the curves curve25519 and curve448 as defined in RFC 7748 or the
curves id-X25519 and id-X448 as defined in RFC 8410. These curves can only be
specified in the CKA_EC_PARAMS attribute of the template for the public
key using the curveName or oId methods. Attempts to generate keys over
these curves using any other EC key pair generation mechanism will fail with
CKR_CURVE_NOT_SUPPORTED.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_EC_POINT attributes to the new public key and the CKA_CLASS,
CKA_KEY_TYPE, CKA_EC_PARAMS and CKA_VALUE attributes to
the new private key. Other attributes supported by the EC public and private
key types (specifically, the flags indicating which functions the keys support)
may also be specified in the templates for the keys, or else are assigned
default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the minimum and
maximum supported number of bits in the field sizes, respectively. For this
mechanism, the only allowed values are 255 and 448 as RFC 7748 only defines
curves of these two sizes. A Cryptoki implementation may support one or both
of these curves and should set the ulMinKeySize and ulMaxKeySize
fields accordingly.
6.3.12 ECDSA without
hashing
Refer section 6.3.1 for signature encoding.
The ECDSA without hashing mechanism, denoted CKM_ECDSA,
is a mechanism for single-part signatures and verification for ECDSA. (This
mechanism corresponds only to the part of ECDSA that processes the hash value,
which should not be longer than 1024 bits; it does not compute the hash value.)
This mechanism does not have a parameter.
Constraints on key types and the length of data are
summarized in the following table:
Table 71, ECDSA without hashing: Key and Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Sign1
|
CKK_EC private key
|
any3
|
2nLen
|
|
C_Verify1
|
CKK_EC public key
|
any3, ≤2nLen
2
|
N/A
|
1
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the minimum and
maximum supported number of bits in the field sizes, respectively. For
example, if a Cryptoki library supports only ECDSA using a field of
characteristic 2 which has between 2200 and 2300 elements
(inclusive), then ulMinKeySize = 201 and ulMaxKeySize = 301 (when
written in binary notation, the number 2200 consists of a 1 bit
followed by 200 0 bits. It is therefore a 201-bit number. Similarly, 2300
is a 301-bit number).
Refer to section 6.3.1 for signature encoding.
The ECDSA with SHA-1, SHA-224, SHA-256, SHA-384, SHA-512,
SHA3-224, SHA3-256, SHA3-384, SHA3-512 mechanism, denoted CKM_ECDSA_[SHA1|SHA224|SHA256|SHA384|SHA512|SHA3_224|SHA3_256|SHA3_384|SHA3_512]
respectively, is a mechanism for single- and multiple-part signatures and
verification for ECDSA. This mechanism computes the entire ECDSA
specification, including the hashing with SHA-1, SHA-224, SHA-256, SHA-384,
SHA-512, SHA3-224, SHA3-256, SHA3-384, SHA3-512 respectively.
This mechanism does not have a parameter.
Constraints on key types and the length of data are
summarized in the following table:
Table 72, ECDSA with hashing: Key and Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Sign
|
CKK_EC private key
|
any
|
2nLen
|
|
C_Verify
|
CKK_EC public key
|
any, ≤2nLen
2
|
N/A
|
2
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the minimum and
maximum supported number of bits in the field sizes, respectively. For
example, if a Cryptoki library supports only ECDSA using a field of
characteristic 2 which has between 2200 and 2300
elements, then ulMinKeySize = 201 and ulMaxKeySize = 301 (when
written in binary notation, the number 2200 consists of a 1 bit
followed by 200 0 bits. It is therefore a 201-bit number. Similarly, 2300
is a 301-bit number).
The EdDSA mechanism, denoted CKM_EDDSA, is a
mechanism for single-part and multipart signatures and verification for EdDSA.
This mechanism implements the five EdDSA signature schemes defined in RFC 8032
and RFC 8410.
For curves according to RFC 8032, this mechanism has an
optional parameter, a CK_EDDSA_PARAMS structure. The absence or
presence of the parameter as well as its content is used to identify which
signature scheme is to be used. The following table enumerates the five
signature schemes defined in RFC 8032 and all supported permutations of the
mechanism parameter and its content.
Table 73, Mapping to RFC 8032
Signature Schemes
|
Signature Scheme
|
Mechanism Param
|
phFlag
|
Context Data
|
|
Ed25519
|
Not Required
|
N/A
|
N/A
|
|
Ed25519ctx
|
Required
|
False
|
Optional
|
|
Ed25519ph
|
Required
|
True
|
Optional
|
|
Ed448
|
Required
|
False
|
Optional
|
|
Ed448ph
|
Required
|
True
|
Optional
|
For curves according to RFC 8410, the mechanism is
implicitly given by the curve, which is EdDSA in pure mode.
Constraints on key types and the length of data are
summarized in the following table:
Table 74, EdDSA: Key and Data
Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Sign
|
CKK_EC_EDWARDS private key
|
any
|
2bLen
|
|
C_Verify
|
CKK_EC_EDWARDS public key
|
any, ≤2bLen 2
|
N/A
|
Note that for EdDSA in pure mode, Ed25519 and Ed448 the data
must be processed twice. Therefore, a token might need to cache all the data,
especially when used with C_SignUpdate/C_VerifyUpdate. If tokens are unable to
do so they can return CKR_TOKEN_RESOURCE_EXCEEDED.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the minimum and maximum
supported number of bits in the field sizes, respectively. For this mechanism,
the only allowed values are 255 and 448 as RFC 8032and RFC 8410 only define
curves of these two sizes. A Cryptoki implementation may support one or both
of these curves and should set the ulMinKeySize and ulMaxKeySize
fields accordingly.
The XEdDSA mechanism, denoted CKM_XEDDSA, is a
mechanism for single-part signatures and verification for XEdDSA. This
mechanism implements the XEdDSA signature scheme defined in [XEDDSA].
CKM_XEDDSA operates on CKK_EC_MONTGOMERY type EC keys, which allows these keys
to be used both for signing/verification and for Diffie-Hellman style
key-exchanges. This double use is necessary for the Extended Triple
Diffie-Hellman where the long-term identity key is used to sign short-term keys
and also contributes to the DH key-exchange.
This mechanism has a parameter, a CK_XEDDSA_PARAMS
structure.
Table 75, XEdDSA: Key and Data
Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Sign1
|
CKK_EC_MONTGOMERY private key
|
any3
|
2b
|
|
C_Verify1
|
CKK_EC_MONTGOMERY public key
|
any3, ≤2b 2
|
N/A
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the minimum and
maximum supported number of bits in the field sizes, respectively. For this
mechanism, the only allowed values are 255 and 448 as [XEDDSA] only
defines curves of these two sizes. A Cryptoki implementation may support one
or both of these curves and should set the ulMinKeySize and ulMaxKeySize
fields accordingly.
6.3.16 EC mechanism parameters
¨
CK_EDDSA_PARAMS, CK_EDDSA_PARAMS_PTR
CK_EDDSA_PARAMS is a structure that provides the
parameters for the CKM_EDDSA signature mechanism. The structure is
defined as follows:
typedef struct CK_EDDSA_PARAMS {
CK_BBOOL phFlag;
CK_ULONG ulContextDataLen;
CK_BYTE_PTR pContextData;
} CK_EDDSA_PARAMS;
The fields of the structure have the following meanings:
phFlag a
Boolean value which indicates if Prehashed variant of EdDSA should used
ulContextDataLen the
length in bytes of the context data where 0 ≤ ulContextDataLen ≤
255.
pContextData context
data shared between the signer and verifier
CK_EDDSA_PARAMS_PTR is a pointer to a CK_EDDSA_PARAMS.
¨
CK_XEDDSA_PARAMS, CK_XEDDSA_PARAMS_PTR
CK_XEDDSA_PARAMS is a structure that provides the
parameters for the CKM_XEDDSA signature mechanism. The structure is
defined as follows:
typedef struct CK_XEDDSA_PARAMS {
CK_XEDDSA_HASH_TYPE hash;
} CK_XEDDSA_PARAMS;
The fields of the structure have the following meanings:
hash a Hash
mechanism to be used by the mechanism.
CK_XEDDSA_PARAMS_PTR is a pointer to a CK_XEDDSA_PARAMS.
¨
CK_XEDDSA_HASH_TYPE, CK_XEDDSA_HASH_TYPE_PTR
CK_XEDDSA_HASH_TYPE is used to indicate the hash
function used in XEDDSA. It is defined as follows:
typedef CK_ULONG CK_XEDDSA_HASH_TYPE;
The following table lists the defined functions.
Table 76, EC: Key Derivation
Functions
|
Source
Identifier
|
|
CKM_BLAKE2B_256
|
|
CKM_BLAKE2B_512
|
|
CKM_SHA3_256
|
|
CKM_SHA3_512
|
|
CKM_SHA256
|
|
CKM_SHA512
|
CK_XEDDSA_HASH_TYPE_PTR is a pointer to a CK_XEDDSA_HASH_TYPE.
¨
CK_EC_KDF_TYPE, CK_EC_KDF_TYPE_PTR
CK_EC_KDF_TYPE is used to
indicate the Key Derivation Function (KDF) applied to derive keying data from a
shared secret. The key derivation function will be used by the EC key
agreement schemes. It is defined as follows:
typedef CK_ULONG CK_EC_KDF_TYPE;
The following table lists the defined functions.
Table 77, EC: Key Derivation Functions
|
Source
Identifier
|
|
CKD_NULL
|
|
CKD_SHA1_KDF
|
|
CKD_SHA224_KDF
|
|
CKD_SHA256_KDF
|
|
CKD_SHA384_KDF
|
|
CKD_SHA512_KDF
|
|
CKD_SHA3_224_KDF
|
|
CKD_SHA3_256_KDF
|
|
CKD_SHA3_384_KDF
|
|
CKD_SHA3_512_KDF
|
|
CKD_SHA1_KDF_SP800
|
|
CKD_SHA224_KDF_SP800
|
|
CKD_SHA256_KDF_SP800
|
|
CKD_SHA384_KDF_SP800
|
|
CKD_SHA512_KDF_SP800
|
|
CKD_SHA3_224_KDF_SP800
|
|
CKD_SHA3_256_KDF_SP800
|
|
CKD_SHA3_384_KDF_SP800
|
|
CKD_SHA3_512_KDF_SP800
|
|
CKD_BLAKE2B_160_KDF
|
|
CKD_BLAKE2B_256_KDF
|
|
CKD_BLAKE2B_384_KDF
|
|
CKD_BLAKE2B_512_KDF
|
The key derivation function CKD_NULL produces a raw
shared secret value without applying any key derivation function.
The key derivation functions CKD_[SHA1|SHA224|SHA384|SHA512|SHA3_224|SHA3_256|SHA3_384|SHA3_512]_KDF,
which are based on SHA-1, SHA-224, SHA-384, SHA-512, SHA3-224, SHA3-256,
SHA3-384, SHA3-512 respectively, derive keying data from the shared secret
value as defined in [ANSI X9.63].
The key derivation functions CKD_[SHA1|SHA224|SHA384|SHA512|SHA3_224|SHA3_256|SHA3_384|SHA3_512]_KDF_SP800,
which are based on SHA-1, SHA-224, SHA-384, SHA-512, SHA3-224, SHA3-256,
SHA3-384, SHA3-512 respectively, derive keying data from the shared secret
value as defined in [FIPS SP800-56A] section 5.8.1.1.
The key derivation functions CKD_BLAKE2B_[160|256|384|512]_KDF,
which are based on the Blake2b family of hashes, derive keying data from
the shared secret value as defined in [FIPS SP800-56A] section 5.8.1.1. CK_EC_KDF_TYPE_PTR
is a pointer to a CK_EC_KDF_TYPE.
¨
CK_ECDH1_DERIVE_PARAMS, CK_ECDH1_DERIVE_PARAMS_PTR
CK_ECDH1_DERIVE_PARAMS is a structure that provides
the parameters for the CKM_ECDH1_DERIVE and CKM_ECDH1_COFACTOR_DERIVE
key derivation mechanisms, where each party contributes one key pair. The
structure is defined as follows:
typedef struct CK_ECDH1_DERIVE_PARAMS {
CK_EC_KDF_TYPE kdf;
CK_ULONG ulSharedDataLen;
CK_BYTE_PTR pSharedData;
CK_ULONG ulPublicDataLen;
CK_BYTE_PTR pPublicData;
} CK_ECDH1_DERIVE_PARAMS;
The fields of the structure have the following meanings:
kdf key
derivation function used on the shared secret value
ulSharedDataLen the length in
bytes of the shared info
pSharedData some data
shared between the two parties
ulPublicDataLen the length
in bytes of the other party’s EC public key
pPublicData[1] pointer to other party’s
EC public key value. For short Weierstrass EC keys: a token MUST be able to
accept this value encoded as a raw octet string (as per section A.5.2 of [ANSI
X9.62]). A token MAY, in addition, support accepting this value as a
DER-encoded ECPoint (as per section E.6 of [ANSI X9.62]) i.e. the same as a
CKA_EC_POINT encoding. The calling application is responsible for converting
the offered public key to the compressed or uncompressed forms of these
encodings if the token does not support the offered form.
For Montgomery keys: the public key is provided as bytes in little endian order
as defined in RFC 7748.
With the key derivation function CKD_NULL, pSharedData
must be NULL and ulSharedDataLen must be zero. With the key derivation
functions CKD_[SHA1|SHA224|SHA384|SHA512|SHA3_224|SHA3_256|SHA3_384|SHA3_512]_KDF,
CKD_[SHA1|SHA224|SHA384|SHA512|SHA3_224|SHA3_256|SHA3_384|SHA3_512]_KDF_SP800,
an optional pSharedData may be supplied, which consists of some data
shared by the two parties intending to share the shared secret. Otherwise, pSharedData
must be NULL and ulSharedDataLen must be zero.
CK_ECDH1_DERIVE_PARAMS_PTR is a pointer to a CK_ECDH1_DERIVE_PARAMS.
¨
CK_ECDH2_DERIVE_PARAMS, CK_ECDH2_DERIVE_PARAMS_PTR
CK_ECDH2_DERIVE_PARAMS
is a structure that provides the parameters to the CKM_ECMQV_DERIVE
key derivation mechanism, where each party contributes two key pairs. The
structure is defined as follows:
typedef struct CK_ECDH2_DERIVE_PARAMS {
CK_EC_KDF_TYPE kdf;
CK_ULONG ulSharedDataLen;
CK_BYTE_PTR pSharedData;
CK_ULONG ulPublicDataLen;
CK_BYTE_PTR pPublicData;
CK_ULONG ulPrivateDataLen;
CK_OBJECT_HANDLE hPrivateData;
CK_ULONG ulPublicDataLen2;
CK_BYTE_PTR pPublicData2;
} CK_ECDH2_DERIVE_PARAMS;
The fields of the structure have the following meanings:
kdf key
derivation function used on the shared secret value
ulSharedDataLen the length in
bytes of the shared info
pSharedData some data
shared between the two parties
ulPublicDataLen the length
in bytes of the other party’s first EC public key
pPublicData pointer to
other party’s first EC public key value. Encoding rules are as per pPublicData
of CK_ECDH1_DERIVE_PARAMS
ulPrivateDataLen the length
in bytes of the second EC private key
hPrivateData key handle
for second EC private key value
ulPublicDataLen2 the length in
bytes of the other party’s second EC public key
pPublicData2 pointer to
other party’s second EC public key value. Encoding rules are as per
pPublicData of CK_ECDH1_DERIVE_PARAMS
With the key derivation function CKD_NULL, pSharedData
must be NULL and ulSharedDataLen must be zero. With the key derivation
function CKD_SHA1_KDF, an optional pSharedData may be supplied,
which consists of some data shared by the two parties intending to share the
shared secret. Otherwise, pSharedData must be NULL and ulSharedDataLen
must be zero.
CK_ECDH2_DERIVE_PARAMS_PTR is a pointer to a CK_ECDH2_DERIVE_PARAMS.
¨
CK_ECMQV_DERIVE_PARAMS, CK_ECMQV_DERIVE_PARAMS_PTR
CK_ECMQV_DERIVE_PARAMS is
a structure that provides the parameters to the CKM_ECMQV_DERIVE key
derivation mechanism, where each party contributes two key pairs. The
structure is defined as follows:
typedef struct CK_ECMQV_DERIVE_PARAMS {
CK_EC_KDF_TYPE kdf;
CK_ULONG ulSharedDataLen;
CK_BYTE_PTR pSharedData;
CK_ULONG ulPublicDataLen;
CK_BYTE_PTR pPublicData;
CK_ULONG ulPrivateDataLen;
CK_OBJECT_HANDLE hPrivateData;
CK_ULONG ulPublicDataLen2;
CK_BYTE_PTR pPublicData2;
CK_OBJECT_HANDLE publicKey;
} CK_ECMQV_DERIVE_PARAMS;
The fields of the structure have the following meanings:
kdf key
derivation function used on the shared secret value
ulSharedDataLen the length in
bytes of the shared info
pSharedData some data
shared between the two parties
ulPublicDataLen the length
in bytes of the other party’s first EC public key
pPublicData pointer to
other party’s first EC public key value. Encoding rules are as per pPublicData
of CK_ECDH1_DERIVE_PARAMS
ulPrivateDataLen the length
in bytes of the second EC private key
hPrivateData key handle
for second EC private key value
ulPublicDataLen2 the length in
bytes of the other party’s second EC public key
pPublicData2 pointer to
other party’s second EC public key value. Encoding rules are as per
pPublicData of CK_ECDH1_DERIVE_PARAMS
publicKey Handle to
the first party’s ephemeral public key
With the key derivation function CKD_NULL, pSharedData
must be NULL and ulSharedDataLen must be zero. With the key derivation
functions CKD_[SHA1|SHA224|SHA384|SHA512|SHA3_224|SHA3_256|SHA3_384|SHA3_512]_KDF,
CKD_[SHA1|SHA224|SHA384|SHA512|SHA3_224|SHA3_256|SHA3_384|SHA3_512]_KDF_SP800,
an optional pSharedData may be supplied, which consists of some data
shared by the two parties intending to share the shared secret. Otherwise, pSharedData
must be NULL and ulSharedDataLen must be zero.
CK_ECMQV_DERIVE_PARAMS_PTR is a pointer to a CK_ECMQV_DERIVE_PARAMS.
The Elliptic Curve Diffie-Hellman (ECDH) key derivation
mechanism, denoted CKM_ECDH1_DERIVE, is a mechanism for key derivation
based on the Diffie-Hellman version of the Elliptic Curve key agreement scheme,
as defined in ANSI X9.63 for short Weierstrass EC keys and RFC 7748 for
Montgomery keys, where each party contributes one key pair all using the same
EC domain parameters.
It has a parameter, a CK_ECDH1_DERIVE_PARAMS
structure.
This mechanism derives a secret value, and truncates the
result according to the CKA_KEY_TYPE attribute of the template and, if
it has one and the key type supports it, the CKA_VALUE_LEN attribute of
the template. (The truncation removes bytes from the leading end of the secret
value.) The mechanism contributes the result as the CKA_VALUE attribute
of the new key; other attributes required by the key type must be specified in
the template.
This mechanism has the following rules about key sensitivity
and extractability:
·
The CKA_SENSITIVE and CKA_EXTRACTABLE attributes in
the template for the new key can both be specified to be either CK_TRUE or
CK_FALSE. If omitted, these attributes each take on some default value.
·
If the base key has its CKA_ALWAYS_SENSITIVE attribute set
to CK_FALSE, then the derived key will as well. If the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE, then the derived key has its CKA_ALWAYS_SENSITIVE
attribute set to the same value as its CKA_SENSITIVE attribute.
·
Similarly, if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_FALSE, then the derived key will, too. If the base key has
its CKA_NEVER_EXTRACTABLE attribute set to CK_TRUE, then the derived key
has its CKA_NEVER_EXTRACTABLE attribute set to the opposite value
from its CKA_EXTRACTABLE attribute.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the minimum and
maximum supported number of bits in the field sizes, respectively. For
example, if a Cryptoki library supports only EC using a field of characteristic
2 which has between 2200 and 2300 elements, then ulMinKeySize
= 201 and ulMaxKeySize = 301 (when written in binary notation, the
number 2200 consists of a 1 bit followed by 200 0 bits. It is
therefore a 201-bit number. Similarly, 2300 is a 301-bit number).
Constraints on key types are summarized in the following
table:
Table 78: ECDH: Allowed Key
Types
|
Function
|
Key type
|
|
C_Derive
|
CKK_EC or CKK_EC_MONTGOMERY
|
The Elliptic Curve Diffie-Hellman (ECDH) with cofactor key
derivation mechanism, denoted CKM_ECDH1_COFACTOR_DERIVE, is a mechanism
for key derivation based on the cofactor Diffie-Hellman version of the Elliptic
Curve key agreement scheme, as defined in ANSI X9.63, where each party
contributes one key pair all using the same EC domain parameters. Cofactor
multiplication is computationally efficient and helps to prevent security
problems like small group attacks.
It has a parameter, a CK_ECDH1_DERIVE_PARAMS
structure.
This mechanism derives a secret value, and truncates the
result according to the CKA_KEY_TYPE attribute of the template and, if
it has one and the key type supports it, the CKA_VALUE_LEN attribute of
the template. (The truncation removes bytes from the leading end of the secret
value.) The mechanism contributes the result as the CKA_VALUE attribute
of the new key; other attributes required by the key type must be specified in
the template.
This mechanism has the following rules about key sensitivity
and extractability:
·
The CKA_SENSITIVE and CKA_EXTRACTABLE attributes in
the template for the new key can both be specified to be either CK_TRUE or
CK_FALSE. If omitted, these attributes each take on some default value.
·
If the base key has its CKA_ALWAYS_SENSITIVE attribute set
to CK_FALSE, then the derived key will as well. If the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE, then the derived key has its CKA_ALWAYS_SENSITIVE
attribute set to the same value as its CKA_SENSITIVE attribute.
·
Similarly, if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_FALSE, then the derived key will, too. If the base key has
its CKA_NEVER_EXTRACTABLE attribute set to CK_TRUE, then the derived key
has its CKA_NEVER_EXTRACTABLE attribute set to the opposite value
from its CKA_EXTRACTABLE attribute.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the minimum and
maximum supported number of bits in the field sizes, respectively. For
example, if a Cryptoki library supports only EC using a field of characteristic
2 which has between 2200 and 2300 elements, then ulMinKeySize
= 201 and ulMaxKeySize = 301 (when written in binary notation, the
number 2200 consists of a 1 bit followed by 200 0 bits. It is
therefore a 201-bit number. Similarly, 2300 is a 301-bit number).
Constraints on key types are summarized in the following
table:
Table 79: ECDH with cofactor:
Allowed Key Types
|
Function
|
Key type
|
|
C_Derive
|
CKK_EC
|
The Elliptic Curve Menezes-Qu-Vanstone (ECMQV) key
derivation mechanism, denoted CKM_ECMQV_DERIVE, is a mechanism for key
derivation based the MQV version of the Elliptic Curve key agreement scheme, as
defined in ANSI X9.63, where each party contributes two key pairs all using the
same EC domain parameters.
It has a parameter, a CK_ECMQV_DERIVE_PARAMS
structure.
This mechanism derives a secret value, and truncates the
result according to the CKA_KEY_TYPE attribute of the template and, if
it has one and the key type supports it, the CKA_VALUE_LEN attribute of
the template. (The truncation removes bytes from the leading end of the secret
value.) The mechanism contributes the result as the CKA_VALUE attribute
of the new key; other attributes required by the key type must be specified in
the template.
This mechanism has the following rules about key sensitivity
and extractability:
·
The CKA_SENSITIVE and CKA_EXTRACTABLE attributes in
the template for the new key can both be specified to be either CK_TRUE or
CK_FALSE. If omitted, these attributes each take on some default value.
·
If the base key has its CKA_ALWAYS_SENSITIVE attribute set
to CK_FALSE, then the derived key will as well. If the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE, then the derived key has its CKA_ALWAYS_SENSITIVE
attribute set to the same value as its CKA_SENSITIVE attribute.
·
Similarly, if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_FALSE, then the derived key will, too. If the base key has
its CKA_NEVER_EXTRACTABLE attribute set to CK_TRUE, then the derived key
has its CKA_NEVER_EXTRACTABLE attribute set to the opposite value
from its CKA_EXTRACTABLE attribute.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the minimum and
maximum supported number of bits in the field sizes, respectively. For example,
if a Cryptoki library supports only EC using a field of characteristic 2 which
has between 2200 and 2300 elements, then ulMinKeySize
= 201 and ulMaxKeySize = 301 (when written in binary notation, the
number 2200 consists of a 1 bit followed by 200 0 bits. It is
therefore a 201-bit number. Similarly, 2300 is a 301-bit number).
Constraints on key types are summarized in the following
table:
Table 80: ECDH MQV: Allowed Key
Types
|
Function
|
Key type
|
|
C_Derive
|
CKK_EC
|
The ECDH AES KEY WRAP mechanism, denoted CKM_ECDH_AES_KEY_WRAP,
is a mechanism based on Elliptic Curve public-key crypto-system and the AES key
wrap mechanism. It supports single-part key wrapping; and key unwrapping.
It has a parameter, a CK_ECDH_AES_KEY_WRAP_PARAMS
structure.
The
mechanism can wrap and unwrap an asymmetric target key of any length and type
using an EC key.
-
A temporary AES key
is derived from a temporary EC key and the wrapping EC key using the CKM_ECDH1_DERIVE
mechanism.
-
The derived AES key
is used for wrapping the target key using the CKM_AES_KEY_WRAP_KWP
mechanism.
For
wrapping, the mechanism -
- Generates
a temporary random EC key (transport key) having the same parameters as
the wrapping EC key (and domain parameters). Saves the transport key
public key material.
- Performs
ECDH operation using CKM_ECDH1_DERIVE with parameters of kdf,
ulSharedDataLen and pSharedData using the private key of the transport EC
key and the public key of wrapping EC key and gets the first ulAESKeyBits
bits of the derived key to be the temporary AES key.
- Wraps
the target key with the temporary AES key using CKM_AES_KEY_WRAP_KWP.
- Zeroizes
the temporary AES key and EC transport private key.
- Concatenates
public key material of the transport key and output the concatenated blob.
The first part is the public key material of the transport key and the
second part is the wrapped target key.
The private
target key will be encoded as defined in section 6.7.
The use of
Attributes in the PrivateKeyInfo structure is OPTIONAL. In case of conflicts
between the object attribute template, and Attributes in the PrivateKeyInfo
structure, an error should be thrown.
For
unwrapping, the mechanism -
- Splits
the input into two parts. The first part is the public key material of the
transport key and the second part is the wrapped target key. The length of
the first part is equal to the length of the public key material of the
unwrapping EC key.
Note: since the transport key and the
wrapping EC key share the same domain, the length of the public key material of
the transport key is the same length of the public key material of the
unwrapping EC key.
- Performs
ECDH operation using CKM_ECDH1_DERIVE with parameters of kdf,
ulSharedDataLen and pSharedData using the private part of unwrapping EC
key and the public part of the transport EC key and gets first
ulAESKeyBits bits of the derived key to be the temporary AES key.
- Un-wraps
the target key from the second part with the temporary AES key using CKM_AES_KEY_WRAP_KWP.
- Zeroizes
the temporary AES key.
Table
81, CKM_ECDH_AES_KEY_WRAP Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_ECDH_AES_KEY_WRAP
|
|
|
|
|
|
ü
|
|
|
1SR =
SignRecover, VR = VerifyRecover
|
Constraints on key types are summarized in the following
table:
Table 82: ECDH AES Key Wrap:
Allowed Key Types
|
Function
|
Key type
|
|
C_Wrap / C_Unwrap
|
CKK_EC or CKK_EC_MONTGOMERY
|
¨ CK_ECDH_AES_KEY_WRAP_PARAMS;
CK_ECDH_AES_KEY_WRAP_PARAMS_PTR
CK_ECDH_AES_KEY_WRAP_PARAMS is a structure that
provides the parameters to the CKM_ECDH_AES_KEY_WRAP mechanism. It is
defined as follows:
typedef struct CK_ECDH_AES_KEY_WRAP_PARAMS {
CK_ULONG ulAESKeyBits;
CK_EC_KDF_TYPE kdf;
CK_ULONG ulSharedDataLen;
CK_BYTE_PTR pSharedData;
} CK_ECDH_AES_KEY_WRAP_PARAMS;
The fields
of the structure have the following meanings:
ulAESKeyBits length of
the temporary AES key in bits. Can be only 128, 192 or 256.
kdf key derivation
function used on the shared secret value to generate AES key.
ulSharedDataLen the length in
bytes of the shared info
pSharedData Some data
shared between the two parties
CK_ECDH_AES_KEY_WRAP_PARAMS_PTR is a pointer to a CK_ECDH_AES_KEY_WRAP_PARAMS.
When CKM_ECDSA is operated in FIPS mode, the curves SHALL
either be NIST recommended curves (with a fixed set of domain parameters) or
curves with domain parameters generated as specified by ANSI X9.64. The NIST
recommended curves are:
P-192, P-224, P-256, P-384, P-521
K-163, B-163, K-233, B-233
K-283, B-283, K-409, B-409
K-571, B-571
Table
83, Diffie-Hellman Mechanisms vs. Functions
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_DH_PKCS_KEY_PAIR_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_DH_PKCS_PARAMETER_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_DH_PKCS_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_X9_42_DH_KEY_PAIR_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_X9_42_DH_PARAMETER_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_X9_42_DH_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_X9_42_DH_HYBRID_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_X9_42_MQV_DERIVE
|
|
|
|
|
|
|
ü
|
This section defines the key type “CKK_DH” for type
CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of [DH] key objects.
Mechanisms:
CKM_DH_PKCS_KEY_PAIR_GEN
CKM_DH_PKCS_PARAMETER_GEN
CKM_DH_PKCS_DERIVE
CKM_X9_42_DH_KEY_PAIR_GEN
CKM_X9_42_DH_PARAMETER_GEN
CKM_X9_42_DH_DERIVE
CKM_X9_42_DH_HYBRID_DERIVE
CKM_X9_42_MQV_DERIVE
Diffie-Hellman public key objects (object class CKO_PUBLIC_KEY,
key type CKK_DH) hold Diffie-Hellman public keys. The following
table defines the Diffie-Hellman public key object attributes, in addition to
the common attributes defined for this object class:
Table 84, Diffie-Hellman Public Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_PRIME1,3
|
Big integer
|
Prime p
|
|
CKA_BASE1,3
|
Big integer
|
Base g
|
|
CKA_VALUE1,4
|
Big integer
|
Public value y
|
- Refer to Table 11 for footnotes
The CKA_PRIME and CKA_BASE attribute values
are collectively the “Diffie-Hellman domain parameters”. Depending on the
token, there may be limits on the length of the key components. See [PKCS #3]
for more information on Diffie-Hellman keys.
The following is a sample template for creating a
Diffie-Hellman public key object:
CK_OBJECT_CLASS class = CKO_PUBLIC_KEY;
CK_KEY_TYPE keyType = CKK_DH;
CK_UTF8CHAR label[] = “A Diffie-Hellman public key object”;
CK_BYTE prime[] = {...};
CK_BYTE base[] = {...};
CK_BYTE value[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_PRIME, prime, sizeof(prime)},
{CKA_BASE, base, sizeof(base)},
{CKA_VALUE, value, sizeof(value)}
};
X9.42 Diffie-Hellman public key objects (object class CKO_PUBLIC_KEY,
key type CKK_X9_42_DH) hold X9.42 Diffie-Hellman public keys. The
following table defines the X9.42 Diffie-Hellman public key object attributes,
in addition to the common attributes defined for this object class:
Table 85, X9.42 Diffie-Hellman Public Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_PRIME1,3
|
Big integer
|
Prime p (³ 1024 bits, in steps of 256 bits)
|
|
CKA_BASE1,3
|
Big integer
|
Base g
|
|
CKA_SUBPRIME1,3
|
Big integer
|
Subprime q (³ 160 bits)
|
|
CKA_VALUE1,4
|
Big integer
|
Public value y
|
- Refer to Table 11 for footnotes
The CKA_PRIME, CKA_BASE and CKA_SUBPRIME
attribute values are collectively the “X9.42 Diffie-Hellman domain
parameters”. See the ANSI X9.42 standard for more information on X9.42
Diffie-Hellman keys.
The following is a sample template for creating a X9.42
Diffie-Hellman public key object:
CK_OBJECT_CLASS class = CKO_PUBLIC_KEY;
CK_KEY_TYPE keyType = CKK_X9_42_DH;
CK_UTF8CHAR label[] = “A X9.42 Diffie-Hellman public key
object”;
CK_BYTE prime[] = {...};
CK_BYTE base[] = {...};
CK_BYTE subprime[] = {...};
CK_BYTE value[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_PRIME, prime, sizeof(prime)},
{CKA_BASE, base, sizeof(base)},
{CKA_SUBPRIME, subprime, sizeof(subprime)},
{CKA_VALUE, value, sizeof(value)}
};
Diffie-Hellman private key objects (object class CKO_PRIVATE_KEY,
key type CKK_DH) hold Diffie-Hellman private keys. The following
table defines the Diffie-Hellman private key object attributes, in addition to
the common attributes defined for this object class:
Table 86, Diffie-Hellman Private Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_PRIME1,4,6
|
Big integer
|
Prime p
|
|
CKA_BASE1,4,6
|
Big integer
|
Base g
|
|
CKA_VALUE1,4,6,7
|
Big integer
|
Private value x
|
|
CKA_VALUE_BITS2,6
|
CK_ULONG
|
Length in bits of private value x
|
- Refer to Table 11 for footnotes
The CKA_PRIME and CKA_BASE attribute values
are collectively the “Diffie-Hellman domain parameters”. Depending on the
token, there may be limits on the length of the key components. See [PKCS #3]
for more information on Diffie-Hellman keys.
Note that when generating a Diffie-Hellman private key, the
Diffie-Hellman parameters are not specified in the key’s template. This
is because Diffie-Hellman private keys are only generated as part of a
Diffie-Hellman key pair, and the Diffie-Hellman parameters for the pair
are specified in the template for the Diffie-Hellman public key.
The following is a sample template for creating a
Diffie-Hellman private key object:
CK_OBJECT_CLASS class = CKO_PRIVATE_KEY;
CK_KEY_TYPE keyType = CKK_DH;
CK_UTF8CHAR label[] = “A Diffie-Hellman private key object”;
CK_BYTE subject[] = {...};
CK_BYTE id[] = {123};
CK_BYTE prime[] = {...};
CK_BYTE base[] = {...};
CK_BYTE value[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_SUBJECT, subject, sizeof(subject)},
{CKA_ID, id, sizeof(id)},
{CKA_SENSITIVE, &true, sizeof(true)},
{CKA_DERIVE, &true, sizeof(true)},
{CKA_PRIME, prime, sizeof(prime)},
{CKA_BASE, base, sizeof(base)},
{CKA_VALUE, value, sizeof(value)}
};
X9.42 Diffie-Hellman private key objects (object class CKO_PRIVATE_KEY,
key type CKK_X9_42_DH) hold X9.42 Diffie-Hellman private keys. The
following table defines the X9.42 Diffie-Hellman private key object attributes,
in addition to the common attributes defined for this object class:
Table 87, X9.42 Diffie-Hellman Private Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_PRIME1,4,6
|
Big integer
|
Prime p (³ 1024 bits, in steps of 256 bits)
|
|
CKA_BASE1,4,6
|
Big integer
|
Base g
|
|
CKA_SUBPRIME1,4,6
|
Big integer
|
Subprime q (³ 160 bits)
|
|
CKA_VALUE1,4,6,7
|
Big integer
|
Private value x
|
- Refer to Table 11 for footnotes
The CKA_PRIME, CKA_BASE and CKA_SUBPRIME
attribute values are collectively the “X9.42 Diffie-Hellman domain
parameters”. Depending on the token, there may be limits on the length of the
key components. See the ANSI X9.42 standard for more information on X9.42
Diffie-Hellman keys.
Note that when generating a X9.42 Diffie-Hellman private
key, the X9.42 Diffie-Hellman domain parameters are not specified in the
key’s template. This is because X9.42 Diffie-Hellman private keys are only
generated as part of a X9.42 Diffie-Hellman key pair, and the X9.42
Diffie-Hellman domain parameters for the pair are specified in the template for
the X9.42 Diffie-Hellman public key.
The following is a sample template for creating a X9.42
Diffie-Hellman private key object:
CK_OBJECT_CLASS class = CKO_PRIVATE_KEY;
CK_KEY_TYPE keyType = CKK_X9_42_DH;
CK_UTF8CHAR label[] = “A X9.42
Diffie-Hellman private key object”;
CK_BYTE subject[] = {...};
CK_BYTE id[] = {123};
CK_BYTE prime[] = {...};
CK_BYTE base[] = {...};
CK_BYTE subprime[] = {...};
CK_BYTE value[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_SUBJECT, subject, sizeof(subject)},
{CKA_ID, id, sizeof(id)},
{CKA_SENSITIVE, &true, sizeof(true)},
{CKA_DERIVE, &true, sizeof(true)},
{CKA_PRIME, prime, sizeof(prime)},
{CKA_BASE, base, sizeof(base)},
{CKA_SUBPRIME, subprime, sizeof(subprime)},
{CKA_VALUE, value, sizeof(value)}
};
Diffie-Hellman domain parameter objects (object class CKO_DOMAIN_PARAMETERS,
key type CKK_DH) hold Diffie-Hellman domain parameters. The
following table defines the Diffie-Hellman domain parameter object attributes,
in addition to the common attributes defined for this object class:
Table 88, Diffie-Hellman Domain Parameter Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_PRIME1,4
|
Big integer
|
Prime p
|
|
CKA_BASE1,4
|
Big integer
|
Base g
|
|
CKA_PRIME_BITS2,3
|
CK_ULONG
|
Length of the prime value.
|
- Refer to Table 11 for footnotes
The CKA_PRIME and CKA_BASE attribute values
are collectively the “Diffie-Hellman domain parameters”. Depending on the
token, there may be limits on the length of the key components. See [PKCS #3]
for more information on Diffie-Hellman domain parameters.
The following is a sample template for creating a
Diffie-Hellman domain parameter object:
CK_OBJECT_CLASS class = CKO_DOMAIN_PARAMETERS;
CK_KEY_TYPE keyType = CKK_DH;
CK_UTF8CHAR label[] = “A Diffie-Hellman domain parameters
object”;
CK_BYTE prime[] = {...};
CK_BYTE base[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_PRIME, prime, sizeof(prime)},
{CKA_BASE, base, sizeof(base)},
};
X9.42 Diffie-Hellman domain parameters objects (object class
CKO_DOMAIN_PARAMETERS, key type CKK_X9_42_DH) hold X9.42
Diffie-Hellman domain parameters. The following table defines the X9.42
Diffie-Hellman domain parameters object attributes, in addition to the common
attributes defined for this object class:
Table 89, X9.42 Diffie-Hellman Domain Parameters Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_PRIME1,4
|
Big integer
|
Prime p (³ 1024 bits, in steps of 256 bits)
|
|
CKA_BASE1,4
|
Big integer
|
Base g
|
|
CKA_SUBPRIME1,4
|
Big integer
|
Subprime q (³ 160 bits)
|
|
CKA_PRIME_BITS2,3
|
CK_ULONG
|
Length of the prime value.
|
|
CKA_SUBPRIME_BITS2,3
|
CK_ULONG
|
Length of the subprime value.
|
- Refer to Table 11 for footnotes
The CKA_PRIME, CKA_BASE and CKA_SUBPRIME
attribute values are collectively the “X9.42 Diffie-Hellman domain
parameters”. Depending on the token, there may be limits on the length of the
domain parameters components. See the ANSI X9.42 standard for more information
on X9.42 Diffie-Hellman domain parameters.
The following is a sample template for creating a X9.42
Diffie-Hellman domain parameters object:
CK_OBJECT_CLASS class = CKO_DOMAIN_PARAMETERS;
CK_KEY_TYPE keyType = CKK_X9_42_DH;
CK_UTF8CHAR label[] = “A X9.42 Diffie-Hellman domain parameters
object”;
CK_BYTE prime[] = {...};
CK_BYTE base[] = {...};
CK_BYTE subprime[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_PRIME, prime, sizeof(prime)},
{CKA_BASE, base, sizeof(base)},
{CKA_SUBPRIME, subprime, sizeof(subprime)},
};
The PKCS #3 Diffie-Hellman key pair generation mechanism,
denoted CKM_DH_PKCS_KEY_PAIR_GEN, is a key pair generation mechanism based
on Diffie-Hellman key agreement, as defined in [PKCS #3]. This is what PKCS #3
calls “phase I”. It does not have a parameter.
The mechanism generates Diffie-Hellman public/private key
pairs with a particular prime and base, as specified in the CKA_PRIME
and CKA_BASE attributes of the template for the public key. If the CKA_VALUE_BITS
attribute of the private key is specified, the mechanism limits the length in
bits of the private value, as described in [PKCS #3].
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new public key and the CKA_CLASS,
CKA_KEY_TYPE, CKA_PRIME, CKA_BASE, and CKA_VALUE
(and the CKA_VALUE_BITS attribute, if it is not already provided in the
template) attributes to the new private key; other attributes required by the
Diffie-Hellman public and private key types must be specified in the templates.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
Diffie-Hellman prime sizes, in bits.
The PKCS #3 Diffie-Hellman domain parameter generation
mechanism, denoted CKM_DH_PKCS_PARAMETER_GEN, is a domain parameter
generation mechanism based on Diffie-Hellman key agreement, as defined in [PKCS
#3].
It does not have a parameter.
The mechanism generates Diffie-Hellman domain parameters
with a particular prime length in bits, as specified in the CKA_PRIME_BITS
attribute of the template.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
CKA_PRIME, CKA_BASE, and CKA_PRIME_BITS attributes to the
new object. Other attributes supported by the Diffie-Hellman domain parameter
types may also be specified in the template, or else are assigned default initial
values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
Diffie-Hellman prime sizes, in bits.
The PKCS #3 Diffie-Hellman key derivation mechanism, denoted
CKM_DH_PKCS_DERIVE, is a mechanism for key derivation based on
Diffie-Hellman key agreement, as defined in [PKCS #3]. This is what PKCS #3
calls “phase II”.
It has a parameter, which is the public value of the other
party in the key agreement protocol, represented as a Cryptoki “Big integer” (i.e.,
a sequence of bytes, most-significant byte first).
This mechanism derives a secret key from a Diffie-Hellman
private key and the public value of the other party. It computes a
Diffie-Hellman secret value from the public value and private key according to [PKCS
#3], and truncates the result according to the CKA_KEY_TYPE attribute of
the template and, if it has one and the key type supports it, the CKA_VALUE_LEN
attribute of the template. (The truncation removes bytes from the leading end
of the secret value.) The mechanism contributes the result as the CKA_VALUE
attribute of the new key; other attributes required by the key type must be
specified in the template.
This mechanism has the following rules about key sensitivity
and extractability:
·
The CKA_SENSITIVE and CKA_EXTRACTABLE attributes in
the template for the new key can both be specified to be either CK_TRUE or
CK_FALSE. If omitted, these attributes each take on some default value.
·
If the base key has its CKA_ALWAYS_SENSITIVE attribute set
to CK_FALSE, then the derived key will as well. If the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE, then the derived key has its CKA_ALWAYS_SENSITIVE
attribute set to the same value as its CKA_SENSITIVE attribute.
·
Similarly, if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_FALSE, then the derived key will, too. If the base key has
its CKA_NEVER_EXTRACTABLE attribute set to CK_TRUE, then the derived key
has its CKA_NEVER_EXTRACTABLE attribute set to the opposite value
from its CKA_EXTRACTABLE attribute.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
Diffie-Hellman prime sizes, in bits.
¨
CK_X9_42_DH_KDF_TYPE, CK_X9_42_DH_KDF_TYPE_PTR
CK_X9_42_DH_KDF_TYPE is used to indicate the Key
Derivation Function (KDF) applied to derive keying data from a shared secret.
The key derivation function will be used by the X9.42 Diffie-Hellman key
agreement schemes. It is defined as follows:
typedef CK_ULONG CK_X9_42_DH_KDF_TYPE;
The following table lists the defined functions.
Table 90, X9.42 Diffie-Hellman Key Derivation Functions
|
Source
Identifier
|
|
CKD_NULL
|
|
CKD_SHA1_KDF_ASN1
|
|
CKD_SHA1_KDF_CONCATENATE
|
The key derivation function CKD_NULL produces a raw
shared secret value without applying any key derivation function whereas the
key derivation functions CKD_SHA1_KDF_ASN1 and CKD_SHA1_KDF_CONCATENATE,
which are both based on SHA-1, derive keying data from the shared secret value
as defined in the ANSI X9.42 standard.
CK_X9_42_DH_KDF_TYPE_PTR is a pointer to a CK_X9_42_DH_KDF_TYPE.
¨ CK_X9_42_DH1_DERIVE_PARAMS,
CK_X9_42_DH1_DERIVE_PARAMS_PTR
CK_X9_42_DH1_DERIVE_PARAMS is a structure that
provides the parameters to the CKM_X9_42_DH_DERIVE key derivation
mechanism, where each party contributes one key pair. The structure is defined
as follows:
typedef struct CK_X9_42_DH1_DERIVE_PARAMS {
CK_X9_42_DH_KDF_TYPE kdf;
CK_ULONG ulOtherInfoLen;
CK_BYTE_PTR pOtherInfo;
CK_ULONG ulPublicDataLen;
CK_BYTE_PTR pPublicData;
} CK_X9_42_DH1_DERIVE_PARAMS;
The fields of the structure have the following meanings:
kdf key
derivation function used on the shared secret value
ulOtherInfoLen the length
in bytes of the other info
pOtherInfo some data
shared between the two parties
ulPublicDataLen the length
in bytes of the other party’s X9.42 Diffie-Hellman public key
pPublicData pointer to
other party’s X9.42 Diffie-Hellman public key value
With the key derivation function CKD_NULL, pOtherInfo
must be NULL and ulOtherInfoLen must be zero. With the key derivation
function CKD_SHA1_KDF_ASN1, pOtherInfo must be supplied, which
contains an octet string, specified in ASN.1 DER encoding, consisting of
mandatory and optional data shared by the two parties intending to share the
shared secret. With the key derivation function CKD_SHA1_KDF_CONCATENATE,
an optional pOtherInfo may be supplied, which consists of some data
shared by the two parties intending to share the shared secret. Otherwise, pOtherInfo
must be NULL and ulOtherInfoLen must be zero.
CK_X9_42_DH1_DERIVE_PARAMS_PTR is a pointer to a CK_X9_42_DH1_DERIVE_PARAMS.
·
CK_X9_42_DH2_DERIVE_PARAMS,
CK_X9_42_DH2_DERIVE_PARAMS_PTR
CK_X9_42_DH2_DERIVE_PARAMS is a structure that
provides the parameters to the CKM_X9_42_DH_HYBRID_DERIVE and CKM_X9_42_MQV_DERIVE
key derivation mechanisms, where each party contributes two key pairs. The
structure is defined as follows:
typedef struct CK_X9_42_DH2_DERIVE_PARAMS {
CK_X9_42_DH_KDF_TYPE kdf;
CK_ULONG ulOtherInfoLen;
CK_BYTE_PTR pOtherInfo;
CK_ULONG ulPublicDataLen;
CK_BYTE_PTR pPublicData;
CK_ULONG ulPrivateDataLen;
CK_OBJECT_HANDLE hPrivateData;
CK_ULONG ulPublicDataLen2;
CK_BYTE_PTR pPublicData2;
} CK_X9_42_DH2_DERIVE_PARAMS;
The fields of the structure have the following meanings:
kdf key
derivation function used on the shared secret value
ulOtherInfoLen the length
in bytes of the other info
pOtherInfo some data
shared between the two parties
ulPublicDataLen the length
in bytes of the other party’s first X9.42 Diffie-Hellman public key
pPublicData pointer to
other party’s first X9.42 Diffie-Hellman public key value
ulPrivateDataLen the length
in bytes of the second X9.42 Diffie-Hellman private key
hPrivateData key handle
for second X9.42 Diffie-Hellman private key value
ulPublicDataLen2 the length in
bytes of the other party’s second X9.42 Diffie-Hellman public key
pPublicData2 pointer to
other party’s second X9.42 Diffie-Hellman public key value
With the key derivation function CKD_NULL, pOtherInfo
must be NULL and ulOtherInfoLen must be zero. With the key derivation
function CKD_SHA1_KDF_ASN1, pOtherInfo must be supplied, which
contains an octet string, specified in ASN.1 DER encoding, consisting of
mandatory and optional data shared by the two parties intending to share the
shared secret. With the key derivation function CKD_SHA1_KDF_CONCATENATE,
an optional pOtherInfo may be supplied, which consists of some data
shared by the two parties intending to share the shared secret. Otherwise, pOtherInfo
must be NULL and ulOtherInfoLen must be zero.
CK_X9_42_DH2_DERIVE_PARAMS_PTR is a pointer to a CK_X9_42_DH2_DERIVE_PARAMS.
·
CK_X9_42_MQV_DERIVE_PARAMS,
CK_X9_42_MQV_DERIVE_PARAMS_PTR
CK_X9_42_MQV_DERIVE_PARAMS is a structure that
provides the parameters to the CKM_X9_42_MQV_DERIVE key derivation
mechanism, where each party contributes two key pairs. The structure is
defined as follows:
typedef struct CK_X9_42_MQV_DERIVE_PARAMS {
CK_X9_42_DH_KDF_TYPE kdf;
CK_ULONG ulOtherInfoLen;
CK_BYTE_PTR pOtherInfo;
CK_ULONG ulPublicDataLen;
CK_BYTE_PTR pPublicData;
CK_ULONG ulPrivateDataLen;
CK_OBJECT_HANDLE hPrivateData;
CK_ULONG ulPublicDataLen2;
CK_BYTE_PTR pPublicData2;
CK_OBJECT_HANDLE publicKey;
} CK_X9_42_MQV_DERIVE_PARAMS;
The fields of the structure
have the following meanings:
kdf key
derivation function used on the shared secret value
ulOtherInfoLen the length
in bytes of the other info
pOtherInfo some data
shared between the two parties
ulPublicDataLen the length
in bytes of the other party’s first X9.42 Diffie-Hellman public key
pPublicData pointer to
other party’s first X9.42 Diffie-Hellman public key value
ulPrivateDataLen the length
in bytes of the second X9.42 Diffie-Hellman private key
hPrivateData key handle
for second X9.42 Diffie-Hellman private key value
ulPublicDataLen2 the length in
bytes of the other party’s second X9.42 Diffie-Hellman public key
pPublicData2 pointer to
other party’s second X9.42 Diffie-Hellman public key value
publicKey Handle to
the first party’s ephemeral public key
With the key derivation function CKD_NULL, pOtherInfo
must be NULL and ulOtherInfoLen must be zero. With the key derivation
function CKD_SHA1_KDF_ASN1, pOtherInfo must be supplied, which
contains an octet string, specified in ASN.1 DER encoding, consisting of
mandatory and optional data shared by the two parties intending to share the
shared secret. With the key derivation function CKD_SHA1_KDF_CONCATENATE,
an optional pOtherInfo may be supplied, which consists of some data
shared by the two parties intending to share the shared secret. Otherwise, pOtherInfo
must be NULL and ulOtherInfoLen must be zero.
CK_X9_42_MQV_DERIVE_PARAMS_PTR is a pointer to a CK_X9_42_MQV_DERIVE_PARAMS.
The X9.42 Diffie-Hellman key pair generation mechanism,
denoted CKM_X9_42_DH_KEY_PAIR_GEN, is a key pair generation mechanism
based on Diffie-Hellman key agreement, as defined in the ANSI X9.42 standard.
It does not have a parameter.
The mechanism generates X9.42 Diffie-Hellman public/private
key pairs with a particular prime, base and subprime, as specified in the CKA_PRIME,
CKA_BASE and CKA_SUBPRIME attributes of the template for the
public key.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new public key and the CKA_CLASS,
CKA_KEY_TYPE, CKA_PRIME, CKA_BASE, CKA_SUBPRIME,
and CKA_VALUE attributes to the new private key; other attributes
required by the X9.42 Diffie-Hellman public and private key types must be
specified in the templates.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
X9.42 Diffie-Hellman prime sizes, in bits, for the CKA_PRIME attribute.
The X9.42 Diffie-Hellman domain parameter generation
mechanism, denoted CKM_X9_42_DH_PARAMETER_GEN, is a domain parameters
generation mechanism based on X9.42 Diffie-Hellman key agreement, as defined in
the ANSI X9.42 standard.
It does not have a parameter.
The mechanism generates X9.42 Diffie-Hellman domain
parameters with particular prime and subprime length in bits, as specified in
the CKA_PRIME_BITS and CKA_SUBPRIME_BITS attributes of the
template for the domain parameters.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
CKA_PRIME, CKA_BASE, CKA_SUBPRIME, CKA_PRIME_BITS and
CKA_SUBPRIME_BITS attributes to the new object. Other attributes supported
by the X9.42 Diffie-Hellman domain parameter types may also be specified in the
template for the domain parameters, or else are assigned default initial
values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
X9.42 Diffie-Hellman prime sizes, in bits.
The X9.42 Diffie-Hellman key derivation mechanism, denoted CKM_X9_42_DH_DERIVE,
is a mechanism for key derivation based on the Diffie-Hellman key agreement
scheme, as defined in the ANSI X9.42 standard, where each party contributes one
key pair, all using the same X9.42 Diffie-Hellman domain parameters.
It has a parameter, a CK_X9_42_DH1_DERIVE_PARAMS
structure.
This mechanism derives a secret value, and truncates the
result according to the CKA_KEY_TYPE attribute of the template and, if
it has one and the key type supports it, the CKA_VALUE_LEN attribute of
the template. (The truncation removes bytes from the leading end of the secret
value.) The mechanism contributes the result as the CKA_VALUE attribute
of the new key; other attributes required by the key type must be specified in
the template. Note that in order to validate this mechanism it may be required
to use the CKA_VALUE attribute as the key of a general-length MAC
mechanism (e.g. CKM_SHA_1_HMAC_GENERAL) over some test data.
This mechanism has the following rules about key sensitivity
and extractability:
·
The CKA_SENSITIVE and CKA_EXTRACTABLE attributes in
the template for the new key can both be specified to be either CK_TRUE or
CK_FALSE. If omitted, these attributes each take on some default value.
·
If the base key has its CKA_ALWAYS_SENSITIVE attribute set
to CK_FALSE, then the derived key will as well. If the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE, then the derived key has its CKA_ALWAYS_SENSITIVE
attribute set to the same value as its CKA_SENSITIVE attribute.
·
Similarly, if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_FALSE, then the derived key will, too. If the base key has
its CKA_NEVER_EXTRACTABLE attribute set to CK_TRUE, then the derived key
has its CKA_NEVER_EXTRACTABLE attribute set to the opposite value
from its CKA_EXTRACTABLE attribute.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
X9.42 Diffie-Hellman prime sizes, in bits, for the CKA_PRIME attribute.
The X9.42 Diffie-Hellman hybrid key derivation mechanism,
denoted CKM_X9_42_DH_HYBRID_DERIVE, is a mechanism for key derivation
based on the Diffie-Hellman hybrid key agreement scheme, as defined in the ANSI
X9.42 standard, where each party contributes two key pair, all using the same
X9.42 Diffie-Hellman domain parameters.
It has a parameter, a CK_X9_42_DH2_DERIVE_PARAMS
structure.
This mechanism derives a secret value, and truncates the
result according to the CKA_KEY_TYPE attribute of the template and, if
it has one and the key type supports it, the CKA_VALUE_LEN attribute of
the template. (The truncation removes bytes from the leading end of the secret
value.) The mechanism contributes the result as the CKA_VALUE attribute
of the new key; other attributes required by the key type must be specified in
the template. Note that in order to validate this mechanism it may be required
to use the CKA_VALUE attribute as the key of a general-length MAC
mechanism (e.g. CKM_SHA_1_HMAC_GENERAL) over some test data.
This mechanism has the following rules about key sensitivity
and extractability:
·
The CKA_SENSITIVE and CKA_EXTRACTABLE attributes in
the template for the new key can both be specified to be either CK_TRUE or
CK_FALSE. If omitted, these attributes each take on some default value.
·
If the base key has its CKA_ALWAYS_SENSITIVE attribute set
to CK_FALSE, then the derived key will as well. If the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE, then the derived key has its CKA_ALWAYS_SENSITIVE
attribute set to the same value as its CKA_SENSITIVE attribute.
·
Similarly, if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_FALSE, then the derived key will, too. If the base key has
its CKA_NEVER_EXTRACTABLE attribute set to CK_TRUE, then the derived key
has its CKA_NEVER_EXTRACTABLE attribute set to the opposite value
from its CKA_EXTRACTABLE attribute.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
X9.42 Diffie-Hellman prime sizes, in bits, for the CKA_PRIME attribute.
The X9.42 Diffie-Hellman Menezes-Qu-Vanstone (MQV) key
derivation mechanism, denoted CKM_X9_42_MQV_DERIVE, is a mechanism for
key derivation based the MQV scheme, as defined in the ANSI X9.42 standard,
where each party contributes two key pairs, all using the same X9.42
Diffie-Hellman domain parameters.
It has a parameter, a CK_X9_42_MQV_DERIVE_PARAMS structure.
This mechanism derives a secret value, and truncates the
result according to the CKA_KEY_TYPE attribute of the template and, if
it has one and the key type supports it, the CKA_VALUE_LEN attribute of
the template. (The truncation removes bytes from the leading end of the secret
value.) The mechanism contributes the result as the CKA_VALUE attribute
of the new key; other attributes required by the key type must be specified in
the template. Note that in order to validate this mechanism it may be required
to use the CKA_VALUE attribute as the key of a general-length MAC
mechanism (e.g. CKM_SHA_1_HMAC_GENERAL) over some test data.
This mechanism has the following rules about key sensitivity
and extractability:
·
The CKA_SENSITIVE and CKA_EXTRACTABLE attributes in
the template for the new key can both be specified to be either CK_TRUE or
CK_FALSE. If omitted, these attributes each take on some default value.
·
If the base key has its CKA_ALWAYS_SENSITIVE attribute set
to CK_FALSE, then the derived key will as well. If the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE, then the derived key has its CKA_ALWAYS_SENSITIVE
attribute set to the same value as its CKA_SENSITIVE attribute.
·
Similarly, if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_FALSE, then the derived key will, too. If the base key has
its CKA_NEVER_EXTRACTABLE attribute set to CK_TRUE, then the derived key
has its CKA_NEVER_EXTRACTABLE attribute set to the opposite value
from its CKA_EXTRACTABLE attribute.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
X9.42 Diffie-Hellman prime sizes, in bits, for the CKA_PRIME attribute.
The Extended Triple
Diffie-Hellman mechanism described here is the one described in [SIGNAL].
Table
91,
Extended Triple Diffie-Hellman Mechanisms vs. Functions
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_X3DH_INITIALIZE
|
|
|
|
|
|
|
ü
|
|
CKM_X3DH_RESPOND
|
|
|
|
|
|
|
ü
|
Mechanisms:
CKM_X3DH_INITIALIZE
CKM_X3DH_RESPOND
Extended Triple Diffie-Hellman uses Elliptic Curve keys in
Montgomery representation (CKK_EC_MONTGOMERY). Three different kinds of
keys are used, they differ in their lifespan:
- identity keys are
long-term keys, which identify the peer,
- prekeys are
short-term keys, which should be rotated often (weekly to hourly)
- onetime prekeys
are keys, which should be used only once.
Any peer intending to be contacted using X3DH must publish
their so-called prekey-bundle, consisting of their:
- public Identity
key,
- current prekey,
signed using XEDDSA with their identity key
- optionally a
batch of One-time public keys.
Initiating an Extended Triple Diffie-Hellman key exchange
starts by retrieving the following required public keys (the so-called
prekey-bundle) of the other peer: the Identity key, the signed public Prekey,
and optionally one One-time public key.
When the necessary key material is available, the initiating
party calls CKM_X3DH_INITIALIZE, also providing the following additional
parameters:
·
the initiators identity key
·
the initiators ephemeral key (a fresh, one-time CKK_EC_MONTGOMERY
type key)
CK_X3DH_INITIATE_PARAMS is a structure that provides
the parameters to the CKM_X3DH_INITIALIZE key exchange mechanism. The
structure is defined as follows:
typedef struct CK_X3DH_INITIATE_PARAMS {
CK_X3DH_KDF_TYPE kdf;
CK_OBJECT_HANDLE pPeer_identity;
CK_OBJECT_HANDLE pPeer_prekey;
CK_BYTE_PTR pPrekey_signature;
CK_BYTE_PTR pOnetime_key;
CK_OBJECT_HANDLE pOwn_identity;
CK_OBJECT_HANDLE pOwn_ephemeral;
} CK_X3DH_INITIATE_PARAMS;
Table
92,
Extended Triple Diffie-Hellman Initiate Message parameters:
|
Parameter
|
Data type
|
Meaning
|
|
kdf
|
CK_X3DH_KDF_TYPE
|
Key derivation function
|
|
pPeer_identity
|
Key handle
|
Peers public Identity key (from the prekey-bundle)
|
|
pPeer_prekey
|
Key Handle
|
Peers public prekey (from the prekey-bundle)
|
|
pPrekey_signature
|
Byte array
|
XEDDSA signature of PEER_PREKEY (from
prekey-bundle)
|
|
pOnetime_key
|
Byte array
|
Optional one-time public prekey of peer (from
the prekey-bundle)
|
|
pOwn_identity
|
Key Handle
|
Initiators Identity key
|
|
pOwn_ephemeral
|
Key Handle
|
Initiators ephemeral key
|
Responding an Extended Triple Diffie-Hellman key exchange is
done by executing a CKM_X3DH_RESPOND mechanism. CK_X3DH_RESPOND_PARAMS
is a structure that provides the parameters to the CKM_X3DH_RESPOND key
exchange mechanism. All these parameter should be supplied by the Initiator in
a message to the responder. The structure is defined as follows:
typedef struct CK_X3DH_RESPOND_PARAMS {
CK_X3DH_KDF_TYPE kdf;
CK_BYTE_PTR pIdentity_id;
CK_BYTE_PTR pPrekey_id;
CK_BYTE_PTR pOnetime_id;
CK_OBJECT_HANDLE pInitiator_identity;
CK_BYTE_PTR pInitiator_ephemeral;
} CK_X3DH_RESPOND_PARAMS;
Table 93,
Extended Triple Diffie-Hellman 1st Message parameters:
|
Parameter
|
Data type
|
Meaning
|
|
kdf
|
CK_X3DH_KDF_TYPE
|
Key derivation function
|
|
pIdentity_id
|
Byte array
|
Peers public Identity key identifier (from
the prekey-bundle)
|
|
pPrekey_id
|
Byte array
|
Peers public prekey identifier (from the
prekey-bundle)
|
|
pOnetime_id
|
Byte array
|
Optional one-time public prekey of peer (from
the prekey-bundle)
|
|
pInitiator_identity
|
Key handle
|
Initiators Identity key
|
|
pInitiator_ephemeral
|
Byte array
|
Initiators ephemeral key
|
Where the *_id fields are identifiers marking which key has
been used from the prekey-bundle, these identifiers could be the keys themselves.
This mechanism has the following rules about key sensitivity
and extractability:
1
The CKA_SENSITIVE and CKA_EXTRACTABLE attributes in the
template for the new key can both be specified to be either CK_TRUE or
CK_FALSE. If omitted, these attributes each take on some default value.
2
If the base key has its CKA_ALWAYS_SENSITIVE attribute set to
CK_FALSE, then the derived key will as well. If the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE, then the derived key has its CKA_ALWAYS_SENSITIVE
attribute set to the same value as its CKA_SENSITIVE attribute.
3
Similarly, if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_FALSE, then the derived key will, too. If the base key has
its CKA_NEVER_EXTRACTABLE attribute set to CK_TRUE, then the derived key
has its CKA_NEVER_EXTRACTABLE attribute set to the opposite value
from its CKA_EXTRACTABLE attribute.
·
CK_X3DH_KDF_TYPE,
CK_X3DH_KDF_TYPE_PTR
CK_X3DH_KDF_TYPE is used to indicate the Key
Derivation Function (KDF) applied to derive keying data from a shared secret.
The key derivation function will be used by the X3DH key agreement schemes. It
is defined as follows:
typedef CK_ULONG CK_X3DH_KDF_TYPE;
The following table lists the defined functions.
Table
94, X3DH:
Key Derivation Functions
|
Source
Identifier
|
|
CKD_NULL
|
|
CKD_BLAKE2B_256_KDF
|
|
CKD_BLAKE2B_512_KDF
|
|
CKD_SHA3_256_KDF
|
|
CKD_SHA256_KDF
|
|
CKD_SHA3_512_KDF
|
|
CKD_SHA512_KDF
|
The Double Ratchet is a key management algorithm managing
the ongoing renewal and maintenance of short-lived session keys providing
forward secrecy and break-in recovery for encrypt/decrypt operations. The
algorithm is described in [DoubleRatchet]. The Signal protocol uses X3DH
to exchange a shared secret in the first step, which is then used to derive a
Double Ratchet secret key.
Table 95, Double
Ratchet Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_X2RATCHET_INITIALIZE
|
|
|
|
|
|
|
✓
|
|
CKM_X2RATCHET_RESPOND
|
|
|
|
|
|
|
✓
|
|
CKM_X2RATCHET_ENCRYPT
|
✓
|
|
|
|
|
✓
|
|
|
CKM_X2RATCHET_DECRYPT
|
✓
|
|
|
|
|
✓
|
|
This section defines the key type “CKK_X2RATCHET” for type
CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of key objects.
Mechanisms:
CKM_X2RATCHET_INITIALIZE
CKM_X2RATCHET_RESPOND
CKM_X2RATCHET_ENCRYPT
CKM_X2RATCHET_DECRYPT
Double Ratchet secret key objects
(object class CKO_SECRET_KEY, key type CKK_X2RATCHET) hold Double Ratchet keys.
Double Ratchet secret keys can only be derived from shared secret keys using
the mechanism CKM_X2RATCHET_INITIALIZE or
CKM_X2RATCHET_RESPOND. In the Signal protocol these are seeded with the shared
secret derived from an Extended Triple Diffie-Hellman [X3DH] key-exchange. The
following table defines the Double Ratchet secret key object attributes, in
addition to the common attributes defined for this object class:
Table
96, Double
Ratchet Secret Key Object Attributes
|
Attribute
|
Data
type
|
Meaning
|
|
CKA_X2RATCHET_RK
|
Byte
array
|
Root
key
|
|
CKA_X2RATCHET_HKS
|
Byte
array
|
Sender
Header key
|
|
CKA_X2RATCHET_HKR
|
Byte
array
|
Receiver
Header key
|
|
CKA_X2RATCHET_NHKS
|
Byte
array
|
Next
Sender Header Key
|
|
CKA_X2RATCHET_NHKR
|
Byte
array
|
Next
Receiver Header Key
|
|
CKA_X2RATCHET_CKS
|
Byte
array
|
Sender
Chain key
|
|
CKA_X2RATCHET_CKR
|
Byte
array
|
Receiver
Chain key
|
|
CKA_X2RATCHET_DHS
|
Byte
array
|
Sender
DH secret key
|
|
CKA_X2RATCHET_DHP
|
Byte
array
|
Sender
DH public key
|
|
CKA_X2RATCHET_DHR
|
Byte
array
|
Receiver
DH public key
|
|
CKA_X2RATCHET_NS
|
ULONG
|
Message
number send
|
|
CKA_X2RATCHET_NR
|
ULONG
|
Message
number receive
|
|
CKA_X2RATCHET_PNS
|
ULONG
|
Previous
message number send
|
|
CKA_X2RATCHET_BOBS1STMSG
|
BOOL
|
Is
this bob and has he ever sent a message?
|
|
CKA_X2RATCHET_ISALICE
|
BOOL
|
Is
this Alice?
|
|
CKA_X2RATCHET_BAGSIZE
|
ULONG
|
How
many out-of-order keys do we store
|
|
CKA_X2RATCHET_BAG
|
Byte
array
|
Out-of-order
keys
|
The Double Ratchet key derivation mechanisms depend on who
is the initiating party, and who the receiving, denoted CKM_X2RATCHET_INITIALIZE
and CKM_X2RATCHET_RESPOND, are the key derivation mechanisms for the
Double Ratchet. Usually the keys are derived from a shared secret by executing
a X3DH key exchange.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key. Additionally the attribute
flags indicating which functions the key supports are also contributed by the
mechanism.
For this mechanism, the only allowed values are 255 and 448
as RFC 8032 only defines curves of these two sizes. A Cryptoki implementation
may support one or both of these curves and should set the ulMinKeySize
and ulMaxKeySize fields accordingly.
·
CK_X2RATCHET_INITIALIZE_PARAMS;
CK_X2RATCHET_INITIALIZE_PARAMS_PTR
CK_X2RATCHET_INITIALIZE_PARAMS provides the
parameters to the CKM_X2RATCHET_INITIALIZE mechanism. It is defined as
follows:
typedef struct CK_X2RATCHET_INITIALIZE_PARAMS {
CK_BYTE_PTR sk;
CK_OBJECT_HANDLE peer_public_prekey;
CK_OBJECT_HANDLE peer_public_identity;
CK_OBJECT_HANDLE own_public_identity;
CK_BBOOL bEncryptedHeader;
CK_ULONG eCurve;
CK_MECHANISM_TYPE aeadMechanism;
CK_X2RATCHET_KDF_TYPE kdfMechanism;
} CK_X2RATCHET_INITIALIZE_PARAMS;
The fields of the structure have the following meanings:
sk the
shared secret with peer (derived using X3DH)
peers_public_prekey Peers public prekey
which the Initiator used in the X3DH
peers_public_identity Peers public
identity which the Initiator used in the X3DH
own_public_identity Initiators public
identity as used in the X3DH
bEncryptedHeader whether the headers
are encrypted
eCurve 255 for
curve 25519 or 448 for curve 448
aeadMechanism a mechanism
supporting AEAD encryption
kdfMechanism a Key Derivation
Mechanism, such as CKD_BLAKE2B_512_KDF
·
CK_X2RATCHET_RESPOND_PARAMS;
CK_X2RATCHET_RESPOND_PARAMS_PTR
CK_X2RATCHET_RESPOND_PARAMS provides the parameters
to the CKM_X2RATCHET_RESPOND mechanism. It is defined as follows:
typedef struct CK_X2RATCHET_RESPOND_PARAMS {
CK_BYTE_PTR sk;
CK_OBJECT_HANDLE own_prekey;
CK_OBJECT_HANDLE initiator_identity;
CK_OBJECT_HANDLE own_public_identity;
CK_BBOOL bEncryptedHeader;
CK_ULONG eCurve;
CK_MECHANISM_TYPE aeadMechanism;
CK_X2RATCHET_KDF_TYPE kdfMechanism;
} CK_X2RATCHET_RESPOND_PARAMS;
The fields of the structure have the following meanings:
sk shared
secret with the Initiator
own_prekey Own Prekey pair
that the Initiator used
initiator_identity Initiators
public identity key used
own_public_identity as used in the
prekey bundle by the initiator in the X3DH
bEncryptedHeader whether the headers
are encrypted
eCurve 255 for
curve 25519 or 448 for curve 448
aeadMechanism a mechanism
supporting AEAD encryption
kdfMechanism a Key Derivation
Mechanism, such as CKD_BLAKE2B_512_KDF
The Double Ratchet encryption mechanism, denoted CKM_X2RATCHET_ENCRYPT
and CKM_X2RATCHET_DECRYPT, are a mechanisms for single part encryption
and decryption based on the Double Ratchet and its underlying AEAD cipher.
·
CK_X2RATCHET_KDF_TYPE,
CK_X2RATCHET_KDF_TYPE_PTR
CK_X2RATCHET_KDF_TYPE is used to indicate the Key
Derivation Function (KDF) applied to derive keying data from a shared secret.
The key derivation function will be used by the X key derivation scheme. It is
defined as follows:
typedef CK_ULONG CK_X2RATCHET_KDF_TYPE;
The following table lists the defined functions.
Table
97,
X2RATCHET: Key Derivation Functions
|
Source
Identifier
|
|
CKD_NULL
|
|
CKD_BLAKE2B_256_KDF
|
|
CKD_BLAKE2B_512_KDF
|
|
CKD_SHA3_256_KDF
|
|
CKD_SHA256_KDF
|
|
CKD_SHA3_512_KDF
|
|
CKD_SHA512_KDF
|
Cryptoki Versions 2.01 and up allow the use of secret keys
for wrapping and unwrapping RSA private keys, Diffie-Hellman private keys,
X9.42 Diffie-Hellman private keys, short Weierstrass EC private keys and DSA
private keys. For wrapping, a private key is BER-encoded according to [PKCS #8]
PrivateKeyInfo ASN.1 type. PKCS #8 requires an algorithm identifier for the
type of the private key. The object identifiers for the required algorithm
identifiers are as follows:
rsaEncryption OBJECT IDENTIFIER ::= { pkcs-1 1 }
dhKeyAgreement OBJECT IDENTIFIER ::= { pkcs-3 1 }
dhpublicnumber OBJECT IDENTIFIER ::= { iso(1) member-body(2)
us(840) ansi-x942(10046) number-type(2) 1 }
id-ecPublicKey OBJECT IDENTIFIER ::= { iso(1) member-body(2) us(840) ansi-x9-62(10045) publicKeyType(2) 1 }
id-dsa OBJECT IDENTIFIER ::= {
iso(1) member-body(2) us(840) x9-57(10040) x9cm(4) 1 }
where
pkcs-1 OBJECT IDENTIFIER ::= {
iso(1) member-body(2) US(840) rsadsi(113549) pkcs(1) 1 }
pkcs-3 OBJECT IDENTIFIER ::= {
iso(1) member-body(2) US(840) rsadsi(113549) pkcs(1) 3 }
These parameters for the algorithm identifiers have the
following types, respectively:
NULL
DHParameter ::= SEQUENCE {
prime INTEGER, -- p
base INTEGER, -- g
privateValueLength INTEGER OPTIONAL
}
DomainParameters ::= SEQUENCE {
prime INTEGER, -- p
base INTEGER, -- g
subprime INTEGER, -- q
cofactor INTEGER OPTIONAL, -- j
validationParms ValidationParms OPTIONAL
}
ValidationParms ::= SEQUENCE {
Seed BIT STRING, -- seed
PGenCounter INTEGER -- parameter verification
}
Parameters ::= CHOICE {
ecParameters ECParameters,
namedCurve CURVES.&id({CurveNames}),
implicitlyCA NULL
}
Dss-Parms ::= SEQUENCE {
p INTEGER,
q INTEGER,
g INTEGER
}
For the X9.42 Diffie-Hellman domain parameters, the cofactor
and the validationParms optional fields should not be used when wrapping
or unwrapping X9.42 Diffie-Hellman private keys since their values are not
stored within the token.
For the EC domain parameters, the use of namedCurve
is recommended over the choice ecParameters. The choice implicitlyCA
must not be used in Cryptoki.
Within the PrivateKeyInfo type:
·
RSA private keys are BER-encoded according to PKCS #1’s
RSAPrivateKey ASN.1 type. This type requires values to be present for all
the attributes specific to Cryptoki’s RSA private key objects. In other words,
if a Cryptoki library does not have values for an RSA private key’s CKA_MODULUS,
CKA_PUBLIC_EXPONENT, CKA_PRIVATE_EXPONENT, CKA_PRIME_1, CKA_PRIME_2,
CKA_EXPONENT_1, CKA_EXPONENT_2, and CKA_COEFFICIENT
values, it must not create an RSAPrivateKey BER-encoding of the key, and so it
must not prepare it for wrapping.
·
Diffie-Hellman private keys are represented as BER-encoded ASN.1 type
INTEGER.
·
X9.42 Diffie-Hellman private keys are represented as BER-encoded
ASN.1 type INTEGER.
·
Short Weierstrass EC private keys are BER-encoded according to
SECG SEC 1 ECPrivateKey ASN.1 type:
ECPrivateKey ::= SEQUENCE {
Version INTEGER { ecPrivkeyVer1(1) } (ecPrivkeyVer1),
privateKey OCTET STRING,
parameters [0] Parameters OPTIONAL,
publicKey [1] BIT STRING OPTIONAL
}
Since the EC domain parameters are
placed in the PKCS #8’s privateKeyAlgorithm field, the optional parameters
field in an ECPrivateKey must be omitted. A Cryptoki application must be able to unwrap an ECPrivateKey that contains the optional publicKey
field; however, what is done with this publicKey
field is outside the scope of Cryptoki.
·
DSA private keys are represented as BER-encoded ASN.1 type
INTEGER.
Once a private key has been BER-encoded as a PrivateKeyInfo
type, the resulting string of bytes is encrypted with the secret key. This
encryption is defined in the section for the respective key wrapping mechanism.
Unwrapping a wrapped private key undoes the above
procedure. The ciphertext is decrypted as defined for the respective key
unwrapping mechanism. The data thereby obtained are parsed as a PrivateKeyInfo
type. An error will result if the original wrapped key does not decrypt
properly, or if the decrypted data does not parse properly, or its type does
not match the key type specified in the template for the new key. The
unwrapping mechanism contributes only those attributes specified in the
PrivateKeyInfo type to the newly-unwrapped key; other attributes must be
specified in the template, or will take their default values.
Earlier drafts of PKCS #11 Version 2.0 and Version 2.01 used
the object identifier
DSA OBJECT IDENTIFIER ::= { algorithm 12 }
algorithm OBJECT IDENTIFIER ::= {
iso(1) identifier-organization(3) oiw(14) secsig(3)
algorithm(2) }
with associated parameters
DSAParameters ::= SEQUENCE {
prime1 INTEGER, -- modulus p
prime2 INTEGER, -- modulus q
base INTEGER -- base g
}
for wrapping DSA private keys. Note that although the two
structures for holding DSA domain parameters appear identical when instances of
them are encoded, the two corresponding object identifiers are different.
Table
98, Generic Secret Key Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_GENERIC_SECRET_KEY_GEN
|
|
|
|
|
ü
|
|
|
This section defines the key type “CKK_GENERIC_SECRET” for
type CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of key objects.
Mechanisms:
CKM_GENERIC_SECRET_KEY_GEN
Generic secret key objects (object class CKO_SECRET_KEY, key
type CKK_GENERIC_SECRET) hold generic secret keys. These keys do not
support encryption or decryption; however, other keys can be derived from them
and they can be used in HMAC operations. The following table defines the
generic secret key object attributes, in addition to the common attributes
defined for this object class:
These key types are used in several of the mechanisms
described in this section.
Table
99, Generic Secret Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_VALUE1,4,6,7
|
Byte array
|
Key value (arbitrary length)
|
|
CKA_VALUE_LEN2,3
|
CK_ULONG
|
Length in bytes of key value
|
- Refer to Table 11 for footnotes
The following is a sample template for creating a generic
secret key object:
CK_OBJECT_CLASS class = CKO_SECRET_KEY;
CK_KEY_TYPE keyType = CKK_GENERIC_SECRET;
CK_UTF8CHAR label[] = “A generic secret key object”;
CK_BYTE value[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_DERIVE, &true, sizeof(true)},
{CKA_VALUE, value, sizeof(value)}
};
CKA_CHECK_VALUE: The value of this attribute is derived from
the key object by taking the first three bytes of the SHA-1 hash of the generic
secret key object’s CKA_VALUE attribute.
The generic secret key generation mechanism, denoted CKM_GENERIC_SECRET_KEY_GEN,
is used to generate generic secret keys. The generated keys take on any
attributes provided in the template passed to the C_GenerateKey call,
and the CKA_VALUE_LEN attribute specifies the length of the key to be
generated.
It does not have a parameter.
The template supplied must specify a value for the CKA_VALUE_LEN
attribute. If the template specifies an object type and a class, they must
have the following values:
CK_OBJECT_CLASS = CKO_SECRET_KEY;
CK_KEY_TYPE = CKK_GENERIC_SECRET;
For this
mechanism, the ulMinKeySize and ulMaxKeySize fields of the CK_MECHANISM_INFO
structure specify the supported range of key sizes, in bits.
Refer to RFC2104 and FIPS 198 for HMAC
algorithm description. The HMAC secret key shall correspond to the PKCS11
generic secret key type or the mechanism specific key types (see mechanism
definition). Such keys, for use with HMAC operations can be created using
C_CreateObject or C_GenerateKey.
The RFC also specifies test vectors for the various hash
function based HMAC mechanisms described in the respective hash mechanism
descriptions. The RFC should be consulted to obtain these test vectors.
·
CK_MAC_GENERAL_PARAMS;
CK_MAC_GENERAL_PARAMS_PTR
CK_MAC_GENERAL_PARAMS provides the parameters to the
general-length MACing mechanisms of the DES, DES3 (triple-DES), AES, Camellia, SEED, and ARIA ciphers. It also
provides the parameters to the general-length HMACing mechanisms (i.e.,SHA-1,
SHA-256, SHA-384, SHA-512, and SHA-512/T family) and the two SSL 3.0 MACing
mechanisms, (i.e., MD5 and SHA-1). It holds the length of the MAC that these
mechanisms produce. It is defined as follows:
typedef CK_ULONG CK_MAC_GENERAL_PARAMS;
CK_MAC_GENERAL_PARAMS_PTR
is a pointer to a CK_MAC_GENERAL_PARAMS.
For the Advanced Encryption Standard (AES) see [FIPS PUB
197].
Table
100, AES Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_AES_KEY_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_AES_ECB
|
ü
|
|
|
|
|
ü
|
|
|
CKM_AES_CBC
|
ü
|
|
|
|
|
ü
|
|
|
CKM_AES_CBC_PAD
|
ü
|
|
|
|
|
ü
|
|
|
CKM_AES_MAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_AES_MAC
|
|
ü
|
|
|
|
|
|
|
CKM_AES_OFB
|
ü
|
|
|
|
|
ü
|
|
|
CKM_AES_CFB64
|
ü
|
|
|
|
|
ü
|
|
|
CKM_AES_CFB8
|
ü
|
|
|
|
|
ü
|
|
|
CKM_AES_CFB128
|
ü
|
|
|
|
|
ü
|
|
|
CKM_AES_CFB1
|
ü
|
|
|
|
|
ü
|
|
|
CKM_AES_XCBC_MAC
|
|
ü
|
|
|
|
|
|
|
CKM_AES_XCBC_MAC_96
|
|
ü
|
|
|
|
|
|
This section defines the key type “CKK_AES” for type
CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of key objects.
Mechanisms:
CKM_AES_KEY_GEN
CKM_AES_ECB
CKM_AES_CBC
CKM_AES_MAC
CKM_AES_MAC_GENERAL
CKM_AES_CBC_PAD
CKM_AES_OFB
CKM_AES_CFB64
CKM_AES_CFB8
CKM_AES_CFB128
CKM_AES_CFB1
CKM_AES_XCBC_MAC
CKM_AES_XCBC_MAC_96
AES secret key objects (object class CKO_SECRET_KEY, key
type CKK_AES) hold AES keys. The following table defines the AES secret
key object attributes, in addition to the common attributes defined for this
object class:
Table 101, AES Secret Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_VALUE1,4,6,7
|
Byte array
|
Key value (16, 24, or 32 bytes)
|
|
CKA_VALUE_LEN2,3,6
|
CK_ULONG
|
Length in bytes of key value
|
- Refer to Table 11 for footnotes
The following is a sample template for creating an AES
secret key object:
CK_OBJECT_CLASS class = CKO_SECRET_KEY;
CK_KEY_TYPE keyType = CKK_AES;
CK_UTF8CHAR label[] = “An AES secret key object”;
CK_BYTE value[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_ENCRYPT, &true, sizeof(true)},
{CKA_VALUE, value, sizeof(value)}
};
CKA_CHECK_VALUE: The value of this attribute is derived from
the key object by taking the first three bytes of the ECB encryption of a
single block of null (0x00) bytes, using the default cipher associated with the
key type of the secret key object.
The AES key generation mechanism, denoted CKM_AES_KEY_GEN,
is a key generation mechanism for NIST’s Advanced Encryption Standard.
It does not have a parameter.
The mechanism generates AES keys with a particular length in
bytes, as specified in the CKA_VALUE_LEN attribute of the template for
the key.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key. Other attributes supported by
the AES key type (specifically, the flags indicating which functions the key
supports) may be specified in the template for the key, or else are assigned
default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
AES key sizes, in bytes.
AES-ECB, denoted CKM_AES_ECB, is a mechanism for
single- and multiple-part encryption and decryption; key wrapping; and key
unwrapping, based on NIST Advanced Encryption Standard and electronic codebook
mode.
It does not have a parameter.
This mechanism can wrap and unwrap any secret key. Of
course, a particular token may not be able to wrap/unwrap every secret key that
it supports. For wrapping, the mechanism encrypts the value of the CKA_VALUE
attribute of the key that is wrapped, padded on the trailing end with up to
block size minus one null bytes so that the resulting length is a multiple of
the block size. The output data is the same length as the padded input data. It
does not wrap the key type, key length, or any other information about the key;
the application must convey these separately.
For unwrapping, the mechanism decrypts the wrapped key, and
truncates the result according to the CKA_KEY_TYPE attribute of the
template and, if it has one, and the key type supports it, the CKA_VALUE_LEN
attribute of the template. The mechanism contributes the result as the CKA_VALUE
attribute of the new key; other attributes required by the key type must be
specified in the template.
Constraints on key types and the length of data are
summarized in the following table:
Table 102, AES-ECB: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
Comments
|
|
C_Encrypt
|
AES
|
multiple of block
size
|
same as input length
|
no final part
|
|
C_Decrypt
|
AES
|
multiple of block
size
|
same as input length
|
no final part
|
|
C_WrapKey
|
AES
|
any
|
input length rounded
up to multiple of block size
|
|
|
C_UnwrapKey
|
AES
|
multiple of block
size
|
determined by type of
key being unwrapped or CKA_VALUE_LEN
|
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
AES key sizes, in bytes.
AES-CBC, denoted CKM_AES_CBC, is a mechanism for
single- and multiple-part encryption and decryption; key wrapping; and key
unwrapping, based on NIST’s Advanced Encryption Standard and cipher-block
chaining mode.
It has a parameter, a 16-byte initialization vector.
This mechanism can wrap and unwrap any secret key. Of
course, a particular token may not be able to wrap/unwrap every secret key that
it supports. For wrapping, the mechanism encrypts the value of the CKA_VALUE
attribute of the key that is wrapped, padded on the trailing end with up to
block size minus one null bytes so that the resulting length is a multiple of
the block size. The output data is the same length as the padded input data. It
does not wrap the key type, key length, or any other information about the key;
the application must convey these separately.
For unwrapping, the mechanism decrypts the wrapped key, and
truncates the result according to the CKA_KEY_TYPE attribute of the
template and, if it has one, and the key type supports it, the CKA_VALUE_LEN
attribute of the template. The mechanism contributes the result as the CKA_VALUE
attribute of the new key; other attributes required by the key type must be
specified in the template.
Constraints on key types and the length of data are
summarized in the following table:
Table 103, AES-CBC: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
Comments
|
|
C_Encrypt
|
AES
|
multiple of block
size
|
same as input length
|
no final part
|
|
C_Decrypt
|
AES
|
multiple of block
size
|
same as input length
|
no final part
|
|
C_WrapKey
|
AES
|
any
|
input length rounded
up to multiple of the block size
|
|
|
C_UnwrapKey
|
AES
|
multiple of block
size
|
determined by type of
key being unwrapped or CKA_VALUE_LEN
|
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
AES key sizes, in bytes.
AES-CBC with PKCS padding, denoted CKM_AES_CBC_PAD,
is a mechanism for single- and multiple-part encryption and decryption; key
wrapping; and key unwrapping, based on NIST’s Advanced Encryption Standard;
cipher-block chaining mode; and the block cipher padding method detailed in [PKCS
#7].
It has a parameter, a 16-byte initialization vector.
The PKCS padding in this mechanism allows the length of the
plaintext value to be recovered from the ciphertext value. Therefore, when
unwrapping keys with this mechanism, no value should be specified for the CKA_VALUE_LEN
attribute.
In addition to being able to wrap and unwrap secret keys,
this mechanism can wrap and unwrap RSA, Diffie-Hellman, X9.42 Diffie-Hellman, short
Weierstrass EC and DSA private keys (see Section 6.7
for details). The entries in the table below for data length constraints when
wrapping and unwrapping keys do not apply to wrapping and unwrapping private
keys.
Constraints on key types and the length of data are
summarized in the following table:
Table 104, AES-CBC with
PKCS Padding: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Encrypt
|
AES
|
any
|
input length rounded
up to multiple of the block size
|
|
C_Decrypt
|
AES
|
multiple of block
size
|
between 1 and block
size bytes shorter than input length
|
|
C_WrapKey
|
AES
|
any
|
input length rounded
up to multiple of the block size
|
|
C_UnwrapKey
|
AES
|
multiple of block
size
|
between 1 and block
length bytes shorter than input length
|
For this mechanism, the ulMinKeySize
and ulMaxKeySize fields of the CK_MECHANISM_INFO structure
specify the supported range of AES key sizes, in bytes.
AES-OFB, denoted CKM_AES_OFB. It is a mechanism for single
and multiple-part encryption and decryption with AES. AES-OFB mode is described
in [NIST sp800-38a].
It has a parameter, an initialization vector for this mode.
The initialization vector has the same length as the block size.
Constraints on key types and the length of data are summarized in the following
table:
Table 105, AES-OFB: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
Comments
|
|
C_Encrypt
|
AES
|
any
|
same as input length
|
no final part
|
|
C_Decrypt
|
AES
|
any
|
same as input length
|
no final part
|
For this mechanism the CK_MECHANISM_INFO structure is as
specified for CBC mode.
Cipher AES has a cipher feedback mode, AES-CFB, denoted
CKM_AES_CFB8, CKM_AES_CFB64, and CKM_AES_CFB128. It is a mechanism for single and
multiple-part encryption and decryption with AES. AES-OFB mode is described
[NIST sp800-38a].
It has a parameter, an initialization vector for this mode.
The initialization vector has the same length as the block size.
Constraints on key types and the length of data are summarized in the following
table:
Table 106, AES-CFB: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
Comments
|
|
C_Encrypt
|
AES
|
any
|
same as input length
|
no final part
|
|
C_Decrypt
|
AES
|
any
|
same as input length
|
no final part
|
For this mechanism the CK_MECHANISM_INFO structure is as
specified for CBC mode.
General-length AES-MAC, denoted CKM_AES_MAC_GENERAL,
is a mechanism for single- and multiple-part signatures and verification, based
on NIST Advanced Encryption Standard as defined in FIPS PUB 197 and data
authentication as defined in FIPS PUB 113.
It has a parameter, a CK_MAC_GENERAL_PARAMS structure,
which specifies the output length desired from the mechanism.
The output bytes from this mechanism are taken from the
start of the final AES cipher block produced in the MACing process.
Constraints on key types and the length of data are
summarized in the following table:
Table 107, General-length AES-MAC: Key And Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
AES
|
any
|
1-block size, as
specified in parameters
|
|
C_Verify
|
AES
|
any
|
1-block size, as
specified in parameters
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
AES key sizes, in bytes.
AES-MAC, denoted by CKM_AES_MAC, is a special case of
the general-length AES-MAC mechanism. AES-MAC always produces and verifies MACs
that are half the block size in length.
It does not have a parameter.
Constraints on key types and the length of data are
summarized in the following table:
Table 108, AES-MAC: Key And Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
AES
|
Any
|
½ block size (8
bytes)
|
|
C_Verify
|
AES
|
Any
|
½ block size (8
bytes)
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
AES key sizes, in bytes.
AES-XCBC-MAC, denoted CKM_AES_XCBC_MAC, is a
mechanism for single and multiple part signatures and verification; based on
NIST’s Advanced Encryption Standard and [RFC 3566].
It does not have a parameter.
Constraints on key types and the length of data are
summarized in the following table:
Table 109, AES-XCBC-MAC: Key And
Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
AES
|
Any
|
16 bytes
|
|
C_Verify
|
AES
|
Any
|
16 bytes
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
AES key sizes, in bytes.
AES-XCBC-MAC-96, denoted CKM_AES_XCBC_MAC_96, is a
mechanism for single and multiple part signatures and verification; based on
NIST’s Advanced Encryption Standard and [RFC 3566].
It does not have a parameter.
Constraints on key types and the length of data are
summarized in the following table:
Table 110, AES-XCBC-MAC: Key And
Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
AES
|
Any
|
12 bytes
|
|
C_Verify
|
AES
|
Any
|
12 bytes
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
AES key sizes, in bytes.
Table
111, AES with Counter Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_AES_CTR
|
ü
|
|
|
|
|
ü
|
|
Mechanisms:
CKM_AES_CTR
¨
CK_AES_CTR_PARAMS; CK_AES_CTR_PARAMS_PTR
CK_AES_CTR_PARAMS is a structure that provides the
parameters to the CKM_AES_CTR mechanism. It is defined as follows:
typedef struct CK_AES_CTR_PARAMS {
CK_ULONG ulCounterBits;
CK_BYTE cb[16];
} CK_AES_CTR_PARAMS;
ulCounterBits specifies the number of bits in the counter
block (cb) that shall be incremented. This number shall be such that 0 < ulCounterBits
≤ 128. For any values outside this range the mechanism shall return CKR_MECHANISM_PARAM_INVALID.
It's up to the caller to initialize all of the bits in the
counter block including the counter bits. The counter bits are the least
significant bits of the counter block (cb). They are a big-endian value usually
starting with 1. The rest of ‘cb’ is for the nonce, and maybe an optional IV.
E.g. as defined in [RFC 3686]:
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Nonce |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Initialization Vector (IV) |
| |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Block Counter |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
This construction permits each packet to consist of up to 232-1
blocks = 4,294,967,295 blocks = 68,719,476,720 octets.
CK_AES_CTR_PARAMS_PTR is a pointer to a CK_AES_CTR_PARAMS.
Generic AES counter mode is described in NIST Special
Publication 800-38A and in RFC 3686. These describe encryption using a counter
block which may include a nonce to guarantee uniqueness of the counter block.
Since the nonce is not incremented, the mechanism parameter must specify the
number of counter bits in the counter block.
The block counter is incremented by 1 after each block of
plaintext is processed. There is no support for any other increment functions
in this mechanism.
If an attempt to encrypt/decrypt is made which will cause an
overflow of the counter block’s counter bits, then the mechanism shall return CKR_DATA_LEN_RANGE.
Note that the mechanism should allow the final post increment of the counter to
overflow (if it implements it this way) but not allow any further processing
after this point. E.g. if ulCounterBits = 2 and the counter bits start as 1
then only 3 blocks of data can be processed.
Ref [NIST AES CTS]
This mode allows unpadded data that has length that is not a
multiple of the block size to be encrypted to the same length of cipher text.
Table
112, AES CBC with Cipher Text Stealing CTS Mechanisms vs.
Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_AES_CTS
|
ü
|
|
|
|
|
ü
|
|
Mechanisms:
CKM_AES_CTS
It has a parameter, a 16-byte initialization vector.
Table 113, AES-CTS: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
Comments
|
|
C_Encrypt
|
AES
|
Any, ≥ block
size (16 bytes)
|
same as input length
|
no final part
|
|
C_Decrypt
|
AES
|
any, ≥ block
size (16 bytes)
|
same as input length
|
no final part
|
Table
114, Additional AES Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_AES_GCM
|
ü
|
|
|
|
|
ü
|
|
|
CKM_AES_CCM
|
ü
|
|
|
|
|
ü
|
|
|
CKM_AES_GMAC
|
|
ü
|
|
|
|
|
|
Mechanisms:
CKM_AES_GCM
CKM_AES_CCM
CKM_AES_GMAC
Generator Functions:
CKG_NO_GENERATE
CKG_GENERATE
CKG_GENERATE_COUNTER
CKG_GENERATE_RANDOM
CKG_GENERATE_COUNTER_XOR
Generic GCM mode is described in [GCM]. To set up for
AES-GCM use the following process, where K (key) and AAD
(additional authenticated data) are as described in [GCM]. AES-GCM uses
CK_GCM_PARAMS for Encrypt, Decrypt and CK_GCM_MESSAGE_PARAMS for MessageEncrypt
and MessageDecrypt.
Encrypt:
- Set
the IV length ulIvLen in the parameter block.
- Set
the IV data pIv in the parameter block.
- Set
the AAD data pAAD and size ulAADLen in the parameter block. pAAD
may be NULL if ulAADLen is 0.
- Set
the tag length ulTagBits in the parameter block.
- Call
C_EncryptInit() for CKM_AES_GCM mechanism with parameters and key K.
- Call
C_Encrypt(), or C_EncryptUpdate()*
C_EncryptFinal(), for the plaintext obtaining ciphertext and
authentication tag output.
Decrypt:
- Set
the IV length ulIvLen in the parameter block.
- Set
the IV data pIv in the parameter block.
- Set
the AAD data pAAD and size ulAADLen in the parameter block. pAAD
may be NULL if ulAADLen is 0.
- Set
the tag length ulTagBits in the parameter block.
- Call
C_DecryptInit() for CKM_AES_GCM mechanism with parameters and key K.
- Call
C_Decrypt(), or C_DecryptUpdate()*1 C_DecryptFinal(), for the
ciphertext, including the appended tag, obtaining plaintext output. Note:
since CKM_AES_GCM is an AEAD cipher, no data should be returned
until C_Decrypt() or C_DecryptFinal().
MessageEncrypt:
- Set
the IV length ulIvLen in the parameter block.
- Set
pIv to hold the IV data returned from C_EncryptMessage() and
C_EncryptMessageBegin(). If ulIvFixedBits is not zero, then the
most significant bits of pIV contain the fixed IV. If ivGenerator
is set to CKG_NO_GENERATE, pIv is an input parameter with the full
IV.
- Set
the ulIvFixedBits and ivGenerator fields in the parameter block.
- Set
the tag length ulTagBits in the parameter block.
- Set pTag to hold the tag data returned
from C_EncryptMessage() or the final C_EncryptMessageNext().
- Call
C_MessageEncryptInit() for CKM_AES_GCM mechanism key K.
- Call
C_EncryptMessage(), or C_EncryptMessageBegin() followed by
C_EncryptMessageNext()*.
The mechanism parameter is passed to all three of these functions.
- Call
C_MessageEncryptFinal() to close the message decryption.
MessageDecrypt:
- Set
the IV length ulIvLen in the parameter block.
- Set
the IV data pIv in the parameter block.
- The ulIvFixedBits and ivGenerator
fields are ignored.
- Set the tag length ulTagBits in the
parameter block.
- Set
the tag data pTag in the parameter block before C_DecryptMessage()
or the final C_DecryptMessageNext().
- Call
C_MessageDecryptInit() for CKM_AES_GCM mechanism key K.
- Call
C_DecryptMessage(), or C_DecryptMessageBegin followed by
C_DecryptMessageNext()*.
The mechanism parameter is passed to all three of these functions.
- Call
C_MessageDecryptFinal() to close the message decryption.
In pIv the least significant bit of the
initialization vector is the rightmost bit. ulIvLen is the length of the
initialization vector in bytes.
On MessageEncrypt, the meaning of ivGenerator is as
follows: CKG_NO_GENERATE means the IV is passed in on MessageEncrypt and no
internal IV generation is done. CKG_GENERATE means that the non-fixed portion
of the IV is generated by the module internally. The generation method is not
defined.
CKG_GENERATE_COUNTER means that the non-fixed portion of the
IV is generated by the module internally by use of an incrementing counter, the
initial IV counter is zero.
CKG_GENERATE_COUNTER_XOR means that the non-fixed portion of
the IV is xored with a counter. The value of the non-fixed portion passed must
not vary from call to call. Like CKG_GENERATE_COUNTER, the counter starts at
zero.
CKG_GENERATE_RANDOM means that the non-fixed portion of the
IV is generated by the module internally using a PRNG. In any case the entire
IV, including the fixed portion, is returned in pIV.
Modules must implement CKG_GENERATE. Modules may also reject
ulIvFixedBits values which are too large. Zero is always an acceptable
value for ulIvFixedBits.
In Encrypt and Decrypt the tag is appended to the cipher
text and the least significant bit of the tag is the rightmost bit and the tag
bits are the rightmost ulTagBits bits. In MessageEncrypt the tag is
returned in the pTag field of CK_GCM_MESSAGE_PARAMS. In MesssageDecrypt
the tag is provided by the pTag field of CK_GCM_MESSAGE_PARAMS.
The key type for K must be compatible with CKM_AES_ECB
and the
C_EncryptInit()/C_DecryptInit()/C_MessageEncryptInit()/C_MessageDecryptInit()
calls shall behave, with respect to K, as if they were called directly
with CKM_AES_ECB, K and NULL parameters.
6.13.3 AES-CCM authenticated
Encryption / Decryption
For IPsec (RFC 4309) and also for use in ZFS encryption.
Generic CCM mode is described in [RFC 3610].
To set up for AES-CCM use the following process, where K
(key), nonce and additional authenticated data are as described in [RFC 3610]. AES-CCM
uses CK_CCM_PARAMS for Encrypt and Decrypt, and CK_CCM_MESSAGE_PARAMS for
MessageEncrypt and MessageDecrypt.
Encrypt:
- Set
the message/data length ulDataLen in the parameter block.
- Set
the nonce length ulNonceLen and the nonce data pNonce in the
parameter block.
- Set
the AAD data pAAD and size ulAADLen in the parameter block. pAAD
may be NULL if ulAADLen is 0.
- Set
the MAC length ulMACLen in the parameter block.
- Call
C_EncryptInit() for CKM_AES_CCM mechanism with parameters and key K.
- Call
C_Encrypt(), C_EncryptUpdate(), or C_EncryptFinal(), for the plaintext
obtaining the final ciphertext output and the MAC. The total length of
data processed must be ulDataLen. The output length will be ulDataLen
+ ulMACLen.
Decrypt:
- Set
the message/data length ulDataLen in the parameter block. This
length must not include the length of the MAC that is appended to the
cipher text.
- Set
the nonce length ulNonceLen and the nonce data pNonce in the
parameter block.
- Set
the AAD data pAAD and size ulAADLen in the parameter block. pAAD
may be NULL if ulAADLen is 0.
- Set
the MAC length ulMACLen in the parameter block.
- Call
C_DecryptInit() for CKM_AES_CCM mechanism with parameters and key K.
- Call
C_Decrypt(), C_DecryptUpdate(), or C_DecryptFinal(), for the ciphertext,
including the appended MAC, obtaining plaintext output. The total length
of data processed must be ulDataLen + ulMACLen. Note: since CKM_AES_CCM
is an AEAD cipher, no data should be returned until C_Decrypt() or
C_DecryptFinal().
MessageEncrypt:
- Set
the message/data length ulDataLen in the parameter block.
- Set
the nonce length ulNonceLen.
- Set
pNonce to hold the nonce data returned from C_EncryptMessage() and
C_EncryptMessageBegin(). If ulNonceFixedBits is not zero, then the
most significant bits of pNonce contain the fixed nonce. If nonceGenerator
is set to CKG_NO_GENERATE, pNonce is an input parameter with the
full nonce.
- Set
the ulNonceFixedBits and nonceGenerator fields in the parameter
block.
- Set
the MAC length ulMACLen in the parameter block.
- Set
pMAC to hold the MAC data returned from C_EncryptMessage() or the
final C_EncryptMessageNext().
- Call
C_MessageEncryptInit() for CKM_AES_CCM mechanism key K.
- Call
C_EncryptMessage(), or C_EncryptMessageBegin() followed by
C_EncryptMessageNext()*. The mechanism parameter is passed to all
three functions.
- Call
C_MessageEncryptFinal() to close the message encryption.
- The
MAC is returned in pMac of the CK_CCM_MESSAGE_PARAMS structure.
MessageDecrypt:
- Set
the message/data length ulDataLen in the parameter block.
- Set
the nonce length ulNonceLen and the nonce data pNonce in the
parameter block
- The
ulNonceFixedBits and nonceGenerator fields in the parameter
block are ignored.
- Set
the MAC length ulMACLen in the parameter block.
- Set
the MAC data pMAC in the parameter block before C_DecryptMessage()
or the final C_DecryptMessageNext().
- Call
C_MessageDecryptInit() for CKM_AES_CCM mechanism key K.
- Call
C_DecryptMessage(), or C_DecryptMessageBegin() followed by
C_DecryptMessageNext()*.
The mechanism parameter is passed to all three functions.
- Call
C_MessageDecryptFinal() to close the message decryption.
In pNonce the least significant bit of the nonce is
the rightmost bit. ulNonceLen is the length of the nonce in bytes.
On MessageEncrypt, the meaning of nonceGenerator is
as follows: CKG_NO_GENERATE means the nonce is passed in on MessageEncrypt and
no internal MAC generation is done. CKG_GENERATE means that the non-fixed
portion of the nonce is generated by the module internally. The generation
method is not defined.
CKG_GENERATE_COUNTER means that the non-fixed portion of the
nonce is generated by the module internally by use of an incrementing counter,
the initial IV counter is zero.
CKG_GENERATE_COUNTER_XOR means that the non-fixed portion of
the IV is xored with a counter. The value of the non-fixed portion passed must
not vary from call to call. Like CKG_GENERATE_COUNTER, the counter starts at
zero.
CKG_GENERATE_RANDOM means that the non-fixed portion of the
nonce is generated by the module internally using a PRNG. In any case the
entire nonce, including the fixed portion, is returned in pNonce.
Modules must implement CKG_GENERATE. Modules may also reject
ulNonceFixedBits values which are too large. Zero is always an
acceptable value for ulNonceFixedBits.
In Encrypt and Decrypt the MAC is appended to the cipher text
and the least significant byte of the MAC is the rightmost byte and the MAC
bytes are the rightmost ulMACLen bytes. In MessageEncrypt the MAC is
returned in the pMAC field of CK_CCM_MESSAGE_PARAMS. In MesssageDecrypt
the MAC is provided by the pMAC field of CK_CCM_MESSAGE_PARAMS.
The key type for K must be compatible with CKM_AES_ECB
and the C_EncryptInit()/C_DecryptInit()/C_MessageEncryptInit()/C_MessageDecryptInit()
calls shall behave, with respect to K, as if they were called directly with CKM_AES_ECB,
K and NULL parameters.
6.13.4 AES-GMAC
AES-GMAC, denoted CKM_AES_GMAC, is a mechanism for
single and multiple-part signatures and verification. It is described in NIST
Special Publication 800-38D [GMAC]. GMAC is a special case of GCM that
authenticates only the Additional Authenticated Data (AAD) part of the GCM
mechanism parameters. When GMAC is used with C_Sign or C_Verify, pData points
to the AAD. GMAC does not use plaintext or ciphertext.
The signature produced by GMAC, also referred to as a Tag,
the tag’s length is determined by the CK_GCM_PARAMS field ulTagBits.
The IV length is determined by the CK_GCM_PARAMS field ulIvLen.
Constraints on key types and the length of data are
summarized in the following table:
Table 115, AES-GMAC: Key And
Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
CKK_AES
|
< 2^64
|
Depends on param’s
ulTagBits
|
|
C_Verify
|
CKK_AES
|
< 2^64
|
Depends on param’s
ulTagBits
|
For this mechanism, the ulMinKeySize and ulMaxKeySize fields
of the CK_MECHANISM_INFO structure specify the supported range of AES
key sizes, in bytes.
¨
CK_GENERATOR_FUNCTION
Functions to generate unique IVs and nonces.
typedef CK_ULONG CK_GENERATOR_FUNCTION;
¨
CK_GCM_PARAMS;
CK_GCM_PARAMS_PTR
CK_GCM_PARAMS is a structure that provides the parameters to
the CKM_AES_GCM mechanism when used for Encrypt or Decrypt. It is defined as
follows:
typedef struct CK_GCM_PARAMS {
CK_BYTE_PTR pIv;
CK_ULONG ulIvLen;
CK_ULONG ulIvBits;
CK_BYTE_PTR pAAD;
CK_ULONG ulAADLen;
CK_ULONG ulTagBits;
} CK_GCM_PARAMS;
The fields of the structure have the following meanings:
pIv pointer
to initialization vector
ulIvLen length
of initialization vector in bytes. The length of the initialization vector can
be any number between 1 and (2^32) - 1. 96-bit (12 byte) IV values can be
processed more efficiently, so that length is recommended for situations in
which efficiency is critical.
ulIvBits length
of initialization vector in bits. Do no use ulIvBits to specify the length of
the initialization vector, but ulIvLen instead.
pAAD pointer
to additional authentication data. This data is authenticated but not
encrypted.
ulAADLen length of
pAAD in bytes. The length of the AAD can be any number between 0 and (2^32) –
1.
ulTagBits length
of authentication tag (output following cipher text) in bits. Can be any value
between 0 and 128.
CK_GCM_PARAMS_PTR is a pointer to a CK_GCM_PARAMS.
¨
CK_GCM_MESSAGE_PARAMS;
CK_GCM_MESSAGE_PARAMS_PTR
CK_GCM_MESSAGE_PARAMS is a structure that provides the
parameters to the CKM_AES_GCM mechanism when used for MessageEncrypt or
MessageDecrypt. It is defined as follows:
typedef struct CK_GCM_MESSAGE_PARAMS {
CK_BYTE_PTR pIv;
CK_ULONG ulIvLen;
CK_ULONG ulIvFixedBits;
CK_GENERATOR_FUNCTION ivGenerator;
CK_BYTE_PTR pTag;
CK_ULONG ulTagBits;
} CK_GCM_MESSAGE_PARAMS;
The fields of the structure have the following meanings:
pIv pointer
to initialization vector
ulIvLen length
of initialization vector in bytes. The length of the initialization vector can
be any number between 1 and (2^32) - 1. 96-bit (12 byte) IV values can be
processed more efficiently, so that length is recommended for situations in
which efficiency is critical.
ulIvFixedBits number of
bits of the original IV to preserve when generating an new IV. These bits are
counted from the Most significant bits (to the right).
ivGenerator Function
used to generate a new IV. Each IV must be unique for a given session.
pTag location
of the authentication tag which is returned on MessageEncrypt, and provided on
MessageDecrypt.
ulTagBits length
of authentication tag in bits. Can be any value between 0 and 128.
CK_GCM_MESSAGE_PARAMS_PTR is a pointer to a CK_GCM_MESSAGE_PARAMS.
¨
CK_CCM_PARAMS;
CK_CCM_PARAMS_PTR
CK_CCM_PARAMS is a structure that provides the
parameters to the CKM_AES_CCM mechanism when used for Encrypt or
Decrypt. It is defined as follows:
typedef struct
CK_CCM_PARAMS {
CK_ULONG ulDataLen;
/*plaintext or ciphertext*/
CK_BYTE_PTR pNonce;
CK_ULONG ulNonceLen;
CK_BYTE_PTR pAAD;
CK_ULONG ulAADLen;
CK_ULONG ulMACLen;
} CK_CCM_PARAMS;
The fields of the structure have the following meanings,
where L is the size in bytes of the data length’s length (2 ≤ L ≤
8):
ulDataLen length of
the data where 0 ≤ ulDataLen < 2^(8L).
pNonce the
nonce.
ulNonceLen length of
pNonce in bytes where 7 ≤ ulNonceLen ≤ 13.
pAAD Additional
authentication data. This data is authenticated but not encrypted.
ulAADLen length of
pAAD in bytes where 0 ≤ ulAADLen ≤ (2^32) - 1.
ulMACLen length of
the MAC (output following cipher text) in bytes. Valid values are 4, 6, 8, 10,
12, 14, and 16.
CK_CCM_PARAMS_PTR is a pointer to a CK_CCM_PARAMS.
¨
CK_CCM_MESSAGE_PARAMS;
CK_CCM_MESSAGE_PARAMS_PTR
CK_CCM_MESSAGE_PARAMS is a structure that provides
the parameters to the CKM_AES_CCM mechanism when used for MessageEncrypt
or MessageDecrypt. It is defined as follows:
typedef struct
CK_CCM_MESSAGE_PARAMS {
CK_ULONG ulDataLen;
/*plaintext or ciphertext*/
CK_BYTE_PTR pNonce;
CK_ULONG ulNonceLen;
CK_ULONG ulNonceFixedBits;
CK_GENERATOR_FUNCTION nonceGenerator;
CK_BYTE_PTR pMAC;
CK_ULONG ulMACLen;
} CK_CCM_MESSAGE_PARAMS;
The fields of the structure have the following meanings,
where L is the size in bytes of the data length’s length (2 ≤ L ≤
8):
ulDataLen length of
the data where 0 ≤ ulDataLen < 2^(8L).
pNonce the
nonce.
ulNonceLen length of
pNonce in bytes where 7 ≤ ulNonceLen ≤ 13.
ulNonceFixedBits number of
bits of the original nonce to preserve when generating a new nonce. These bits
are counted from the Most significant bits (to the right).
nonceGenerator Function
used to generate a new nonce. Each nonce must be unique for a given session.
pMAC location
of the CCM MAC returned on MessageEncrypt, provided on MessageDecrypt
ulMACLen length of
the MAC (output following cipher text) in bytes. Valid values are 4, 6, 8, 10,
12, 14, and 16.
CK_CCM_MESSAGE_PARAMS_PTR is a pointer to a CK_CCM_MESSAGE_PARAMS.
6.14 AES CMAC
Table 116, Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_AES_CMAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_AES_CMAC
|
|
ü
|
|
|
|
|
|
1 .
Mechanisms:
CKM_AES_CMAC_GENERAL
CKM_AES_CMAC
CKM_AES_CMAC_GENERAL uses the existing CK_MAC_GENERAL_PARAMS
structure. CKM_AES_CMAC does not use a mechanism parameter.
General-length AES-CMAC, denoted CKM_AES_CMAC_GENERAL,
is a mechanism for single- and multiple-part signatures and verification, based
on [NIST SP800-38B] and [RFC 4493].
It has a parameter, a CK_MAC_GENERAL_PARAMS structure,
which specifies the output length desired from the mechanism.
The output bytes from this mechanism are taken from the
start of the final AES cipher block produced in the MACing process.
Constraints on key types and the length of data are
summarized in the following table:
Table 117, General-length AES-CMAC: Key And Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
CKK_AES
|
any
|
1-block size, as
specified in parameters
|
|
C_Verify
|
CKK_AES
|
any
|
1-block size, as
specified in parameters
|
References [NIST SP800-38B] and [RFC 4493] recommend that
the output MAC is not truncated to less than 64 bits. The MAC length must be
specified before the communication starts, and must not be changed during the
lifetime of the key. It is the caller’s responsibility to follow these rules.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
AES key sizes, in bytes.
AES-CMAC, denoted CKM_AES_CMAC, is a special case of
the general-length AES-CMAC mechanism. AES-MAC always produces and verifies
MACs that are a full block size in length, the default output length specified
by [RFC 4493].
Constraints on key types and the length of data are
summarized in the following table:
Table 118, AES-CMAC: Key And Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
CKK_AES
|
any
|
Block size (16 bytes)
|
|
C_Verify
|
CKK_AES
|
any
|
Block size (16 bytes)
|
References [NIST SP800-38B] and [RFC 4493] recommend that
the output MAC is not truncated to less than 64 bits. The MAC length must be
specified before the communication starts, and must not be changed during the
lifetime of the key. It is the caller’s responsibility to follow these rules.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
AES key sizes, in bytes.
Table 119, Mechanisms vs.
Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_AES_XTS
|
ü
|
|
|
|
|
ü
|
|
|
CKM_AES_XTS_KEY_GEN
|
|
|
|
|
ü
|
|
|
This section defines the key type “CKK_AES_XTS” for type
CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of key objects.
Mechanisms:
CKM_AES_XTS
CKM_AES_XTS_KEY_GEN
Table 120, AES-XTS Secret Key
Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_VALUE1,4,6,7
|
Byte array
|
Key value (32 or 64 bytes)
|
|
CKA_VALUE_LEN2,3,6
|
CK_ULONG
|
Length in bytes of key value
|
- Refer
to Table 11 for footnotes
The double-length AES-XTS key generation mechanism, denoted CKM_AES_XTS_KEY_GEN,
is a key generation mechanism for double-length AES-XTS keys.
The mechanism generates AES-XTS keys with a particular
length in bytes as specified in the CKA_VALUE_LEN attributes of the template
for the key.
This mechanism contributes the CKA_CLASS, CKA_KEY_TYPE, and
CKA_VALUE attributes to the new key. Other attributes supported by the
double-length AES-XTS key type (specifically, the flags indicating which
functions the key supports) may be specified in the template for the key, or
else are assigned default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize fields
of the CK_MECHANISM_INFO structure specify the supported range of AES-XTS key
sizes, in bytes.
AES-XTS (XEX-based Tweaked CodeBook mode with CipherText
Stealing), denoted CKM_AES_XTS, isa mechanism for single- and
multiple-part encryption and decryption. It is specified in NIST SP800-38E.
Its single parameter is a Data Unit Sequence Number 16 bytes
long. Supported key lengths are 32 and 64 bytes. Keys are internally split into
half-length sub-keys of 16 and 32 bytes respectively. Constraintson key types
and the length of data are summarized in the following table:
Table 121, AES-XTS: Key And Data
Length
|
Function
|
Key type
|
Input length
|
Output length
|
Comments
|
|
C_Encrypt
|
CKK_AES_XTS
|
Any, ≥ block
size (16 bytes)
|
Same as input length
|
No final part
|
|
C_Decrypt
|
CKK_AES_XTS
|
Any, ≥ block
size (16 bytes)
|
Same as input length
|
No final part
|
Table
122, AES Key Wrap Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_AES_KEY_WRAP
|
ü
|
|
|
|
|
ü
|
|
|
CKM_AES_KEY_WRAP_PAD
|
ü
|
|
|
|
|
ü
|
|
|
CKM_AES_KEY_WRAP_KWP
|
ü
|
|
|
|
|
ü
|
|
|
CKM_AES_KEY_WRAP_PKCS7
|
ü
|
|
|
|
|
ü
|
|
|
1SR = SignRecover, VR = VerifyRecover
|
Mechanisms:
CKM_AES_KEY_WRAP
CKM_AES_KEY_WRAP_PAD
CKM_AES_KEY_WRAP_KWP
CKM_AES_KEY_WRAP_PKCS7
The mechanisms will accept an optional
mechanism parameter as the Initialization vector which, if present, must be a
fixed size array of 8 bytes for CKM_AES_KEY_WRAP and CKM_AES_KEY_WRAP_PKCS7,
resp. 4 bytes for CKM_AES_KEY_WRAP_KWP; and, if NULL, will use the default
initial value defined in Section 4.3 resp. 6.2 / 6.3 of [AES KEYWRAP].
The type of this parameter is CK_BYTE_PTR and the pointer
points to the array of bytes to be used as the initial value. The length shall
be either 0 and the pointer NULL; or 8 for CKM_AES_KEY_WRAP and CKM_AES_KEY_WRAP_PKCS7,
resp. 4 for CKM_AES_KEY_WRAP_KWP, and the pointer non-NULL.
The mechanisms support only single-part operations, i.e.
single part wrapping and unwrapping, and single-part encryption and decryption.
¨
CKM_AES_KEY_WRAP
The
CKM_AES_KEY_WRAP mechanism can wrap
a key of any length. A secret key whose length is not a multiple of the AES Key
Wrap semiblock size (8 bytes) will be zero padded to fit. Semiblock size
is defined in Section 5.2 of [AES KEYWRAP].
A private key will be encoded as defined in section 6.7;
the encoded private key will be zero padded to fit if necessary.
The
CKM_AES_KEY_WRAP mechanism can only encrypt a block of data whose size is an
exact multiple of the AES Key Wrap algorithm semiblock size.
For unwrapping, the mechanism decrypts the wrapped key. In
case of a secret key, it truncates the result according to the CKA_KEY_TYPE
attribute of the template and, if it has one and the key type supports it, the
CKA_VALUE_LEN attribute of the template. The length specified in the template
must not be less than n-7 bytes, where n is the length of the wrapped key. In
case of a private key, the mechanism parses the encoding as defined in section 6.7
and ignores trailing zero bytes.
¨
CKM_AES_KEY_WRAP_PAD
The CKM_AES_KEY_WRAP_PAD mechanism is deprecated.
CKM_AES_KEY_WRAP_KWP resp. CKM_AES_KEY_WRAP_PKCS7 shall be used instead.
¨
CKM_AES_KEY_WRAP_KWP
The CKM_AES_KEY_WRAP_KWP mechanism can wrap a key or encrypt
block of data of any length. The input is zero-padded and wrapped / encrypted
as defined in Section 6.3 of [AES KEYWRAP], which produces same results as RFC
5649.
¨
CKM_AES_KEY_WRAP_PKCS7
The CKM_AES_KEY_WRAP_PKCS7 mechanism can wrap a key or
encrypt a block of data of any length. It does the padding detailed in [PKCS #7]
of inputs (keys or data blocks) up to a semiblock size to make it an exact
multiple of AES Key Wrap algorithum semiblock size (8bytes), always producing
wrapped output that is larger than the input key/data to be wrapped. This
padding is done by the token before being passed to the AES key wrap algorithm,
which then wraps / encrypts the padded block of data as defined in Section 6.2
of [AES KEYWRAP].
These mechanisms allow derivation of keys using the result
of an encryption operation as the key value. They are for use with the
C_DeriveKey function.
Table
123, Key derivation by data encryption Mechanisms vs.
Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_DES_ECB_ENCRYPT_DATA
|
|
|
|
|
|
|
ü
|
|
CKM_DES_CBC_ENCRYPT_DATA
|
|
|
|
|
|
|
ü
|
|
CKM_DES3_ECB_ENCRYPT_DATA
|
|
|
|
|
|
|
ü
|
|
CKM_DES3_CBC_ENCRYPT_DATA
|
|
|
|
|
|
|
ü
|
|
CKM_AES_ECB_ENCRYPT_DATA
|
|
|
|
|
|
|
ü
|
|
CKM_AES_CBC_ENCRYPT_DATA
|
|
|
|
|
|
|
ü
|
Mechanisms:
CKM_DES_ECB_ENCRYPT_DATA
CKM_DES_CBC_ENCRYPT_DATA
CKM_DES3_ECB_ENCRYPT_DATA
CKM_DES3_CBC_ENCRYPT_DATA
CKM_AES_ECB_ENCRYPT_DATA
CKM_AES_CBC_ENCRYPT_DATA
typedef struct CK_DES_CBC_ENCRYPT_DATA_PARAMS {
CK_BYTE iv[8];
CK_BYTE_PTR pData;
CK_ULONG length;
} CK_DES_CBC_ENCRYPT_DATA_PARAMS;
typedef CK_DES_CBC_ENCRYPT_DATA_PARAMS CK_PTR
CK_DES_CBC_ENCRYPT_DATA_PARAMS_PTR;
typedef struct CK_AES_CBC_ENCRYPT_DATA_PARAMS {
CK_BYTE iv[16];
CK_BYTE_PTR pData;
CK_ULONG length;
} CK_AES_CBC_ENCRYPT_DATA_PARAMS;
typedef CK_AES_CBC_ENCRYPT_DATA_PARAMS CK_PTR
CK_AES_CBC_ENCRYPT_DATA_PARAMS_PTR;
Uses CK_KEY_DERIVATION_STRING_DATA as defined in section 6.43.2
Table 124, Mechanism Parameters
|
CKM_DES_ECB_ENCRYPT_DATA
CKM_DES3_ECB_ENCRYPT_DATA
|
Uses
CK_KEY_DERIVATION_STRING_DATA structure. Parameter is the data to be
encrypted and must be a multiple of 8 bytes long.
|
|
CKM_AES_ECB_ENCRYPT_DATA
|
Uses
CK_KEY_DERIVATION_STRING_DATA structure. Parameter is the data to be
encrypted and must be a multiple of 16 long.
|
|
CKM_DES_CBC_ENCRYPT_DATA
CKM_DES3_CBC_ENCRYPT_DATA
|
Uses
CK_DES_CBC_ENCRYPT_DATA_PARAMS. Parameter is an 8 byte IV value followed by
the data. The data value part must be a multiple of 8 bytes long.
|
|
CKM_AES_CBC_ENCRYPT_DATA
|
Uses
CK_AES_CBC_ENCRYPT_DATA_PARAMS. Parameter is an 16 byte IV value followed by
the data. The data value part
must
be a multiple of 16 bytes long.
|
The mechanisms will function by performing the encryption
over the data provided using the base key. The resulting cipher text shall be
used to create the key value of the resulting key. If not all the cipher text
is used then the part discarded will be from the trailing end (least
significant bytes) of the cipher text data. The derived key shall be defined by
the attribute template supplied but constrained by the length of cipher text
available for the key value and other normal PKCS11 derivation constraints.
Attribute template handling, attribute defaulting and key
value preparation will operate as per the SHA-1 Key Derivation mechanism in
section 6.20.5.
If the data is too short to make the requested key then the
mechanism returns CKR_DATA_LEN_RANGE.
Table
125, Double and Triple-Length DES Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_DES2_KEY_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_DES3_KEY_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_DES3_ECB
|
ü
|
|
|
|
|
ü
|
|
|
CKM_DES3_CBC
|
ü
|
|
|
|
|
ü
|
|
|
CKM_DES3_CBC_PAD
|
ü
|
|
|
|
|
ü
|
|
|
CKM_DES3_MAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_DES3_MAC
|
|
ü
|
|
|
|
|
|
This section defines the key type “CKK_DES2” and “CKK_DES3”
for type CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of key objects.
Mechanisms:
CKM_DES2_KEY_GEN
CKM_DES3_KEY_GEN
CKM_DES3_ECB
CKM_DES3_CBC
CKM_DES3_MAC
CKM_DES3_MAC_GENERAL
CKM_DES3_CBC_PAD
DES2 secret key objects (object class CKO_SECRET_KEY, key
type CKK_DES2) hold double-length DES keys. The following table defines
the DES2 secret key object attributes, in addition to the common attributes
defined for this object class:
Table 126, DES2 Secret Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_VALUE1,4,6,7
|
Byte array
|
Key value (always 16 bytes long)
|
- Refer to Table 11 for footnotes
DES2 keys must always have their parity bits properly set as
described in FIPS PUB 46-3 (i.e., each of the DES keys comprising a DES2
key must have its parity bits properly set). Attempting to create or unwrap a
DES2 key with incorrect parity will return an error.
The following is a sample template for creating a
double-length DES secret key object:
CK_OBJECT_CLASS class = CKO_SECRET_KEY;
CK_KEY_TYPE keyType = CKK_DES2;
CK_UTF8CHAR label[] = “A DES2 secret key object”;
CK_BYTE value[16] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_ENCRYPT, &true, sizeof(true)},
{CKA_VALUE, value, sizeof(value)}
};
CKA_CHECK_VALUE: The value of this attribute is derived from
the key object by taking the first three bytes of the ECB encryption of a single
block of null (0x00) bytes, using the default cipher associated with the key
type of the secret key object.
DES3 secret key objects (object class CKO_SECRET_KEY, key
type CKK_DES3) hold triple-length DES keys. The following table defines
the DES3 secret key object attributes, in addition to the common attributes
defined for this object class:
Table 127, DES3 Secret Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_VALUE1,4,6,7
|
Byte array
|
Key value (always 24 bytes long)
|
- Refer to Table 11 for footnotes
DES3 keys must always have their parity bits properly set as
described in FIPS PUB 46-3 (i.e., each of the DES keys comprising a DES3
key must have its parity bits properly set). Attempting to create or unwrap a
DES3 key with incorrect parity will return an error.
The following is a sample template for creating a
triple-length DES secret key object:
CK_OBJECT_CLASS class = CKO_SECRET_KEY;
CK_KEY_TYPE keyType = CKK_DES3;
CK_UTF8CHAR label[] = “A DES3 secret key object”;
CK_BYTE value[24] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_ENCRYPT, &true, sizeof(true)},
{CKA_VALUE, value, sizeof(value)}
};
CKA_CHECK_VALUE: The value of this attribute is derived from
the key object by taking the first three bytes of the ECB encryption of a single
block of null (0x00) bytes, using the default cipher associated with the key
type of the secret key object.
The double-length DES key generation mechanism, denoted CKM_DES2_KEY_GEN,
is a key generation mechanism for double-length DES keys. The DES keys making
up a double-length DES key both have their parity bits set properly, as
specified in FIPS PUB 46-3.
It does not have a parameter.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key. Other attributes supported by
the double-length DES key type (specifically, the flags indicating which
functions the key supports) may be specified in the template for the key, or
else are assigned default initial values.
Double-length DES keys can be used with all the same
mechanisms as triple-DES keys: CKM_DES3_ECB, CKM_DES3_CBC, CKM_DES3_CBC_PAD,
CKM_DES3_MAC_GENERAL, and CKM_DES3_MAC. Triple-DES encryption
with a double-length DES key is equivalent to encryption with a triple-length
DES key with K1=K3 as specified in FIPS PUB 46-3.
When
double-length DES keys are generated, it is token-dependent whether or not it
is possible for either of the component DES keys to be “weak” or “semi-weak”
keys.
Triple-length DES encryptions are carried out as specified
in FIPS PUB 46-3: encrypt, decrypt, encrypt. Decryptions are carried out with
the opposite three steps: decrypt, encrypt, decrypt. The mathematical
representations of the encrypt and decrypt operations are as follows:
DES3-E({K1,K2,K3}, P) = E(K3, D(K2, E(K1, P)))
DES3-D({K1,K2,K3}, C) = D(K1, E(K2, D(K3, P)))
Triple-length DES operations in CBC mode, with double or
triple-length keys, are performed using outer CBC as defined in X9.52. X9.52
describes this mode as TCBC. The mathematical representations of the CBC
encrypt and decrypt operations are as follows:
DES3-CBC-E({K1,K2,K3}, P) = E(K3, D(K2, E(K1, P +
I)))
DES3-CBC-D({K1,K2,K3}, C) = D(K1, E(K2, D(K3, P)))
+ I
The value I is either an 8-byte initialization vector
or the previous block of cipher text that is added to the current input block.
The addition operation is used is addition modulo-2 (XOR).
Table
128, DES and Triple Length DES in OFB Mode Mechanisms vs.
Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_DES_OFB64
|
ü
|
|
|
|
|
|
|
|
CKM_DES_OFB8
|
ü
|
|
|
|
|
|
|
|
CKM_DES_CFB64
|
ü
|
|
|
|
|
|
|
|
CKM_DES_CFB8
|
ü
|
|
|
|
|
|
|
Cipher DES has a output feedback mode, DES-OFB, denoted CKM_DES_OFB8
and CKM_DES_OFB64. It is a mechanism for single and multiple-part
encryption and decryption with DES.
It has a parameter, an initialization vector for this mode.
The initialization vector has the same length as the block size.
Constraints on key types and the length of data are
summarized in the following table:
Table 129, OFB: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
Comments
|
|
C_Encrypt
|
CKK_DES, CKK_DES2, CKK_DES3
|
any
|
same as input length
|
no final part
|
|
C_Decrypt
|
CKK_DES, CKK_DES2, CKK_DES3
|
any
|
same as input length
|
no final part
|
For this mechanism the CK_MECHANISM_INFO structure is
as specified for CBC mode.
Cipher DES has a cipher feedback mode, DES-CFB, denoted CKM_DES_CFB8
and CKM_DES_CFB64. It is a mechanism for single and multiple-part
encryption and decryption with DES.
It has a parameter, an initialization vector for this mode.
The initialization vector has the same length as the block size.
Constraints on key types and the length of data are
summarized in the following table:
Table 130, CFB: Key And Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
Comments
|
|
C_Encrypt
|
CKK_DES, CKK_DES2, CKK_DES3
|
any
|
same as input length
|
no final part
|
|
C_Decrypt
|
CKK_DES, CKK_DES2, CKK_DES3
|
any
|
same as input length
|
no final part
|
For this mechanism the CK_MECHANISM_INFO structure is
as specified for CBC mode.
Table
131, Double and Triple-length DES CMAC Mechanisms vs.
Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_DES3_CMAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_DES3_CMAC
|
|
ü
|
|
|
|
|
|
1
Mechanisms:
CKM_DES3_CMAC_GENERAL
CKM_DES3_CMAC
CKM_DES3_CMAC_GENERAL uses the existing CK_MAC_GENERAL_PARAMS
structure. CKM_DES3_CMAC does not use a mechanism parameter.
General-length DES3-CMAC, denoted CKM_DES3_CMAC_GENERAL,
is a mechanism for single- and multiple-part signatures and verification with
DES3 or DES2 keys, based on [NIST sp800-38b].
It has a parameter, a CK_MAC_GENERAL_PARAMS structure,
which specifies the output length desired from the mechanism.
The output bytes from this mechanism are taken from the start
of the final DES3 cipher block produced in the MACing process.
Constraints on key types and the length of data are
summarized in the following table:
Table 132, General-length DES3-CMAC: Key And Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
CKK_DES3
CKK_DES2
|
any
|
1-block size, as
specified in parameters
|
|
C_Verify
|
CKK_DES3
CKK_DES2
|
any
|
1-block size, as
specified in parameters
|
Reference [NIST sp800-38b] recommends that the output MAC is
not truncated to less than 64 bits (which means using the entire block for
DES). The MAC length must be specified before the communication starts, and
must not be changed during the lifetime of the key. It is the caller’s
responsibility to follow these rules.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure are not used
DES3-CMAC, denoted CKM_DES3_CMAC, is a special case
of the general-length DES3-CMAC mechanism. DES3-MAC always produces and
verifies MACs that are a full block size in length, since the DES3 block length
is the minimum output length recommended by [NIST sp800-38b].
Constraints on key types and the length of data are
summarized in the following table:
Table 133, DES3-CMAC: Key And Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
CKK_DES3
CKK_DES2
|
any
|
Block size (8 bytes)
|
|
C_Verify
|
CKK_DES3
CKK_DES2
|
any
|
Block size (8 bytes)
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure are not used.
Table
134, SHA-1 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_SHA_1
|
|
|
|
ü
|
|
|
|
|
CKM_SHA_1_HMAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_SHA_1_HMAC
|
|
ü
|
|
|
|
|
|
|
CKM_SHA1_KEY_DERIVATION
|
|
|
|
|
|
|
ü
|
|
CKM_SHA_1_KEY_GEN
|
|
|
|
|
ü
|
|
|
This section defines the key type “CKK_SHA_1_HMAC” for type
CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of key objects.
Mechanisms:
CKM_SHA_1
CKM_SHA_1_HMAC
CKM_SHA_1_HMAC_GENERAL
CKM_SHA1_KEY_DERIVATION
CKM_SHA_1_KEY_GEN
The SHA-1 mechanism, denoted CKM_SHA_1, is a
mechanism for message digesting, following the Secure Hash Algorithm with a
160-bit message digest defined in FIPS PUB 180-2.
It does not have a parameter.
Constraints on the length of input and output data are
summarized in the following table. For single-part digesting, the data and the
digest may begin at the same location in memory.
Table
135, SHA-1: Data Length
|
Function
|
Input length
|
Digest length
|
|
C_Digest
|
any
|
20
|
The general-length SHA-1-HMAC mechanism, denoted CKM_SHA_1_HMAC_GENERAL,
is a mechanism for signatures and verification. It uses the HMAC construction,
based on the SHA-1 hash function. The keys it uses are generic secret keys and
CKK_SHA_1_HMAC.
It has a parameter, a CK_MAC_GENERAL_PARAMS, which
holds the length in bytes of the desired output. This length should be in the
range 1-20 (the output size of SHA-1 is 20 bytes). Signatures (MACs) produced
by this mechanism will be taken from the start of the full 20-byte HMAC output.
Table 136, General-length SHA-1-HMAC: Key And Data
Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
generic secret
CKK_SHA_1_HMAC
|
any
|
1-20, depending on
parameters
|
|
C_Verify
|
generic secret
CKK_SHA_1_HMAC
|
any
|
1-20, depending on
parameters
|
The SHA-1-HMAC mechanism, denoted CKM_SHA_1_HMAC, is
a special case of the general-length SHA-1-HMAC mechanism in Section 6.20.3.
It has no parameter, and always produces an output of length
20.
SHA-1 key derivation, denoted CKM_SHA1_KEY_DERIVATION,
is a mechanism which provides the capability of deriving a secret key by
digesting the value of another secret key with SHA-1.
The value of the base key is digested once, and the result
is used to make the value of derived secret key.
·
If no length or key type is provided in the template, then the
key produced by this mechanism will be a generic secret key. Its length will
be 20 bytes (the output size of SHA-1).
·
If no key type is provided in the template, but a length is, then
the key produced by this mechanism will be a generic secret key of the
specified length.
·
If no length was provided in the template, but a key type is,
then that key type must have a well-defined length. If it does, then the key
produced by this mechanism will be of the type specified in the template. If
it doesn’t, an error will be returned.
·
If both a key type and a length are provided in the template, the
length must be compatible with that key type. The key produced by this
mechanism will be of the specified type and length.
If a DES, DES2, or CDMF key is derived with this mechanism,
the parity bits of the key will be set properly.
If the requested type of key requires more than 20 bytes,
such as DES3, an error is generated.
This mechanism has the following rules about key sensitivity
and extractability:
·
The CKA_SENSITIVE and CKA_EXTRACTABLE attributes in
the template for the new key can both be specified to be either CK_TRUE or
CK_FALSE. If omitted, these attributes each take on some default value.
·
If the base key has its CKA_ALWAYS_SENSITIVE attribute set
to CK_FALSE, then the derived key will as well. If the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE, then the derived key has its CKA_ALWAYS_SENSITIVE
attribute set to the same value as its CKA_SENSITIVE attribute.
·
Similarly, if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_FALSE, then the derived key will, too. If the base key has
its CKA_NEVER_EXTRACTABLE attribute set to CK_TRUE, then the derived key
has its CKA_NEVER_EXTRACTABLE attribute set to the opposite value
from its CKA_EXTRACTABLE attribute.
The SHA-1-HMAC key generation
mechanism, denoted CKM_SHA_1_KEY_GEN, is a key generation mechanism for
NIST’s SHA-1-HMAC.
It does not have a parameter.
The mechanism generates SHA-1-HMAC
keys with a particular length in bytes, as specified in the CKA_VALUE_LEN
attribute of the template for the key.
The mechanism contributes the CKA_CLASS,
CKA_KEY_TYPE, and CKA_VALUE attributes to the new key. Other
attributes supported by the SHA-1-HMAC key type (specifically, the flags
indicating which functions the key supports) may be specified in the template
for the key, or else are assigned default initial values.
For this mechanism, the ulMinKeySize
and ulMaxKeySize fields of the CK_MECHANISM_INFO structure
specify the supported range of CKM_SHA_1_HMAC key sizes, in bytes.
Table
137, SHA-224 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_SHA224
|
|
|
|
ü
|
|
|
|
|
CKM_SHA224_HMAC
|
|
ü
|
|
|
|
|
|
|
CKM_SHA224_HMAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_SHA224_KEY_DERIVATION
|
|
|
|
|
|
|
ü
|
|
CKM_SHA224_KEY_GEN
|
|
|
|
|
ü
|
|
|
This section defines the key type
“CKK_SHA224_HMAC” for type CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of
key objects.
Mechanisms:
CKM_SHA224
CKM_SHA224_HMAC
CKM_SHA224_HMAC_GENERAL
CKM_SHA224_KEY_DERIVATION
CKM_SHA224_KEY_GEN
The SHA-224 mechanism, denoted CKM_SHA224, is a
mechanism for message digesting, following the Secure Hash Algorithm with a
224-bit message digest defined in FIPS PUB 180-4.
It does not have a parameter.
Constraints on the length of input and output data are
summarized in the following table. For single-part digesting, the data and the
digest may begin at the same location in memory.
Table 138, SHA-224: Data Length
|
Function
|
Input length
|
Digest length
|
|
C_Digest
|
any
|
28
|
The general-length SHA-224-HMAC mechanism, denoted CKM_SHA224_HMAC_GENERAL,
is the same as the general-length SHA-1-HMAC mechanism except that it uses the
HMAC construction based on the SHA-224 hash function and length of the output
should be in the range 1-28. The keys it uses are generic secret keys and
CKK_SHA224_HMAC. FIPS-198 compliant tokens may require the key length to be at
least 14 bytes; that is, half the size of the SHA-224 hash output.
It has a parameter, a CK_MAC_GENERAL_PARAMS, which
holds the length in bytes of the desired output. This length should be in the
range 1-28 (the output size of SHA-224 is 28 bytes). FIPS-198 compliant tokens
may constrain the output length to be at least 4 or 14 (half the maximum
length). Signatures (MACs) produced by this mechanism will be taken from the
start of the full 28-byte HMAC output.
Table 139, General-length SHA-224-HMAC: Key And Data
Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
generic secret
CKK_SHA224_HMAC
|
Any
|
1-28, depending on
parameters
|
|
C_Verify
|
generic secret
CKK_SHA224_HMAC
|
Any
|
1-28, depending on
parameters
|
The SHA-224-HMAC mechanism, denoted CKM_SHA224_HMAC,
is a special case of the general-length SHA-224-HMAC mechanism.
It has no parameter, and always produces an output of length
28.
SHA-224 key derivation, denoted CKM_SHA224_KEY_DERIVATION,
is the same as the SHA-1 key derivation mechanism in Section 6.20.5
except that it uses the SHA-224 hash function and the relevant length is 28
bytes.
The SHA-224-HMAC key generation
mechanism, denoted CKM_SHA224_KEY_GEN, is a key generation mechanism for
NIST’s SHA224-HMAC.
It does not have a parameter.
The mechanism generates
SHA224-HMAC keys with a particular length in bytes, as specified in the CKA_VALUE_LEN
attribute of the template for the key.
The mechanism contributes the CKA_CLASS,
CKA_KEY_TYPE, and CKA_VALUE attributes to the new key. Other
attributes supported by the SHA224-HMAC key type (specifically, the flags
indicating which functions the key supports) may be specified in the template
for the key, or else are assigned default initial values.
For this mechanism, the ulMinKeySize
and ulMaxKeySize fields of the CK_MECHANISM_INFO structure
specify the supported range of CKM_SHA224_HMAC key sizes, in bytes.
Table
140, SHA-256 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_SHA256
|
|
|
|
ü
|
|
|
|
|
CKM_SHA256_HMAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_SHA256_HMAC
|
|
ü
|
|
|
|
|
|
|
CKM_SHA256_KEY_DERIVATION
|
|
|
|
|
|
|
ü
|
|
CKM_SHA256_KEY_GEN
|
|
|
|
|
ü
|
|
|
This section defines the key type
“CKK_SHA256_HMAC” for type CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of
key objects.
Mechanisms:
CKM_SHA256
CKM_SHA256_HMAC
CKM_SHA256_HMAC_GENERAL
CKM_SHA256_KEY_DERIVATION
CKM_SHA256_KEY_GEN
The SHA-256 mechanism, denoted CKM_SHA256, is a
mechanism for message digesting, following the Secure Hash Algorithm with a
256-bit message digest defined in FIPS PUB 180-2.
It does not have a parameter.
Constraints on the length of input and output data are
summarized in the following table. For single-part digesting, the data and the
digest may begin at the same location in memory.
Table 141, SHA-256: Data Length
|
Function
|
Input length
|
Digest length
|
|
C_Digest
|
any
|
32
|
The general-length SHA-256-HMAC mechanism, denoted CKM_SHA256_HMAC_GENERAL,
is the same as the general-length SHA-1-HMAC mechanism in Section 6.20.3, except that it uses the HMAC construction based on the SHA-256 hash function
and length of the output should be in the range 1-32. The keys it uses are
generic secret keys and CKK_SHA256_HMAC. FIPS-198 compliant tokens may require
the key length to be at least 16 bytes; that is, half the size of the SHA-256
hash output.
It has a parameter, a CK_MAC_GENERAL_PARAMS, which
holds the length in bytes of the desired output. This length should be in the
range 1-32 (the output size of SHA-256 is 32 bytes). FIPS-198 compliant tokens
may constrain the output length to be at least 4 or 16 (half the maximum
length). Signatures (MACs) produced by this mechanism will be taken from the
start of the full 32-byte HMAC output.
Table 142, General-length SHA-256-HMAC: Key And Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
generic secret,
CKK_SHA256_HMAC
|
Any
|
1-32, depending on
parameters
|
|
C_Verify
|
generic secret,
CKK_SHA256_HMAC
|
Any
|
1-32, depending on parameters
|
The SHA-256-HMAC mechanism, denoted CKM_SHA256_HMAC,
is a special case of the general-length SHA-256-HMAC mechanism in Section 6.22.3.
It has no parameter, and always produces an output of length
32.
SHA-256 key derivation, denoted CKM_SHA256_KEY_DERIVATION,
is the same as the SHA-1 key derivation mechanism in Section 6.20.5,
except that it uses the SHA-256 hash function and the relevant length is 32
bytes.
The SHA-256-HMAC key generation
mechanism, denoted CKM_SHA256_KEY_GEN, is a key generation mechanism for
NIST’s SHA256-HMAC.
It does not have a parameter.
The mechanism generates
SHA256-HMAC keys with a particular length in bytes, as specified in the CKA_VALUE_LEN
attribute of the template for the key.
The mechanism contributes the CKA_CLASS,
CKA_KEY_TYPE, and CKA_VALUE attributes to the new key. Other
attributes supported by the SHA256-HMAC key type (specifically, the flags
indicating which functions the key supports) may be specified in the template
for the key, or else are assigned default initial values.
For this mechanism, the ulMinKeySize
and ulMaxKeySize fields of the CK_MECHANISM_INFO structure
specify the supported range of CKM_SHA256_HMAC key sizes, in bytes.
Table
143, SHA-384 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_SHA384
|
|
|
|
ü
|
|
|
|
|
CKM_SHA384_HMAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_SHA384_HMAC
|
|
ü
|
|
|
|
|
|
|
CKM_SHA384_KEY_DERIVATION
|
|
|
|
|
|
|
ü
|
|
CKM_SHA384_KEY_GEN
|
|
|
|
|
ü
|
|
|
This section defines the key type
“CKK_SHA384_HMAC” for type CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of
key objects.
CKM_SHA384
CKM_SHA384_HMAC
CKM_SHA384_HMAC_GENERAL
CKM_SHA384_KEY_DERIVATION
CKM_SHA384_KEY_GEN
The SHA-384 mechanism, denoted CKM_SHA384, is a
mechanism for message digesting, following the Secure Hash Algorithm with a
384-bit message digest defined in FIPS PUB 180-2.
It does not have a parameter.
Constraints on the length of input and output data are
summarized in the following table. For single-part digesting, the data and the
digest may begin at the same location in memory.
Table 144, SHA-384: Data Length
|
Function
|
Input length
|
Digest length
|
|
C_Digest
|
any
|
48
|
The general-length SHA-384-HMAC mechanism, denoted CKM_SHA384_HMAC_GENERAL,
is the same as the general-length SHA-1-HMAC mechanism in Section 6.20.3, except that it uses the HMAC construction based on the SHA-384 hash function
and length of the output should be in the range 1-48.
The keys it uses are generic
secret keys and CKK_SHA384_HMAC. FIPS-198 compliant tokens may require the key
length to be at least 24 bytes; that is, half the size of the SHA-384 hash
output.
It has a parameter, a CK_MAC_GENERAL_PARAMS, which holds the length in bytes of the desired output.
This length should be in the range 0-48 (the output size of SHA-384 is 48
bytes). FIPS-198 compliant tokens may constrain the output length to be at
least 4 or 24 (half the maximum length). Signatures (MACs) produced by this
mechanism will be taken from the start of the full 48-byte HMAC output.
Table 145, General-length
SHA-384-HMAC: Key And Data Length
|
Function
|
Key type
|
Data
length
|
Signature
length
|
|
C_Sign
|
generic
secret, CKK_SHA384_HMAC
|
Any
|
1-48,
depending on parameters
|
|
C_Verify
|
generic
secret,
CKK_SHA384_HMAC
|
Any
|
1-48, depending
on parameters
|
The SHA-384-HMAC mechanism, denoted CKM_SHA384_HMAC,
is a special case of the general-length SHA-384-HMAC mechanism.
It has no parameter, and always produces an output of length
48.
SHA-384 key derivation, denoted CKM_SHA384_KEY_DERIVATION,
is the same as the SHA-1 key derivation mechanism in Section 6.20.5,
except that it uses the SHA-384 hash function and the relevant length is 48
bytes.
The SHA-384-HMAC key generation
mechanism, denoted CKM_SHA384_KEY_GEN, is a key generation mechanism for
NIST’s SHA384-HMAC.
It does not have a parameter.
The mechanism generates
SHA384-HMAC keys with a particular length in bytes, as specified in the CKA_VALUE_LEN
attribute of the template for the key.
The mechanism contributes the CKA_CLASS,
CKA_KEY_TYPE, and CKA_VALUE attributes to the new key. Other
attributes supported by the SHA384-HMAC key type (specifically, the flags
indicating which functions the key supports) may be specified in the template
for the key, or else are assigned default initial values.
For this mechanism, the ulMinKeySize
and ulMaxKeySize fields of the CK_MECHANISM_INFO structure
specify the supported range of CKM_SHA384_HMAC key sizes, in bytes.
Table
146, SHA-512 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_SHA512
|
|
|
|
ü
|
|
|
|
|
CKM_SHA512_HMAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_SHA512_HMAC
|
|
ü
|
|
|
|
|
|
|
CKM_SHA512_KEY_DERIVATION
|
|
|
|
|
|
|
ü
|
|
CKM_SHA512_KEY_GEN
|
|
|
|
|
ü
|
|
|
This section defines the key type
“CKK_SHA512_HMAC” for type CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of
key objects.
Mechanisms:
CKM_SHA512
CKM_SHA512_HMAC
CKM_SHA512_HMAC_GENERAL
CKM_SHA512_KEY_DERIVATION
CKM_SHA512_KEY_GEN
The SHA-512 mechanism, denoted CKM_SHA512, is a
mechanism for message digesting, following the Secure Hash Algorithm with a
512-bit message digest defined in FIPS PUB 180-2.
It does not have a parameter.
Constraints on the length of input and output data are
summarized in the following table. For single-part digesting, the data and the
digest may begin at the same location in memory.
Table 147, SHA-512: Data Length
|
Function
|
Input length
|
Digest length
|
|
C_Digest
|
any
|
64
|
The general-length SHA-512-HMAC mechanism, denoted CKM_SHA512_HMAC_GENERAL,
is the same as the general-length SHA-1-HMAC mechanism in Section 6.20.3, except that it uses the HMAC construction based on the SHA-512 hash function
and length of the output should be in the range 1-64.
The keys it uses are generic
secret keys and CKK_SHA512_HMAC. FIPS-198 compliant tokens may require the key
length to be at least 32 bytes; that is, half the size of the SHA-512 hash
output.
It has a parameter, a CK_MAC_GENERAL_PARAMS, which holds the length in bytes of the desired output.
This length should be in the range 0-64 (the output size of SHA-512 is 64 bytes).
FIPS-198 compliant tokens may constrain the output length to be at least 4 or
32 (half the maximum length). Signatures (MACs) produced by this mechanism will
be taken from the start of the full 64-byte HMAC output.
Table 148, General-length
SHA-384-HMAC: Key And Data Length
|
Function
|
Key type
|
Data
length
|
Signature
length
|
|
C_Sign
|
generic
secret, CKK_SHA512_HMAC
|
Any
|
1-64,
depending on parameters
|
|
C_Verify
|
generic
secret,
CKK_SHA512_HMAC
|
Any
|
1-64,
depending on parameters
|
The SHA-512-HMAC mechanism, denoted CKM_SHA512_HMAC,
is a special case of the general-length SHA-512-HMAC mechanism.
It has no parameter, and always produces an output of length
64.
SHA-512 key derivation, denoted CKM_SHA512_KEY_DERIVATION,
is the same as the SHA-1 key derivation mechanism in Section 6.20.5,
except that it uses the SHA-512 hash function and the relevant length is 64
bytes.
The SHA-512-HMAC key generation
mechanism, denoted CKM_SHA512_KEY_GEN, is a key generation mechanism for
NIST’s SHA512-HMAC.
It does not have a parameter.
The mechanism generates
SHA512-HMAC keys with a particular length in bytes, as specified in the CKA_VALUE_LEN
attribute of the template for the key.
The mechanism contributes the CKA_CLASS,
CKA_KEY_TYPE, and CKA_VALUE attributes to the new key. Other
attributes supported by the SHA512-HMAC key type (specifically, the flags
indicating which functions the key supports) may be specified in the template
for the key, or else are assigned default initial values.
For this mechanism, the ulMinKeySize
and ulMaxKeySize fields of the CK_MECHANISM_INFO structure
specify the supported range of CKM_SHA512_HMAC key sizes, in bytes.
Table
149, SHA-512/224 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_SHA512_224
|
|
|
|
ü
|
|
|
|
|
CKM_SHA512_224_HMAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_SHA512_224_HMAC
|
|
ü
|
|
|
|
|
|
|
CKM_SHA512_224_KEY_DERIVATION
|
|
|
|
|
|
|
ü
|
|
CKM_SHA512_224_KEY_GEN
|
|
|
|
|
ü
|
|
|
This section defines the key type
“CKK_SHA512_224_HMAC” for type CK_KEY_TYPE as used in the CKA_KEY_TYPE
attribute of key objects.
Mechanisms:
CKM_SHA512_224
CKM_SHA512_224_HMAC
CKM_SHA512_224_HMAC_GENERAL
CKM_SHA512_224_KEY_DERIVATION
CKM_SHA512_224_KEY_GEN
The SHA-512/224 mechanism, denoted CKM_SHA512_224, is
a mechanism for message digesting, following the Secure Hash Algorithm defined
in FIPS PUB 180-4, section 5.3.6. It is based on a 512-bit message digest with
a distinct initial hash value and truncated to 224 bits. CKM_SHA512_224
is the same as CKM_SHA512_T with a parameter value of 224.
It does not have a parameter.
Constraints on the length of input and output data are
summarized in the following table. For single-part digesting, the data and the
digest may begin at the same location in memory.
Table 150, SHA-512/224: Data
Length
|
Function
|
Input length
|
Digest length
|
|
C_Digest
|
any
|
28
|
The general-length SHA-512/224-HMAC mechanism, denoted CKM_SHA512_224_HMAC_GENERAL,
is the same as the general-length SHA-1-HMAC mechanism in Section 6.20.3, except that it uses the HMAC construction based on the SHA-512/224 hash
function and length of the output should be in the range 1-28. The keys it uses are generic secret keys and
CKK_SHA512_224_HMAC. FIPS-198 compliant tokens may require the key length to
be at least 14 bytes; that is, half the size of the SHA-512/224 hash output.
It has a parameter, a CK_MAC_GENERAL_PARAMS, which holds the length in bytes of the desired output.
This length should be in the range 0-28 (the output size of SHA-512/224 is 28
bytes). FIPS-198 compliant tokens may constrain the output length to be at
least 4 or 14 (half the maximum length). Signatures (MACs) produced by this
mechanism will be taken from the start of the full 28-byte HMAC output.
Table 151, General-length
SHA-384-HMAC: Key And Data Length
|
Function
|
Key type
|
Data
length
|
Signature
length
|
|
C_Sign
|
generic
secret, CKK_SHA512_224_HMAC
|
Any
|
1-28,
depending on parameters
|
|
C_Verify
|
generic
secret,
CKK_SHA512_224_HMAC
|
Any
|
1-28,
depending on parameters
|
The SHA-512-HMAC mechanism, denoted CKM_SHA512_224_HMAC,
is a special case of the general-length SHA-512/224-HMAC mechanism.
It has no parameter, and always produces an output of length
28.
The SHA-512/224 key derivation, denoted CKM_SHA512_224_KEY_DERIVATION,
is the same as the SHA-512 key derivation mechanism in section 6.24.5, except that it uses the SHA-512/224 hash function and the relevant length is
28 bytes.
The SHA-512/224-HMAC key
generation mechanism, denoted CKM_SHA512_224_KEY_GEN, is a key
generation mechanism for NIST’s SHA512/224-HMAC.
It does not have a parameter.
The mechanism generates
SHA512/224-HMAC keys with a particular length in bytes, as specified in the CKA_VALUE_LEN
attribute of the template for the key.
The mechanism contributes the CKA_CLASS,
CKA_KEY_TYPE, and CKA_VALUE attributes to the new key. Other
attributes supported by the SHA512/224-HMAC key type (specifically, the flags
indicating which functions the key supports) may be specified in the template
for the key, or else are assigned default initial values.
For this mechanism, the ulMinKeySize
and ulMaxKeySize fields of the CK_MECHANISM_INFO structure
specify the supported range of CKM_SHA512_224_HMAC key sizes, in bytes.
Table
152, SHA-512/256 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_SHA512_256
|
|
|
|
ü
|
|
|
|
|
CKM_SHA512_256_HMAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_SHA512_256_HMAC
|
|
ü
|
|
|
|
|
|
|
CKM_SHA512_256_KEY_DERIVATION
|
|
|
|
|
|
|
ü
|
|
CKM_SHA512_256_KEY_GEN
|
|
|
|
|
ü
|
|
|
This section defines the key type
“CKK_SHA512_256_HMAC” for type CK_KEY_TYPE as used in the CKA_KEY_TYPE
attribute of key objects.
Mechanisms:
CKM_SHA512_256
CKM_SHA512_256_HMAC
CKM_SHA512_256_HMAC_GENERAL
CKM_SHA512_256_KEY_DERIVATION
CKM_SHA512_256_KEY_GEN
The SHA-512/256 mechanism, denoted CKM_SHA512_256, is
a mechanism for message digesting, following the Secure Hash Algorithm defined
in FIPS PUB 180-4, section 5.3.6. It is based on a 512-bit message digest with
a distinct initial hash value and truncated to 256 bits. CKM_SHA512_256
is the same as CKM_SHA512_T with a parameter value of 256.
It does not have a parameter.
Constraints on the length of input and output data are
summarized in the following table. For single-part digesting, the data and the
digest may begin at the same location in memory.
Table 153, SHA-512/256: Data
Length
|
Function
|
Input length
|
Digest length
|
|
C_Digest
|
any
|
32
|
The general-length SHA-512/256-HMAC mechanism, denoted CKM_SHA512_256_HMAC_GENERAL,
is the same as the general-length SHA-1-HMAC mechanism in Section 6.20.3, except that it uses the HMAC construction based on the SHA-512/256 hash
function and length of the output should be in the range 1-32. The keys it uses are generic secret keys and
CKK_SHA512_256_HMAC. FIPS-198 compliant tokens may require the key length to
be at least 16 bytes; that is, half the size of the SHA-512/256 hash output.
It has a parameter, a CK_MAC_GENERAL_PARAMS, which holds the length in bytes of the desired output.
This length should be in the range 1-32 (the output size of SHA-512/256 is 32
bytes). FIPS-198 compliant tokens may constrain the output length to be at
least 4 or 16 (half the maximum length). Signatures (MACs) produced by this
mechanism will be taken from the start of the full 32-byte HMAC output.
Table 154, General-length
SHA-384-HMAC: Key And Data Length
|
Function
|
Key type
|
Data
length
|
Signature
length
|
|
C_Sign
|
generic
secret, CKK_SHA512_256_HMAC
|
Any
|
1-32,
depending on parameters
|
|
C_Verify
|
generic
secret,
CKK_SHA512_256_HMAC
|
Any
|
1-32,
depending on parameters
|
The SHA-512-HMAC mechanism, denoted CKM_SHA512_256_HMAC,
is a special case of the general-length SHA-512/256-HMAC mechanism.
It has no parameter, and always produces an output of length
32.
The SHA-512/256 key derivation, denoted CKM_SHA512_256_KEY_DERIVATION,
is the same as the SHA-512 key derivation mechanism in section 6.24.5, except that it uses the SHA-512/256 hash function and the relevant length is
32 bytes.
The SHA-512/256-HMAC key
generation mechanism, denoted CKM_SHA512_256_KEY_GEN, is a key
generation mechanism for NIST’s SHA512/256-HMAC.
It does not have a parameter.
The mechanism generates
SHA512/256-HMAC keys with a particular length in bytes, as specified in the CKA_VALUE_LEN
attribute of the template for the key.
The mechanism contributes the CKA_CLASS,
CKA_KEY_TYPE, and CKA_VALUE attributes to the new key. Other
attributes supported by the SHA512/256-HMAC key type (specifically, the flags
indicating which functions the key supports) may be specified in the template
for the key, or else are assigned default initial values.
For this mechanism, the ulMinKeySize
and ulMaxKeySize fields of the CK_MECHANISM_INFO structure
specify the supported range of CKM_SHA512_256_HMAC key sizes, in bytes.
Table
155, SHA-512 / t Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_SHA512_T
|
|
|
|
ü
|
|
|
|
|
CKM_SHA512_T_HMAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_SHA512_T_HMAC
|
|
ü
|
|
|
|
|
|
|
CKM_SHA512_T_KEY_DERIVATION
|
|
|
|
|
|
|
ü
|
|
CKM_SHA512_T_KEY_GEN
|
|
|
|
|
ü
|
|
|
This section defines the key type
“CKK_SHA512_T_HMAC” for type CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute
of key objects.
Mechanisms:
CKM_SHA512_T
CKM_SHA512_T_HMAC
CKM_SHA512_T_HMAC_GENERAL
CKM_SHA512_T_KEY_DERIVATION
CKM_SHA512_T_KEY_GEN
The SHA-512/t mechanism, denoted CKM_SHA512_T, is a
mechanism for message digesting, following the Secure Hash Algorithm defined in
FIPS PUB 180-4, section 5.3.6. It is based on a 512-bit message digest with a
distinct initial hash value and truncated to t bits.
It has a parameter, a CK_MAC_GENERAL_PARAMS, which
holds the value of t in bits. The length in bytes of the desired output should
be in the range of 0-⌈
t/8⌉,
where 0 < t < 512, and t <> 384.
Constraints on the length of input and output data are
summarized in the following table. For single-part digesting, the data and the
digest may begin at the same location in memory.
Table 156, SHA-512/256: Data
Length
|
Function
|
Input length
|
Digest length
|
|
C_Digest
|
any
|
⌈t/8⌉, where 0 < t <
512, and t <> 384
|
The general-length SHA-512/t-HMAC mechanism, denoted CKM_SHA512_T_HMAC_GENERAL,
is the same as the general-length SHA-1-HMAC mechanism in Section 6.20.3, except that it uses the HMAC construction based on the SHA-512/t hash
function and length of the output should be in the range 0 – ⌈t/8⌉, where 0 <
t < 512, and t <> 384.
The SHA-512/t-HMAC mechanism, denoted CKM_SHA512_T_HMAC,
is a special case of the general-length SHA-512/t-HMAC mechanism.
It has a parameter, a CK_MAC_GENERAL_PARAMS, which
holds the value of t in bits. The length in bytes of the desired output should
be in the range of 0-⌈t/8⌉, where 0 <
t < 512, and t <> 384.
The SHA-512/t key derivation, denoted CKM_SHA512_T_KEY_DERIVATION,
is the same as the SHA-512 key derivation mechanism in section 6.24.5, except that it uses the SHA-512/t hash function and the relevant length is ⌈t/8⌉ bytes, where 0
< t < 512, and t <> 384.
The SHA-512/t-HMAC key generation
mechanism, denoted CKM_SHA512_T_KEY_GEN, is a key generation mechanism
for NIST’s SHA512/t-HMAC.
It does not have a parameter.
The mechanism generates
SHA512/t-HMAC keys with a particular length in bytes, as specified in the CKA_VALUE_LEN
attribute of the template for the key.
The mechanism contributes the CKA_CLASS,
CKA_KEY_TYPE, and CKA_VALUE attributes to the new key. Other
attributes supported by the SHA512/t-HMAC key type (specifically, the flags
indicating which functions the key supports) may be specified in the template
for the key, or else are assigned default initial values.
For this mechanism, the ulMinKeySize
and ulMaxKeySize fields of the CK_MECHANISM_INFO structure
specify the supported range of CKM_SHA512_T_HMAC key sizes, in bytes.
Table
157, SHA3-224
Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_SHA3_224
|
|
|
|
ü
|
|
|
|
|
CKM_SHA3_224_HMAC
|
|
ü
|
|
|
|
|
|
|
CKM_SHA3_224_HMAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_SHA3_224_KEY_DERIVATION
|
|
|
|
|
|
|
ü
|
|
CKM_SHA3_224_KEY_GEN
|
|
|
|
|
ü
|
|
|
Mechanisms:
CKM_SHA3_224
CKM_SHA3_224_HMAC
CKM_SHA3_224_HMAC_GENERAL
CKM_SHA3_224_KEY_DERIVATION
CKM_SHA3_224_KEY_GEN
CKK_SHA3_224_HMAC
The SHA3-224 mechanism, denoted CKM_SHA3_224, is a
mechanism for message digesting, following the Secure Hash 3 Algorithm with a
224-bit message digest defined in FIPS Pub 202.
It does not have a parameter.
Constraints on the length of input and output data are
summarized in the following table. For single-part digesting, the data and the
digest may begin at the same location in memory.
Table 158,
SHA3-224: Data Length
|
Function
|
Input length
|
Digest length
|
|
C_Digest
|
any
|
28
|
The general-length SHA3-224-HMAC mechanism, denoted CKM_SHA3_224_HMAC_GENERAL,
is the same as the general-length SHA-1-HMAC mechanism in section 6.20.4 except that it uses the HMAC construction based on the SHA3-224 hash function
and length of the output should be in the range 1-28. The keys it uses are
generic secret keys and CKK_SHA3_224_HMAC. FIPS-198 compliant tokens may
require the key length to be at least 14 bytes; that is, half the size of the
SHA3-224 hash output.
It has a parameter, a CK_MAC_GENERAL_PARAMS, which
holds the length in bytes of the desired output. This length should be in the
range 1-28 (the output size of SHA3-224 is 28 bytes). FIPS-198 compliant tokens
may constrain the output length to be at least 4 or 14 (half the maximum
length). Signatures (MACs) produced by this mechanism shall be taken from the
start of the full 28-byte HMAC output.
Table 159,
General-length SHA3-224-HMAC: Key And Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
generic secret or
CKK_SHA3_224_HMAC
|
Any
|
1-28, depending on
parameters
|
|
C_Verify
|
generic secret or
CKK_SHA3_224_HMAC
|
Any
|
1-28, depending on parameters
|
The SHA3-224-HMAC mechanism, denoted CKM_SHA3_224_HMAC,
is a special case of the general-length SHA3-224-HMAC mechanism.
It has no parameter, and always produces an output of length
28.
SHA-224 key derivation, denoted CKM_SHA3_224_KEY_DERIVATION,
is the same as the SHA-1 key derivation mechanism in Section 6.20.5
except that it uses the SHA3-224 hash function and the relevant length is 28
bytes.
The SHA3-224-HMAC key generation mechanism, denoted CKM_SHA3_224_KEY_GEN,
is a key generation mechanism for NIST’s SHA3-224-HMAC.
It does not have a parameter.
The mechanism generates SHA3-224-HMAC keys with a particular
length in bytes, as specified in the CKA_VALUE_LEN attribute of the
template for the key.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key. Other attributes supported by
the SHA3-224-HMAC key type (specifically, the flags indicating which functions
the key supports) may be specified in the template for the key, or else are
assigned default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
CKM_SHA3_224_HMAC key sizes, in bytes.
Table
160,
SHA3-256 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_SHA3_256
|
|
|
|
ü
|
|
|
|
|
CKM_SHA3_256_HMAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_SHA3_256_HMAC
|
|
ü
|
|
|
|
|
|
|
CKM_SHA3_256_KEY_DERIVATION
|
|
|
|
|
|
|
ü
|
|
CKM_SHA3_256_KEY_GEN
|
|
|
|
|
ü
|
|
|
Mechanisms:
CKM_SHA3_256
CKM_SHA3_256_HMAC
CKM_SHA3_256_HMAC_GENERAL
CKM_SHA3_256_KEY_DERIVATION
CKM_SHA3_256_KEY_GEN
CKK_SHA3_256_HMAC
The SHA3-256 mechanism, denoted CKM_SHA3_256, is a
mechanism for message digesting, following the Secure Hash 3 Algorithm with a
256-bit message digest defined in FIPS PUB 202.
It does not have a parameter.
Constraints on the length of input and output data are
summarized in the following table. For single-part digesting, the data and the
digest may begin at the same location in memory.
Table 161,
SHA3-256: Data Length
|
Function
|
Input length
|
Digest length
|
|
C_Digest
|
any
|
32
|
The general-length SHA3-256-HMAC mechanism, denoted CKM_SHA3_256_HMAC_GENERAL,
is the same as the general-length SHA-1-HMAC mechanism in Section 6.20.4, except that it uses the HMAC construction based on the SHA3-256 hash function
and length of the output should be in the range 1-32. The keys it uses are
generic secret keys and CKK_SHA3_256_HMAC. FIPS-198 compliant tokens may
require the key length to be at least 16 bytes; that is, half the size of the
SHA3-256 hash output.
It has a parameter, a CK_MAC_GENERAL_PARAMS, which
holds the length in bytes of the desired output. This length should be in the
range 1-32 (the output size of SHA3-256 is 32 bytes). FIPS-198 compliant tokens
may constrain the output length to be at least 4 or 16 (half the maximum
length). Signatures (MACs) produced by this mechanism shall be taken from the
start of the full 32-byte HMAC output.
Table 162,
General-length SHA3-256-HMAC: Key And Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
generic secret or
CKK_SHA3_256_HMAC
|
Any
|
1-32, depending on
parameters
|
|
C_Verify
|
generic secret or
CKK_SHA3_256_HMAC
|
Any
|
1-32, depending on
parameters
|
The SHA-256-HMAC mechanism, denoted CKM_SHA3_256_HMAC,
is a special case of the general-length SHA-256-HMAC mechanism.
It has no parameter, and always produces an output of length
32.
SHA-256 key derivation, denoted CKM_SHA3_256_KEY_DERIVATION,
is the same as the SHA-1 key derivation mechanism in Section 6.20.5,
except that it uses the SHA3-256 hash function and the relevant length is 32
bytes.
The SHA3-256-HMAC key generation mechanism, denoted CKM_SHA3_256_KEY_GEN,
is a key generation mechanism for NIST’s SHA3-256-HMAC.
It does not have a parameter.
The mechanism generates SHA3-256-HMAC keys with a particular
length in bytes, as specified in the CKA_VALUE_LEN attribute of the
template for the key.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key. Other attributes supported by
the SHA3-256-HMAC key type (specifically, the flags indicating which functions
the key supports) may be specified in the template for the key, or else are
assigned default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
CKM_SHA3_256_HMAC key sizes, in bytes.
Table
163,
SHA3-384 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_SHA3_384
|
|
|
|
ü
|
|
|
|
|
CKM_SHA3_384_HMAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_SHA3_384_HMAC
|
|
ü
|
|
|
|
|
|
|
CKM_SHA3_384_KEY_DERIVATION
|
|
|
|
|
|
|
ü
|
|
CKM_SHA3_384_KEY_GEN
|
|
|
|
ü
|
|
|
|
CKM_SHA3_384
CKM_SHA3_384_HMAC
CKM_SHA3_384_HMAC_GENERAL
CKM_SHA3_384_KEY_DERIVATION
CKM_SHA3_384_KEY_GEN
CKK_SHA3_384_HMAC
The SHA3-384 mechanism, denoted CKM_SHA3_384, is a mechanism
for message digesting, following the Secure Hash 3 Algorithm with a 384-bit
message digest defined in FIPS PUB 202.
It does not have a parameter.
Constraints on the length of input and output data are
summarized in the following table. For single-part digesting, the data and the
digest may begin at the same location in memory.
Table 164,
SHA3-384: Data Length
|
Function
|
Input length
|
Digest length
|
|
C_Digest
|
any
|
48
|
The general-length SHA3-384-HMAC mechanism, denoted CKM_SHA3_384_HMAC_GENERAL,
is the same as the general-length SHA-1-HMAC mechanism in Section 6.20.4, except that it uses the HMAC construction based on the SHA-384 hash function
and length of the output should be in the range 1-48.The keys it uses are
generic secret keys and CKK_SHA3_384_HMAC. FIPS-198 compliant tokens may
require the key length to be at least 24 bytes; that is, half the size of the
SHA3-384 hash output.
It has a parameter, a CK_MAC_GENERAL_PARAMS, which
holds the length in bytes of the desired output. This length should be in the
range 1-48 (the output size of SHA3-384 is 48 bytes). FIPS-198 compliant tokens
may constrain the output length to be at least 4 or 24 (half the maximum
length). Signatures (MACs) produced by this mechanism shall be taken from the
start of the full 48-byte HMAC output.
Table 165,
General-length SHA3-384-HMAC: Key And Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
generic secret or
CKK_SHA3_384_HMAC
|
Any
|
1-48, depending on
parameters
|
|
C_Verify
|
generic secret or
CKK_SHA3_384_HMAC
|
Any
|
1-48, depending on parameters
|
The SHA3-384-HMAC mechanism, denoted CKM_SHA3_384_HMAC,
is a special case of the general-length SHA3-384-HMAC mechanism.
It has no parameter, and always produces an output of length
48.
SHA3-384 key derivation, denoted CKM_SHA3_384_KEY_DERIVATION,
is the same as the SHA-1 key derivation mechanism in Section 6.20.5,
except that it uses the SHA-384 hash function and the relevant length is 48
bytes.
The SHA3-384-HMAC key generation mechanism, denoted CKM_SHA3_384_KEY_GEN,
is a key generation mechanism for NIST’s SHA3-384-HMAC.
It does not have a parameter.
The mechanism generates SHA3-384-HMAC keys with a particular
length in bytes, as specified in the CKA_VALUE_LEN attribute of the
template for the key.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key. Other attributes supported by
the SHA3-384-HMAC key type (specifically, the flags indicating which functions
the key supports) may be specified in the template for the key, or else are
assigned default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
CKM_SHA3_384_HMAC key sizes, in bytes.
Table
166,
SHA-512 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_SHA3_512
|
|
|
|
ü
|
|
|
|
|
CKM_SHA3_512_HMAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_SHA3_512_HMAC
|
|
ü
|
|
|
|
|
|
|
CKM_SHA3_512_KEY_DERIVATION
|
|
|
|
|
|
|
ü
|
|
CKM_SHA3_512_KEY_GEN
|
|
|
|
ü
|
|
|
|
CKM_SHA3_512
CKM_SHA3_512_HMAC
CKM_SHA3_512_HMAC_GENERAL
CKM_SHA3_512_KEY_DERIVATION
CKM_SHA3_512_KEY_GEN
CKK_SHA3_512_HMAC
The SHA3-512 mechanism, denoted CKM_SHA3_512, is a mechanism
for message digesting, following the Secure Hash 3 Algorithm with a 512-bit
message digest defined in FIPS PUB 202.
It does not have a parameter.
Constraints on the length of input and output data are
summarized in the following table. For single-part digesting, the data and the
digest may begin at the same location in memory.
Table 167,
SHA3-512: Data Length
|
Function
|
Input length
|
Digest length
|
|
C_Digest
|
any
|
64
|
The general-length SHA3-512-HMAC mechanism, denoted CKM_SHA3_512_HMAC_GENERAL,
is the same as the general-length SHA-1-HMAC mechanism in Section 6.20.4, except that it uses the HMAC construction based on the SHA3-512 hash function
and length of the output should be in the range 1-64.The keys it uses are
generic secret keys and CKK_SHA3_512_HMAC. FIPS-198 compliant tokens may
require the key length to be at least 32 bytes; that is, half the size of the
SHA3-512 hash output.
It has a parameter, a CK_MAC_GENERAL_PARAMS, which
holds the length in bytes of the desired output. This length should be in the
range 1-64 (the output size of SHA3-512 is 64 bytes). FIPS-198 compliant tokens
may constrain the output length to be at least 4 or 32 (half the maximum
length). Signatures (MACs) produced by this mechanism shall be taken from the
start of the full 64-byte HMAC output.
Table 168,
General-length SHA3-512-HMAC: Key And Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
generic secret or
CKK_SHA3_512_HMAC
|
Any
|
1-64, depending on
parameters
|
|
C_Verify
|
generic secret or
CKK_SHA3_512_HMAC
|
Any
|
1-64, depending on parameters
|
The SHA3-512-HMAC mechanism, denoted CKM_SHA3_512_HMAC,
is a special case of the general-length SHA3-512-HMAC mechanism.
It has no parameter, and always produces an output of length
64.
SHA3-512 key derivation, denoted CKM_SHA3_512_KEY_DERIVATION,
is the same as the SHA-1 key derivation mechanism in Section 6.20.5,
except that it uses the SHA-512 hash function and the relevant length is 64
bytes.
The SHA3-512-HMAC key generation mechanism, denoted CKM_SHA3_512_KEY_GEN,
is a key generation mechanism for NIST’s SHA3-512-HMAC.
It does not have a parameter.
The mechanism generates SHA3-512-HMAC keys with a particular
length in bytes, as specified in the CKA_VALUE_LEN attribute of the
template for the key.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key. Other attributes supported by
the SHA3-512-HMAC key type (specifically, the flags indicating which functions
the key supports) may be specified in the template for the key, or else are
assigned default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
CKM_SHA3_512_HMAC key sizes, in bytes.
Table
169,
SHA-512 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_SHAKE_128_KEY_DERIVATION
|
|
|
|
|
|
|
ü
|
|
CKM_SHAKE_256_KEY_DERIVATION
|
|
|
|
|
|
|
ü
|
CKM_SHAKE_128_KEY_DERIVATION
CKM_SHAKE_256_KEY_DERIVATION
SHAKE-128 and SHAKE-256 key derivation, denoted CKM_SHAKE_128_KEY_DERIVATION
and CKM_SHAKE_256_KEY_DERIVATION, implements the SHAKE expansion
function defined in FIPS 202 on the input key.
·
If no length or key type is provided in the template a CKR_TEMPLATE_INCOMPLETE
error is generated.
·
If no key type is provided in the template, but a length is, then
the key produced by this mechanism shall be a generic secret key of the
specified length.
·
If no length was provided in the template, but a key type is,
then that key type must have a well-defined length. If it does, then the key
produced by this mechanism shall be of the type specified in the template. If
it doesn’t, an error shall be returned.
·
If both a key type and a length are provided in the template, the
length must be compatible with that key type. The key produced by this
mechanism shall be of the specified type and length.
If a DES, DES2, or CDMF key is derived with this mechanism,
the parity bits of the key shall be set properly.
This mechanism has the following rules about key sensitivity
and extractability:
·
The CKA_SENSITIVE and CKA_EXTRACTABLE attributes in
the template for the new key can both be specified to be either CK_TRUE or CK_FALSE.
If omitted, these attributes each take on some default value.
·
If the base key has its CKA_ALWAYS_SENSITIVE attribute set
to CK_FALSE, then the derived key shall as well. If the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE, then the derived key has its CKA_ALWAYS_SENSITIVE
attribute set to the same value as its CKA_SENSITIVE attribute.
·
Similarly, if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_FALSE, then the derived key shall, too. If the base key
has its CKA_NEVER_EXTRACTABLE attribute set to CK_TRUE, then the derived
key has its CKA_NEVER_EXTRACTABLE attribute set to the opposite
value from its CKA_EXTRACTABLE attribute.
Table
170,
BLAKE2B-160 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_BLAKE2B_160
|
|
|
|
ü
|
|
|
|
|
CKM_BLAKE2B_160_HMAC
|
|
ü
|
|
|
|
|
|
|
CKM_BLAKE2B_160_HMAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_BLAKE2B_160_KEY_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_BLAKE2B_160_KEY_GEN
|
|
|
|
|
ü
|
|
|
Mechanisms:
CKM_BLAKE2B_160
CKM_BLAKE2B_160_HMAC
CKM_BLAKE2B_160_HMAC_GENERAL
CKM_BLAKE2B_160_KEY_DERIVE
CKM_BLAKE2B_160_KEY_GEN
CKK_BLAKE2B_160_HMAC
The BLAKE2B-160 mechanism, denoted CKM_BLAKE2B_160,
is a mechanism for message digesting, following the Blake2b Algorithm with a
160-bit message digest without a key as defined in RFC 7693.
It does not have a parameter.
Constraints on the length of input and output data are
summarized in the following table. For single-part digesting, the data and the
digest may begin at the same location in memory.
Table 171,
BLAKE2B-160: Data Length
|
Function
|
Input length
|
Digest length
|
|
C_Digest
|
any
|
20
|
The general-length BLAKE2B-160-HMAC mechanism, denoted CKM_BLAKE2B_160_HMAC_GENERAL,
is the keyed variant of BLAKE2b-160 and length of the output should be in the
range 1-20. The keys it uses are generic secret keys and CKK_BLAKE2B_160_HMAC.
It has a parameter, a CK_MAC_GENERAL_PARAMS, which
holds the length in bytes of the desired output. This length should be in the
range 1-20 (the output size of BLAKE2B-160 is 20 bytes). Signatures (MACs)
produced by this mechanism shall be taken from the start of the full 20-byte
HMAC output.
Table 172,
General-length BLAKE2B-160-HMAC: Key And Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
generic secret or
CKK_BLAKE2B_160_HMAC
|
Any
|
1-20, depending on
parameters
|
|
C_Verify
|
generic secret or
CKK_BLAKE2B_160_HMAC
|
Any
|
1-20, depending on
parameters
|
The BLAKE2B-160-HMAC mechanism, denoted CKM_BLAKE2B_160_HMAC,
is a special case of the general-length BLAKE2B-160-HMAC mechanism.
It has no parameter, and always produces an output of length
20.
BLAKE2B-160 key derivation, denoted CKM_BLAKE2B_160_KEY_DERIVE,
is the same as the SHA-1 key derivation mechanism in Section 6.20.5
except that it uses the BLAKE2B-160 hash function and the relevant length is 20
bytes.
The BLAKE2B-160-HMAC key generation mechanism, denoted CKM_BLAKE2B_160_KEY_GEN,
is a key generation mechanism for BLAKE2B-160-HMAC.
It does not have a parameter.
The mechanism generates BLAKE2B-160-HMAC keys with a
particular length in bytes, as specified in the CKA_VALUE_LEN attribute
of the template for the key.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key. Other attributes supported by
the BLAKE2B-160-HMAC key type (specifically, the flags indicating which
functions the key supports) may be specified in the template for the key, or
else are assigned default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
CKM_BLAKE2B_160_HMAC key sizes, in bytes.
Table
173,
BLAKE2B-256 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_BLAKE2B_256
|
|
|
|
ü
|
|
|
|
|
CKM_BLAKE2B_256_HMAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_BLAKE2B_256_HMAC
|
|
ü
|
|
|
|
|
|
|
CKM_BLAKE2B_256_KEY_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_BLAKE2B_256_KEY_GEN
|
|
|
|
|
ü
|
|
|
Mechanisms:
CKM_BLAKE2B_256
CKM_BLAKE2B_256_HMAC
CKM_BLAKE2B_256_HMAC_GENERAL
CKM_BLAKE2B_256_KEY_DERIVE
CKM_BLAKE2B_256_KEY_GEN
CKK_BLAKE2B_256_HMAC
The BLAKE2B-256 mechanism, denoted CKM_BLAKE2B_256,
is a mechanism for message digesting, following the Blake2b Algorithm with a
256-bit message digest without a key as defined in RFC 7693.
It does not have a parameter.
Constraints on the length of input and output data are
summarized in the following table. For single-part digesting, the data and the
digest may begin at the same location in memory.
Table 174,
BLAKE2B-256: Data Length
|
Function
|
Input length
|
Digest length
|
|
C_Digest
|
any
|
32
|
The general-length BLAKE2B-256-HMAC mechanism, denoted CKM_BLAKE2B_256_HMAC_GENERAL,
is the keyed variant of Blake2b-256 and length of the output should be in the
range 1-32. The keys it uses are generic secret keys and CKK_BLAKE2B_256_HMAC.
It has a parameter, a CK_MAC_GENERAL_PARAMS, which
holds the length in bytes of the desired output. This length should be in the
range 1-32 (the output size of BLAKE2B-256 is 32 bytes). Signatures (MACs)
produced by this mechanism shall be taken from the start of the full 32-byte
HMAC output.
Table 175,
General-length BLAKE2B-256-HMAC: Key And Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
generic secret or
CKK_BLAKE2B_256_HMAC
|
Any
|
1-32, depending on
parameters
|
|
C_Verify
|
generic secret or
CKK_BLAKE2B_256_HMAC
|
Any
|
1-32, depending on
parameters
|
The BLAKE2B-256-HMAC mechanism, denoted CKM_BLAKE2B_256_HMAC,
is a special case of the general-length BLAKE2B-256-HMAC mechanism in Section6.34.3.
It has no parameter, and always produces an output of length
32.
BLAKE2B-256 key derivation, denoted CKM_BLAKE2B_256_KEY_DERIVE,
is the same as the SHA-1 key derivation mechanism in Section 6.20.5,
except that it uses the BLAKE2B-256 hash function and the relevant length is 32
bytes.
The BLAKE2B-256-HMAC key generation mechanism, denoted CKM_BLAKE2B_256_KEY_GEN,
is a key generation mechanism for BLAKE2B-256-HMAC.
It does not have a parameter.
The mechanism generates BLAKE2B-256-HMAC keys with a
particular length in bytes, as specified in the CKA_VALUE_LEN attribute
of the template for the key.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key. Other attributes supported by
the BLAKE2B-256-HMAC key type (specifically, the flags indicating which
functions the key supports) may be specified in the template for the key, or
else are assigned default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
CKM_BLAKE2B_256_HMAC key sizes, in bytes.
Table
176,
BLAKE2B-384 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_BLAKE2B_384
|
|
|
|
ü
|
|
|
|
|
CKM_BLAKE2B_384_HMAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_BLAKE2B_384_HMAC
|
|
ü
|
|
|
|
|
|
|
CKM_BLAKE2B_384_KEY_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_BLAKE2B_384_KEY_GEN
|
|
|
|
ü
|
|
|
|
CKM_BLAKE2B_384
CKM_BLAKE2B_384_HMAC
CKM_BLAKE2B_384_HMAC_GENERAL
CKM_BLAKE2B_384_KEY_DERIVE
CKM_BLAKE2B_384_KEY_GEN
CKK_BLAKE2B_384_HMAC
The BLAKE2B-384 mechanism, denoted CKM_BLAKE2B_384,
is a mechanism for message digesting, following the Blake2b Algorithm with a
384-bit message digest without a key as defined in RFC 7693.
It does not have a parameter.
Constraints on the length of input and output data are
summarized in the following table. For single-part digesting, the data and the
digest may begin at the same location in memory.
Table 177,
BLAKE2B-384: Data Length
|
Function
|
Input length
|
Digest length
|
|
C_Digest
|
any
|
48
|
The general-length BLAKE2B-384-HMAC mechanism, denoted CKM_BLAKE2B_384_HMAC_GENERAL,
is the keyed variant of the BLAKE2B-384 hash function and length of the output
should be in the range 1-48.The keys it uses are generic secret keys and
CKK_BLAKE2B_384_HMAC.
It has a parameter, a CK_MAC_GENERAL_PARAMS,
which holds the length in bytes of the desired output. This length should be in
the range 1-48 (the output size of BLAKE2B-384 is 48 bytes). Signatures (MACs)
produced by this mechanism shall be taken from the start of the full 48-byte
HMAC output.
Table 178,
General-length BLAKE2B-384-HMAC: Key And Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
generic secret or
CKK_BLAKE2B_384_HMAC
|
Any
|
1-48, depending on
parameters
|
|
C_Verify
|
generic secret or
CKK_BLAKE2B_384_HMAC
|
Any
|
1-48, depending on parameters
|
The BLAKE2B-384-HMAC mechanism, denoted CKM_BLAKE2B_384_HMAC,
is a special case of the general-length BLAKE2B-384-HMAC mechanism.
It has no parameter, and always produces an output of length
48.
BLAKE2B-384 key derivation, denoted CKM_BLAKE2B_384_KEY_DERIVE,
is the same as the SHA-1 key derivation mechanism in Section 6.20.5,
except that it uses the BLAKE2B-384 hash function and the relevant length is 48
bytes.
The BLAKE2B-384-HMAC key generation mechanism, denoted CKM_BLAKE2B_384_KEY_GEN,
is a key generation mechanism for NIST’s BLAKE2B-384-HMAC.
It does not have a parameter.
The mechanism generates BLAKE2B-384-HMAC keys with a
particular length in bytes, as specified in the CKA_VALUE_LEN attribute
of the template for the key.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key. Other attributes supported by
the BLAKE2B-384-HMAC key type (specifically, the flags indicating which
functions the key supports) may be specified in the template for the key, or
else are assigned default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
CKM_BLAKE2B_384_HMAC key sizes, in bytes.
Table
179,
SHA-512 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_BLAKE2B_512
|
|
|
|
ü
|
|
|
|
|
CKM_BLAKE2B_512_HMAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_BLAKE2B_512_HMAC
|
|
ü
|
|
|
|
|
|
|
CKM_BLAKE2B_512_KEY_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_BLAKE2B_512_KEY_GEN
|
|
|
|
ü
|
|
|
|
CKM_BLAKE2B_512
CKM_BLAKE2B_512_HMAC
CKM_BLAKE2B_512_HMAC_GENERAL
CKM_BLAKE2B_512_KEY_DERIVE
CKM_BLAKE2B_512_KEY_GEN
CKK_BLAKE2B_512_HMAC
The BLAKE2B-512 mechanism, denoted CKM_BLAKE2B_512,
is a mechanism for message digesting, following the Blake2b Algorithm with a
512-bit message digest defined in RFC 7693.
It does not have a parameter.
Constraints on the length of input and output data are
summarized in the following table. For single-part digesting, the data and the
digest may begin at the same location in memory.
Table 180,
BLAKE2B-512: Data Length
|
Function
|
Input length
|
Digest length
|
|
C_Digest
|
any
|
64
|
The general-length BLAKE2B-512-HMAC mechanism, denoted CKM_BLAKE2B_512_HMAC_GENERAL,
is the keyed variant of the BLAKE2B-512 hash function and length of the output
should be in the range 1-64.The keys it uses are generic secret keys and
CKK_BLAKE2B_512_HMAC.
It has a parameter, a CK_MAC_GENERAL_PARAMS, which
holds the length in bytes of the desired output. This length should be in the
range 1-64 (the output size of BLAKE2B-512 is 64 bytes). Signatures (MACs)
produced by this mechanism shall be taken from the start of the full 64-byte
HMAC output.
Table 181,
General-length BLAKE2B-512-HMAC: Key And Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
generic secret or CKK_BLAKE2B_512_HMAC
|
Any
|
1-64, depending on
parameters
|
|
C_Verify
|
generic secret or
CKK_BLAKE2B_512_HMAC
|
Any
|
1-64, depending on
parameters
|
The BLAKE2B-512-HMAC mechanism, denoted CKM_BLAKE2B_512_HMAC,
is a special case of the general-length BLAKE2B-512-HMAC mechanism.
It has no parameter, and always produces an output of length
64.
BLAKE2B-512 key derivation, denoted CKM_BLAKE2B_512_KEY_DERIVE,
is the same as the SHA-1 key derivation mechanism in Section6.20.5,
except that it uses the BLAKE2B-512 hash function and the relevant length is 64
bytes.
The BLAKE2B-512-HMAC key generation mechanism, denoted CKM_BLAKE2B_512_KEY_GEN,
is a key generation mechanism for NIST’s BLAKE2B-512-HMAC.
It does not have a parameter.
The mechanism generates BLAKE2B-512-HMAC keys with a
particular length in bytes, as specified in the CKA_VALUE_LEN attribute
of the template for the key.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key. Other attributes supported by
the BLAKE2B-512-HMAC key type (specifically, the flags indicating which functions
the key supports) may be specified in the template for the key, or else are
assigned default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
CKM_BLAKE2B_512_HMAC key sizes, in bytes.
The mechanisms in this section are for generating keys and
IVs for performing password-based encryption. The method used to generate keys
and IVs is specified in [PKCS #5].
Table
182, PKCS 5 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_PBE_SHA1_DES3_EDE_CBC
|
|
|
|
|
ü
|
|
|
|
CKM_PBE_SHA1_DES2_EDE_CBC
|
|
|
|
|
ü
|
|
|
|
CKM_PBA_SHA1_WITH_SHA1_HMAC
|
|
|
|
|
ü
|
|
|
|
CKM_PKCS5_PBKD2
|
|
|
|
|
ü
|
|
|
Mechanisms:
CKM_PBE_SHA1_DES3_EDE_CBC
CKM_PBE_SHA1_DES2_EDE_CBC
CKM_PKCS5_PBKD2
CKM_PBA_SHA1_WITH_SHA1_HMAC
¨
CK_PBE_PARAMS;
CK_PBE_PARAMS_PTR
CK_PBE_PARAMS is a structure which provides all of
the necessary information required by the CKM_PBE mechanisms (see [PKCS #5] and
[PKCS #12] for information on the PBE generation mechanisms) and the
CKM_PBA_SHA1_WITH_SHA1_HMAC mechanism. It is defined as follows:
typedef struct CK_PBE_PARAMS {
CK_BYTE_PTR pInitVector;
CK_UTF8CHAR_PTR pPassword;
CK_ULONG ulPasswordLen;
CK_BYTE_PTR pSalt;
CK_ULONG ulSaltLen;
CK_ULONG ulIteration;
} CK_PBE_PARAMS;
The fields of the structure
have the following meanings:
pInitVector pointer
to the location that receives the 8-byte initialization vector (IV), if an IV
is required;
pPassword points to
the password to be used in the PBE key generation;
ulPasswordLen length in
bytes of the password information;
pSalt points
to the salt to be used in the PBE key generation;
ulSaltLen length
in bytes of the salt information;
ulIteration number
of iterations required for the generation.
CK_PBE_PARAMS_PTR is a pointer to a CK_PBE_PARAMS.
¨
CK_PKCS5_PBKD2_PSEUDO_RANDOM_FUNCTION_TYPE;
CK_PKCS5_PBKD2_PSEUDO_RANDOM_FUNCTION_TYPE_PTR
CK_PKCS5_PBKD2_PSEUDO_RANDOM_FUNCTION_TYPE is used to
indicate the Pseudo-Random Function (PRF) used to generate key bits using PKCS
#5 PBKDF2. It is defined as follows:
typedef CK_ULONG CK_PKCS5_PBKD2_PSEUDO_RANDOM_FUNCTION_TYPE;
The following PRFs are defined in PKCS #5 v2.1. The
following table lists the defined functions.
Table 183, PKCS #5 PBKDF2 Key Generation: Pseudo-random functions
|
PRF Identifier
|
Value
|
Parameter Type
|
|
CKP_PKCS5_PBKD2_HMAC_SHA1
|
0x00000001UL
|
No
Parameter. pPrfData must be NULL and ulPrfDataLen must be zero.
|
|
CKP_PKCS5_PBKD2_HMAC_GOSTR3411
|
0x00000002UL
|
This PRF uses GOST R34.11-94 hash to produce secret key
value. pPrfData should point to DER-encoded OID, indicating
GOSTR34.11-94 parameters. ulPrfDataLen holds encoded OID length in
bytes. If pPrfData is set to NULL_PTR, then id-GostR3411-94-CryptoProParamSet
parameters will be used (RFC 4357, 11.2), and ulPrfDataLen must be 0.
|
|
CKP_PKCS5_PBKD2_HMAC_SHA224
|
0x00000003UL
|
No Parameter. pPrfData must be
NULL and ulPrfDataLen must be zero.
|
|
CKP_PKCS5_PBKD2_HMAC_SHA256
|
0x00000004UL
|
No Parameter. pPrfData must be
NULL and ulPrfDataLen must be zero.
|
|
CKP_PKCS5_PBKD2_HMAC_SHA384
|
0x00000005UL
|
No Parameter. pPrfData must be
NULL and ulPrfDataLen must be zero.
|
|
CKP_PKCS5_PBKD2_HMAC_SHA512
|
0x00000006UL
|
No Parameter. pPrfData must be
NULL and ulPrfDataLen must be zero.
|
|
CKP_PKCS5_PBKD2_HMAC_SHA512_224
|
0x00000007UL
|
No Parameter. pPrfData must be
NULL and ulPrfDataLen must be zero.
|
|
CKP_PKCS5_PBKD2_HMAC_SHA512_256
|
0x00000008UL
|
No Parameter. pPrfData must be
NULL and ulPrfDataLen must be zero.
|
CK_PKCS5_PBKD2_PSEUDO_RANDOM_FUNCTION_TYPE_PTR is a
pointer to a CK_PKCS5_PBKD2_PSEUDO_RANDOM_FUNCTION_TYPE.
¨
CK_PKCS5_PBKDF2_SALT_SOURCE_TYPE;
CK_PKCS5_PBKDF2_SALT_SOURCE_TYPE_PTR
CK_PKCS5_PBKDF2_SALT_SOURCE_TYPE is used to indicate
the source of the salt value when deriving a key using PKCS #5 PBKDF2. It is
defined as follows:
typedef CK_ULONG CK_PKCS5_PBKDF2_SALT_SOURCE_TYPE;
The following salt value sources are defined in PKCS #5
v2.1. The following table lists the defined sources along with the
corresponding data type for the pSaltSourceData field in the CK_PKCS5_PBKD2_PARAMS2
structure defined below.
Table 184, PKCS #5 PBKDF2 Key Generation: Salt sources
|
Source
Identifier
|
Value
|
Data
Type
|
|
CKZ_SALT_SPECIFIED
|
0x00000001
|
Array
of CK_BYTE containing the value of the salt value.
|
CK_PKCS5_PBKDF2_SALT_SOURCE_TYPE_PTR is a pointer to
a CK_PKCS5_PBKDF2_SALT_SOURCE_TYPE.
¨
CK_PKCS5_PBKD2_PARAMS2;
CK_PKCS5_PBKD2_PARAMS2_PTR
CK_PKCS5_PBKD2_PARAMS2 is a structure that provides
the parameters to the CKM_PKCS5_PBKD2 mechanism. The structure is
defined as follows:
typedef struct CK_PKCS5_PBKD2_PARAMS2 {
CK_PKCS5_PBKDF2_SALT_SOURCE_TYPE saltSource;
CK_VOID_PTR pSaltSourceData;
CK_ULONG ulSaltSourceDataLen;
CK_ULONG iterations;
CK_PKCS5_PBKD2_PSEUDO_RANDOM_FUNCTION_TYPE prf;
CK_VOID_PTR pPrfData;
CK_ULONG ulPrfDataLen;
CK_UTF8CHAR_PTR pPassword;
CK_ULONG ulPasswordLen;
} CK_PKCS5_PBKD2_PARAMS2;
The fields of the structure have the following meanings:
saltSource source of
the salt value
pSaltSourceData data used as
the input for the salt source
ulSaltSourceDataLen length of the
salt source input
iterations number
of iterations to perform when generating each block of random data
prf pseudo-random
function used to generate the key
pPrfData data
used as the input for PRF in addition to the salt value
ulPrfDataLen length of
the input data for the PRF
pPassword points to
the password to be used in the PBE key generation
ulPasswordLen length in
bytes of the password information
CK_PKCS5_PBKD2_PARAMS2_PTR is a pointer to a CK_PKCS5_PBKD2_PARAMS2.
PKCS #5 PBKDF2 key generation, denoted CKM_PKCS5_PBKD2,
is a mechanism used for generating a secret key from a password and a salt
value. This functionality is defined in PKCS#5 as PBKDF2.
It has a parameter, a CK_PKCS5_PBKD2_PARAMS2
structure. The parameter specifies the salt value source, pseudo-random
function, and iteration count used to generate the new key.
Since this mechanism can be used to generate any type of
secret key, new key templates must contain the CKA_KEY_TYPE and CKA_VALUE_LEN
attributes. If the key type has a fixed length the CKA_VALUE_LEN
attribute may be omitted.
The mechanisms in this section are for generating keys and
IVs for performing password-based encryption or authentication. The method
used to generate keys and IVs is based on a method that was specified in [PKCS
#12].
We specify here a general method for producing various types
of pseudo-random bits from a password, p; a string of salt bits, s;
and an iteration count, c. The “type” of pseudo-random bits to be
produced is identified by an identification byte, ID, the meaning of
which will be discussed later.
Let H be
a hash function built around a compression function f: Z2u
´ Z2v
® Z2u
(that is, H has a chaining variable and output of length u bits, and the
message input to the compression function of H is v bits). For MD2 and
MD5, u=128 and v=512; for SHA-1, u=160 and v=512.
We assume here that u and v are both multiples
of 8, as are the lengths in bits of the password and salt strings and the
number n of pseudo-random bits required. In addition, u and v
are of course nonzero.
1. Construct
a string, D (the “diversifier”), by concatenating v/8 copies of ID.
2. Concatenate
copies of the salt together to create a string S of length v×és/vù
bits (the final copy of the salt may be truncated to create S). Note
that if the salt is the empty string, then so is S.
3. Concatenate
copies of the password together to create a string P of length v×ép/vù
bits (the final copy of the password may be truncated to create P).
Note that if the password is the empty string, then so is P.
4. Set I=S||P
to be the concatenation of S and P.
5. Set j=én/uù.
6. For i=1,
2, …, j, do the following:
a. Set Ai=Hc(D||I),
the cth hash of D||I. That is, compute the
hash of D||I; compute the hash of that hash; etc.; continue in
this fashion until a total of c hashes have been computed, each on the
result of the previous hash.
b. Concatenate
copies of Ai to create a string B of length v
bits (the final copy of Ai may be truncated to create B).
c. Treating I
as a concatenation I0, I1, …, Ik-1
of v-bit blocks, where k=és/vù+ép/vù, modify I by setting Ij=(Ij+B+1)
mod 2v for each j. To perform this addition, treat
each v-bit block as a binary number represented most-significant bit
first.
7. Concatenate
A1, A2, …, Aj together to
form a pseudo-random bit string, A.
8. Use the
first n bits of A as the output of this entire process.
When the password-based encryption mechanisms presented in
this section are used to generate a key and IV (if needed) from a password,
salt, and an iteration count, the above algorithm is used. To generate a key,
the identifier byte ID is set to the value 1; to generate an IV, the
identifier byte ID is set to the value 2.
When the password based authentication mechanism presented
in this section is used to generate a key from a password, salt, and an
iteration count, the above algorithm is used. The identifier byte ID is
set to the value 3.
SHA-1-PBE for 3-key triple-DES-CBC, denoted CKM_PBE_SHA1_DES3_EDE_CBC,
is a mechanism used for generating a 3-key triple-DES secret key and IV from a
password and a salt value by using the SHA-1 digest algorithm and an iteration
count. The method used to generate the key and IV is described above. Each
byte of the key produced will have its low-order bit adjusted, if necessary, so
that a valid 3-key triple-DES key with proper parity bits is obtained.
It has a parameter, a CK_PBE_PARAMS structure. The
parameter specifies the input information for the key generation process and
the location of the application-supplied buffer which will receive the 8-byte
IV generated by the mechanism.
The key and IV produced by this
mechanism will typically be used for performing password-based encryption.
SHA-1-PBE for 2-key triple-DES-CBC, denoted CKM_PBE_SHA1_DES2_EDE_CBC,
is a mechanism used for generating a 2-key triple-DES secret key and IV from a
password and a salt value by using the SHA-1 digest algorithm and an iteration
count. The method used to generate the key and IV is described above. Each
byte of the key produced will have its low-order bit adjusted, if necessary, so
that a valid 2-key triple-DES key with proper parity bits is obtained.
It has a parameter, a CK_PBE_PARAMS structure. The
parameter specifies the input information for the key generation process and
the location of the application-supplied buffer which will receive the 8-byte
IV generated by the mechanism.
The key and IV produced by this
mechanism will typically be used for performing password-based encryption.
SHA-1-PBA for SHA-1-HMAC, denoted CKM_PBA_SHA1_WITH_SHA1_HMAC,
is a mechanism used for generating a 160-bit generic secret key from a password
and a salt value by using the SHA-1 digest algorithm and an iteration count.
The method used to generate the key is described above.
It has a parameter, a CK_PBE_PARAMS structure. The
parameter specifies the input information for the key generation process. The
parameter also has a field to hold the location of an application-supplied
buffer which will receive an IV; for this mechanism, the contents of this field
are ignored, since authentication with SHA-1-HMAC does not require an IV.
The key generated by this mechanism will typically be used
for computing a SHA-1 HMAC to perform password-based authentication (not password-based
encryption). At the time of this writing, this is primarily done to ensure
the integrity of a PKCS #12 PDU.
Table
185,SSL Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_SSL3_PRE_MASTER_KEY_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_TLS_PRE_MASTER_KEY_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_SSL3_MASTER_KEY_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_SSL3_MASTER_KEY_DERIVE_DH
|
|
|
|
|
|
|
ü
|
|
CKM_SSL3_KEY_AND_MAC_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_SSL3_MD5_MAC
|
|
ü
|
|
|
|
|
|
|
CKM_SSL3_SHA1_MAC
|
|
ü
|
|
|
|
|
|
Mechanisms:
CKM_SSL3_PRE_MASTER_KEY_GEN
CKM_TLS_PRE_MASTER_KEY_GEN
CKM_SSL3_MASTER_KEY_DERIVE
CKM_SSL3_KEY_AND_MAC_DERIVE
CKM_SSL3_MASTER_KEY_DERIVE_DH
CKM_SSL3_MD5_MAC
CKM_SSL3_SHA1_MAC
¨ CK_SSL3_RANDOM_DATA
CK_SSL3_RANDOM_DATA is a structure which provides
information about the random data of a client and a server in an SSL context.
This structure is used by both the CKM_SSL3_MASTER_KEY_DERIVE and the CKM_SSL3_KEY_AND_MAC_DERIVE
mechanisms. It is defined as follows:
typedef struct CK_SSL3_RANDOM_DATA {
CK_BYTE_PTR pClientRandom;
CK_ULONG ulClientRandomLen;
CK_BYTE_PTR pServerRandom;
CK_ULONG ulServerRandomLen;
} CK_SSL3_RANDOM_DATA;
The fields of the structure have the following meanings:
pClientRandom pointer to
the client’s random data
ulClientRandomLen length in
bytes of the client’s random data
pServerRandom pointer to the
server’s random data
ulServerRandomLen length in bytes
of the server’s random data
¨
CK_SSL3_MASTER_KEY_DERIVE_PARAMS;
CK_SSL3_MASTER_KEY_DERIVE_PARAMS_PTR
CK_SSL3_MASTER_KEY_DERIVE_PARAMS is a structure that
provides the parameters to the CKM_SSL3_MASTER_KEY_DERIVE mechanism. It
is defined as follows:
typedef struct CK_SSL3_MASTER_KEY_DERIVE_PARAMS {
CK_SSL3_RANDOM_DATA RandomInfo;
CK_VERSION_PTR pVersion;
} CK_SSL3_MASTER_KEY_DERIVE_PARAMS;
The fields of the structure have the following meanings:
RandomInfo client’s
and server’s random data information.
pVersion pointer
to a CK_VERSION structure which receives the SSL protocol version
information
CK_SSL3_MASTER_KEY_DERIVE_PARAMS_PTR is a pointer to
a CK_SSL3_MASTER_KEY_DERIVE_PARAMS.
¨
CK_SSL3_KEY_MAT_OUT;
CK_SSL3_KEY_MAT_OUT_PTR
CK_SSL3_KEY_MAT_OUT is
a structure that contains the resulting key handles and initialization vectors
after performing a C_DeriveKey function with the CKM_SSL3_KEY_AND_MAC_DERIVE
mechanism. It is defined as follows:
typedef struct CK_SSL3_KEY_MAT_OUT {
CK_OBJECT_HANDLE hClientMacSecret;
CK_OBJECT_HANDLE hServerMacSecret;
CK_OBJECT_HANDLE hClientKey;
CK_OBJECT_HANDLE hServerKey;
CK_BYTE_PTR pIVClient;
CK_BYTE_PTR pIVServer;
} CK_SSL3_KEY_MAT_OUT;
The fields of the structure have the following meanings:
hClientMacSecret key handle
for the resulting Client MAC Secret key
hServerMacSecret key handle for
the resulting Server MAC Secret key
hClientKey key
handle for the resulting Client Secret key
hServerKey key handle
for the resulting Server Secret key
pIVClient pointer
to a location which receives the initialization vector (IV) created for the
client (if any)
pIVServer pointer
to a location which receives the initialization vector (IV) created for the
server (if any)
CK_SSL3_KEY_MAT_OUT_PTR is a pointer to a CK_SSL3_KEY_MAT_OUT.
¨
CK_SSL3_KEY_MAT_PARAMS;
CK_SSL3_KEY_MAT_PARAMS_PTR
CK_SSL3_KEY_MAT_PARAMS is a structure that provides
the parameters to the CKM_SSL3_KEY_AND_MAC_DERIVE mechanism. It is
defined as follows:
typedef struct CK_SSL3_KEY_MAT_PARAMS {
CK_ULONG ulMacSizeInBits;
CK_ULONG ulKeySizeInBits;
CK_ULONG ulIVSizeInBits;
CK_BBOOL bIsExport;
CK_SSL3_RANDOM_DATA RandomInfo;
CK_SSL3_KEY_MAT_OUT_PTR pReturnedKeyMaterial;
} CK_SSL3_KEY_MAT_PARAMS;
The fields of the structure
have the following meanings:
ulMacSizeInBits the length
(in bits) of the MACing keys agreed upon during the protocol handshake phase
ulKeySizeInBits the length
(in bits) of the secret keys agreed upon during the protocol handshake phase
ulIVSizeInBits the length
(in bits) of the IV agreed upon during the protocol handshake phase. If no IV
is required, the length should be set to 0
bIsExport a
Boolean value which indicates whether the keys have to be derived for an export
version of the protocol
RandomInfo client’s
and server’s random data information.
pReturnedKeyMaterial points to a
CK_SSL3_KEY_MAT_OUT structures which receives the handles for the keys
generated and the IVs
CK_SSL3_KEY_MAT_PARAMS_PTR is a pointer to a CK_SSL3_KEY_MAT_PARAMS.
Pre-master key generation in SSL 3.0, denoted CKM_SSL3_PRE_MASTER_KEY_GEN,
is a mechanism which generates a 48-byte generic secret key. It is used to
produce the "pre_master" key used in SSL version 3.0 for RSA-like
cipher suites.
It has one parameter, a CK_VERSION structure, which
provides the client’s SSL version number.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key (as well as the CKA_VALUE_LEN
attribute, if it is not supplied in the template). Other attributes may be
specified in the template, or else are assigned default values.
The template sent along with this mechanism during a C_GenerateKey
call may indicate that the object class is CKO_SECRET_KEY, the key type
is CKK_GENERIC_SECRET, and the CKA_VALUE_LEN attribute has value
48. However, since these facts are all implicit in the mechanism, there is no
need to specify any of them.
For this mechanism, the ulMinKeySize and ulMaxKeySize fields
of the CK_MECHANISM_INFO structure both indicate 48 bytes.
CKM_TLS_PRE_MASTER_KEY_GEN has identical
functionality as CKM_SSL3_PRE_MASTER_KEY_GEN. It exists only for
historical reasons, please use CKM_SSL3_PRE_MASTER_KEY_GEN instead.
Master key derivation in SSL 3.0, denoted CKM_SSL3_MASTER_KEY_DERIVE,
is a mechanism used to derive one 48-byte generic secret key from another
48-byte generic secret key. It is used to produce the
"master_secret" key used in the SSL protocol from the
"pre_master" key. This mechanism returns the value of the client
version, which is built into the "pre_master" key as well as a handle
to the derived "master_secret" key.
It has a parameter, a CK_SSL3_MASTER_KEY_DERIVE_PARAMS
structure, which allows for the passing of random data to the token as well as
the returning of the protocol version number which is part of the pre-master
key. This structure is defined in Section 6.39.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key (as well as the CKA_VALUE_LEN
attribute, if it is not supplied in the template). Other attributes may be
specified in the template; otherwise they are assigned default values.
The template sent along with this mechanism during a C_DeriveKey
call may indicate that the object class is CKO_SECRET_KEY, the key type
is CKK_GENERIC_SECRET, and the CKA_VALUE_LEN attribute has value
48. However, since these facts are all implicit in the mechanism, there is no
need to specify any of them.
This mechanism has the following rules about key sensitivity
and extractability:
·
The CKA_SENSITIVE and CKA_EXTRACTABLE attributes in
the template for the new key can both be specified to be either CK_TRUE or
CK_FALSE. If omitted, these attributes each take on some default value.
·
If the base key has its CKA_ALWAYS_SENSITIVE attribute set
to CK_FALSE, then the derived key will as well. If the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE, then the derived key has its CKA_ALWAYS_SENSITIVE
attribute set to the same value as its CKA_SENSITIVE attribute.
·
Similarly, if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_FALSE, then the derived key will, too. If the base key has
its CKA_NEVER_EXTRACTABLE attribute set to CK_TRUE, then the derived key
has its CKA_NEVER_EXTRACTABLE attribute set to the opposite value
from its CKA_EXTRACTABLE attribute.
For this mechanism, the ulMinKeySize and ulMaxKeySize fields
of the CK_MECHANISM_INFO structure both indicate 48 bytes.
Note that the CK_VERSION structure pointed to by the CK_SSL3_MASTER_KEY_DERIVE_PARAMS
structure’s pVersion field will be modified by the C_DeriveKey
call. In particular, when the call returns, this structure will hold the SSL
version associated with the supplied pre_master key.
Note that this mechanism is only useable for cipher suites
that use a 48-byte “pre_master” secret with an embedded version number. This
includes the RSA cipher suites, but excludes the Diffie-Hellman cipher suites.
Master key derivation for Diffie-Hellman in SSL 3.0, denoted
CKM_SSL3_MASTER_KEY_DERIVE_DH, is a mechanism used to derive one 48-byte
generic secret key from another arbitrary length generic secret key. It is
used to produce the "master_secret" key used in the SSL protocol from
the "pre_master" key.
It has a parameter, a CK_SSL3_MASTER_KEY_DERIVE_PARAMS
structure, which allows for the passing of random data to the token. This
structure is defined in Section 6.39. The pVersion field of the
structure must be set to NULL_PTR since the version number is not embedded in
the "pre_master" key as it is for RSA-like cipher suites.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key (as well as the CKA_VALUE_LEN
attribute, if it is not supplied in the template). Other attributes may be
specified in the template, or else are assigned default values.
The template sent along with this mechanism during a C_DeriveKey
call may indicate that the object class is CKO_SECRET_KEY, the key type
is CKK_GENERIC_SECRET, and the CKA_VALUE_LEN attribute has value
48. However, since these facts are all implicit in the mechanism, there is no
need to specify any of them.
This mechanism has the following rules about key sensitivity
and extractability:
·
The CKA_SENSITIVE and CKA_EXTRACTABLE attributes in
the template for the new key can both be specified to be either CK_TRUE or
CK_FALSE. If omitted, these attributes each take on some default value.
·
If the base key has its CKA_ALWAYS_SENSITIVE attribute set
to CK_FALSE, then the derived key will as well. If the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE, then the derived key has its CKA_ALWAYS_SENSITIVE
attribute set to the same value as its CKA_SENSITIVE attribute.
·
Similarly, if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_FALSE, then the derived key will, too. If the base key has
its CKA_NEVER_EXTRACTABLE attribute set to CK_TRUE, then the derived key
has its CKA_NEVER_EXTRACTABLE attribute set to the opposite value
from its CKA_EXTRACTABLE attribute.
For this mechanism, the ulMinKeySize and ulMaxKeySize fields
of the CK_MECHANISM_INFO structure both indicate 48 bytes.
Note that this mechanism is only useable for cipher suites
that do not use a fixed length 48-byte “pre_master” secret with an embedded
version number. This includes the Diffie-Hellman cipher suites, but excludes
the RSA cipher suites.
Key, MAC and IV derivation in SSL 3.0, denoted CKM_SSL3_KEY_AND_MAC_DERIVE,
is a mechanism used to derive the appropriate cryptographic keying material
used by a "CipherSuite" from the "master_secret" key and
random data. This mechanism returns the key handles for the keys generated in
the process, as well as the IVs created.
It has a parameter, a CK_SSL3_KEY_MAT_PARAMS
structure, which allows for the passing of random data as well as the
characteristic of the cryptographic material for the given CipherSuite and a
pointer to a structure which receives the handles and IVs which were generated.
This structure is defined in Section 6.39.
This mechanism contributes to the creation of four distinct
keys on the token and returns two IVs (if IVs are requested by the caller) back
to the caller. The keys are all given an object class of CKO_SECRET_KEY.
The two MACing keys ("client_write_MAC_secret" and
"server_write_MAC_secret") are always given a type of CKK_GENERIC_SECRET.
They are flagged as valid for signing, verification, and derivation operations.
The other two keys ("client_write_key" and
"server_write_key") are typed according to information found in the
template sent along with this mechanism during a C_DeriveKey function
call. By default, they are flagged as valid for encryption, decryption, and
derivation operations.
IVs will be generated and returned if the ulIVSizeInBits
field of the CK_SSL3_KEY_MAT_PARAMS field has a nonzero value. If they
are generated, their length in bits will agree with the value in the ulIVSizeInBits
field.
All four keys inherit the values of the CKA_SENSITIVE,
CKA_ALWAYS_SENSITIVE, CKA_EXTRACTABLE, and CKA_NEVER_EXTRACTABLE
attributes from the base key. The template provided to C_DeriveKey may
not specify values for any of these attributes which differ from those held by
the base key.
Note that the CK_SSL3_KEY_MAT_OUT structure pointed
to by the CK_SSL3_KEY_MAT_PARAMS structure’s pReturnedKeyMaterial
field will be modified by the C_DeriveKey call. In particular, the four
key handle fields in the CK_SSL3_KEY_MAT_OUT structure will be modified
to hold handles to the newly-created keys; in addition, the buffers pointed to
by the CK_SSL3_KEY_MAT_OUT structure’s pIVClient and pIVServer
fields will have IVs returned in them (if IVs are requested by the caller).
Therefore, these two fields must point to buffers with sufficient space to hold
any IVs that will be returned.
This mechanism departs from the other key derivation
mechanisms in Cryptoki in its returned information. For most key-derivation
mechanisms, C_DeriveKey returns a single key handle as a result of a
successful completion. However, since the CKM_SSL3_KEY_AND_MAC_DERIVE
mechanism returns all of its key handles in the CK_SSL3_KEY_MAT_OUT
structure pointed to by the CK_SSL3_KEY_MAT_PARAMS structure specified
as the mechanism parameter, the parameter phKey passed to C_DeriveKey
is unnecessary, and should be a NULL_PTR.
If a call to C_DeriveKey with this mechanism fails,
then none of the four keys will be created on the token.
MD5 MACing in SSL3.0, denoted CKM_SSL3_MD5_MAC, is a
mechanism for single- and multiple-part signatures (data authentication) and
verification using MD5, based on the SSL 3.0 protocol. This technique is very
similar to the HMAC technique.
It has a parameter, a CK_MAC_GENERAL_PARAMS, which
specifies the length in bytes of the signatures produced by this mechanism.
Constraints on key types and the length of input and output
data are summarized in the following table:
Table 186, MD5 MACing in SSL 3.0: Key And Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
generic secret
|
any
|
4-8, depending on
parameters
|
|
C_Verify
|
generic secret
|
any
|
4-8, depending on
parameters
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
generic secret key sizes, in bits.
SHA-1 MACing in SSL3.0, denoted CKM_SSL3_SHA1_MAC, is
a mechanism for single- and multiple-part signatures (data authentication) and
verification using SHA-1, based on the SSL 3.0 protocol. This technique is very
similar to the HMAC technique.
It has a parameter, a CK_MAC_GENERAL_PARAMS, which
specifies the length in bytes of the signatures produced by this mechanism.
Constraints on key types and the length of input and output
data are summarized in the following table:
Table 187, SHA-1 MACing in SSL 3.0: Key And Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
generic secret
|
any
|
4-8, depending on
parameters
|
|
C_Verify
|
generic secret
|
any
|
4-8, depending on
parameters
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
generic secret key sizes, in bits.
Details for TLS 1.2 and its key derivation and MAC
mechanisms can be found in [TLS12]. TLS 1.2 mechanisms differ from TLS 1.0 and
1.1 mechanisms in that the base hash used in the underlying TLS PRF
(pseudo-random function) can be negotiated. Therefore each mechanism parameter
for the TLS 1.2 mechanisms contains a new value in the parameters structure to
specify the hash function.
This section also specifies CKM_TLS12_MAC which should be
used in place of CKM_TLS_PRF to calculate the verify_data in the TLS
"finished" message.
This section also specifies CKM_TLS_KDF that can be
used in place of CKM_TLS_PRF to implement key material exporters.
Table
188, TLS 1.2 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_TLS12_MASTER_KEY_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_TLS12_MASTER_KEY_DERIVE_DH
|
|
|
|
|
|
|
ü
|
|
CKM_TLS12_KEY_AND_MAC_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_TLS12_KEY_SAFE_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_TLS_KDF
|
|
|
|
|
|
|
ü
|
|
CKM_TLS12_MAC
|
|
ü
|
|
|
|
|
|
|
CKM_TLS12_KDF
|
|
|
|
|
|
|
ü
|
Mechanisms:
CKM_TLS12_MASTER_KEY_DERIVE
CKM_TLS12_MASTER_KEY_DERIVE_DH
CKM_TLS12_KEY_AND_MAC_DERIVE
CKM_TLS12_KEY_SAFE_DERIVE
CKM_TLS_KDF
CKM_TLS12_MAC
CKM_TLS12_KDF
¨
CK_TLS12_MASTER_KEY_DERIVE_PARAMS;
CK_TLS12_MASTER_KEY_DERIVE_PARAMS_PTR
CK_TLS12_MASTER_KEY_DERIVE_PARAMS is a structure that
provides the parameters to the CKM_TLS12_MASTER_KEY_DERIVE mechanism.
It is defined as follows:
typedef struct CK_TLS12_MASTER_KEY_DERIVE_PARAMS {
CK_SSL3_RANDOM_DATA RandomInfo;
CK_VERSION_PTR pVersion;
CK_MECHANISM_TYPE prfHashMechanism;
} CK_TLS12_MASTER_KEY_DERIVE_PARAMS;
The fields of the structure have the following meanings:
RandomInfo client’s
and server’s random data information.
pVersion pointer
to a CK_VERSION structure which receives the SSL protocol version
information
prfHashMechanism base hash used
in the underlying TLS1.2 PRF operation used to derive the master key.
CK_TLS12_MASTER_KEY_DERIVE_PARAMS_PTR is a pointer to
a CK_TLS12_MASTER_KEY_DERIVE_PARAMS.
¨
CK_TLS12_KEY_MAT_PARAMS;
CK_TLS12_KEY_MAT_PARAMS_PTR
CK_TLS12_KEY_MAT_PARAMS is a structure that provides
the parameters to the CKM_TLS12_KEY_AND_MAC_DERIVE mechanism. It is
defined as follows:
typedef struct CK_TLS12_KEY_MAT_PARAMS {
CK_ULONG ulMacSizeInBits;
CK_ULONG ulKeySizeInBits;
CK_ULONG ulIVSizeInBits;
CK_BBOOL bIsExport;
CK_SSL3_RANDOM_DATA RandomInfo;
CK_SSL3_KEY_MAT_OUT_PTR pReturnedKeyMaterial;
CK_MECHANISM_TYPE prfHashMechanism;
} CK_TLS12_KEY_MAT_PARAMS;
The fields of the structure have the following meanings:
ulMacSizeInBits the length
(in bits) of the MACing keys agreed upon during the protocol handshake phase.
If no MAC key is required, the length should be set to 0.
ulKeySizeInBits the length
(in bits) of the secret keys agreed upon during the protocol handshake phase
ulIVSizeInBits the
length (in bits) of the IV agreed upon during the protocol handshake phase. If
no IV is required, the length should be set to 0
bIsExport must be
set to CK_FALSE because export cipher suites must not be used in TLS 1.1 and
later.
RandomInfo client’s
and server’s random data information.
pReturnedKeyMaterial points to a
CK_SSL3_KEY_MAT_OUT structures which receives the handles for the keys
generated and the IVs
prfHashMechanism base hash used
in the underlying TLS1.2 PRF operation used to derive the master key.
CK_TLS12_KEY_MAT_PARAMS_PTR is a pointer to a CK_TLS12_KEY_MAT_PARAMS.
¨
CK_TLS_KDF_PARAMS;
CK_TLS_KDF_PARAMS_PTR
CK_TLS_KDF_PARAMS is a structure that provides the
parameters to the CKM_TLS_KDF mechanism. It is defined as follows:
typedef
struct CK_TLS_KDF_PARAMS {
CK_MECHANISM_TYPE prfMechanism;
CK_BYTE_PTR pLabel;
CK_ULONG ulLabelLength;
CK_SSL3_RANDOM_DATA RandomInfo;
CK_BYTE_PTR pContextData;
CK_ULONG ulContextDataLength;
}
CK_TLS_KDF_PARAMS;
The fields of the structure have the following meanings:
prfMechanism the hash
mechanism used in the TLS1.2 PRF construct or CKM_TLS_PRF to use with the
TLS1.0 and 1.1 PRF construct.
pLabel a
pointer to the label for this key derivation
ulLabelLength length of
the label in bytes
RandomInfo the random
data for the key derivation
pContextData a pointer
to the context data for this key derivation. NULL_PTR if not present
ulContextDataLength length of the
context data in bytes. 0 if not present.
CK_TLS_KDF_PARAMS_PTR is a pointer to a CK_TLS_KDF_PARAMS.
¨
CK_TLS_MAC_PARAMS;
CK_TLS_MAC_PARAMS_PTR
CK_TLS_MAC_PARAMS is a
structure that provides the parameters to the CKM_TLS_MAC mechanism. It
is defined as follows:
typedef struct CK_TLS_MAC_PARAMS {
CK_MECHANISM_TYPE prfHashMechanism;
CK_ULONG ulMacLength;
CK_ULONG ulServerOrClient;
} CK_TLS_MAC_PARAMS;
The fields of the structure
have the following meanings:
prfHashMechanism the hash
mechanism used in the TLS12 PRF construct or CKM_TLS_PRF to use with the TLS1.0
and 1.1 PRF construct.
ulMacLength the length
of the MAC tag required or offered. Always 12 octets in TLS 1.0 and 1.1. Generally
12 octets, but may be negotiated to a longer value in TLS1.2.
ulServerOrClient 1 to use
the label "server finished", 2 to use the label "client
finished". All other values are invalid.
CK_TLS_MAC_PARAMS_PTR is a pointer to a CK_TLS_MAC_PARAMS.
¨ CK_TLS_PRF_PARAMS; CK_TLS_PRF_PARAMS_PTR
CK_TLS_PRF_PARAMS is a structure, which provides the
parameters to the CKM_TLS_PRF mechanism. It is defined as follows:
typedef struct CK_TLS_PRF_PARAMS {
CK_BYTE_PTR pSeed;
CK_ULONG ulSeedLen;
CK_BYTE_PTR pLabel;
CK_ULONG ulLabelLen;
CK_BYTE_PTR pOutput;
CK_ULONG_PTR pulOutputLen;
} CK_TLS_PRF_PARAMS;
The fields of the structure have the following meanings:
pSeed pointer
to the input seed
ulSeedLen length
in bytes of the input seed
pLabel pointer
to the identifying label
ulLabelLen length
in bytes of the identifying label
pOutput pointer
receiving the output of the operation
pulOutputLen pointer
to the length in bytes that the output to be created shall have, has to hold
the desired length as input and will receive the calculated length as output
CK_TLS_PRF_PARAMS_PTR is a pointer to a CK_TLS_PRF_PARAMS.
6.40.3 TLS MAC
The TLS MAC mechanism is used to generate integrity tags for
the TLS "finished" message. It replaces the use of the CKM_TLS_PRF
function for TLS1.0 and 1.1 and that mechanism is deprecated.
CKM_TLS_MAC takes a parameter of CK_TLS_MAC_PARAMS.
To use this mechanism with TLS1.0 and TLS1.1, use CKM_TLS_PRF as the
value for prfMechanism in place of a hash mechanism. Note: Although CKM_TLS_PRF
is deprecated as a mechanism for C_DeriveKey, the manifest value is retained
for use with this mechanism to indicate the use of the TLS1.0/1.1 pseudo-random
function.
In TLS1.0 and 1.1 the "finished" message verify_data
(i.e. the output signature from the MAC mechanism) is always 12 bytes. In
TLS1.2 the "finished" message verify_data is a minimum of 12 bytes,
defaults to 12 bytes, but may be negotiated to longer length.
Table 189,
General-length TLS MAC: Key And Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
generic secret
|
any
|
≥ 12 bytes
|
|
C_Verify
|
generic secret
|
any
|
≥ 12 bytes
|
Master key derivation in TLS 1.0, denoted CKM_TLS_MASTER_KEY_DERIVE,
is a mechanism used to derive one 48-byte generic secret key from another
48-byte generic secret key. It is used to produce the
"master_secret" key used in the TLS protocol from the "pre_master"
key. This mechanism returns the value of the client version, which is built
into the "pre_master" key as well as a handle to the derived
"master_secret" key.
It has a parameter, a CK_SSL3_MASTER_KEY_DERIVE_PARAMS
structure, which allows for the passing of random data to the token as well as
the returning of the protocol version number which is part of the pre-master
key. This structure is defined in Section 6.39.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key (as well as the CKA_VALUE_LEN
attribute, if it is not supplied in the template). Other attributes may be
specified in the template, or else are assigned default values.
The mechanism also contributes the CKA_ALLOWED_MECHANISMS
attribute consisting only of CKM_TLS12_KEY_AND_MAC_DERIVE,
CKM_TLS12_KEY_SAFE_DERIVE, CKM_TLS12_KDF and CKM_TLS12_MAC.
The template sent along with this mechanism during a C_DeriveKey
call may indicate that the object class is CKO_SECRET_KEY, the key type
is CKK_GENERIC_SECRET, and the CKA_VALUE_LEN attribute has value
48. However, since these facts are all implicit in the mechanism, there is no
need to specify any of them.
This mechanism has the following rules about key sensitivity
and extractability:
·
The CKA_SENSITIVE and CKA_EXTRACTABLE attributes in
the template for the new key can both be specified to be either CK_TRUE or
CK_FALSE. If omitted, these attributes each take on some default value.
·
If the base key has its CKA_ALWAYS_SENSITIVE attribute set
to CK_FALSE, then the derived key will as well. If the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE, then the derived key has its CKA_ALWAYS_SENSITIVE
attribute set to the same value as its CKA_SENSITIVE attribute.
·
Similarly, if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_FALSE, then the derived key will, too. If the base key has
its CKA_NEVER_EXTRACTABLE attribute set to CK_TRUE, then the derived key
has its CKA_NEVER_EXTRACTABLE attribute set to the opposite value
from its CKA_EXTRACTABLE attribute.
For this mechanism, the ulMinKeySize and ulMaxKeySize fields
of the CK_MECHANISM_INFO structure both indicate 48 bytes.
Note that the CK_VERSION structure pointed to by the CK_SSL3_MASTER_KEY_DERIVE_PARAMS
structure’s pVersion field will be modified by the C_DeriveKey
call. In particular, when the call returns, this structure will hold the SSL
version associated with the supplied pre_master key.
Note that this mechanism is only useable for cipher suites
that use a 48-byte “pre_master” secret with an embedded version number. This
includes the RSA cipher suites, but excludes the Diffie-Hellman cipher suites.
Master key derivation for Diffie-Hellman in TLS 1.0, denoted
CKM_TLS_MASTER_KEY_DERIVE_DH, is a mechanism used to derive one 48-byte
generic secret key from another arbitrary length generic secret key. It is
used to produce the "master_secret" key used in the TLS protocol from
the "pre_master" key.
It has a parameter, a CK_SSL3_MASTER_KEY_DERIVE_PARAMS
structure, which allows for the passing of random data to the token. This
structure is defined in Section 6.39. The pVersion field of the structure
must be set to NULL_PTR since the version number is not embedded in the
"pre_master" key as it is for RSA-like cipher suites.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key (as well as the CKA_VALUE_LEN
attribute, if it is not supplied in the template). Other attributes may be
specified in the template, or else are assigned default values.
The mechanism also contributes the CKA_ALLOWED_MECHANISMS
attribute consisting only of CKM_TLS12_KEY_AND_MAC_DERIVE,
CKM_TLS12_KEY_SAFE_DERIVE, CKM_TLS12_KDF and CKM_TLS12_MAC.
The template sent along with this mechanism during a C_DeriveKey
call may indicate that the object class is CKO_SECRET_KEY, the key type
is CKK_GENERIC_SECRET, and the CKA_VALUE_LEN attribute has value
48. However, since these facts are all implicit in the mechanism, there is no
need to specify any of them.
This mechanism has the following rules about key sensitivity
and extractability:
·
The CKA_SENSITIVE and CKA_EXTRACTABLE attributes in
the template for the new key can both be specified to be either CK_TRUE or
CK_FALSE. If omitted, these attributes each take on some default value.
·
If the base key has its CKA_ALWAYS_SENSITIVE attribute set
to CK_FALSE, then the derived key will as well. If the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE, then the derived key has its CKA_ALWAYS_SENSITIVE
attribute set to the same value as its CKA_SENSITIVE attribute.
·
Similarly, if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_FALSE, then the derived key will, too. If the base key has
its CKA_NEVER_EXTRACTABLE attribute set to CK_TRUE, then the derived key
has its CKA_NEVER_EXTRACTABLE attribute set to the opposite value
from its CKA_EXTRACTABLE attribute.
For this mechanism, the ulMinKeySize and ulMaxKeySize fields
of the CK_MECHANISM_INFO structure both indicate 48 bytes.
Note that this mechanism is only useable for cipher suites
that do not use a fixed length 48-byte “pre_master” secret with an embedded version
number. This includes the Diffie-Hellman cipher suites, but excludes the RSA
cipher suites.
Key, MAC and IV derivation in TLS 1.0, denoted CKM_TLS_KEY_AND_MAC_DERIVE,
is a mechanism used to derive the appropriate cryptographic keying material
used by a "CipherSuite" from the "master_secret" key and
random data. This mechanism returns the key handles for the keys generated in
the process, as well as the IVs created.
It has a parameter, a CK_SSL3_KEY_MAT_PARAMS
structure, which allows for the passing of random data as well as the
characteristic of the cryptographic material for the given CipherSuite and a
pointer to a structure which receives the handles and IVs which were generated.
This structure is defined in Section 6.39.
This mechanism contributes to the creation of four distinct
keys on the token and returns two IVs (if IVs are requested by the caller) back
to the caller. The keys are all given an object class of CKO_SECRET_KEY.
The two MACing keys ("client_write_MAC_secret" and
"server_write_MAC_secret") (if present) are always given a type of CKK_GENERIC_SECRET.
They are flagged as valid for signing and verification.
The other two keys ("client_write_key" and
"server_write_key") are typed according to information found in the
template sent along with this mechanism during a C_DeriveKey function
call. By default, they are flagged as valid for encryption, decryption, and
derivation operations.
For CKM_TLS12_KEY_AND_MAC_DERIVE, IVs will be
generated and returned if the ulIVSizeInBits field of the CK_SSL3_KEY_MAT_PARAMS
field has a nonzero value. If they are generated, their length in bits will
agree with the value in the ulIVSizeInBits field.
Note Well:
CKM_TLS12_KEY_AND_MAC_DERIVE produces both private (key) and public (IV) data.
It is possible to "leak" private data by the simple expedient of
decreasing the length of private data requested. E.g. Setting ulMacSizeInBits
and ulKeySizeInBits to 0 (or other lengths less than the key size) will result
in the private key data being placed in the destination designated for the
IV's. Repeated calls with the same master key and same RandomInfo but with
differing lengths for the private key material will result in different data
being leaked.<
All four keys inherit the values of the CKA_SENSITIVE,
CKA_ALWAYS_SENSITIVE, CKA_EXTRACTABLE, and CKA_NEVER_EXTRACTABLE
attributes from the base key. The template provided to C_DeriveKey may
not specify values for any of these attributes which differ from those held by
the base key.
Note that the CK_SSL3_KEY_MAT_OUT structure pointed
to by the CK_SSL3_KEY_MAT_PARAMS structure’s pReturnedKeyMaterial
field will be modified by the C_DeriveKey call. In particular, the four
key handle fields in the CK_SSL3_KEY_MAT_OUT structure will be modified
to hold handles to the newly-created keys; in addition, the buffers pointed to
by the CK_SSL3_KEY_MAT_OUT structure’s pIVClient and pIVServer
fields will have IVs returned in them (if IVs are requested by the caller).
Therefore, these two fields must point to buffers with sufficient space to hold
any IVs that will be returned.
This mechanism departs from the other key derivation
mechanisms in Cryptoki in its returned information. For most key-derivation
mechanisms, C_DeriveKey returns a single key handle as a result of a
successful completion. However, since the CKM_SSL3_KEY_AND_MAC_DERIVE
mechanism returns all of its key handles in the CK_SSL3_KEY_MAT_OUT
structure pointed to by the CK_SSL3_KEY_MAT_PARAMS structure specified
as the mechanism parameter, the parameter phKey passed to C_DeriveKey
is unnecessary, and should be a NULL_PTR.
If a call to C_DeriveKey with this mechanism fails,
then none of the four keys will be created on
the token.
CKM_TLS12_KEY_SAFE_DERIVE is identical to CKM_TLS12_KEY_AND_MAC_DERIVE
except that it shall never produce IV data, and the ulIvSizeInBits field of CK_TLS12_KEY_MAT_PARAMS
is ignored and treated as 0. All of the other conditions and behavior
described for CKM_TLS12_KEY_AND_MAC_DERIVE, with the exception of the black box
warning, apply to this mechanism.
CKM_TLS12_KEY_SAFE_DERIVE is provided as a separate
mechanism to allow a client to control the export of IV material (and possible
leaking of key material) through the use of the CKA_ALLOWED_MECHANISMS key
attribute.
CKM_TLS_KDF is the mechanism defined in [RFC 5705].
It uses the TLS key material and TLS PRF function to produce additional key
material for protocols that want to leverage the TLS key negotiation
mechanism. CKM_TLS_KDF has a parameter of CK_TLS_KDF_PARAMS. If
the protocol using this mechanism does not use context information, the pContextData
field shall be set to NULL_PTR and the ulContextDataLength field
shall be set to 0.
To use this mechanism with TLS1.0 and TLS1.1, use CKM_TLS_PRF
as the value for prfMechanism in place of a hash mechanism. Note:
Although CKM_TLS_PRF is deprecated as a mechanism for C_DeriveKey, the
manifest value is retained for use with this mechanism to indicate the use of
the TLS1.0/1.1 Pseudo-random function.
This mechanism can be used to derive multiple keys (e.g.
similar to CKM_TLS12_KEY_AND_MAC_DERIVE) by first deriving the key
stream as a CKK_GENERIC_SECRET of the necessary length and doing
subsequent derives against that derived key using the CKM_EXTRACT_KEY_FROM_KEY
mechanism to split the key stream into the actual operational keys.
The mechanism should not be used with the labels defined for
use with TLS, but the token does not enforce this behavior.
This mechanism has the following rules about key sensitivity
and extractability:
·
If the original key has its CKA_SENSITIVE attribute set to
CK_TRUE, so does the derived key. If not, then the derived key’s CKA_SENSITIVE
attribute is set either from the supplied template or from the original key.
·
Similarly, if the original key has its CKA_EXTRACTABLE
attribute set to CK_FALSE, so does the derived key. If not, then the derived
key’s CKA_EXTRACTABLE attribute is set either from the supplied template
or from the original key.
·
The derived key’s CKA_ALWAYS_SENSITIVE attribute is set to
CK_TRUE if and only if the original key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE.
·
Similarly, the derived key’s CKA_NEVER_EXTRACTABLE
attribute is set to CK_TRUE if and only if the original key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_TRUE.
CKM_TLS12_KDF is the mechanism defined in [RFC 5705].
It uses the TLS key material and TLS PRF function to produce additional key
material for protocols that want to leverage the TLS key negotiation
mechanism. CKM_TLS12_KDF has a parameter of CK_TLS_KDF_PARAMS.
If the protocol using this mechanism does not use context information, the pContextData
field shall be set to NULL_PTR and the ulContextDataLength field
shall be set to 0.
To use this mechanism with TLS1.0 and TLS1.1, use CKM_TLS_PRF
as the value for prfMechanism in place of a hash mechanism. Note:
Although CKM_TLS_PRF is deprecated as a mechanism for C_DeriveKey, the
manifest value is retained for use with this mechanism to indicate the use of
the TLS1.0/1.1 Pseudo-random function.
This mechanism can be used to derive multiple keys (e.g.
similar to CKM_TLS12_KEY_AND_MAC_DERIVE) by first deriving the key
stream as a CKK_GENERIC_SECRET of the necessary length and doing
subsequent derives against that derived key stream using the CKM_EXTRACT_KEY_FROM_KEY
mechanism to split the key stream into the actual operational keys.
The mechanism should not be used with the labels defined for
use with TLS, but the token does not enforce this behavior.
This mechanism has the following rules about key sensitivity
and extractability:
·
If the original key has its CKA_SENSITIVE attribute set to
CK_TRUE, so does the derived key. If not, then the derived key’s CKA_SENSITIVE
attribute is set either from the supplied template or from the original key.
·
Similarly, if the original key has its CKA_EXTRACTABLE
attribute set to CK_FALSE, so does the derived key. If not, then the derived
key’s CKA_EXTRACTABLE attribute is set either from the supplied template
or from the original key.
·
The derived key’s CKA_ALWAYS_SENSITIVE attribute is set to
CK_TRUE if and only if the original key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE.
·
Similarly, the derived key’s CKA_NEVER_EXTRACTABLE
attribute is set to CK_TRUE if and only if the original key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_TRUE.
Details can be found in [WTLS].
When comparing the existing TLS mechanisms with these
extensions to support WTLS one could argue that there would be no need to have
distinct handling of the client and server side of the handshake. However,
since in WTLS the server and client use different sequence numbers, there could
be instances (e.g. when WTLS is used to protect asynchronous protocols) where
sequence numbers on the client and server side differ, and hence this motivates
the introduced split.
Table
190, WTLS Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_WTLS_PRE_MASTER_KEY_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_WTLS_MASTER_KEY_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_WTLS_MASTER_KEY_DERIVE_DH_ECC
|
|
|
|
|
|
|
ü
|
|
CKM_WTLS_SERVER_KEY_AND_MAC_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_WTLS_CLIENT_KEY_AND_MAC_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_WTLS_PRF
|
|
|
|
|
|
|
ü
|
Mechanisms:
CKM_WTLS_PRE_MASTER_KEY_GEN
CKM_WTLS_MASTER_KEY_DERIVE
CKM_WTLS_MASTER_KEY_DERIVE_DH_ECC
CKM_WTLS_PRF
CKM_WTLS_SERVER_KEY_AND_MAC_DERIVE
CKM_WTLS_CLIENT_KEY_AND_MAC_DERIVE
¨ CK_WTLS_RANDOM_DATA; CK_WTLS_RANDOM_DATA_PTR
CK_WTLS_RANDOM_DATA is a structure, which provides
information about the random data of a client and a server in a WTLS context.
This structure is used by the CKM_WTLS_MASTER_KEY_DERIVE mechanism. It
is defined as follows:
typedef struct CK_WTLS_RANDOM_DATA {
CK_BYTE_PTR pClientRandom;
CK_ULONG ulClientRandomLen;
CK_BYTE_PTR pServerRandom;
CK_ULONG ulServerRandomLen;
} CK_WTLS_RANDOM_DATA;
The fields of the structure have the following meanings:
pClientRandom pointer to
the client’s random data
pClientRandomLen length in bytes
of the client’s random data
pServerRaondom pointer to the
server’s random data
ulServerRandomLen length in bytes
of the server’s random data
CK_WTLS_RANDOM_DATA_PTR is a pointer to a CK_WTLS_RANDOM_DATA.
¨
CK_WTLS_MASTER_KEY_DERIVE_PARAMS;
CK_WTLS_MASTER_KEY_DERIVE_PARAMS _PTR
CK_WTLS_MASTER_KEY_DERIVE_PARAMS is a structure,
which provides the parameters to the CKM_WTLS_MASTER_KEY_DERIVE
mechanism. It is defined as follows:
typedef struct CK_WTLS_MASTER_KEY_DERIVE_PARAMS {
CK_MECHANISM_TYPE DigestMechanism;
CK_WTLS_RANDOM_DATA RandomInfo;
CK_BYTE_PTR pVersion;
} CK_WTLS_MASTER_KEY_DERIVE_PARAMS;
The fields of the structure have the following meanings:
DigestMechanism the mechanism
type of the digest mechanism to be used (possible types can be found in [WTLS])
RandomInfo Client’s
and server’s random data information
pVersion pointer
to a CK_BYTE which receives the WTLS protocol version information
CK_WTLS_MASTER_KEY_DERIVE_PARAMS_PTR is a pointer to
a CK_WTLS_MASTER_KEY_DERIVE_PARAMS.
¨ CK_WTLS_PRF_PARAMS;
CK_WTLS_PRF_PARAMS_PTR
CK_WTLS_PRF_PARAMS is a structure, which provides the
parameters to the CKM_WTLS_PRF mechanism. It is defined as follows:
typedef struct CK_WTLS_PRF_PARAMS {
CK_MECHANISM_TYPE DigestMechanism;
CK_BYTE_PTR pSeed;
CK_ULONG ulSeedLen;
CK_BYTE_PTR pLabel;
CK_ULONG ulLabelLen;
CK_BYTE_PTR pOutput;
CK_ULONG_PTR pulOutputLen;
} CK_WTLS_PRF_PARAMS;
The fields of the structure have the following meanings:
Digest Mechanism the mechanism
type of the digest mechanism to be used (possible types can be found in [WTLS])
pSeed pointer
to the input seed
ulSeedLen length in
bytes of the input seed
pLabel pointer
to the identifying label
ulLabelLen length in
bytes of the identifying label
pOutput pointer
receiving the output of the operation
pulOutputLen pointer to
the length in bytes that the output to be created shall have, has to hold the
desired length as input and will receive the calculated length as output
CK_WTLS_PRF_PARAMS_PTR is a pointer to a CK_WTLS_PRF_PARAMS.
¨
CK_WTLS_KEY_MAT_OUT;
CK_WTLS_KEY_MAT_OUT_PTR
CK_WTLS_KEY_MAT_OUT is a structure that contains the
resulting key handles and initialization vectors after performing a C_DeriveKey
function with the CKM_WTLS_SERVER_KEY_AND_MAC_DERIVE or with the CKM_WTLS_CLIENT_KEY_AND_MAC_DERIVE
mechanism. It is defined as follows:
typedef struct CK_WTLS_KEY_MAT_OUT {
CK_OBJECT_HANDLE hMacSecret;
CK_OBJECT_HANDLE hKey;
CK_BYTE_PTR pIV;
} CK_WTLS_KEY_MAT_OUT;
The fields of the structure have the following meanings:
hMacSecret Key handle
for the resulting MAC secret key
hKey Key
handle for the resulting secret key
pIV Pointer
to a location which receives the initialization vector (IV) created (if any)
CK_WTLS_KEY_MAT_OUT _PTR is a pointer to a CK_WTLS_KEY_MAT_OUT.
¨
CK_WTLS_KEY_MAT_PARAMS;
CK_WTLS_KEY_MAT_PARAMS_PTR
CK_WTLS_KEY_MAT_PARAMS is a structure that provides
the parameters to the CKM_WTLS_SERVER_KEY_AND_MAC_DERIVE and the CKM_WTLS_CLIENT_KEY_AND_MAC_DERIVE
mechanisms. It is defined as follows:
typedef struct CK_WTLS_KEY_MAT_PARAMS {
CK_MECHANISM_TYPE DigestMechanism;
CK_ULONG ulMacSizeInBits;
CK_ULONG ulKeySizeInBits;
CK_ULONG ulIVSizeInBits;
CK_ULONG ulSequenceNumber;
CK_BBOOL bIsExport;
CK_WTLS_RANDOM_DATA RandomInfo;
CK_WTLS_KEY_MAT_OUT_PTR pReturnedKeyMaterial;
} CK_WTLS_KEY_MAT_PARAMS;
The fields of the structure have the following meanings:
Digest Mechanism the mechanism
type of the digest mechanism to be used (possible types can be found in [WTLS])
ulMaxSizeInBits the length
(in bits) of the MACing key agreed upon during the protocol handshake phase
ulKeySizeInBits the length
(in bits) of the secret key agreed upon during the handshake phase
ulIVSizeInBits the
length (in bits) of the IV agreed upon during the handshake phase. If no IV is
required, the length should be set to 0.
ulSequenceNumber the current
sequence number used for records sent by the client and server respectively
bIsExport a
boolean value which indicates whether the keys have to be derives for an export
version of the protocol. If this value is true (i.e., the keys are exportable)
then ulKeySizeInBits is the length of the key in bits before expansion. The
length of the key after expansion is determined by the information found in the
template sent along with this mechanism during a C_DeriveKey function call
(either the CKA_KEY_TYPE or the CKA_VALUE_LEN attribute).
RandomInfo client’s
and server’s random data information
pReturnedKeyMaterial points to a CK_WTLS_KEY_MAT_OUT
structure which receives the handles for the keys generated and the IV
CK_WTLS_KEY_MAT_PARAMS_PTR is a pointer to a CK_WTLS_KEY_MAT_PARAMS.
Pre master secret key generation for the RSA key exchange
suite in WTLS denoted CKM_WTLS_PRE_MASTER_KEY_GEN, is a mechanism, which
generates a variable length secret key. It is used to produce the pre master
secret key for RSA key exchange suite used in WTLS. This mechanism returns a
handle to the pre master secret key.
It has one parameter, a CK_BYTE, which provides the
client’s WTLS version.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE
and CKA_VALUE attributes to the new key (as well as the CKA_VALUE_LEN
attribute, if it is not supplied in the template). Other attributes may be
specified in the template, or else are assigned default values.
The template sent along with this mechanism during a C_GenerateKey
call may indicate that the object class is CKO_SECRET_KEY, the key type
is CKK_GENERIC_SECRET, and the CKA_VALUE_LEN attribute indicates
the length of the pre master secret key.
For this mechanism, the ulMinKeySize field of the CK_MECHANISM_INFO
structure shall indicate 20 bytes.
Master secret derivation in WTLS, denoted CKM_WTLS_MASTER_KEY_DERIVE,
is a mechanism used to derive a 20 byte generic secret key from variable length
secret key. It is used to produce the master secret key used in WTLS from the
pre master secret key. This mechanism returns the value of the client version,
which is built into the pre master secret key as well as a handle to the
derived master secret key.
It has a parameter, a CK_WTLS_MASTER_KEY_DERIVE_PARAMS
structure, which allows for passing the mechanism type of the digest mechanism
to be used as well as the passing of random data to the token as well as the
returning of the protocol version number which is part of the pre master secret
key.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key (as well as the CKA_VALUE_LEN
attribute, if it is not supplied in the template). Other attributes may be
specified in the template, or else are assigned default values.
The template sent along with this mechanism during a C_DeriveKey
call may indicate that the object class is CKO_SECRET_KEY, the key type
is CKK_GENERIC_SECRET, and the CKA_VALUE_LEN attribute has value
20. However, since these facts are all implicit in the mechanism, there is no
need to specify any of them.
This mechanism has the following rules about key sensitivity
and extractability:
The CKA_SENSITIVE and CKA_EXTRACTABLE
attributes in the template for the new key can both be specified to be either
CK_TRUE or CK_FALSE. If omitted, these attributes each take on some default
value.
If the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_FALSE, then the derived key will as well. If the base key
has its CKA_ALWAYS_SENSITIVE attribute set to CK_TRUE, then the derived
key has its CKA_ALWAYS_SENSITIVE attribute set to the same value as its CKA_SENSITIVE
attribute.
Similarly, if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_FALSE, then the derived key will, too. If the base key has
its CKA_NEVER_EXTRACTABLE attribute set to CK_TRUE, then the derived key
has its CKA_NEVER_EXTRACTABLE attribute set to the opposite value
from its CKA_EXTRACTABLE attribute.
For this mechanism, the ulMinKeySize and ulMaxKeySize fields
of the CK_MECHANISM_INFO structure both indicate 20 bytes.
Note that the CK_BYTE pointed to by the CK_WTLS_MASTER_KEY_DERIVE_PARAMS
structure’s pVersion field will be modified by the C_DeriveKey call.
In particular, when the call returns, this byte will hold the WTLS version
associated with the supplied pre master secret key.
Note that this mechanism is only useable for key exchange
suites that use a 20-byte pre master secret key with an embedded version
number. This includes the RSA key exchange suites, but excludes the
Diffie-Hellman and Elliptic Curve Cryptography key exchange suites.
Master secret derivation for Diffie-Hellman and Elliptic
Curve Cryptography in WTLS, denoted CKM_WTLS_MASTER_KEY_DERIVE_DH_ECC,
is a mechanism used to derive a 20 byte generic secret key from variable length
secret key. It is used to produce the master secret key used in WTLS from the
pre master secret key. This mechanism returns a handle to the derived master
secret key.
It has a parameter, a CK_WTLS_MASTER_KEY_DERIVE_PARAMS
structure, which allows for the passing of the mechanism type of the digest
mechanism to be used as well as random data to the token. The pVersion field
of the structure must be set to NULL_PTR since the version number is not
embedded in the pre master secret key as it is for RSA-like key exchange
suites.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key (as well as the CKA_VALUE_LEN
attribute, if it is not supplied in the template). Other attributes may be
specified in the template, or else are assigned default values.
The template sent along with this mechanism during a C_DeriveKey
call may indicate that the object class is CKO_SECRET_KEY, the key type
is CKK_GENERIC_SECRET, and the CKA_VALUE_LEN attribute has value
20. However, since these facts are all implicit in the mechanism, there is no
need to specify any of them.
This mechanism has the following rules about key sensitivity
and extractability:
The CKA_SENSITIVE and CKA_EXTRACTABLE
attributes in the template for the new key can both be specified to be either
CK_TRUE or CK_FALSE. If omitted, these attributes each take on some default
value.
If the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_FALSE, then the derived key will as well. If the base key
has its CKA_ALWAYS_SENSITIVE attribute set to CK_TRUE, then the derived
key has its CKA_ALWAYS_SENSITIVE attribute set to the same value as its CKA_SENSITIVE
attribute.
Similarly, if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_FALSE, then the derived key will, too. If the base key has
its CKA_NEVER_EXTRACTABLE attribute set to CK_TRUE, then the derived key
has its CKA_NEVER_EXTRACTABLE attribute set to the opposite value
from its CKA_EXTRACTABLE attribute.
For this mechanism, the ulMinKeySize and ulMaxKeySize fields
of the CK_MECHANISM_INFO structure both indicate 20 bytes.
Note that this mechanism is only useable for key exchange
suites that do not use a fixed length 20-byte pre master secret key with an
embedded version number. This includes the Diffie-Hellman and Elliptic Curve
Cryptography key exchange suites, but excludes the RSA key exchange suites.
PRF (pseudo random function) in WTLS, denoted CKM_WTLS_PRF,
is a mechanism used to produce a securely generated pseudo-random output of arbitrary
length. The keys it uses are generic secret keys.
It has a parameter, a CK_WTLS_PRF_PARAMS structure,
which allows for passing the mechanism type of the digest mechanism to be used,
the passing of the input seed and its length, the passing of an identifying
label and its length and the passing of the length of the output to the token
and for receiving the output.
This mechanism produces securely generated pseudo-random
output of the length specified in the parameter.
This mechanism departs from the other
key derivation mechanisms in Cryptoki in not using the template sent along with
this mechanism during a C_DeriveKey function call, which means the
template shall be a NULL_PTR. For most key-derivation mechanisms, C_DeriveKey
returns a single key handle as a result of a successful completion. However,
since the CKM_WTLS_PRF mechanism returns the requested number of output
bytes in the CK_WTLS_PRF_PARAMS structure specified as the mechanism
parameter, the parameter phKey passed to C_DeriveKey is
unnecessary, and should be a NULL_PTR.
If a call to C_DeriveKey with this mechanism fails,
then no output will be generated.
Server key, MAC and IV derivation in WTLS, denoted CKM_WTLS_SERVER_KEY_AND_MAC_DERIVE,
is a mechanism used to derive the appropriate cryptographic keying material
used by a cipher suite from the master secret key and random data. This
mechanism returns the key handles for the keys generated in the process, as
well as the IV created.
It has a parameter, a CK_WTLS_KEY_MAT_PARAMS
structure, which allows for the passing of the mechanism type of the digest
mechanism to be used, random data, the characteristic of the cryptographic
material for the given cipher suite, and a pointer to a structure which
receives the handles and IV which were generated.
This mechanism contributes to the creation of two distinct
keys and returns one IV (if an IV is requested by the caller) back to the
caller. The keys are all given an object class of CKO_SECRET_KEY.
The MACing key (server write MAC secret) is always given a
type of CKK_GENERIC_SECRET. It is flagged as valid for signing,
verification and derivation operations.
The other key (server write key) is typed according to
information found in the template sent along with this mechanism during a C_DeriveKey
function call. By default, it is flagged as valid for encryption, decryption,
and derivation operations.
An IV (server write IV) will be generated and returned if
the ulIVSizeInBits field of the CK_WTLS_KEY_MAT_PARAMS field has
a nonzero value. If it is generated, its length in bits will agree with the
value in the ulIVSizeInBits field
Both keys inherit the values of the CKA_SENSITIVE, CKA_ALWAYS_SENSITIVE,
CKA_EXTRACTABLE, and CKA_NEVER_EXTRACTABLE attributes from the
base key. The template provided to C_DeriveKey may not specify values
for any of these attributes that differ from those held by the base key.
Note that the CK_WTLS_KEY_MAT_OUT structure pointed
to by the CK_WTLS_KEY_MAT_PARAMS structure’s pReturnedKeyMaterial
field will be modified by the C_DeriveKey call. In particular, the two
key handle fields in the CK_WTLS_KEY_MAT_OUT structure will be modified
to hold handles to the newly-created keys; in addition, the buffer pointed to
by the CK_WTLS_KEY_MAT_OUT structure’s pIV field will have the IV
returned in them (if an IV is requested by the caller). Therefore, this field
must point to a buffer with sufficient space to hold any IV that will be
returned.
This mechanism departs from the other key derivation
mechanisms in Cryptoki in its returned information. For most key-derivation
mechanisms, C_DeriveKey returns a single key handle as a result of a
successful completion. However, since the CKM_WTLS_SERVER_KEY_AND_MAC_DERIVE
mechanism returns all of its key handles in the CK_WTLS_KEY_MAT_OUT
structure pointed to by the CK_WTLS_KEY_MAT_PARAMS structure specified
as the mechanism parameter, the parameter phKey passed to C_DeriveKey
is unnecessary, and should be a NULL_PTR.
If a call to C_DeriveKey with this mechanism fails,
then none of the two keys will be created.
Client key, MAC and IV derivation in WTLS, denoted CKM_WTLS_CLIENT_KEY_AND_MAC_DERIVE,
is a mechanism used to derive the appropriate cryptographic keying material
used by a cipher suite from the master secret key and random data. This
mechanism returns the key handles for the keys generated in the process, as
well as the IV created.
It has a parameter, a CK_WTLS_KEY_MAT_PARAMS
structure, which allows for the passing of the mechanism type of the digest
mechanism to be used, random data, the characteristic of the cryptographic
material for the given cipher suite, and a pointer to a structure which
receives the handles and IV which were generated.
This mechanism contributes to the creation of two distinct
keys and returns one IV (if an IV is requested by the caller) back to the
caller. The keys are all given an object class of CKO_SECRET_KEY.
The MACing key (client write MAC secret) is always given a
type of CKK_GENERIC_SECRET. It is flagged as valid for signing,
verification and derivation operations.
The other key (client write key) is typed according to
information found in the template sent along with this mechanism during a C_DeriveKey
function call. By default, it is flagged as valid for encryption, decryption,
and derivation operations.
An IV (client write IV) will be generated and returned if
the ulIVSizeInBits field of the CK_WTLS_KEY_MAT_PARAMS field has
a nonzero value. If it is generated, its length in bits will agree with the
value in the ulIVSizeInBits field
Both keys inherit the values of the CKA_SENSITIVE, CKA_ALWAYS_SENSITIVE,
CKA_EXTRACTABLE, and CKA_NEVER_EXTRACTABLE attributes from the
base key. The template provided to C_DeriveKey may not specify values
for any of these attributes that differ from those held by the base key.
Note that the CK_WTLS_KEY_MAT_OUT structure pointed
to by the CK_WTLS_KEY_MAT_PARAMS structure’s pReturnedKeyMaterial
field will be modified by the C_DeriveKey call. In particular, the two
key handle fields in the CK_WTLS_KEY_MAT_OUT structure will be modified
to hold handles to the newly-created keys; in addition, the buffer pointed to
by the CK_WTLS_KEY_MAT_OUT structure’s pIV field will have the IV
returned in them (if an IV is requested by the caller). Therefore, this field
must point to a buffer with sufficient space to hold any IV that will be
returned.
This mechanism departs from the other key derivation
mechanisms in Cryptoki in its returned information. For most key-derivation
mechanisms, C_DeriveKey returns a single key handle as a result of a
successful completion. However, since the CKM_WTLS_CLIENT_KEY_AND_MAC_DERIVE
mechanism returns all of its key handles in the CK_WTLS_KEY_MAT_OUT
structure pointed to by the CK_WTLS_KEY_MAT_PARAMS structure specified
as the mechanism parameter, the parameter phKey passed to C_DeriveKey
is unnecessary, and should be a NULL_PTR.
If a call to C_DeriveKey with this mechanism fails,
then none of the two keys will be created.
NIST SP800-108 defines three types of key derivation
functions (KDF); a Counter Mode KDF, a Feedback Mode KDF and a Double Pipeline
Mode KDF.
This section defines a unique mechanism for each type of
KDF. These mechanisms can be used to derive one or more symmetric keys from a
single base symmetric key.
The KDFs defined in SP800-108 are all built upon pseudo
random functions (PRF). In general terms, the PRFs accepts two pieces of
input; a base key and some input data. The base key is taken from the hBaseKey
parameter to C_Derive. The input data is constructed from an iteration
variable (internally defined by the KDF/PRF) and the data provided in the CK_ PRF_DATA_PARAM
array that is part of the mechanism parameter.
Table
191, SP800-108 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_SP800_108_COUNTER_KDF
|
|
|
|
|
|
|
ü
|
|
CKM_SP800_108_FEEDBACK_KDF
|
|
|
|
|
|
|
ü
|
|
CKM_SP800_108_DOUBLE_PIPELINE_KDF
|
|
|
|
|
|
|
ü
|
For these mechanisms, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the minimum and
maximum supported base key size in bits. Note, these mechanisms support
multiple PRF types and key types; as such the values reported by ulMinKeySize
and ulMaxKeySize specify the minimum and maximum supported base key size when
all PRF and keys types are considered. For example, a Cryptoki implementation
may support CKK_GENERIC_SECRET keys that can be as small as 8-bits in length
and therefore ulMinKeySize could report 8-bits. However, for an AES-CMAC PRF
the base key must be of type CKK_AES and must be either 16-bytes, 24-bytes or
32-bytes in lengths and therefore the value reported by ulMinKeySize could be
misleading. Depending on the PRF type selected, additional key size
restrictions may apply.
Mechanisms:
CKM_SP800_108_COUNTER_KDF
CKM_SP800_108_FEEDBACK_KDF
CKM_SP800_108_DOUBLE_PIPELINE_KDF
Data Field Types:
CK_SP800_108_ITERATION_VARIABLE
CK_SP800_108_COUNTER
CK_SP800_108_DKM_LENGTH
CK_SP800_108_BYTE_ARRAY
DKM Length Methods:
CK_SP800_108_DKM_LENGTH_SUM_OF_KEYS
CK_SP800_108_DKM_LENGTH_SUM_OF_SEGMENTS
¨
CK_SP800_108_PRF_TYPE
The CK_SP800_108_PRF_TYPE field of the mechanism
parameter is used to specify the type of PRF that is to be used. It is defined
as follows:
typedef CK_MECHANISM_TYPE
CK_SP800_108_PRF_TYPE;
The CK_SP800_108_PRF_TYPE field reuses the existing
mechanisms definitions. The following table lists the supported PRF types:
Table 192, SP800-108 Pseudo
Random Functions
|
Pseudo Random Function
Identifiers
|
|
CKM_SHA_1_HMAC
|
|
CKM_SHA224_HMAC
|
|
CKM_SHA256_HMAC
|
|
CKM_SHA384_HMAC
|
|
CKM_SHA512_HMAC
|
|
CKM_SHA3_224_HMAC
|
|
CKM_SHA3_256_HMAC
|
|
CKM_SHA3_384_HMAC
|
|
CKM_SHA3_512_HMAC
|
|
CKM_DES3_CMAC
|
|
CKM_AES_CMAC
|
¨
CK_PRF_DATA_TYPE
Each mechanism parameter contains an array of CK_PRF_DATA_PARAM
structures. The CK_PRF_DATA_PARAM structure contains CK_PRF_DATA_TYPE
field. The CK_PRF_DATA_TYPE field is used to identify the type of data
identified by each CK_PRF_DATA_PARAM element in the array. Depending on
the type of KDF used, some data field types are mandatory, some data field
types are optional and some data field types are not allowed. These
requirements are defined on a per-mechanism basis in the sections below. The CK_PRF_DATA_TYPE
is defined as follows:
typedef CK_ULONG CK_PRF_DATA_TYPE;
The following table lists all of the supported data field
types:
Table 193, SP800-108 PRF Data
Field Types
|
Data
Field Identifier
|
Description
|
|
CK_SP800_108_ITERATION_VARIABLE
|
Identifies
the iteration variable defined internally by the KDF.
|
|
CK_SP800_108_COUNTER
|
Identifies
an optional counter value represented as a binary string. Exact formatting
of the counter value is defined by the CK_SP800_108_COUNTER_FORMAT
structure. The value of the counter is defined by the KDF’s internal loop
counter.
|
|
CK_SP800_108_DKM_LENGTH
|
Identifies
the length in bits of the derived keying material (DKM) represented as a
binary string. Exact formatting of the length value is defined by the
CK_SP800_108_DKM_LENGTH_FORMAT structure.
|
|
CK_SP800_108_BYTE_ARRAY
|
Identifies a generic byte
array of data. This data type can be used to provide “context”, “label”,
“separator bytes” as well as any other type of encoding information required
by the higher level protocol.
|
¨
CK_PRF_DATA_PARAM
CK_PRF_DATA_PARAM is used to define a segment of
input for the PRF. Each mechanism parameter supports an array of CK_PRF_DATA_PARAM
structures. The CK_PRF_DATA_PARAM is defined as follows:
typedef struct CK_PRF_DATA_PARAM
{
CK_PRF_DATA_TYPE type;
CK_VOID_PTR pValue;
CK_ULONG ulValueLen;
} CK_PRF_DATA_PARAM;
typedef
CK_PRF_DATA_PARAM CK_PTR CK_PRF_DATA_PARAM_PTR
The fields of
the CK_PRF_DATA_PARAM structure have the following meaning:
type defines
the type of data pointed to by pValue
pValue pointer
to the data defined by type
ulValueLen size of
the data pointed to by pValue
If the type
field of the CK_PRF_DATA_PARAM structure is set to
CK_SP800_108_ITERATION_VARIABLE, then pValue must be set the appropriate
value for the KDF’s iteration variable type. For the Counter Mode KDF, pValue
must be assigned a valid CK_SP800_108_COUNTER_FORMAT_PTR and ulValueLen
must be set to sizeof(CK_SP800_108_COUNTER_FORMAT). For all other KDF types, pValue
must be set to NULL_PTR and ulValueLen must be set to 0.
If the type
field of the CK_PRF_DATA_PARAM structure is set to CK_SP800_108_COUNTER,
then pValue must be assigned a valid CK_SP800_108_COUNTER_FORMAT_PTR and
ulValueLen must be set to sizeof(CK_SP800_108_COUNTER_FORMAT).
If the type
field of the CK_PRF_DATA_PARAM structure is set to
CK_SP800_108_DKM_LENGTH then pValue must be assigned a valid
CK_SP800_108_DKM_LENGTH_FORMAT_PTR and ulValueLen must be set to
sizeof(CK_SP800_108_DKM_LENGTH_FORMAT).
If the type
field of the CK_PRF_DATA_PARAM structure is set to
CK_SP800_108_BYTE_ARRAY, then pValue must be assigned a valid
CK_BYTE_PTR value and ulValueLen must be set to a non-zero length.
¨
CK_SP800_108_COUNTER_FORMAT
CK_SP800_108_COUNTER_FORMAT is used to define the
encoding format for a counter value. The CK_SP800_108_COUNTER_FORMAT is
defined as follows:
typedef struct CK_SP800_108_COUNTER_FORMAT
{
CK_BBOOL bLittleEndian;
CK_ULONG ulWidthInBits;
} CK_SP800_108_COUNTER_FORMAT;
typedef
CK_SP800_108_COUNTER_FORMAT CK_PTR CK_SP800_108_COUNTER_FORMAT_PTR
The fields of
the CK_SP800_108_COUNTER_FORMAT structure have the following meaning:
bLittleEndian defines
if the counter should be represented in Big Endian or Little Endian format
ulWidthInBits defines
the number of bits used to represent the counter value
¨
CK_SP800_108_DKM_LENGTH_METHOD
CK_SP800_108_DKM_LENGTH_METHOD is used to define how
the DKM length value is calculated. The CK_SP800_108_DKM_LENGTH_METHOD
type is defined as follows:
typedef CK_ULONG
CK_SP800_108_DKM_LENGTH_METHOD;
The following table lists all of the supported DKM Length
Methods:
Table 194, SP800-108 DKM Length
Methods
|
DKM Length Method Identifier
|
Description
|
|
CK_SP800_108_DKM_LENGTH_SUM_OF_KEYS
|
Specifies that the DKM
length should be set to the sum of the length of all keys derived by this
invocation of the KDF.
|
|
CK_SP800_108_DKM_LENGTH_SUM_OF_SEGMENTS
|
Specifies that the DKM
length should be set to the sum of the length of all segments of output
produced by the PRF by this invocation of the KDF.
|
¨
CK_SP800_108_DKM_LENGTH_FORMAT
CK_SP800_108_DKM_LENGTH_FORMAT is used to define the
encoding format for the DKM length value. The CK_SP800_108_DKM_LENGTH_FORMAT
is defined as follows:
typedef struct CK_SP800_108_DKM_LENGTH_FORMAT
{
CK_SP800_108_DKM_LENGTH_METHOD dkmLengthMethod;
CK_BBOOL bLittleEndian;
CK_ULONG ulWidthInBits;
} CK_SP800_108_DKM_LENGTH_FORMAT;
typedef
CK_SP800_108_DKM_LENGTH_FORMAT CK_PTR CK_SP800_108_DKM_LENGTH_FORMAT_PTR
The fields of
the CK_SP800_108_DKM_LENGTH_FORMAT structure have the following meaning:
dkmLengthMethod defines the
method used to calculate the DKM length value
bLittleEndian defines
if the DKM length value should be represented in Big Endian or Little Endian
format
ulWidthInBits defines
the number of bits used to represent the DKM length value
¨
CK_DERIVED_KEY
CK_DERIVED_KEY is used to define an additional key to
be derived as well as provide a CK_OBJECT_HANDLE_PTR to receive the handle for
the derived keys. The CK_DERIVED_KEY is defined as follows:
typedef struct CK_DERIVED_KEY
{
CK_ATTRIBUTE_PTR pTemplate;
CK_ULONG ulAttributeCount;
CK_OBJECT_HANDLE_PTR phKey;
} CK_DERIVED_KEY;
typedef
CK_DERIVED_KEY CK_PTR CK_DERIVED_KEY_PTR
The fields of
the CK_DERIVED_KEY structure have the following meaning:
pTemplate pointer
to a template that defines a key to derive
ulAttributeCount number of
attributes in the template pointed to by pTemplate
phKey pointer
to receive the handle for a derived key
¨
CK_SP800_108_KDF_PARAMS,
CK_SP800_108_KDF_PARAMS_PTR
CK_SP800_108_KDF_PARAMS
is a structure that provides the parameters for the CKM_SP800_108_COUNTER_KDF
and CKM_SP800_108_DOUBLE_PIPELINE_KDF mechanisms.
typedef struct CK_SP800_108_KDF_PARAMS
{
CK_SP800_108_PRF_TYPE prfType;
CK_ULONG ulNumberOfDataParams;
CK_PRF_DATA_PARAM_PTR pDataParams;
CK_ULONG ulAdditionalDerivedKeys;
CK_DERIVED_KEY_PTR pAdditionalDerivedKeys;
}
CK_SP800_108_KDF_PARAMS;
typedef CK_SP800_108_KDF_PARAMS CK_PTR CK_SP800_108_KDF_PARAMS_PTR;
The fields of
the CK_SP800_108_KDF_PARAMS structure have the following meaning:
prfType type of
PRF
ulNumberOfDataParams number of elements
in the array pointed to by pDataParams
pDataParams an array of
CK_PRF_DATA_PARAM structures. The array defines input parameters that are used
to construct the “data” input to the PRF.
ulAdditionalDerivedKeys number of
additional keys that will be derived and the number of elements in the array
pointed to by pAdditionalDerivedKeys. If pAdditionalDerivedKeys is set to
NULL_PTR, this parameter must be set to 0.
pAdditionalDerivedKeys an array of
CK_DERIVED_KEY structures. If ulAdditionalDerivedKeys is set to 0, this
parameter must be set to NULL_PTR
¨
CK_SP800_108_FEEDBACK_KDF_PARAMS,
CK_SP800_108_FEEDBACK_KDF_PARAMS_PTR
The CK_SP800_108_FEEDBACK_KDF_PARAMS structure
provides the parameters for the CKM_SP800_108_FEEDBACK_KDF mechanism. It is
defined as follows:
typedef
struct CK_SP800_108_FEEDBACK_KDF_PARAMS
{
CK_SP800_108_PRF_TYPE prfType;
CK_ULONG ulNumberOfDataParams;
CK_PRF_DATA_PARAM_PTR pDataParams;
CK_ULONG ulIVLen;
CK_BYTE_PTR pIV;
CK_ULONG ulAdditionalDerivedKeys;
CK_DERIVED_KEY_PTR pAdditionalDerivedKeys;
} CK_SP800_108_FEEDBACK_KDF_PARAMS;
typedef CK_SP800_108_FEEDBACK_KDF_PARAMS CK_PTR
CK_SP800_108_FEEDBACK_KDF_PARAMS_PTR;
The fields of
the CK_SP800_108_FEEDBACK_KDF_PARAMS structure have the following
meaning:
prfType type of
PRF
ulNumberOfDataParams number of elements
in the array pointed to by pDataParams
pDataParams an array of
CK_PRF_DATA_PARAM structures. The array defines input parameters that are used
to construct the “data” input to the PRF.
ulIVLen the
length in bytes of the IV. If pIV is set to NULL_PTR, this parameter must be
set to 0.
pIV an
array of bytes to be used as the IV for the feedback mode KDF. This parameter
is optional and can be set to NULL_PTR. If ulIVLen is set to 0, this parameter
must be set to NULL_PTR.
ulAdditionalDerivedKeys number of
additional keys that will be derived and the number of elements in the array
pointed to by pAdditionalDerivedKeys. If pAdditionalDerivedKeys is set to
NULL_PTR, this parameter must be set to 0.
pAdditionalDerivedKeys an array of
CK_DERIVED_KEY structures. If ulAdditionalDerivedKeys is set to 0, this
parameter must be set to NULL_PTR.
The SP800-108 Counter Mode KDF mechanism, denoted CKM_SP800_108_COUNTER_KDF,
represents the KDF defined SP800-108 section 5.1. CKM_SP800_108_COUNTER_KDF
is a mechanism for deriving one or more symmetric keys from a symmetric base
key.
It has a parameter, a CK_SP800_108_KDF_PARAMS
structure.
The following table lists the data field types that are supported
for this KDF type and their meaning:
Table 195, Counter Mode data
field requirements
|
Data Field Identifier
|
Description
|
|
CK_SP800_108_ITERATION_VARIABLE
|
This data field type is
mandatory.
This data field type
identifies the location of the iteration variable in the constructed PRF
input data.
The iteration variable for
this KDF type is a counter.
Exact formatting of the
counter value is defined by the CK_SP800_108_COUNTER_FORMAT structure.
|
|
CK_SP800_108_COUNTER
|
This data field type is
invalid for this KDF type.
|
|
CK_SP800_108_DKM_LENGTH
|
This data field type is
optional.
This data field type
identifies the location of the DKM length in the constructed PRF input data.
Exact formatting of the DKM
length is defined by the CK_SP800_108_DKM_LENGTH_FORMAT structure.
If specified, only one
instance of this type may be specified.
|
|
CK_SP800_108_BYTE_ARRAY
|
This data field type is
optional.
This data field type
identifies the location and value of a byte array of data in the constructed
PRF input data.
This standard does not restrict the number of instances of
this data type.
|
SP800-108 limits the amount of derived keying material that
can be produced by a Counter Mode KDF by limiting the internal loop counter to
(2r−1), where “r” is the number of bits used to represent the
counter. Therefore the maximum number of bits that can be produced is (2r−1)h,
where “h” is the length in bits of the output of the selected PRF.
6.42.4
Feedback Mode KDF
The SP800-108 Feedback Mode KDF mechanism, denoted CKM_SP800_108_FEEDBACK_KDF,
represents the KDF defined SP800-108 section 5.2. CKM_SP800_108_FEEDBACK_KDF
is a mechanism for deriving one or more symmetric keys from a symmetric base
key.
It has a parameter, a CK_SP800_108_FEEDBACK_KDF_PARAMS
structure.
The following table lists the data field types that are
supported for this KDF type and their meaning:
Table 196, Feedback Mode data
field requirements
|
Data Field Identifier
|
Description
|
|
CK_SP800_108_ITERATION_VARIABLE
|
This data field type is
mandatory.
This data field type
identifies the location of the iteration variable in the constructed PRF
input data.
The iteration variable is
defined as K(i-1) in section 5.2 of SP800-108.
The size, format and value
of this data input is defined by the internal KDF structure and PRF output.
Exact formatting of the
counter value is defined by the CK_SP800_108_COUNTER_FORMAT structure.
|
|
CK_SP800_108_COUNTER
|
This data field type is optional.
This data field type
identifies the location of the counter in the constructed PRF input data.
Exact formatting of the
counter value is defined by the CK_SP800_108_COUNTER_FORMAT structure.
If specified, only one
instance of this type may be specified.
|
|
CK_SP800_108_DKM_LENGTH
|
This data field type is
optional.
This data field type
identifies the location of the DKM length in the constructed PRF input data.
Exact formatting of the DKM
length is defined by the CK_SP800_108_DKM_LENGTH_FORMAT structure.
If specified, only one
instance of this type may be specified.
|
|
CK_SP800_108_BYTE_ARRAY
|
This data field type is
optional.
This data field type
identifies the location and value of a byte array of data in the constructed
PRF input data.
This standard does not restrict the number of instances of
this data type.
|
SP800-108 limits the amount of derived keying material that
can be produced by a Feedback Mode KDF by limiting the internal loop counter to
(232−1). Therefore the maximum number of bits that can be
produced is (232−1)h, where “h” is the length in bits of the
output of the selected PRF.
The SP800-108 Double Pipeline Mode KDF mechanism, denoted CKM_SP800_108_DOUBLE_PIPELINE_KDF,
represents the KDF defined SP800-108 section 5.3. CKM_SP800_108_DOUBLE_PIPELINE_KDF
is a mechanism for deriving one or more symmetric keys from a symmetric base
key.
It has a parameter, a CK_SP800_108_KDF_PARAMS structure.
The following table lists the data field types that are
supported for this KDF type and their meaning:
Table 197, Double Pipeline Mode
data field requirements
|
Data Field Identifier
|
Description
|
|
CK_SP800_108_ITERATION_VARIABLE
|
This data field type is mandatory.
This data field type
identifies the location of the iteration variable in the constructed PRF
input data.
The iteration variable is
defined as A(i) in section 5.3 of SP800-108.
The size, format and value
of this data input is defined by the internal KDF structure and PRF output.
Exact formatting of the
counter value is defined by the CK_SP800_108_COUNTER_FORMAT structure.
|
|
CK_SP800_108_COUNTER
|
This data field type is
optional.
This data field type
identifies the location of the counter in the constructed PRF input data.
Exact formatting of the
counter value is defined by the CK_SP800_108_COUNTER_FORMAT structure.
If specified, only one
instance of this type may be specified.
|
|
CK_SP800_108_DKM_LENGTH
|
This data field type is
optional.
This data field type
identifies the location of the DKM length in the constructed PRF input data.
Exact formatting of the DKM
length is defined by the CK_SP800_108_DKM_LENGTH_FORMAT structure.
If specified, only one
instance of this type may be specified.
|
|
CK_SP800_108_BYTE_ARRAY
|
This data field type is
optional.
This data field type
identifies the location and value of a byte array of data in the constructed
PRF input data.
This standard does not restrict the number of instances of
this data type.
|
SP800-108 limits the amount of derived keying material that
can be produced by a Double-Pipeline Mode KDF by limiting the internal loop
counter to (232−1). Therefore the maximum number of bits that
can be produced is (232−1)h, where “h” is the length in bits
of the output of the selected PRF.
The Double Pipeline KDF requires an internal IV value. The
IV is constructed using the same method used to construct the PRF input data;
the data/values identified by the array of CK_PRF_DATA_PARAM structures
are concatenated in to a byte array that is used as the IV. As shown in
SP800-108 section 5.3, the CK_SP800_108_ITERATION_VARIABLE and
CK_SP800_108_COUNTER data field types are not included in IV construction
process. All other data field types are included in the construction process.
The KDFs defined in this section can be used to derive more
than one symmetric key from the base key. The C_Derive function accepts
one CK_ATTRIBUTE_PTR to define a single derived key and one CK_OBJECT_HANDLE_PTR
to receive the handle for the derived key.
To derive additional keys, the mechanism parameter structure
can be filled in with one or more CK_DERIVED_KEY structures. Each structure
contains a CK_ATTRIBUTE_PTR to define a derived key and a CK_OBJECT_HANDLE_PTR
to receive the handle for the additional derived keys. The key defined by the C_Derive
function parameters is always derived before the keys defined by the
CK_DERIVED_KEY array that is part of the mechanism parameter. The additional
keys that are defined by the CK_DERIVED_KEY array are derived in the order they
are defined in the array. That is to say that the derived keying material
produced by the KDF is processed from left to right, and bytes are assigned
first to the key defined by the C_Derive function parameters, and then
bytes are assigned to the keys that are defined by the CK_DERIVED_KEY array in
the order they are defined in the array.
Each internal iteration of a KDF produces a unique segment
of PRF output. Sometimes, a single iteration will produce enough keying
material for the key being derived. Other times, additional internal
iterations are performed to produce multiple segments which are concatenated
together to produce enough keying material for the derived key(s).
When deriving multiple keys, no key can be created using
part of a segment that was used for another key. All keys must be created from
disjoint segments. For example, if the parameters are defined such that a
48-byte key (defined by the C_Derive function parameters) and a 16-byte
key (defined by the content of CK_DERIVED_KEY) are to be derived using CKM_SHA256_HMAC
as a PRF, three internal iterations of the KDF will be performed and three
segments of PRF output will be produced. The first segment and half of the
second segment will be used to create the 48-byte key and the third segment
will be used to create the 16-byte key.
In the above example, if the CK_SP800_108_DKM_LENGTH data
field type is specified with method CK_SP800_108_DKM_LENGTH_SUM_OF_KEYS, then
the DKM length value will be 512 bits. If the CK_SP800_108_DKM_LENGTH data
field type is specified with method CK_SP800_108_DKM_LENGTH_SUM_OF_SEGMENTS,
then the DKM length value will be 768 bits.
When deriving multiple keys, if any of the keys cannot be
derived for any reason, none of the keys shall be derived. If the failure was
caused by the content of a specific key’s template (ie the template defined by
the content of pTemplate), the corresponding phKey value will be
set to CK_INVALID_HANDLE to identify the offending template.
The CKM_SP800_108_COUNTER_KDF, CKM_SP800_108_FEEDBACK_KDF
and CKM_SP800_108_DOUBLE_PIPELINE_KDF mechanisms have the following
rules about key sensitivity and extractability:
·
The CKA_SENSITIVE and CKA_EXTRACTABLE attributes in
the template for the new key(s) can both be specified to be either CK_TRUE or
CK_FALSE. If omitted, these attributes each take on some default value.
·
If the base key has its CKA_ALWAYS_SENSITIVE attribute set
to CK_FALSE, then the derived key will as well. If the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE, then the derived key has its CKA_ALWAYS_SENSITIVE
attribute set to the same value as its CKA_SENSITIVE attribute.
·
Similarly, if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_FALSE, then the derived key will, too. If the base key has
its CKA_NEVER_EXTRACTABLE attribute set to CK_TRUE, then the derived key
has its CKA_NEVER_EXTRACTABLE attribute set to the opposite value
from its CKA_EXTRACTABLE attribute.
6.42.8
Constructing PRF Input Data
SP800-108 defines the PRF input data for each KDF at a high
level using terms like “label”, “context”, “separator”, “counter”…etc. The
value, formatting and order of the input data is not strictly defined by
SP800-108, instead it is described as being defined by the “encoding scheme”.
To support any encoding scheme, these mechanisms construct
the PRF input data from from the array of CK_PRF_DATA_PARAM structures in the
mechanism parameter. All of the values defined by the CK_PRF_DATA_PARAM array
are concatenated in the order they are defined and passed in to the PRF as the
data parameter.
SP800-108 section 5.1 outlines a sample Counter Mode KDF
which defines the following PRF input:
PRF (KI, [i]2 || Label || 0x00 || Context || [L]2)
Section 5.1 does not define the number of bits used to
represent the counter (the “r” value) or the DKM length (the “L” value), so
16-bits is assumed for both cases. The following sample code shows how to
define this PRF input data using an array of CK_PRF_DATA_PARAM structures.
#define DIM(a) (sizeof((a))/sizeof((a)[0]))
CK_OBJECT_HANDLE hBaseKey;
CK_OBJECT_HANDLE hDerivedKey;
CK_ATTRIBUTE derivedKeyTemplate =
{ … };
CK_BYTE baLabel[] = {0xde, 0xad,
0xbe , 0xef};
CK_ULONG ulLabelLen = sizeof(baLabel);
CK_BYTE baContext[] = {0xfe,
0xed, 0xbe , 0xef};
CK_ULONG ulContextLen = sizeof(baContext);
CK_SP800_108_COUNTER_FORMAT
counterFormat = {0, 16};
CK_SP800_108_DKM_LENGTH_FORMAT
dkmFormat
= {CK_SP800_108_DKM_LENGTH_SUM_OF_KEYS, 0, 16};
CK_PRF_DATA_PARAM dataParams[] =
{
{
CK_SP800_108_ITERATION_VARIABLE,
&counterFormat, sizeof(counterFormat)
},
{ CK_SP800_108_BYTE_ARRAY,
baLabel, ulLabelLen },
{ CK_SP800_108_BYTE_ARRAY,
{0x00}, 1 },
{ CK_SP800_108_BYTE_ARRAY,
baContext, ulContextLen },
{ CK_SP800_108_DKM_LENGTH,
dkmFormat, sizeof(dkmFormat) }
};
CK_SP800_108_KDF_PARAMS kdfParams
=
{
CKM_AES_CMAC,
DIM(dataParams),
&dataParams,
0, /* no addition
derived keys */
NULL /* no addition
derived keys */
};
CK_MECHANISM = mechanism
{
CKM_SP800_108_COUNTER_KDF,
&kdfParams,
sizeof(kdfParams)
};
hBaseKey = GetBaseKeyHandle(.....);
rv = C_DeriveKey(
hSession,
&mechanism,
hBaseKey,
&derivedKeyTemplate,
DIM(derivedKeyTemplate),
&hDerivedKey);
The SCP03 standard defines a variation of a counter mode KDF
which defines the following PRF input:
PRF (KI, Label || 0x00 || [L]2 || [i]2 ||
Context)
SCP03 defines the number of bits used to represent the
counter (the “r” value) and number of bits used to represent the DKM length
(the “L” value) as 16-bits. The following sample code shows how to define this
PRF input data using an array of CK_PRF_DATA_PARAM structures.
#define DIM(a) (sizeof((a))/sizeof((a)[0]))
CK_OBJECT_HANDLE hBaseKey;
CK_OBJECT_HANDLE hDerivedKey;
CK_ATTRIBUTE derivedKeyTemplate =
{ … };
CK_BYTE baLabel[] = {0xde, 0xad,
0xbe , 0xef};
CK_ULONG ulLabelLen = sizeof(baLabel);
CK_BYTE baContext[] = {0xfe,
0xed, 0xbe , 0xef};
CK_ULONG ulContextLen = sizeof(baContext);
CK_SP800_108_COUNTER_FORMAT
counterFormat = {0, 16};
CK_SP800_108_DKM_LENGTH_FORMAT
dkmFormat
=
{CK_SP800_108_DKM_LENGTH_SUM_OF_KEYS, 0, 16};
CK_PRF_DATA_PARAM dataParams[] =
{
{ CK_SP800_108_BYTE_ARRAY,
baLabel, ulLabelLen },
{ CK_SP800_108_BYTE_ARRAY,
{0x00}, 1 },
{ CK_SP800_108_DKM_LENGTH,
dkmFormat, sizeof(dkmFormat) },
{
CK_SP800_108_ITERATION_VARIABLE,
&counterFormat, sizeof(counterFormat)
},
{ CK_SP800_108_BYTE_ARRAY,
baContext, ulContextLen }
};
CK_SP800_108_KDF_PARAMS kdfParams
=
{
CKM_AES_CMAC,
DIM(dataParams),
&dataParams,
0, /* no addition
derived keys */
NULL /* no addition
derived keys */
};
CK_MECHANISM = mechanism
{
CKM_SP800_108_COUNTER_KDF,
&kdfParams,
sizeof(kdfParams)
};
hBaseKey = GetBaseKeyHandle(.....);
rv = C_DeriveKey(
hSession,
&mechanism,
hBaseKey,
&derivedKeyTemplate,
DIM(derivedKeyTemplate),
&hDerivedKey);
SP800-108 section 5.2 outlines a sample Feedback Mode KDF
which defines the following PRF input:
PRF (KI, K(i-1) {|| [i]2 }|| Label ||
0x00 || Context || [L]2)
Section 5.2 does not define the number of bits used to
represent the counter (the “r” value) or the DKM length (the “L” value), so
16-bits is assumed for both cases. The counter is defined as being optional
and is included in this example. The following sample code shows how to define
this PRF input data using an array of CK_PRF_DATA_PARAM structures.
#define DIM(a) (sizeof((a))/sizeof((a)[0]))
CK_OBJECT_HANDLE hBaseKey;
CK_OBJECT_HANDLE hDerivedKey;
CK_ATTRIBUTE derivedKeyTemplate =
{ … };
CK_BYTE baFeedbackIV[] =
{0x01, 0x02, 0x03, 0x04};
CK_ULONG ulFeedbackIVLen =
sizeof(baFeedbackIV);
CK_BYTE baLabel[] = {0xde,
0xad, 0xbe, 0xef};
CK_ULONG ulLabelLen = sizeof(baLabel);
CK_BYTE baContext[] =
{0xfe, 0xed, 0xbe, 0xef};
CK_ULONG ulContextLen = sizeof(baContext);
CK_SP800_108_COUNTER_FORMAT
counterFormat = {0, 16};
CK_SP800_108_DKM_LENGTH_FORMAT
dkmFormat
=
{CK_SP800_108_DKM_LENGTH_SUM_OF_KEYS, 0, 16};
CK_PRF_DATA_PARAM dataParams[] =
{
{
CK_SP800_108_ITERATION_VARIABLE,
&counterFormat, sizeof(counterFormat)
},
{ CK_SP800_108_BYTE_ARRAY,
baLabel, ulLabelLen },
{ CK_SP800_108_BYTE_ARRAY,
{0x00}, 1 },
{ CK_SP800_108_BYTE_ARRAY,
baContext, ulContextLen },
{ CK_SP800_108_DKM_LENGTH,
dkmFormat, sizeof(dkmFormat) }
};
CK_SP800_108_FEEDBACK_KDF_PARAMS
kdfParams =
{
CKM_AES_CMAC,
DIM(dataParams),
&dataParams,
ulFeedbackIVLen,
baFeedbackIV,
0, /* no addition
derived keys */
NULL /* no addition
derived keys */
};
CK_MECHANISM = mechanism
{
CKM_SP800_108_FEEDBACK_KDF,
&kdfParams,
sizeof(kdfParams)
};
hBaseKey = GetBaseKeyHandle(.....);
rv = C_DeriveKey(
hSession,
&mechanism,
hBaseKey,
&derivedKeyTemplate,
DIM(derivedKeyTemplate),
&hDerivedKey);
SP800-108 section 5.3 outlines a sample Double-Pipeline Mode
KDF which defines the two following PRF inputs:
PRF (KI, A(i-1))
PRF (KI, K(i-1) {|| [i]2 }|| Label ||
0x00 || Context || [L]2)
Section 5.3 does not define the number of bits used to represent
the counter (the “r” value) or the DKM length (the “L” value), so 16-bits is
assumed for both cases. The counter is defined as being optional so it is left
out in this example. The following sample code shows how to define this PRF
input data using an array of CK_PRF_DATA_PARAM structures.
#define DIM(a) (sizeof((a))/sizeof((a)[0]))
CK_OBJECT_HANDLE hBaseKey;
CK_OBJECT_HANDLE hDerivedKey;
CK_ATTRIBUTE derivedKeyTemplate =
{ … };
CK_BYTE baLabel[] = {0xde,
0xad, 0xbe , 0xef};
CK_ULONG ulLabelLen = sizeof(baLabel);
CK_BYTE baContext[] =
{0xfe, 0xed, 0xbe , 0xef};
CK_ULONG ulContextLen = sizeof(baContext);
CK_SP800_108_DKM_LENGTH_FORMAT
dkmFormat
=
{CK_SP800_108_DKM_LENGTH_SUM_OF_KEYS, 0, 16};
CK_PRF_DATA_PARAM dataParams[] =
{
{ CK_SP800_108_BYTE_ARRAY,
baLabel, ulLabelLen },
{ CK_SP800_108_BYTE_ARRAY,
{0x00}, 1 },
{ CK_SP800_108_BYTE_ARRAY,
baContext, ulContextLen },
{ CK_SP800_108_DKM_LENGTH,
dkmFormat, sizeof(dkmFormat) }
};
CK_SP800_108_KDF_PARAMS kdfParams
=
{
CKM_AES_CMAC,
DIM(dataParams),
&dataParams,
0, /* no addition
derived keys */
NULL /* no addition
derived keys */
};
CK_MECHANISM = mechanism
{
CKM_SP800_108_DOUBLE_PIPELINE_KDF,
&kdfParams,
sizeof(kdfParams)
};
hBaseKey = GetBaseKeyHandle(.....);
rv = C_DeriveKey(
hSession,
&mechanism,
hBaseKey,
&derivedKeyTemplate,
DIM(derivedKeyTemplate),
&hDerivedKey);
Table
198, Miscellaneous simple key derivation Mechanisms vs.
Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_CONCATENATE_BASE_AND_KEY
|
|
|
|
|
|
|
ü
|
|
CKM_CONCATENATE_BASE_AND_DATA
|
|
|
|
|
|
|
ü
|
|
CKM_CONCATENATE_DATA_AND_BASE
|
|
|
|
|
|
|
ü
|
|
CKM_XOR_BASE_AND_DATA
|
|
|
|
|
|
|
ü
|
|
CKM_EXTRACT_KEY_FROM_KEY
|
|
|
|
|
|
|
ü
|
Mechanisms:
CKM_CONCATENATE_BASE_AND_DATA
CKM_CONCATENATE_DATA_AND_BASE
CKM_XOR_BASE_AND_DATA
CKM_EXTRACT_KEY_FROM_KEY
CKM_CONCATENATE_BASE_AND_KEY
¨
CK_KEY_DERIVATION_STRING_DATA;
CK_KEY_DERIVATION_STRING_DATA_PTR
CK_KEY_DERIVATION_STRING_DATA provides the parameters for
the CKM_CONCATENATE_BASE_AND_DATA, CKM_CONCATENATE_DATA_AND_BASE, and
CKM_XOR_BASE_AND_DATA mechanisms. It is defined as follows:
typedef struct CK_KEY_DERIVATION_STRING_DATA {
CK_BYTE_PTR pData;
CK_ULONG ulLen;
} CK_KEY_DERIVATION_STRING_DATA;
The fields of the structure have the following meanings:
pData pointer
to the byte string
ulLen length
of the byte string
CK_KEY_DERIVATION_STRING_DATA_PTR is a pointer to a CK_KEY_DERIVATION_STRING_DATA.
¨
CK_EXTRACT_PARAMS;
CK_EXTRACT_PARAMS_PTR
CK_EXTRACT_PARAMS provides the parameter to the CKM_EXTRACT_KEY_FROM_KEY
mechanism. It specifies which bit of the base key should be used as the first
bit of the derived key. It is defined as follows:
typedef CK_ULONG CK_EXTRACT_PARAMS;
CK_EXTRACT_PARAMS_PTR is a pointer to a CK_EXTRACT_PARAMS.
This mechanism, denoted CKM_CONCATENATE_BASE_AND_KEY,
derives a secret key from the concatenation of two existing secret keys. The
two keys are specified by handles; the values of the keys specified are
concatenated together in a buffer.
This mechanism takes a parameter, a CK_OBJECT_HANDLE.
This handle produces the key value information which is appended to the end of
the base key’s value information (the base key is the key whose handle is
supplied as an argument to C_DeriveKey).
For example, if the value of the base key is 0x01234567, and
the value of the other key is 0x89ABCDEF, then the value of the derived key
will be taken from a buffer containing the string 0x0123456789ABCDEF.
·
If no length or key type is provided in the template, then the
key produced by this mechanism will be a generic secret key. Its length will
be equal to the sum of the lengths of the values of the two original keys.
·
If no key type is provided in the template, but a length is, then
the key produced by this mechanism will be a generic secret key of the
specified length.
·
If no length is provided in the template, but a key type is, then
that key type must have a well-defined length. If it does, then the key
produced by this mechanism will be of the type specified in the template. If
it doesn’t, an error will be returned.
·
If both a key type and a length are provided in the template, the
length must be compatible with that key type. The key produced by this
mechanism will be of the specified type and length.
If a DES, DES2, DES3, or CDMF key is derived with this
mechanism, the parity bits of the key will be set properly.
If the requested type of key requires more bytes than are
available by concatenating the two original keys’ values, an error is
generated.
This mechanism has the following rules about key sensitivity
and extractability:
·
If either of the two original keys has its CKA_SENSITIVE
attribute set to CK_TRUE, so does the derived key. If not, then the derived
key’s CKA_SENSITIVE attribute is set either from the supplied template
or from a default value.
·
Similarly, if either of the two original keys has its CKA_EXTRACTABLE
attribute set to CK_FALSE, so does the derived key. If not, then the derived
key’s CKA_EXTRACTABLE attribute is set either from the supplied template
or from a default value.
·
The derived key’s CKA_ALWAYS_SENSITIVE attribute is set to
CK_TRUE if and only if both of the original keys have their CKA_ALWAYS_SENSITIVE
attributes set to CK_TRUE.
·
Similarly, the derived key’s CKA_NEVER_EXTRACTABLE
attribute is set to CK_TRUE if and only if both of the original keys have their
CKA_NEVER_EXTRACTABLE attributes set to CK_TRUE.
This mechanism, denoted CKM_CONCATENATE_BASE_AND_DATA,
derives a secret key by concatenating data onto the end of a specified secret
key.
This mechanism takes a parameter, a CK_KEY_DERIVATION_STRING_DATA
structure, which specifies the length and value of the data which will be
appended to the base key to derive another key.
For example, if the value of the base key is 0x01234567, and
the value of the data is 0x89ABCDEF, then the value of the derived key will be
taken from a buffer containing the string 0x0123456789ABCDEF.
·
If no length or key type is provided in the template, then the
key produced by this mechanism will be a generic secret key. Its length will
be equal to the sum of the lengths of the value of the original key and the
data.
·
If no key type is provided in the template, but a length is, then
the key produced by this mechanism will be a generic secret key of the
specified length.
·
If no length is provided in the template, but a key type is, then
that key type must have a well-defined length. If it does, then the key
produced by this mechanism will be of the type specified in the template. If
it doesn’t, an error will be returned.
·
If both a key type and a length are provided in the template, the
length must be compatible with that key type. The key produced by this
mechanism will be of the specified type and length.
If a DES, DES2, DES3, or CDMF key is derived with this
mechanism, the parity bits of the key will be set properly.
If the requested type of key requires more bytes than are
available by concatenating the original key’s value and the data, an error is
generated.
This mechanism has the following rules about key sensitivity
and extractability:
·
If the base key has its CKA_SENSITIVE attribute set to
CK_TRUE, so does the derived key. If not, then the derived key’s CKA_SENSITIVE
attribute is set either from the supplied template or from a default value.
·
Similarly, if the base key has its CKA_EXTRACTABLE
attribute set to CK_FALSE, so does the derived key. If not, then the derived
key’s CKA_EXTRACTABLE attribute is set either from the supplied template
or from a default value.
·
The derived key’s CKA_ALWAYS_SENSITIVE attribute is set to
CK_TRUE if and only if the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE.
·
Similarly, the derived key’s CKA_NEVER_EXTRACTABLE
attribute is set to CK_TRUE if and only if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_TRUE.
This mechanism, denoted CKM_CONCATENATE_DATA_AND_BASE,
derives a secret key by prepending data to the start of a specified secret key.
This mechanism takes a parameter, a CK_KEY_DERIVATION_STRING_DATA
structure, which specifies the length and value of the data which will be
prepended to the base key to derive another key.
For example, if the value of the base key is 0x01234567, and
the value of the data is 0x89ABCDEF, then the value of the derived key will be
taken from a buffer containing the string 0x89ABCDEF01234567.
·
If no length or key type is provided in the template, then the
key produced by this mechanism will be a generic secret key. Its length will
be equal to the sum of the lengths of the data and the value of the original
key.
·
If no key type is provided in the template, but a length is, then
the key produced by this mechanism will be a generic secret key of the
specified length.
·
If no length is provided in the template, but a key type is, then
that key type must have a well-defined length. If it does, then the key
produced by this mechanism will be of the type specified in the template. If
it doesn’t, an error will be returned.
·
If both a key type and a length are provided in the template, the
length must be compatible with that key type. The key produced by this
mechanism will be of the specified type and length.
If a DES, DES2, DES3, or CDMF key is derived with this
mechanism, the parity bits of the key will be set properly.
If the requested type of key requires more bytes than are
available by concatenating the data and the original key’s value, an error is
generated.
This mechanism has the following rules about key sensitivity
and extractability:
·
If the base key has its CKA_SENSITIVE attribute set to
CK_TRUE, so does the derived key. If not, then the derived key’s CKA_SENSITIVE
attribute is set either from the supplied template or from a default value.
·
Similarly, if the base key has its CKA_EXTRACTABLE
attribute set to CK_FALSE, so does the derived key. If not, then the derived
key’s CKA_EXTRACTABLE attribute is set either from the supplied template
or from a default value.
·
The derived key’s CKA_ALWAYS_SENSITIVE attribute is set to
CK_TRUE if and only if the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE.
·
Similarly, the derived key’s CKA_NEVER_EXTRACTABLE
attribute is set to CK_TRUE if and only if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_TRUE.
XORing key derivation, denoted CKM_XOR_BASE_AND_DATA,
is a mechanism which provides the capability of deriving a secret key by
performing a bit XORing of a key pointed to by a base key handle and some data.
This mechanism takes a parameter, a CK_KEY_DERIVATION_STRING_DATA
structure, which specifies the data with which to XOR the original key’s value.
For example, if the value of the base key is 0x01234567, and
the value of the data is 0x89ABCDEF, then the value of the derived key will be
taken from a buffer containing the string 0x88888888.
·
If no length or key type is provided in the template, then the
key produced by this mechanism will be a generic secret key. Its length will
be equal to the minimum of the lengths of the data and the value of the
original key.
·
If no key type is provided in the template, but a length is, then
the key produced by this mechanism will be a generic secret key of the
specified length.
·
If no length is provided in the template, but a key type is, then
that key type must have a well-defined length. If it does, then the key
produced by this mechanism will be of the type specified in the template. If
it doesn’t, an error will be returned.
·
If both a key type and a length are provided in the template, the
length must be compatible with that key type. The key produced by this
mechanism will be of the specified type and length.
If a DES, DES2, DES3, or CDMF key is derived with this
mechanism, the parity bits of the key will be set properly.
If the requested type of key requires more bytes than are
available by taking the shorter of the data and the original key’s value, an
error is generated.
This mechanism has the following rules about key sensitivity
and extractability:
·
If the base key has its CKA_SENSITIVE attribute set to
CK_TRUE, so does the derived key. If not, then the derived key’s CKA_SENSITIVE
attribute is set either from the supplied template or from a default value.
·
Similarly, if the base key has its CKA_EXTRACTABLE
attribute set to CK_FALSE, so does the derived key. If not, then the derived
key’s CKA_EXTRACTABLE attribute is set either from the supplied template
or from a default value.
·
The derived key’s CKA_ALWAYS_SENSITIVE attribute is set to
CK_TRUE if and only if the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE.
·
Similarly, the derived key’s CKA_NEVER_EXTRACTABLE
attribute is set to CK_TRUE if and only if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_TRUE.
Extraction of one key from another key, denoted CKM_EXTRACT_KEY_FROM_KEY,
is a mechanism which provides the capability of creating one secret key from
the bits of another secret key.
This mechanism has a parameter, a CK_EXTRACT_PARAMS, which
specifies which bit of the original key should be used as the first bit of the
newly-derived key.
We give an example of how this mechanism works. Suppose a
token has a secret key with the 4-byte value 0x329F84A9. We will derive a
2-byte secret key from this key, starting at bit position 21 (i.e., the value
of the parameter to the CKM_EXTRACT_KEY_FROM_KEY mechanism is 21).
1. We write
the key’s value in binary: 0011 0010 1001 1111 1000 0100 1010 1001. We regard
this binary string as holding the 32 bits of the key, labeled as b0, b1, …,
b31.
2. We then
extract 16 consecutive bits (i.e., 2 bytes) from this binary string, starting
at bit b21. We obtain the binary string 1001 0101 0010 0110.
3. The value
of the new key is thus 0x9526.
Note that when constructing the value of the derived key, it
is permissible to wrap around the end of the binary string representing the
original key’s value.
If the original key used in this process is sensitive, then
the derived key must also be sensitive for the derivation to succeed.
·
If no length or key type is provided in the template, then an
error will be returned.
·
If no key type is provided in the template, but a length is, then
the key produced by this mechanism will be a generic secret key of the
specified length.
·
If no length is provided in the template, but a key type is, then
that key type must have a well-defined length. If it does, then the key
produced by this mechanism will be of the type specified in the template. If
it doesn’t, an error will be returned.
·
If both a key type and a length are provided in the template, the
length must be compatible with that key type. The key produced by this
mechanism will be of the specified type and length.
If a DES, DES2, DES3, or CDMF key is derived with this
mechanism, the parity bits of the key will be set properly.
If the requested type of key requires more bytes than the
original key has, an error is generated.
This mechanism has the following rules about key sensitivity
and extractability:
·
If the base key has its CKA_SENSITIVE attribute set to
CK_TRUE, so does the derived key. If not, then the derived key’s CKA_SENSITIVE
attribute is set either from the supplied template or from a default value.
·
Similarly, if the base key has its CKA_EXTRACTABLE
attribute set to CK_FALSE, so does the derived key. If not, then the derived
key’s CKA_EXTRACTABLE attribute is set either from the supplied template
or from a default value.
·
The derived key’s CKA_ALWAYS_SENSITIVE attribute is set to
CK_TRUE if and only if the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE.
·
Similarly, the derived key’s CKA_NEVER_EXTRACTABLE
attribute is set to CK_TRUE if and only if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_TRUE.
Table
199, CMS Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_CMS_SIG
|
|
ü
|
ü
|
|
|
|
|
Mechanisms:
CKM_CMS_SIG
These objects provide information relating to the
CKM_CMS_SIG mechanism. CKM_CMS_SIG mechanism object attributes represent
information about supported CMS signature attributes in the token. They are
only present on tokens supporting the CKM_CMS_SIG mechanism, but must be
present on those tokens.
Table 200, CMS Signature Mechanism Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_REQUIRED_CMS_ATTRIBUTES
|
Byte array
|
Attributes the token always will include in
the set of CMS signed attributes
|
|
CKA_DEFAULT_CMS_ATTRIBUTES
|
Byte array
|
Attributes the token will include in the set
of CMS signed attributes in the absence of any attributes specified by the
application
|
|
CKA_SUPPORTED_CMS_ATTRIBUTES
|
Byte array
|
Attributes the token may include in the set
of CMS signed attributes upon request by the application
|
The contents of each byte array will be a DER-encoded list
of CMS Attributes with
optional accompanying values. Any attributes in the list shall be identified
with its object identifier, and any values shall be DER-encoded. The list of
attributes is defined in ASN.1 as:
Attributes ::= SET SIZE (1..MAX) OF Attribute
Attribute ::= SEQUENCE {
attrType OBJECT IDENTIFIER,
attrValues SET OF ANY DEFINED BY OBJECT IDENTIFIER OPTIONAL
}
The client may not set any of the attributes.
·
CK_CMS_SIG_PARAMS,
CK_CMS_SIG_PARAMS_PTR
CK_CMS_SIG_PARAMS is a structure that provides the
parameters to the CKM_CMS_SIG mechanism. It is defined as follows:
typedef struct CK_CMS_SIG_PARAMS {
CK_OBJECT_HANDLE certificateHandle;
CK_MECHANISM_PTR pSigningMechanism;
CK_MECHANISM_PTR pDigestMechanism;
CK_UTF8CHAR_PTR pContentType;
CK_BYTE_PTR pRequestedAttributes;
CK_ULONG ulRequestedAttributesLen;
CK_BYTE_PTR pRequiredAttributes;
CK_ULONG ulRequiredAttributesLen;
} CK_CMS_SIG_PARAMS;
The fields of the structure have the following meanings:
certificateHandle Object
handle for a certificate associated with the signing key. The token may use
information from this certificate to identify the signer in the SignerInfo result value. CertificateHandle
may be NULL_PTR if the certificate is not available as a PKCS #11 object or if
the calling application leaves the choice of certificate completely to the
token.
pSigningMechanism Mechanism to use
when signing a constructed CMS SignedAttributes
value. E.g. CKM_SHA1_RSA_PKCS.
pDigestMechanism Mechanism to
use when digesting the data. Value shall be NULL_PTR when the digest mechanism
to use follows from the pSigningMechanism parameter.
pContentType NULL-terminated
string indicating complete MIME Content-type of message to be signed; or the
value NULL_PTR if the message is a MIME object (which the token can parse to
determine its MIME Content-type if required). Use the value
“application/octet-stream“ if the MIME type for the message is unknown or
undefined. Note that the pContentType string shall conform to the syntax
specified in RFC 2045, i.e. any parameters needed for correct presentation of
the content by the token (such as, for example, a non-default “charset”) must
be present. The token must follow rules and procedures defined in RFC 2045 when
presenting the content.
pRequestedAttributes Pointer to
DER-encoded list of CMS Attributes
the caller requests to be included in the signed attributes. Token may freely
ignore this list or modify any supplied values.
ulRequestedAttributesLen Length in bytes
of the value pointed to by pRequestedAttributes
pRequiredAttributes Pointer to
DER-encoded list of CMS Attributes
(with accompanying values) required to be included in the resulting signed
attributes. Token must not modify any supplied values. If the token does not
support one or more of the attributes, or does not accept provided values, the
signature operation will fail. The token will use its own default attributes
when signing if both the pRequestedAttributes and pRequiredAttributes field are
set to NULL_PTR.
ulRequiredAttributesLen Length in
bytes, of the value pointed to by pRequiredAttributes.
The CMS mechanism, denoted CKM_CMS_SIG, is a
multi-purpose mechanism based on the structures defined in [PKCS #7] and RFC
2630. It supports single- or multiple-part signatures with and without message
recovery. The mechanism is intended for use with, e.g., PTDs (see MeT-PTD) or
other capable tokens. The token will construct a CMS SignedAttributes value and compute a signature on this value.
The content of the SignedAttributes
value is decided by the token, however the caller can suggest some attributes
in the parameter pRequestedAttributes. The caller can also require some
attributes to be present through the parameters pRequiredAttributes. The
signature is computed in accordance with the parameter pSigningMechanism.
When this mechanism is used in successful calls to C_Sign
or C_SignFinal, the pSignature return value will point to a
DER-encoded value of type SignerInfo.
SignerInfo is defined in ASN.1 as
follows (for a complete definition of all fields and types, see RFC 2630):
SignerInfo ::= SEQUENCE {
version CMSVersion,
sid SignerIdentifier,
digestAlgorithm DigestAlgorithmIdentifier,
signedAttrs [0] IMPLICIT SignedAttributes OPTIONAL,
signatureAlgorithm SignatureAlgorithmIdentifier,
signature SignatureValue,
unsignedAttrs [1] IMPLICIT UnsignedAttributes OPTIONAL }
The certificateHandle parameter, when set, helps the
token populate the sid field of the
SignerInfo value. If certificateHandle
is NULL_PTR the choice of a suitable certificate reference in the SignerInfo result value is left to the token
(the token could, e.g., interact with the user).
This mechanism shall not be used in calls to C_Verify
or C_VerifyFinal (use the pSigningMechanism mechanism instead).
For the pRequiredAttributes field, the token may have
to interact with the user to find out whether to accept a proposed value or
not. The token should never accept any proposed attribute values without some
kind of confirmation from its owner (but this could be through, e.g., configuration
or policy settings and not direct interaction). If a user rejects proposed
values, or the signature request as such, the value CKR_FUNCTION_REJECTED shall
be returned.
When possible, applications should use the CKM_CMS_SIG
mechanism when generating CMS-compatible signatures rather than lower-level
mechanisms such as CKM_SHA1_RSA_PKCS. This is especially true when the
signatures are to be made on content that the token is able to present to a
user. Exceptions may include those cases where the token does not support a
particular signing attribute. Note however that the token may refuse usage of a
particular signature key unless the content to be signed is known (i.e. the CKM_CMS_SIG
mechanism is used).
When a token does not have presentation capabilities, the
PKCS #11-aware application may avoid sending the whole message to the token by
electing to use a suitable signature mechanism (e.g. CKM_RSA_PKCS) as
the pSigningMechanism value in the CK_CMS_SIG_PARAMS structure,
and digesting the message itself before passing it to the token.
PKCS #11-aware applications making use of tokens with
presentation capabilities, should attempt to provide messages to be signed by
the token in a format possible for the token to present to the user. Tokens
that receive multipart MIME-messages for which only certain parts are possible
to present may fail the signature operation with a return value of CKR_DATA_INVALID,
but may also choose to add a signing attribute indicating which parts of the
message were possible to present.
Blowfish, a secret-key block cipher. It is a Feistel
network, iterating a simple encryption function 16 times. The block size is 64
bits, and the key can be any length up to 448 bits. Although there is a complex
initialization phase required before any encryption can take place, the actual
encryption of data is very efficient on large microprocessors. See [BLOWFISH]
for details.
Table
201, Blowfish Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_BLOWFISH_CBC
|
✓
|
|
|
|
|
✓
|
|
|
CKM_BLOWFISH_CBC_PAD
|
✓
|
|
|
|
|
✓
|
|
This section defines the key type “CKK_BLOWFISH” for type
CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of key objects.
Mechanisms:
CKM_BLOWFISH_KEY_GEN
CKM_BLOWFISH_CBC
CKM_BLOWFISH_CBC_PAD
Blowfish secret key objects
(object class CKO_SECRET_KEY, key type CKK_BLOWFISH) hold Blowfish keys. The following
table defines the Blowfish secret key object attributes, in addition to the
common attributes defined for this object class:
Table 202, BLOWFISH Secret Key Object
|
Attribute
|
Data type
|
Meaning
|
|
CKA_VALUE1,4,6,7
|
Byte array
|
Key value the key can be any length up to 448
bits. Bit length restricted to a byte array.
|
|
CKA_VALUE_LEN2,3
|
CK_ULONG
|
Length in bytes of key value
|
- Refer to Table 11 for footnotes
The following is a sample template for creating an Blowfish
secret key object:
CK_OBJECT_CLASS class = CKO_SECRET_KEY;
CK_KEY_TYPE keyType = CKK_BLOWFISH;
CK_UTF8CHAR label[] = “A blowfish secret key object”;
CK_BYTE value[16] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_ENCRYPT, &true, sizeof(true)},
{CKA_VALUE, value, sizeof(value)}
};
The Blowfish key generation mechanism, denoted CKM_BLOWFISH_KEY_GEN,
is a key generation mechanism Blowfish.
It does not have a parameter.
The mechanism generates Blowfish keys with a particular
length, as specified in the CKA_VALUE_LEN attribute of the template for
the key.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key. Other attributes supported by
the key type (specifically, the flags indicating which functions the key
supports) may be specified in the template for the key, or else are assigned
default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
key sizes in bytes.
Blowfish-CBC, denoted CKM_BLOWFISH_CBC, is a
mechanism for single- and multiple-part encryption and decryption; key
wrapping; and key unwrapping.
It has a parameter, a 8-byte initialization vector.
This
mechanism can wrap and unwrap any secret key. For wrapping, the mechanism
encrypts the value of the CKA_VALUE attribute of the key that is
wrapped, padded on the trailing end with up to block size minus one null bytes
so that the resulting length is a multiple of the block size. The output data
is the same length as the padded input data. It does not wrap the key type, key
length, or any other information about the key; the application must convey
these separately.
For unwrapping, the mechanism
decrypts the wrapped key, and truncates the result according to the CKA_KEY_TYPE
attribute of the template and, if it has one, and the key type supports it, the
CKA_VALUE_LEN attribute of the template. The mechanism contributes the
result as the CKA_VALUE attribute of the new key; other attributes
required by the key type must be specified in the template.
Constraints on key types and the
length of data are summarized in the following table:
Table 203, BLOWFISH-CBC: Key and Data Length
|
Function
|
Key type
|
Input Length
|
Output Length
|
|
C_Encrypt
|
BLOWFISH
|
Multiple of block size
|
Same as input length
|
|
C_Decrypt
|
BLOWFISH
|
Multiple of block size
|
Same as input length
|
|
C_WrapKey
|
BLOWFISH
|
Any
|
Input length rounded up to multiple of the
block size
|
|
C_UnwrapKey
|
BLOWFISH
|
Multiple of block size
|
Determined by type of key being unwrapped or
CKA_VALUE_LEN
|
For this mechanism, the ulMinKeySize and ulMaxKeySize fields
of the CK_MECHANISM_INFO structure specify the supported range of
BLOWFISH key sizes, in bytes.
Blowfish-CBC-PAD, denoted CKM_BLOWFISH_CBC_PAD, is a
mechanism for single- and multiple-part encryption and decryption, key wrapping
and key unwrapping, cipher-block chaining mode and the block cipher padding
method detailed in [PKCS #7].
It has a parameter, a 8-byte initialization vector.
The PKCS padding in this mechanism allows the length of the
plaintext value to be recovered from the ciphertext value. Therefore, when
unwrapping keys with this mechanism, no value should be specified for the CKA_VALUE_LEN
attribute.
The entries in the table below for data length constraints
when wrapping and unwrapping keys do not apply to wrapping and unwrapping
private keys.
Constraints on key types and the length of data are
summarized in the following table:
Table
204, BLOWFISH-CBC with PKCS Padding: Key and Data Length
|
Function
|
Key type
|
Input Length
|
Output Length
|
|
C_Encrypt
|
BLOWFISH
|
Any
|
Input length rounded up to multiple of the
block size
|
|
C_Decrypt
|
BLOWFISH
|
Multiple of block size
|
Between 1 and block length block size bytes
shorter than input length
|
|
C_WrapKey
|
BLOWFISH
|
Any
|
Input length rounded up to multiple of the
block size
|
|
C_UnwrapKey
|
BLOWFISH
|
Multiple of block size
|
Between 1 and block length block size bytes
shorter than input length
|
Twofish is a secret key block cipher. See [TWOFISH] for
details.
This section defines the key type “CKK_TWOFISH” for type
CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of key objects.
Mechanisms:
CKM_TWOFISH_KEY_GEN
CKM_TWOFISH_CBC
CKM_TWOFISH_CBC_PAD
Twofish secret key objects (object class CKO_SECRET_KEY, key
type CKK_TWOFISH) hold Twofish keys. The following table defines the
Twofish secret key object attributes, in addition to the common attributes defined
for this object class:
Table 205, Twofish Secret Key Object
|
Attribute
|
Data type
|
Meaning
|
|
CKA_VALUE1,4,6,7
|
Byte array
|
Key value 128-, 192-, or 256-bit key
|
|
CKA_VALUE_LEN2,3
|
CK_ULONG
|
Length in bytes of key value
|
- Refer to Table 11 for footnotes
The following is a sample template for creating an TWOFISH
secret key object:
CK_OBJECT_CLASS class = CKO_SECRET_KEY;
CK_KEY_TYPE keyType = CKK_TWOFISH;
CK_UTF8CHAR label[] = “A twofish secret key object”;
CK_BYTE value[16] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_ENCRYPT, &true, sizeof(true)},
{CKA_VALUE, value, sizeof(value)}
};
The Twofish key generation mechanism, denoted CKM_TWOFISH_KEY_GEN,
is a key generation mechanism Twofish.
It does not have a parameter.
The mechanism generates Blowfish keys with a particular
length, as specified in the CKA_VALUE_LEN attribute of the template for
the key.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key. Other attributes supported by
the key type (specifically, the flags indicating which functions the key
supports) may be specified in the template for the key, or else are assigned
default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
key sizes, in bytes.
Twofish-CBC, denoted CKM_TWOFISH_CBC, is a mechanism
for single- and multiple-part encryption and decryption; key wrapping; and key
unwrapping.
It has a parameter, a 16-byte initialization vector.
Twofish-CBC-PAD, denoted CKM_TWOFISH_CBC_PAD, is a mechanism
for single- and multiple-part encryption and decryption, key wrapping and key
unwrapping, cipher-block chaining mode and the block cipher padding method
detailed in [PKCS #7].
It has a parameter, a 16-byte initialization vector.
The PKCS padding in this mechanism allows the length of the
plaintext value to be recovered from the ciphertext value. Therefore, when
unwrapping keys with this mechanism, no value should be specified for the CKA_VALUE_LEN
attribute.
Camellia is a block cipher with 128-bit block size and 128-,
192-, and 256-bit keys, similar to AES. Camellia is described e.g. in IETF RFC
3713.
Table
206, Camellia Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_CAMELLIA_KEY_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_CAMELLIA_ECB
|
ü
|
|
|
|
|
ü
|
|
|
CKM_CAMELLIA_CBC
|
ü
|
|
|
|
|
ü
|
|
|
CKM_CAMELLIA_CBC_PAD
|
ü
|
|
|
|
|
ü
|
|
|
CKM_CAMELLIA_MAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_CAMELLIA_MAC
|
|
ü
|
|
|
|
|
|
|
CKM_CAMELLIA_ECB_ENCRYPT_DATA
|
|
|
|
|
|
|
ü
|
|
CKM_CAMELLIA_CBC_ENCRYPT_DATA
|
|
|
|
|
|
|
ü
|
This section defines the key type “CKK_CAMELLIA” for type
CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of key objects.
Mechanisms:
CKM_CAMELLIA_KEY_GEN
CKM_CAMELLIA_ECB
CKM_CAMELLIA_CBC
CKM_CAMELLIA_MAC
CKM_CAMELLIA_MAC_GENERAL
CKM_CAMELLIA_CBC_PAD
Camellia secret key objects (object class CKO_SECRET_KEY,
key type CKK_CAMELLIA) hold Camellia keys. The following table
defines the Camellia secret key object attributes, in addition to the common
attributes defined for this object class:
Table 207, Camellia Secret
Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_VALUE1,4,6,7
|
Byte array
|
Key value (16, 24, or 32 bytes)
|
|
CKA_VALUE_LEN2,3,6
|
CK_ULONG
|
Length in bytes of key value
|
- Refer to Table 11 for footnotes
The following is a sample template for creating a Camellia
secret key object:
CK_OBJECT_CLASS class = CKO_SECRET_KEY;
CK_KEY_TYPE keyType = CKK_CAMELLIA;
CK_UTF8CHAR label[] = “A Camellia secret key object”;
CK_BYTE value[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_ENCRYPT, &true, sizeof(true)},
{CKA_VALUE, value, sizeof(value)}
};
6.47.3
Camellia key generation
The Camellia key generation mechanism, denoted
CKM_CAMELLIA_KEY_GEN, is a key generation mechanism for Camellia.
It does not have a parameter.
The mechanism generates Camellia keys with a particular
length in bytes, as specified in the CKA_VALUE_LEN attribute of the
template for the key.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key. Other attributes supported by
the Camellia key type (specifically, the flags indicating which functions the
key supports) may be specified in the template for the key, or else are
assigned default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
Camellia key sizes, in bytes.
Camellia-ECB, denoted CKM_CAMELLIA_ECB, is a
mechanism for single- and multiple-part encryption and decryption; key
wrapping; and key unwrapping, based on Camellia and electronic codebook mode.
It does not have a parameter.
This mechanism can wrap and unwrap any secret key. Of
course, a particular token may not be able to wrap/unwrap every secret key that
it supports. For wrapping, the mechanism encrypts the value of the CKA_VALUE
attribute of the key that is wrapped, padded on the trailing end with up to
block size minus one null bytes so that the resulting length is a multiple of
the block size. The output data is the same length as the padded input data. It
does not wrap the key type, key length, or any other information about the key;
the application must convey these separately.
For unwrapping, the mechanism decrypts the wrapped key, and
truncates the result according to the CKA_KEY_TYPE attribute of the
template and, if it has one, and the key type supports it, the CKA_VALUE_LEN
attribute of the template. The mechanism contributes the result as the CKA_VALUE
attribute of the new key; other attributes required by the key type must be
specified in the template.
Constraints on key types and the length of data are
summarized in the following table:
Table 208, Camellia-ECB: Key
and Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
Comments
|
|
C_Encrypt
|
CKK_CAMELLIA
|
multiple of block
size
|
same as input length
|
no final part
|
|
C_Decrypt
|
CKK_CAMELLIA
|
multiple of block
size
|
same as input length
|
no final part
|
|
C_WrapKey
|
CKK_CAMELLIA
|
any
|
input length rounded
up to multiple of block size
|
|
|
C_UnwrapKey
|
CKK_CAMELLIA
|
multiple of block
size
|
determined by type of
key being unwrapped or CKA_VALUE_LEN
|
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
Camellia key sizes, in bytes.
Camellia-CBC, denoted CKM_CAMELLIA_CBC, is a
mechanism for single- and multiple-part encryption and decryption; key
wrapping; and key unwrapping, based on Camellia and cipher-block chaining mode.
It has a parameter, a 16-byte initialization vector.
This mechanism can wrap and unwrap any secret key. Of
course, a particular token may not be able to wrap/unwrap every secret key that
it supports. For wrapping, the mechanism encrypts the value of the CKA_VALUE
attribute of the key that is wrapped, padded on the trailing end with up to
block size minus one null bytes so that the resulting length is a multiple of
the block size. The output data is the same length as the padded input data. It
does not wrap the key type, key length, or any other information about the key;
the application must convey these separately.
For unwrapping, the mechanism decrypts the wrapped key, and
truncates the result according to the CKA_KEY_TYPE attribute of the
template and, if it has one, and the key type supports it, the CKA_VALUE_LEN
attribute of the template. The mechanism contributes the result as the CKA_VALUE
attribute of the new key; other attributes required by the key type must be
specified in the template.
Constraints on key types and the length of data are
summarized in the following table:
Table 209, Camellia-CBC: Key
and Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
Comments
|
|
C_Encrypt
|
CKK_CAMELLIA
|
multiple of block
size
|
same as input length
|
no final part
|
|
C_Decrypt
|
CKK_CAMELLIA
|
multiple of block
size
|
same as input length
|
no final part
|
|
C_WrapKey
|
CKK_CAMELLIA
|
any
|
input length rounded
up to multiple of the block size
|
|
|
C_UnwrapKey
|
CKK_CAMELLIA
|
multiple of block
size
|
determined by type of
key being unwrapped or CKA_VALUE_LEN
|
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
Camellia key sizes, in bytes.
Camellia-CBC with PKCS padding, denoted CKM_CAMELLIA_CBC_PAD,
is a mechanism for single- and multiple-part encryption and decryption; key
wrapping; and key unwrapping, based on Camellia; cipher-block chaining mode;
and the block cipher padding method detailed in [PKCS #7].
It has a parameter, a 16-byte initialization vector.
The PKCS padding in this mechanism allows the length of the
plaintext value to be recovered from the ciphertext value. Therefore, when
unwrapping keys with this mechanism, no value should be specified for the CKA_VALUE_LEN
attribute.
In addition to being able to wrap and unwrap secret keys,
this mechanism can wrap and unwrap RSA, Diffie-Hellman, X9.42 Diffie-Hellman, short
Weierstrass EC and DSA private keys (see Section 6.7
for details). The entries in the table below for data length constraints when
wrapping and unwrapping keys do not apply to wrapping and unwrapping private
keys.
Constraints on key types and the length of data are
summarized in the following table:
Table 210, Camellia-CBC with
PKCS Padding: Key and Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Encrypt
|
CKK_CAMELLIA
|
any
|
input length rounded
up to multiple of the block size
|
|
C_Decrypt
|
CKK_CAMELLIA
|
multiple of block
size
|
between 1 and block
size bytes shorter than input length
|
|
C_WrapKey
|
CKK_CAMELLIA
|
any
|
input length rounded
up to multiple of the block size
|
|
C_UnwrapKey
|
CKK_CAMELLIA
|
multiple of block
size
|
between 1 and block
length bytes shorter than input length
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
Camellia key sizes, in bytes.
¨
CK_CAMELLIA_CTR_PARAMS;
CK_CAMELLIA_CTR_PARAMS_PTR
CK_CAMELLIA_CTR_PARAMS is a structure that provides
the parameters to the CKM_CAMELLIA_CTR mechanism. It is defined as
follows:
typedef struct CK_CAMELLIA_CTR_PARAMS {
CK_ULONG ulCounterBits;
CK_BYTE cb[16];
} CK_CAMELLIA_CTR_PARAMS;
ulCounterBits specifies the number of bits in the counter
block (cb) that shall be incremented. This number shall be such that 0 < ulCounterBits
≤ 128. For any values outside this range the mechanism shall return CKR_MECHANISM_PARAM_INVALID.
It's up to the caller to initialize all of the bits in the
counter block including the counter bits. The counter bits are the least
significant bits of the counter block (cb). They are a big-endian value usually
starting with 1. The rest of ‘cb’ is for the nonce, and maybe an optional IV.
E.g. as defined in [RFC 3686]:
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Nonce |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Initialization Vector (IV) |
| |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Block Counter |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
This construction permits each packet to consist of up to 232-1
blocks = 4,294,967,295 blocks = 68,719,476,720 octets.
CK_CAMELLIA_CTR_PARAMS_PTR is a pointer to a CK_CAMELLIA_CTR_PARAMS.
General-length Camellia -MAC, denoted
CKM_CAMELLIA_MAC_GENERAL, is a mechanism for single- and multiple-part
signatures and verification, based on Camellia and data authentication as
defined in.[CAMELLIA]
It has a parameter, a CK_MAC_GENERAL_PARAMS structure,
which specifies the output length desired from the mechanism.
The output bytes from this mechanism are taken from the
start of the final Camellia cipher block produced in the MACing process.
Constraints on key types and the length of data are
summarized in the following table:
Table 211, General-length Camellia-MAC:
Key and Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
CKK_CAMELLIA
|
any
|
1-block size, as
specified in parameters
|
|
C_Verify
|
CKK_CAMELLIA
|
any
|
1-block size, as
specified in parameters
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
Camellia key sizes, in bytes.
Camellia-MAC, denoted by CKM_CAMELLIA_MAC, is a
special case of the general-length Camellia-MAC mechanism. Camellia-MAC always
produces and verifies MACs that are half the block size in length.
It does not have a parameter.
Constraints on key types and the length of data are
summarized in the following table:
Table 212, Camellia-MAC: Key
and Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
CKK_CAMELLIA
|
any
|
½ block size (8
bytes)
|
|
C_Verify
|
CKK_CAMELLIA
|
any
|
½ block size (8
bytes)
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
Camellia key sizes, in bytes.
These mechanisms allow derivation of keys using the result
of an encryption operation as the key value. They are for use with the
C_DeriveKey function.
Mechanisms:
CKM_CAMELLIA_ECB_ENCRYPT_DATA
CKM_CAMELLIA_CBC_ENCRYPT_DATA
typedef struct CK_CAMELLIA_CBC_ENCRYPT_DATA_PARAMS {
CK_BYTE iv[16];
CK_BYTE_PTR pData;
CK_ULONG length;
} CK_CAMELLIA_CBC_ENCRYPT_DATA_PARAMS;
typedef CK_CAMELLIA_CBC_ENCRYPT_DATA_PARAMS CK_PTR
CK_CAMELLIA_CBC_ENCRYPT_DATA_PARAMS_PTR;
Uses CK_CAMELLIA_CBC_ENCRYPT_DATA_PARAMS, and CK_KEY_DERIVATION_STRING_DATA.
Table 213, Mechanism
Parameters for Camellia-based key derivation
|
CKM_CAMELLIA_ECB_ENCRYPT_DATA
|
Uses
CK_KEY_DERIVATION_STRING_DATA structure. Parameter is the data to be
encrypted and must be a multiple of 16 long.
|
|
CKM_CAMELLIA_CBC_ENCRYPT_DATA
|
Uses
CK_CAMELLIA_CBC_ENCRYPT_DATA_PARAMS.
Parameter is an 16 byte IV value followed by the data. The data value part
must be a multiple of 16 bytes long.
|
ARIA is a block cipher with 128-bit block size and 128-,
192-, and 256-bit keys, similar to AES. ARIA is described in NSRI
“Specification of ARIA”.
Table
214, ARIA Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_ARIA_KEY_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_ARIA_ECB
|
ü
|
|
|
|
|
ü
|
|
|
CKM_ARIA_CBC
|
ü
|
|
|
|
|
ü
|
|
|
CKM_ARIA_CBC_PAD
|
ü
|
|
|
|
|
ü
|
|
|
CKM_ARIA_MAC_GENERAL
|
|
ü
|
|
|
|
|
|
|
CKM_ARIA_MAC
|
|
ü
|
|
|
|
|
|
|
CKM_ARIA_ECB_ENCRYPT_DATA
|
|
|
|
|
|
|
ü
|
|
CKM_ARIA_CBC_ENCRYPT_DATA
|
|
|
|
|
|
|
ü
|
This section defines the key type “CKK_ARIA” for type
CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of key objects.
Mechanisms:
CKM_ARIA_KEY_GEN
CKM_ARIA_ECB
CKM_ARIA_CBC
CKM_ARIA_MAC
CKM_ARIA_MAC_GENERAL
CKM_ARIA_CBC_PAD
ARIA secret key objects (object class CKO_SECRET_KEY, key
type CKK_ARIA) hold ARIA keys. The following table defines the ARIA
secret key object attributes, in addition to the common attributes defined for
this object class:
Table 215, ARIA Secret Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_VALUE1,4,6,7
|
Byte array
|
Key value (16, 24, or 32 bytes)
|
|
CKA_VALUE_LEN2,3,6
|
CK_ULONG
|
Length in bytes of key value
|
- Refer to Table 11 for footnotes
The following is a sample template for creating an ARIA
secret key object:
CK_OBJECT_CLASS class = CKO_SECRET_KEY;
CK_KEY_TYPE keyType = CKK_ARIA;
CK_UTF8CHAR label[] = “An ARIA secret key object”;
CK_BYTE value[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_ENCRYPT, &true, sizeof(true)},
{CKA_VALUE, value, sizeof(value)}
};
The ARIA key generation mechanism, denoted CKM_ARIA_KEY_GEN,
is a key generation mechanism for Aria.
It does not have a parameter.
The mechanism generates ARIA keys with a particular length
in bytes, as specified in the CKA_VALUE_LEN attribute of the template
for the key.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key. Other attributes supported by
the ARIA key type (specifically, the flags indicating which functions the key
supports) may be specified in the template for the key, or else are assigned
default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
ARIA key sizes, in bytes.
ARIA-ECB, denoted CKM_ARIA_ECB, is a mechanism for
single- and multiple-part encryption and decryption; key wrapping; and key
unwrapping, based on Aria and electronic codebook mode.
It does not have a parameter.
This mechanism can wrap and unwrap any secret key. Of
course, a particular token may not be able to wrap/unwrap every secret key that
it supports. For wrapping, the mechanism encrypts the value of the CKA_VALUE
attribute of the key that is wrapped, padded on the trailing end with up to
block size minus one null bytes so that the resulting length is a multiple of
the block size. The output data is the same length as the padded input data. It
does not wrap the key type, key length, or any other information about the key;
the application must convey these separately.
For unwrapping, the mechanism decrypts the wrapped key, and
truncates the result according to the CKA_KEY_TYPE attribute of the
template and, if it has one, and the key type supports it, the CKA_VALUE_LEN
attribute of the template. The mechanism contributes the result as the CKA_VALUE
attribute of the new key; other attributes required by the key type must be
specified in the template.
Constraints on key types and the length of data are
summarized in the following table:
Table 216, ARIA-ECB: Key and Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
Comments
|
|
C_Encrypt
|
CKK_ARIA
|
multiple of block
size
|
same as input length
|
no final part
|
|
C_Decrypt
|
CKK_ARIA
|
multiple of block
size
|
same as input length
|
no final part
|
|
C_WrapKey
|
CKK_ARIA
|
any
|
input length rounded
up to multiple of block size
|
|
|
C_UnwrapKey
|
CKK_ARIA
|
multiple of block
size
|
determined by type of
key being unwrapped or CKA_VALUE_LEN
|
|
For this mechanism, the ulMinKeySize
and ulMaxKeySize fields of the CK_MECHANISM_INFO structure
specify the supported range of ARIA key sizes, in bytes.
ARIA-CBC, denoted CKM_ARIA_CBC, is a mechanism for
single- and multiple-part encryption and decryption; key wrapping; and key
unwrapping, based on ARIA and cipher-block chaining mode.
It has a parameter, a 16-byte initialization vector.
This mechanism can wrap and unwrap any secret key. Of
course, a particular token may not be able to wrap/unwrap every secret key that
it supports. For wrapping, the mechanism encrypts the value of the CKA_VALUE
attribute of the key that is wrapped, padded on the trailing end with up to
block size minus one null bytes so that the resulting length is a multiple of
the block size. The output data is the same length as the padded input data. It
does not wrap the key type, key length, or any other information about the key;
the application must convey these separately.
For unwrapping, the mechanism decrypts the wrapped key, and
truncates the result according to the CKA_KEY_TYPE attribute of the
template and, if it has one, and the key type supports it, the CKA_VALUE_LEN
attribute of the template. The mechanism contributes the result as the CKA_VALUE
attribute of the new key; other attributes required by the key type must be
specified in the template.
Constraints on key types and the length of data are
summarized in the following table:
Table 217, ARIA-CBC: Key and Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
Comments
|
|
C_Encrypt
|
CKK_ARIA
|
multiple of block
size
|
same as input length
|
no final part
|
|
C_Decrypt
|
CKK_ARIA
|
multiple of block
size
|
same as input length
|
no final part
|
|
C_WrapKey
|
CKK_ARIA
|
any
|
input length rounded
up to multiple of the block size
|
|
|
C_UnwrapKey
|
CKK_ARIA
|
multiple of block
size
|
determined by type of
key being unwrapped or CKA_VALUE_LEN
|
|
For this mechanism, the
ulMinKeySize and ulMaxKeySize fields of the CK_MECHANISM_INFO structure specify
the supported range of Aria key sizes, in bytes.
ARIA-CBC with PKCS padding, denoted CKM_ARIA_CBC_PAD,
is a mechanism for single- and multiple-part encryption and decryption; key
wrapping; and key unwrapping, based on ARIA; cipher-block chaining mode; and
the block cipher padding method detailed in [PKCS #7].
It has a parameter, a 16-byte initialization vector.
The PKCS padding in this mechanism allows the length of the
plaintext value to be recovered from the ciphertext value. Therefore, when
unwrapping keys with this mechanism, no value should be specified for the CKA_VALUE_LEN
attribute.
In addition to being able to wrap and unwrap secret keys,
this mechanism can wrap and unwrap RSA, Diffie-Hellman, X9.42 Diffie-Hellman, short
Weierstrass EC and DSA private keys (see Section 6.7
for details). The entries in the table below for data length constraints when
wrapping and unwrapping keys do not apply to wrapping and unwrapping private
keys.
Constraints on key types and the length of data are
summarized in the following table:
Table 218, ARIA-CBC with PKCS Padding: Key and Data
Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Encrypt
|
CKK_ARIA
|
any
|
input length rounded
up to multiple of the block size
|
|
C_Decrypt
|
CKK_ARIA
|
multiple of block
size
|
between 1 and block
size bytes shorter than input length
|
|
C_WrapKey
|
CKK_ARIA
|
any
|
input length rounded
up to multiple of the block size
|
|
C_UnwrapKey
|
CKK_ARIA
|
multiple of block
size
|
between 1 and block
length bytes shorter than input length
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
ARIA key sizes, in bytes.
General-length ARIA -MAC, denoted CKM_ARIA_MAC_GENERAL,
is a mechanism for single- and multiple-part signatures and verification, based
on ARIA and data authentication as defined in [FIPS 113].
It has a parameter, a CK_MAC_GENERAL_PARAMS structure,
which specifies the output length desired from the mechanism.
The output bytes from this mechanism are taken from the
start of the final ARIA cipher block produced in the MACing process.
Constraints on key types and the length of data are
summarized in the following table:
Table 219, General-length ARIA-MAC: Key and Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
CKK_ARIA
|
any
|
1-block size, as
specified in parameters
|
|
C_Verify
|
CKK_ARIA
|
any
|
1-block size, as
specified in parameters
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
ARIA key sizes, in bytes.
ARIA-MAC, denoted by CKM_ARIA_MAC, is a special case
of the general-length ARIA-MAC mechanism. ARIA-MAC always produces and verifies
MACs that are half the block size in length.
It does not have a parameter.
Constraints on key types and the length of data are
summarized in the following table:
Table 220, ARIA-MAC: Key and Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
CKK_ARIA
|
any
|
½ block size (8
bytes)
|
|
C_Verify
|
CKK_ARIA
|
any
|
½ block size (8
bytes)
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
ARIA key sizes, in bytes.
These mechanisms allow derivation of keys using the result
of an encryption operation as the key value. They are for use with the
C_DeriveKey function.
Mechanisms:
CKM_ARIA_ECB_ENCRYPT_DATA
CKM_ARIA_CBC_ENCRYPT_DATA
typedef struct CK_ARIA_CBC_ENCRYPT_DATA_PARAMS {
CK_BYTE iv[16];
CK_BYTE_PTR pData;
CK_ULONG length;
} CK_ARIA_CBC_ENCRYPT_DATA_PARAMS;
typedef CK_ARIA_CBC_ENCRYPT_DATA_PARAMS CK_PTR
CK_ARIA_CBC_ENCRYPT_DATA_PARAMS_PTR;
Uses CK_ARIA_CBC_ENCRYPT_DATA_PARAMS, and
CK_KEY_DERIVATION_STRING_DATA.
Table 221, Mechanism Parameters for Aria-based key
derivation
|
CKM_ARIA_ECB_ENCRYPT_DATA
|
Uses
CK_KEY_DERIVATION_STRING_DATA structure. Parameter is the data to be
encrypted and must be a multiple of 16 long.
|
|
CKM_ARIA_CBC_ENCRYPT_DATA
|
Uses
CK_ARIA_CBC_ENCRYPT_DATA_PARAMS.
Parameter is an 16 byte IV value followed by the data. The data value part
must be a multiple of 16 bytes long.
|
SEED is a symmetric block cipher developed by the South
Korean Information Security Agency (KISA). It has a 128-bit key size and a
128-bit block size.
Its specification has been published as Internet [RFC 4269].
RFCs have been published defining
the use of SEED in
TLS ftp://ftp.rfc-editor.org/in-notes/rfc4162.txt
IPsec ftp://ftp.rfc-editor.org/in-notes/rfc4196.txt
CMS ftp://ftp.rfc-editor.org/in-notes/rfc4010.txt
TLS cipher suites that use SEED
include:
CipherSuite
TLS_RSA_WITH_SEED_CBC_SHA = { 0x00, 0x96};
CipherSuite
TLS_DH_DSS_WITH_SEED_CBC_SHA = { 0x00, 0x97};
CipherSuite
TLS_DH_RSA_WITH_SEED_CBC_SHA = { 0x00, 0x98};
CipherSuite
TLS_DHE_DSS_WITH_SEED_CBC_SHA = { 0x00, 0x99};
CipherSuite
TLS_DHE_RSA_WITH_SEED_CBC_SHA = { 0x00, 0x9A};
CipherSuite
TLS_DH_anon_WITH_SEED_CBC_SHA = { 0x00, 0x9B};
As with any block cipher, it can be used in the ECB, CBC,
OFB and CFB modes of operation, as well as in a MAC algorithm such as HMAC.
OIDs have been published for all these uses. A list may be
seen at http://www.alvestrand.no/objectid/1.2.410.200004.1.html
Table
222, SEED Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_SEED_KEY_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_SEED_ECB
|
|
|
ü
|
|
|
|
|
|
CKM_SEED_CBC
|
|
|
ü
|
|
|
|
|
|
CKM_SEED_CBC_PAD
|
ü
|
|
|
|
|
ü
|
|
|
CKM_SEED_MAC_GENERAL
|
|
|
ü
|
|
|
|
|
|
CKM_SEED_MAC
|
|
|
|
ü
|
|
|
|
|
CKM_SEED_ECB_ENCRYPT_DATA
|
|
|
|
|
|
|
ü
|
|
CKM_SEED_CBC_ENCRYPT_DATA
|
|
|
|
|
|
|
ü
|
This section defines the key type “CKK_SEED” for type
CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of key objects.
Mechanisms:
CKM_SEED_KEY_GEN
CKM_SEED_ECB
CKM_SEED_CBC
CKM_SEED_MAC
CKM_SEED_MAC_GENERAL
CKM_SEED_CBC_PAD
For all of these mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO are always 16.
SEED secret key objects (object class CKO_SECRET_KEY, key
type CKK_SEED) hold SEED keys. The following table defines the secret
key object attributes, in addition to the common attributes defined for this
object class:
Table 223, SEED Secret Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_VALUE1,4,6,7
|
Byte array
|
Key value (always 16 bytes long)
|
- Refer to Table 11 for footnotes
The following is a sample template for creating a SEED
secret key object:
CK_OBJECT_CLASS class = CKO_SECRET_KEY;
CK_KEY_TYPE keyType = CKK_SEED;
CK_UTF8CHAR label[] = “A SEED secret key object”;
CK_BYTE value[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_ENCRYPT, &true, sizeof(true)},
{CKA_VALUE, value, sizeof(value)}
};
The SEED key generation mechanism, denoted CKM_SEED_KEY_GEN,
is a key generation mechanism for SEED.
It does not have a parameter.
The mechanism generates SEED keys.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key. Other attributes supported by
the SEED key type (specifically, the flags indicating which functions the key
supports) may be specified in the template for the key, or else are assigned
default initial values.
SEED-ECB, denoted CKM_SEED_ECB, is a mechanism for
single- and multiple-part encryption and decryption; key wrapping; and key
unwrapping, based on SEED and electronic codebook mode.
It does not have a parameter.
SEED-CBC, denoted CKM_SEED_CBC, is a mechanism for
single- and multiple-part encryption and decryption; key wrapping; and key
unwrapping, based on SEED and cipher-block chaining mode.
It has a parameter, a 16-byte initialization vector.
SEED-CBC with PKCS padding, denoted CKM_SEED_CBC_PAD,
is a mechanism for single- and multiple-part encryption and decryption; key
wrapping; and key unwrapping, based on SEED; cipher-block chaining mode; and
the block cipher padding method detailed in [PKCS #7].
It has a parameter, a 16-byte initialization vector.
General-length SEED-MAC, denoted CKM_SEED_MAC_GENERAL,
is a mechanism for single- and multiple-part signatures and verification, based
on SEED and data authentication.
It has a parameter, a CK_MAC_GENERAL_PARAMS structure,
which specifies the output length desired from the mechanism.
The output bytes from this mechanism are taken from the
start of the final cipher block produced in the MACing process.
SEED-MAC, denoted by CKM_SEED_MAC, is a special case
of the general-length SEED-MAC mechanism. SEED-MAC always produces and verifies
MACs that are half the block size in length.
It does not have a parameter.
These mechanisms allow derivation of keys using the result
of an encryption operation as the key value. They are for use with the
C_DeriveKey function.
Mechanisms:
CKM_SEED_ECB_ENCRYPT_DATA
CKM_SEED_CBC_ENCRYPT_DATA
typedef struct CK_SEED_CBC_ENCRYPT_DATA_PARAMS {
CK_BYTE iv[16];
CK_BYTE_PTR pData;
CK_ULONG length;
} CK_SEED_CBC_ENCRYPT_DATA_PARAMS;
typedef CK_SEED_CBC_ENCRYPT_DATA_PARAMS CK_PTR
CK_SEED_CBC_ENCRYPT_DATA_PARAMS_PTR;
Table 224, Mechanism Parameters for SEED-based key derivation
|
CKM_SEED_ECB_ENCRYPT_DATA
|
Uses
CK_KEY_DERIVATION_STRING_DATA structure. Parameter is the data to be
encrypted and must be a multiple of 16 long.
|
|
CKM_SEED_CBC_ENCRYPT_DATA
|
Uses
CK_SEED_CBC_ENCRYPT_DATA_PARAMS. Parameter is an 16 byte IV value followed by
the data. The data value part must be a multiple of 16 bytes long.
|
OTP tokens represented as PKCS #11 mechanisms may be used in
a variety of ways. The usage cases can be categorized according to the type of
sought functionality.
.
Figure 2: Retrieving OTP values
through C_Sign
Figure 2 shows an integration of PKCS #11 into an
application that needs to authenticate users holding OTP tokens. In this
particular example, a connected hardware token is used, but a software token is
equally possible. The application invokes C_Sign to retrieve the OTP
value from the token. In the example, the application then passes the retrieved
OTP value to a client API that sends it via the network to an authentication
server. The client API may implement a standard authentication protocol such as
RADIUS [RFC 2865] or EAP [RFC 3748], or a proprietary protocol such as that
used by RSA Security's ACE/Agent® software.
Figure 3: Server-side
verification of OTP values
Figure 3 illustrates the server-side equivalent of the
scenario depicted in Figure 2. In this case, a server application invokes C_Verify
with the received OTP value as the signature value to be verified.
Figure 4: Generation of an OTP
key
Figure 4 shows an integration of PKCS #11 into an
application that generates OTP keys. The application invokes C_GenerateKey
to generate an OTP key of a particular type on the token. The key may
subsequently be used as a basis to generate OTP values.
OTP key objects (object class CKO_OTP_KEY) hold
secret keys used by OTP tokens. The following table defines the attributes
common to all OTP keys, in addition to the attributes defined for secret keys,
all of which are inherited by this class:
Table 225: Common OTP key
attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_OTP_FORMAT
|
CK_ULONG
|
Format of OTP values produced with this key:
CK_OTP_FORMAT_DECIMAL = Decimal (default)
(UTF8-encoded)
CK_OTP_FORMAT_HEXADECIMAL = Hexadecimal
(UTF8-encoded)
CK_OTP_FORMAT_ALPHANUMERIC = Alphanumeric
(UTF8-encoded)
CK_OTP_FORMAT_BINARY = Only binary values.
|
|
CKA_OTP_LENGTH9
|
CK_ULONG
|
Default length of OTP values (in the
CKA_OTP_FORMAT) produced with this key.
|
|
CKA_OTP_USER_FRIENDLY_MODE9
|
CK_BBOOL
|
Set to CK_TRUE when the token is capable of
returning OTPs suitable for human consumption. See the description of
CKF_USER_FRIENDLY_OTP below.
|
|
CKA_OTP_CHALLENGE_REQUIREMENT9
|
CK_ULONG
|
Parameter requirements when generating or
verifying OTP values with this key:
CK_OTP_PARAM_MANDATORY = A challenge must be
supplied.
CK_OTP_PARAM_OPTIONAL
= A challenge may be supplied but need not be.
CK_OTP_PARAM_IGNORED
= A challenge, if supplied, will be ignored.
|
|
CKA_OTP_TIME_REQUIREMENT9
|
CK_ULONG
|
Parameter requirements when generating or
verifying OTP values with this key:
CK_OTP_PARAM_MANDATORY = A time value must
be supplied.
CK_OTP_PARAM_OPTIONAL
= A time value may be supplied but need not be.
CK_OTP_PARAM_IGNORED = A time value, if
supplied, will be ignored.
|
|
CKA_OTP_COUNTER_REQUIREMENT9
|
CK_ULONG
|
Parameter requirements when generating or
verifying OTP values with this key:
CK_OTP_PARAM_MANDATORY = A counter value
must be supplied.
CK_OTP_PARAM_OPTIONAL
= A counter value may be supplied but need not be.
CK_OTP_PARAM_IGNORED = A counter value, if
supplied, will be ignored.
|
|
CKA_OTP_PIN_REQUIREMENT9
|
CK_ULONG
|
Parameter requirements when generating or
verifying OTP values with this key:
CK_OTP_PARAM_MANDATORY = A PIN value must be
supplied.
CK_OTP_PARAM_OPTIONAL = A PIN value may be
supplied but need not be (if not supplied, then library will be responsible
for collecting it)
CK_OTP_PARAM_IGNORED = A PIN value, if
supplied, will be ignored.
|
|
CKA_OTP_COUNTER
|
Byte array
|
Value of the associated internal counter.
Default value is empty (i.e. ulValueLen = 0).
|
|
CKA_OTP_TIME
|
RFC 2279 string
|
Value of the associated internal UTC time in
the form YYYYMMDDhhmmss. Default value is empty (i.e. ulValueLen= 0).
|
|
CKA_OTP_USER_IDENTIFIER
|
RFC 2279 string
|
Text string that identifies a user
associated with the OTP key (may be used to enhance the user experience).
Default value is empty (i.e. ulValueLen = 0).
|
|
CKA_OTP_SERVICE_IDENTIFIER
|
RFC 2279 string
|
Text string that identifies a service that
may validate OTPs generated by this key. Default value is empty (i.e. ulValueLen
= 0).
|
|
CKA_OTP_SERVICE_LOGO
|
Byte array
|
Logotype image that identifies a service
that may validate OTPs generated by this key. Default value is empty (i.e. ulValueLen
= 0).
|
|
CKA_OTP_SERVICE_LOGO_TYPE
|
RFC 2279 string
|
MIME type of the CKA_OTP_SERVICE_LOGO
attribute value. Default value is empty (i.e. ulValueLen = 0).
|
|
CKA_VALUE1, 4, 6, 7
|
Byte array
|
Value of the key.
|
|
CKA_VALUE_LEN2, 3
|
CK_ULONG
|
Length in bytes of key value.
|
- Refer to Table 11 for footnotes
Note: A Cryptoki library may support PIN-code caching in
order to reduce user interactions. An OTP-PKCS #11 application should therefore
always consult the state of the CKA_OTP_PIN_REQUIREMENT attribute before each
call to C_SignInit, as the value of this attribute may change
dynamically.
For OTP tokens with multiple keys, the keys may be
enumerated using C_FindObjects. The CKA_OTP_SERVICE_IDENTIFIER
and/or the CKA_OTP_SERVICE_LOGO attribute may be used to distinguish
between keys. The actual choice of key for a particular operation is however
application-specific and beyond the scope of this document.
For all OTP keys, the CKA_ALLOWED_MECHANISMS attribute
should be set as required.
This document extends the set of defined notifications as
follows:
CKN_OTP_CHANGED Cryptoki is informing
the application that the OTP for a key on a connected token just changed. This
notification is particularly useful when applications wish to display the
current OTP value for time-based mechanisms.
The following table shows, for the OTP mechanisms defined in
this document, their support by different cryptographic operations. For any
particular token, of course, a particular operation may well support only a
subset of the mechanisms listed. There is also no guarantee that a token that
supports one mechanism for some operation supports any other mechanism for any
other operation (or even supports that same mechanism for any other operation).
Table 226: OTP mechanisms vs.
applicable functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_SECURID_KEY_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_SECURID
|
|
ü
|
|
|
|
|
|
|
CKM_HOTP_KEY_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_HOTP
|
|
ü
|
|
|
|
|
|
|
CKM_ACTI_KEY_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_ACTI
|
|
ü
|
|
|
|
|
|
The remainder of this section will present in detail the OTP
mechanisms and the parameters that are supplied to them.
¨
CK_OTP_PARAM_TYPE
CK_OTP_PARAM_TYPE is a value that identifies an OTP
parameter type. It is defined as follows:
typedef CK_ULONG CK_OTP_PARAM_TYPE;
The following CK_OTP_PARAM_TYPE types are defined:
Table 227, OTP parameter types
|
Parameter
|
Data type
|
Meaning
|
|
CK_OTP_PIN
|
RFC 2279 string
|
A UTF8 string containing a PIN for use when
computing or verifying PIN-based OTP values.
|
|
CK_OTP_CHALLENGE
|
Byte array
|
Challenge to use when computing or verifying
challenge-based OTP values.
|
|
CK_OTP_TIME
|
RFC 2279 string
|
UTC time value in the form YYYYMMDDhhmmss to
use when computing or verifying time-based OTP values.
|
|
CK_OTP_COUNTER
|
Byte array
|
Counter value to use when computing or
verifying counter-based OTP values.
|
|
CK_OTP_FLAGS
|
CK_FLAGS
|
Bit flags indicating the characteristics of
the sought OTP as defined below.
|
|
CK_OTP_OUTPUT_LENGTH
|
CK_ULONG
|
Desired output length (overrides any default
value). A Cryptoki library will return CKR_MECHANISM_PARAM_INVALID if a
provided length value is not supported.
|
|
CK_OTP_OUTPUT_FORMAT
|
CK_ULONG
|
Returned OTP format (allowed values are the
same as for CKA_OTP_FORMAT). This parameter is only intended for C_Sign
output, see paragraphs below. When not present, the returned OTP format will
be the same as the value of the CKA_OTP_FORMAT attribute for the key in
question.
|
|
CK_OTP_VALUE
|
Byte array
|
An actual OTP value. This parameter type is
intended for C_Sign output, see paragraphs below.
|
The following table defines the possible values for the
CK_OTP_FLAGS type:
Table 228: OTP Mechanism Flags
|
Bit flag
|
Mask
|
Meaning
|
|
CKF_NEXT_OTP
|
0x00000001
|
True (i.e. set) if the OTP computation shall
be for the next OTP, rather than the current one (current being interpreted
in the context of the algorithm, e.g. for the current counter value or
current time window). A Cryptoki library shall return
CKR_MECHANISM_PARAM_INVALID if the CKF_NEXT_OTP flag is set and the OTP
mechanism in question does not support the concept of “next” OTP or the
library is not capable of generating the next OTP.
|
|
CKF_EXCLUDE_TIME
|
0x00000002
|
True (i.e. set) if the OTP computation must
not include a time value. Will have an effect only on mechanisms that do
include a time value in the OTP computation and then only if the mechanism
(and token) allows exclusion of this value. A Cryptoki library shall return
CKR_MECHANISM_PARAM_INVALID if exclusion of the value is not allowed.
|
|
CKF_EXCLUDE_COUNTER
|
0x00000004
|
True (i.e. set) if the OTP computation must
not include a counter value. Will have an effect only on mechanisms that do
include a counter value in the OTP computation and then only if the mechanism
(and token) allows exclusion of this value. A Cryptoki library shall return
CKR_MECHANISM_PARAM_INVALID if exclusion of the value is not allowed.
|
|
CKF_EXCLUDE_CHALLENGE
|
0x00000008
|
True (i.e. set) if the OTP computation must
not include a challenge. Will have an effect only on mechanisms that do
include a challenge in the OTP computation and then only if the mechanism
(and token) allows exclusion of this value. A Cryptoki library shall return
CKR_MECHANISM_PARAM_INVALID if exclusion of the value is not allowed.
|
|
CKF_EXCLUDE_PIN
|
0x00000010
|
True (i.e. set) if the OTP computation must
not include a PIN value. Will have an effect only on mechanisms that do
include a PIN in the OTP computation and then only if the mechanism (and
token) allows exclusion of this value. A Cryptoki library shall return
CKR_MECHANISM_PARAM_INVALID if exclusion of the value is not allowed.
|
|
CKF_USER_FRIENDLY_OTP
|
0x00000020
|
True (i.e. set) if the OTP returned shall be
in a form suitable for human consumption. If this flag is set, and the call
is successful, then the returned CK_OTP_VALUE shall be a UTF8-encoded
printable string. A Cryptoki library shall return CKR_MECHANISM_PARAM_INVALID
if this flag is set when CKA_OTP_USER_FRIENDLY_MODE for the key in question
is CK_FALSE.
|
Note: Even if CKA_OTP_FORMAT is not set to
CK_OTP_FORMAT_BINARY, then there may still be value in setting the CKF_USER_FRIENDLY_OTP
flag (assuming CKA_OTP_USER_FRIENDLY_MODE is CK_TRUE, of course) if the intent
is for a human to read the generated OTP value, since it may become shorter or
otherwise better suited for a user. Applications that do not intend to provide
a returned OTP value to a user should not set the CKF_USER_FRIENDLY_OTP flag.
¨
CK_OTP_PARAM;
CK_OTP_PARAM_PTR
CK_OTP_PARAM is a structure that includes the type,
value, and length of an OTP parameter. It is defined as follows:
typedef struct CK_OTP_PARAM {
CK_OTP_PARAM_TYPE type;
CK_VOID_PTR pValue;
CK_ULONG ulValueLen;
} CK_OTP_PARAM;
The fields of the structure have the following meanings:
type the
parameter type
pValue pointer
to the value of the parameter
ulValueLen length in
bytes of the value
If a parameter has no value, then ulValueLen = 0, and
the value of pValue is irrelevant. Note that pValue is a “void”
pointer, facilitating the passing of arbitrary values. Both the application
and the Cryptoki library must ensure that the pointer can be safely cast to the
expected type (i.e., without word-alignment errors).
CK_OTP_PARAM_PTR is a pointer to a CK_OTP_PARAM.
¨
CK_OTP_PARAMS;
CK_OTP_PARAMS_PTR
CK_OTP_PARAMS is a structure that is used to provide
parameters for OTP mechanisms in a generic fashion. It is defined as follows:
typedef struct CK_OTP_PARAMS {
CK_OTP_PARAM_PTR pParams;
CK_ULONG ulCount;
} CK_OTP_PARAMS;
The fields of the structure have the following meanings:
pParams pointer
to an array of OTP parameters
ulCount the
number of parameters in the array
CK_OTP_PARAMS_PTR is a pointer to a CK_OTP_PARAMS.
When calling C_SignInit or C_VerifyInit with a mechanism
that takes a CK_OTP_PARAMS structure as a parameter, the CK_OTP_PARAMS
structure shall be populated in accordance with the CKA_OTP_X_REQUIREMENT
key attributes for the identified key, where X is PIN, CHALLENGE, TIME,
or COUNTER.
For example, if CKA_OTP_TIME_REQUIREMENT =
CK_OTP_PARAM_MANDATORY, then the CK_OTP_TIME parameter shall be present. If
CKA_OTP_TIME_REQUIREMENT = CK_OTP_PARAM_OPTIONAL, then a CK_OTP_TIME parameter
may be present. If it is not present, then the library may collect it (during
the C_Sign call). If CKA_OTP_TIME_REQUIREMENT = CK_OTP_PARAM_IGNORED, then a
provided CK_OTP_TIME parameter will always be ignored. Additionally, a provided
CK_OTP_TIME parameter will always be ignored if CKF_EXCLUDE_TIME is set in a
CK_OTP_FLAGS parameter. Similarly, if this flag is set, a library will not
attempt to collect the value itself, and it will also instruct the token not to
make use of any internal value, subject to token policies. It is an error
(CKR_MECHANISM_PARAM_INVALID) to set the CKF_EXCLUDE_TIME flag when the
CKA_OTP_TIME_REQUIREMENT attribute is CK_OTP_PARAM_MANDATORY.
The above discussion holds for all CKA_OTP_X_REQUIREMENT
attributes (i.e., CKA_OTP_PIN_REQUIREMENT,
CKA_OTP_CHALLENGE_REQUIREMENT, CKA_OTP_COUNTER_REQUIREMENT,
CKA_OTP_TIME_REQUIREMENT). A library may set a particular CKA_OTP_X_REQUIREMENT
attribute to CK_OTP_PARAM_OPTIONAL even if it is required by the mechanism as
long as the token (or the library itself) has the capability of providing the
value to the computation. One example of this is a token with an on-board
clock.
In addition, applications may use the CK_OTP_FLAGS, the
CK_OTP_OUTPUT_FORMAT and the CKA_OTP_LENGTH parameters to set additional
parameters.
¨
CK_OTP_SIGNATURE_INFO,
CK_OTP_SIGNATURE_INFO_PTR
CK_OTP_SIGNATURE_INFO is a structure that is returned
by all OTP mechanisms in successful calls to C_Sign (C_SignFinal).
The structure informs applications of actual parameter values used in
particular OTP computations in addition to the OTP value itself. It is used by
all mechanisms for which the key belongs to the class CKO_OTP_KEY and is
defined as follows:
typedef struct CK_OTP_SIGNATURE_INFO {
CK_OTP_PARAM_PTR pParams;
CK_ULONG ulCount;
} CK_OTP_SIGNATURE_INFO;
The fields of the structure have the following meanings:
pParams pointer
to an array of OTP parameter values
ulCount the
number of parameters in the array
After successful calls to C_Sign or C_SignFinal
with an OTP mechanism, the pSignature parameter will be set to point to
a CK_OTP_SIGNATURE_INFO structure. One of the parameters in this
structure will be the OTP value itself, identified with the CK_OTP_VALUE
tag. Other parameters may be present for informational purposes, e.g. the
actual time used in the OTP calculation. In order to simplify OTP validations,
authentication protocols may permit authenticating parties to send some or all
of these parameters in addition to OTP values themselves. Applications should
therefore check for their presence in returned CK_OTP_SIGNATURE_INFO values
whenever such circumstances apply.
Since C_Sign and C_SignFinal follows the
convention described in Section 5.2 on producing output, a call to C_Sign (or
C_SignFinal) with pSignature set to NULL_PTR will return (in the pulSignatureLen
parameter) the required number of bytes to hold the CK_OTP_SIGNATURE_INFO
structure as well as all the data in all its CK_OTP_PARAM components. If
an application allocates a memory block based on this information, it shall
therefore not subsequently de-allocate components of such a received value but
rather de-allocate the complete CK_OTP_PARAMS structure itself. A
Cryptoki library that is called with a non-NULL pSignature pointer will
assume that it points to a contiguous memory block of the size indicated
by the pulSignatureLen parameter.
When verifying an OTP value using an OTP mechanism, pSignature
shall be set to the OTP value itself, e.g. the value of the CK_OTP_VALUE
component of a CK_OTP_PARAM structure returned by a call to C_Sign.
The CK_OTP_PARAM value supplied in the C_VerifyInit call sets the
values to use in the verification operation.
CK_OTP_SIGNATURE_INFO_PTR points to a CK_OTP_SIGNATURE_INFO.
RSA SecurID secret key objects (object class CKO_OTP_KEY,
key type CKK_SECURID) hold RSA SecurID secret keys. The following table
defines the RSA SecurID secret key object attributes, in addition to the common
attributes defined for this object class:
Table 229, RSA SecurID secret
key object attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_OTP_TIME_INTERVAL1
|
CK_ULONG
|
Interval between OTP values produced with
this key, in seconds. Default is 60.
|
- Refer to Table 11 for footnotes
The following is a sample template for creating an RSA
SecurID secret key object:
CK_OBJECT_CLASS class = CKO_OTP_KEY;
CK_KEY_TYPE keyType = CKK_SECURID;
CK_DATE endDate = {...};
CK_UTF8CHAR label[] = “RSA SecurID secret key object”;
CK_BYTE keyId[]= {...};
CK_ULONG outputFormat = CK_OTP_FORMAT_DECIMAL;
CK_ULONG outputLength = 6;
CK_ULONG needPIN = CK_OTP_PARAM_MANDATORY;
CK_ULONG timeInterval = 60;
CK_BYTE value[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_END_DATE, &endDate, sizeof(endDate)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_SENSITIVE, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_SIGN, &true, sizeof(true)},
{CKA_VERIFY, &true, sizeof(true)},
{CKA_ID, keyId, sizeof(keyId)},
{CKA_OTP_FORMAT, &outputFormat, sizeof(outputFormat)},
{CKA_OTP_LENGTH, &outputLength, sizeof(outputLength)},
{CKA_OTP_PIN_REQUIREMENT, &needPIN, sizeof(needPIN)},
{CKA_OTP_TIME_INTERVAL, &timeInterval,
sizeof(timeInterval)},
{CKA_VALUE, value, sizeof(value)}
};
The RSA SecurID key generation mechanism, denoted CKM_SECURID_KEY_GEN,
is a key generation mechanism for the RSA SecurID algorithm.
It does not have a parameter.
The mechanism generates RSA SecurID keys with a particular
set of attributes as specified in the template for the key.
The mechanism contributes at least the CKA_CLASS, CKA_KEY_TYPE,
CKA_VALUE_LEN, and CKA_VALUE attributes to the new key. Other
attributes supported by the RSA SecurID key type may be specified in the
template for the key, or else are assigned default initial values
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
SecurID key sizes, in bytes.
CKM_SECURID is the mechanism for the retrieval and
verification of RSA SecurID OTP values.
The mechanism takes a pointer to a CK_OTP_PARAMS
structure as a parameter.
When signing or verifying using the CKM_SECURID
mechanism, pData shall be set to NULL_PTR and ulDataLen shall be
set to 0.
Support for the CKM_SECURID mechanism extends the set of return
values for C_Verify with the following values:
·
CKR_NEW_PIN_MODE: The supplied OTP was not accepted and the
library requests a new OTP computed using a new PIN. The new PIN is set through
means out of scope for this document.
·
CKR_NEXT_OTP: The supplied OTP was correct but indicated a larger
than normal drift in the token's internal state (e.g. clock, counter). To
ensure this was not due to a temporary problem, the application should provide
the next one-time password to the library for verification.
HOTP secret key objects (object class CKO_OTP_KEY, key
type CKK_HOTP) hold generic secret keys and associated counter values.
The CKA_OTP_COUNTER value may be set at key
generation; however, some tokens may set it to a fixed initial value. Depending
on the token’s security policy, this value may not be modified and/or may not
be revealed if the object has its CKA_SENSITIVE attribute set to CK_TRUE
or its CKA_EXTRACTABLE attribute set to CK_FALSE.
For HOTP keys, the CKA_OTP_COUNTER value shall
be an 8 bytes unsigned integer in big endian (i.e. network byte order) form.
The same holds true for a CK_OTP_COUNTER value in a CK_OTP_PARAM structure.
The following is a sample template for creating a HOTP
secret key object:
CK_OBJECT_CLASS class = CKO_OTP_KEY;
CK_KEY_TYPE keyType = CKK_HOTP;
CK_UTF8CHAR label[] = “HOTP secret key object”;
CK_BYTE keyId[]= {...};
CK_ULONG outputFormat = CK_OTP_FORMAT_DECIMAL;
CK_ULONG outputLength = 6;
CK_DATE endDate = {...};
CK_BYTE counterValue[8] = {0};
CK_BYTE value[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_END_DATE, &endDate, sizeof(endDate)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_SENSITIVE, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_SIGN, &true, sizeof(true)},
{CKA_VERIFY, &true, sizeof(true)},
{CKA_ID, keyId, sizeof(keyId)},
{CKA_OTP_FORMAT, &outputFormat, sizeof(outputFormat)},
{CKA_OTP_LENGTH, &outputLength, sizeof(outputLength)},
{CKA_OTP_COUNTER, counterValue, sizeof(counterValue)},
{CKA_VALUE, value, sizeof(value)}
};
The HOTP key generation mechanism, denoted CKM_HOTP_KEY_GEN,
is a key generation mechanism for the HOTP algorithm.
It does not have a parameter.
The mechanism generates HOTP keys with a particular set of
attributes as specified in the template for the key.
The mechanism contributes at least the CKA_CLASS, CKA_KEY_TYPE,
CKA_OTP_COUNTER, CKA_VALUE and CKA_VALUE_LEN attributes to
the new key. Other attributes supported by the HOTP key type may be specified
in the template for the key, or else are assigned default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
HOTP key sizes, in bytes.
CKM_HOTP is the mechanism for the retrieval and
verification of HOTP OTP values based on the current internal counter, or a
provided counter.
The mechanism takes a pointer to a CK_OTP_PARAMS
structure as a parameter.
As for the CKM_SECURID mechanism, when signing or
verifying using the CKM_HOTP mechanism, pData shall be set to
NULL_PTR and ulDataLen shall be set to 0.
For verify operations, the counter value CK_OTP_COUNTER
must be provided as a CK_OTP_PARAM parameter to C_VerifyInit.
When verifying an OTP value using the CKM_HOTP mechanism, pSignature
shall be set to the OTP value itself, e.g. the value of the CK_OTP_VALUE
component of a CK_OTP_PARAM structure in the case of an earlier call to C_Sign.
ACTI secret key objects (object class CKO_OTP_KEY, key
type CKK_ACTI) hold ActivIdentity ACTI secret keys.
For ACTI keys, the CKA_OTP_COUNTER value shall be an
8 bytes unsigned integer in big endian (i.e. network byte order) form. The same
holds true for the CK_OTP_COUNTER value in the CK_OTP_PARAM structure.
The CKA_OTP_COUNTER value may be set at key
generation; however, some tokens may set it to a fixed initial value. Depending
on the token’s security policy, this value may not be modified and/or may not
be revealed if the object has its CKA_SENSITIVE attribute set to CK_TRUE
or its CKA_EXTRACTABLE attribute set to CK_FALSE.
The CKA_OTP_TIME value may be set at key generation;
however, some tokens may set it to a fixed initial value. Depending on the
token’s security policy, this value may not be modified and/or may not be
revealed if the object has its CKA_SENSITIVE attribute set to CK_TRUE or
its CKA_EXTRACTABLE attribute set to CK_FALSE.
The following is a sample template for creating an ACTI
secret key object:
CK_OBJECT_CLASS class = CKO_OTP_KEY;
CK_KEY_TYPE keyType = CKK_ACTI;
CK_UTF8CHAR label[] = “ACTI secret key object”;
CK_BYTE keyId[]= {...};
CK_ULONG outputFormat = CK_OTP_FORMAT_DECIMAL;
CK_ULONG outputLength = 6;
CK_DATE endDate = {...};
CK_BYTE counterValue[8] = {0};
CK_BYTE value[] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_END_DATE, &endDate, sizeof(endDate)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_SENSITIVE, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_SIGN, &true, sizeof(true)},
{CKA_VERIFY, &true, sizeof(true)},
{CKA_ID, keyId, sizeof(keyId)},
{CKA_OTP_FORMAT, &outputFormat,
sizeof(outputFormat)},
{CKA_OTP_LENGTH, &outputLength,
sizeof(outputLength)},
{CKA_OTP_COUNTER, counterValue,
sizeof(counterValue)},
{CKA_VALUE, value, sizeof(value)}
};
The ACTI key generation mechanism, denoted CKM_ACTI_KEY_GEN,
is a key generation mechanism for the ACTI algorithm.
It does not have a parameter.
The mechanism generates ACTI keys with a particular set of
attributes as specified in the template for the key.
The mechanism contributes at least the CKA_CLASS, CKA_KEY_TYPE,
CKA_VALUE and CKA_VALUE_LEN attributes to the new key. Other
attributes supported by the ACTI key type may be specified in the template for
the key, or else are assigned default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported
range of ACTI key sizes, in bytes.
CKM_ACTI is the mechanism for the retrieval and
verification of ACTI OTP values.
The mechanism takes a pointer to a CK_OTP_PARAMS structure
as a parameter.
When signing or verifying using the CKM_ACTI mechanism,
pData shall be set to NULL_PTR and ulDataLen shall be set to 0.
When verifying an OTP value using the CKM_ACTI mechanism,
pSignature shall be set to the OTP value itself, e.g. the value of the CK_OTP_VALUE
component of a CK_OTP_PARAM structure in the case of an earlier call
to C_Sign.
Figure 5: PKCS #11 and CT-KIP
integration
Figure 5 shows an integration of PKCS #11 into an
application that generates cryptographic keys through the use of CT-KIP. The
application invokes C_DeriveKey to derive a key of a particular type on
the token. The key may subsequently be used as a basis to e.g., generate
one-time password values. The application communicates with a CT-KIP server
that participates in the key derivation and stores a copy of the key in its
database. The key is transferred to the server in wrapped form, after a call to
C_WrapKey. The server authenticates itself to the client and the client
verifies the authentication by calls to C_Verify.
The following table shows, for the mechanisms defined in
this document, their support by different cryptographic operations. For any
particular token, of course, a particular operation may well support only a
subset of the mechanisms listed. There is also no guarantee that a token that
supports one mechanism for some operation supports any other mechanism for any
other operation (or even supports that same mechanism for any other operation).
Table 230: CT-KIP Mechanisms vs. applicable functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_KIP_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_KIP_WRAP
|
|
|
|
|
|
ü
|
|
|
CKM_KIP_MAC
|
|
ü
|
|
|
|
|
|
The remainder of this section will present in detail the
mechanisms and the parameters that are supplied to them.
Mechanisms:
CKM_KIP_DERIVE
CKM_KIP_WRAP
CKM_KIP_MAC
¨ CK_KIP_PARAMS; CK_KIP_PARAMS_PTR
CK_KIP_PARAMS is a structure that provides the
parameters to all the CT-KIP related mechanisms: The CKM_KIP_DERIVE key
derivation mechanism, the CKM_KIP_WRAP key wrap and key unwrap
mechanism, and the CKM_KIP_MAC signature mechanism. The structure is
defined as follows:
typedef struct CK_KIP_PARAMS {
CK_MECHANISM_PTR pMechanism;
CK_OBJECT_HANDLE hKey;
CK_BYTE_PTR pSeed;
CK_ULONG ulSeedLen;
} CK_KIP_PARAMS;
The fields of the structure have the following meanings:
pMechanism pointer to
the underlying cryptographic mechanism (e.g. AES, SHA-256)
hKey handle
to a key that will contribute to the entropy of the derived key
(CKM_KIP_DERIVE) or will be used in the MAC operation (CKM_KIP_MAC)
pSeed pointer
to an input seed
ulSeedLen length in
bytes of the input seed
CK_KIP_PARAMS_PTR is a pointer to a CK_KIP_PARAMS
structure.
The CT-KIP key derivation mechanism, denoted CKM_KIP_DERIVE,
is a key derivation mechanism that is capable of generating secret keys of
potentially any type, subject to token limitations.
It takes a parameter of type CK_KIP_PARAMS which
allows for the passing of the desired underlying cryptographic mechanism as
well as some other data. In particular, when the hKey parameter is a
handle to an existing key, that key will be used in the key derivation in
addition to the hBaseKey of C_DeriveKey. The pSeed
parameter may be used to seed the key derivation operation.
The mechanism derives a secret key with a particular set of
attributes as specified in the attributes of the template for the key.
The mechanism contributes the CKA_CLASS and CKA_VALUE
attributes to the new key. Other attributes supported by the key type may be
specified in the template for the key, or else will be assigned default initial
values. Since the mechanism is generic, the CKA_KEY_TYPE attribute
should be set in the template, if the key is to be used with a particular
mechanism.
The CT-KIP key wrap and unwrap mechanism, denoted CKM_KIP_WRAP,
is a key wrap mechanism that is capable of wrapping and unwrapping generic
secret keys.
It takes a parameter of type CK_KIP_PARAMS, which
allows for the passing of the desired underlying cryptographic mechanism as
well as some other data. It does not make use of the hKey parameter of CK_KIP_PARAMS.
The CT-KIP signature (MAC) mechanism, denoted CKM_KIP_MAC,
is a mechanism used to produce a message authentication code of arbitrary
length. The keys it uses are secret keys.
It takes a parameter of type CK_KIP_PARAMS, which
allows for the passing of the desired underlying cryptographic mechanism as
well as some other data. The mechanism does not make use of the pSeed
and the ulSeedLen parameters of CT_KIP_PARAMS.
This mechanism produces a MAC of the length specified by pulSignatureLen
parameter in calls to C_Sign.
If a call to C_Sign with this mechanism fails, then
no output will be generated.
GOST 28147-89 is a block cipher with 64-bit block size and
256-bit keys.
Table
231, GOST 28147-89 Mechanisms vs. Functions
|
Mechanism
|
Functions
|
|
Encrypt &
Decrypt
|
Sign &
Verify
|
SR & VR
|
Digest
|
Gen. Key/ Key
Pair
|
Wrap &
Unwrap
|
Derive
|
|
CKM_GOST28147_KEY_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_GOST28147_ECB
|
ü
|
|
|
|
|
ü
|
|
|
CKM_GOST28147
|
ü
|
|
|
|
|
ü
|
|
|
CKM_GOST28147_MAC
|
|
ü
|
|
|
|
|
|
|
CKM_GOST28147_KEY_WRAP
|
|
|
|
|
|
ü
|
|
This section defines the key type “CKK_GOST28147” for type
CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of key objects and domain
parameter objects.
Mechanisms:
CKM_GOST28147_KEY_GEN
CKM_GOST28147_ECB
CKM_GOST28147
CKM_GOST28147_MAC
CKM_GOST28147_KEY_WRAP
GOST 28147‑89 secret key objects (object class CKO_SECRET_KEY,
key type CKK_GOST28147) hold GOST 28147‑89 keys. The
following table defines the GOST 28147‑89 secret key object
attributes, in addition to the common attributes defined for this object class:
Table
232, GOST 28147-89 Secret Key Object Attributes
|
Attribute
|
Data type
|
Meaning
|
|
CKA_VALUE1,4,6,7
|
Byte array
|
32 bytes in little endian order
|
|
CKA_GOST28147_PARAMS1,3,5
|
Byte array
|
DER-encoding of the object identifier indicating the data
object type of GOST 28147‑89.
When key is used the domain parameter object of key type
CKK_GOST28147 must be specified with the same attribute CKA_OBJECT_ID
|
- Refer to Table 11 for footnotes
The following is a sample template for creating a
GOST 28147‑89 secret key object:
CK_OBJECT_CLASS class = CKO_SECRET_KEY;
CK_KEY_TYPE keyType = CKK_GOST28147;
CK_UTF8CHAR label[] = “A GOST 28147-89 secret key object”;
CK_BYTE value[32] = {...};
CK_BYTE params_oid[] = {0x06, 0x07, 0x2a, 0x85, 0x03, 0x02,
0x02, 0x1f, 0x00};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_ENCRYPT, &true, sizeof(true)},
{CKA_GOST28147_PARAMS, params_oid, sizeof(params_oid)},
{CKA_VALUE, value, sizeof(value)}
};
GOST 28147‑89 domain parameter objects (object
class CKO_DOMAIN_PARAMETERS, key type CKK_GOST28147) hold GOST 28147‑89
domain parameters.
The following table defines the GOST 28147‑89 domain
parameter object attributes, in addition to the common attributes defined for
this object class:
Table 233, GOST 28147-89 Domain Parameter Object
Attributes
|
Attribute
|
Data Type
|
Meaning
|
|
CKA_VALUE1
|
Byte array
|
DER-encoding of the domain parameters as it
was introduced in [4] section 8.1 (type Gost28147-89-ParamSetParameters)
|
|
CKA_OBJECT_ID1
|
Byte array
|
DER-encoding of the object identifier
indicating the domain parameters
|
- Refer to Table 11 for footnotes
For any particular token, there is no guarantee that a token
supports domain parameters loading up and/or fetching out. Furthermore,
applications, that make direct use of domain parameters objects, should take in
account that CKA_VALUE attribute may be inaccessible.
The following is a sample template for creating a
GOST 28147‑89 domain parameter object:
CK_OBJECT_CLASS class = CKO_DOMAIN_PARAMETERS;
CK_KEY_TYPE keyType = CKK_GOST28147;
CK_UTF8CHAR label[] = “A GOST 28147-89 cryptographic parameters
object”;
CK_BYTE oid[] = {0x06, 0x07, 0x2a, 0x85, 0x03, 0x02, 0x02, 0x1f,
0x00};
CK_BYTE value[] = {
0x30,0x62,0x04,0x40,0x4c,0xde,0x38,0x9c,0x29,0x89,0xef,0xb6,
0xff,0xeb,0x56,0xc5,0x5e,0xc2,0x9b,0x02,0x98,0x75,0x61,0x3b,
0x11,0x3f,0x89,0x60,0x03,0x97,0x0c,0x79,0x8a,0xa1,0xd5,0x5d,
0xe2,0x10,0xad,0x43,0x37,0x5d,0xb3,0x8e,0xb4,0x2c,0x77,0xe7,
0xcd,0x46,0xca,0xfa,0xd6,0x6a,0x20,0x1f,0x70,0xf4,0x1e,0xa4,
0xab,0x03,0xf2,0x21,0x65,0xb8,0x44,0xd8,0x02,0x01,0x00,0x02,
0x01,0x40,0x30,0x0b,0x06,0x07,0x2a,0x85,0x03,0x02,0x02,0x0e,
0x00,0x05,0x00
};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_OBJECT_ID, oid, sizeof(oid)},
{CKA_VALUE, value, sizeof(value)}
};
The GOST 28147‑89 key generation mechanism,
denoted CKM_GOST28147_KEY_GEN, is a key generation mechanism for
GOST 28147‑89.
It does not have a parameter.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key. Other attributes supported by
the GOST 28147‑89 key type may be specified for objects of object
class CKO_SECRET_KEY.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO are not used.
GOST 28147‑89-ECB, denoted CKM_GOST28147_ECB,
is a mechanism for single and multiple-part encryption and decryption; key
wrapping; and key unwrapping, based on GOST 28147‑89 and electronic
codebook mode.
It does not have a parameter.
This mechanism can wrap and unwrap any secret key. Of
course, a particular token may not be able to wrap/unwrap every secret key that
it supports.
For wrapping (C_WrapKey), the mechanism encrypts the
value of the CKA_VALUE attribute of the key that is wrapped, padded on
the trailing end with up to block size so that the resulting length is a
multiple of the block size.
For unwrapping (C_UnwrapKey), the mechanism decrypts
the wrapped key, and truncates the result according to the CKA_KEY_TYPE attribute
of the template and, if it has one, and the key type supports it, the CKA_VALUE_LEN
attribute of the template. The mechanism contributes the result as the CKA_VALUE
attribute of the new key.
Constraints on key types and the length of data are
summarized in the following table:
Table
234, GOST 28147-89-ECB: Key and Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Encrypt
|
CKK_GOST28147
|
Multiple of block
size
|
Same as input length
|
|
C_Decrypt
|
CKK_GOST28147
|
Multiple of block
size
|
Same as input length
|
|
C_WrapKey
|
CKK_GOST28147
|
Any
|
Input length rounded up to multiple of block size
|
|
C_UnwrapKey
|
CKK_GOST28147
|
Multiple of block
size
|
Determined by type of key being unwrapped
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure are not used.
GOST 28147‑89 encryption mode except ECB, denoted
CKM_GOST28147, is a mechanism for single and multiple-part encryption
and decryption; key wrapping; and key unwrapping, based on [GOST 28147‑89]
and CFB, counter mode, and additional CBC mode defined in [RFC 4357] section 2.
Encryption’s parameters are specified in object identifier of attribute CKA_GOST28147_PARAMS.
It has a parameter, which is an 8-byte initialization
vector. This parameter may be omitted then a zero initialization vector is
used.
This mechanism can wrap and unwrap any secret key. Of
course, a particular token may not be able to wrap/unwrap every secret key that
it supports.
For wrapping (C_WrapKey), the mechanism encrypts the
value of the CKA_VALUE attribute of the key that is wrapped.
For unwrapping (C_UnwrapKey), the mechanism decrypts
the wrapped key, and contributes the result as the CKA_VALUE attribute
of the new key.
Constraints on key types and the length of data are
summarized in the following table:
Table
235, GOST 28147-89 encryption modes except ECB: Key and
Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Encrypt
|
CKK_GOST28147
|
Any
|
For counter mode and CFB is the same as input length. For
CBC is the same as input length padded on the trailing end with up to block
size so that the resulting length is a multiple of the block size
|
|
C_Decrypt
|
CKK_GOST28147
|
Any
|
|
C_WrapKey
|
CKK_GOST28147
|
Any
|
|
C_UnwrapKey
|
CKK_GOST28147
|
Any
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure are not used.
GOST 28147-89-MAC, denoted CKM_GOST28147_MAC, is a mechanism
for data integrity and authentication based on GOST 28147-89 and key meshing
algorithms [RFC 4357] section 2.3.
MACing parameters are specified in object identifier of
attribute CKA_GOST28147_PARAMS.
The output bytes from this mechanism are taken from the
start of the final GOST 28147‑89 cipher block produced in the MACing
process.
It has a parameter, which is an 8-byte MAC initialization
vector. This parameter may be omitted then a zero initialization vector is
used.
Constraints on key types and the length of data are
summarized in the following table:
Table
236, GOST28147-89-MAC: Key and Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
CKK_GOST28147
|
Any
|
4 bytes
|
|
C_Verify
|
CKK_GOST28147
|
Any
|
4 bytes
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure are not used.
GOST 28147‑89 keys as a KEK (key encryption keys)
for encryption GOST 28147‑89 keys, denoted by CKM_GOST28147_KEY_WRAP,
is a mechanism for key wrapping; and key unwrapping, based on GOST 28147‑89.
Its purpose is to encrypt and decrypt keys have been generated by key generation
mechanism for GOST 28147‑89.
For wrapping (C_WrapKey), the mechanism first
computes MAC from the value of the CKA_VALUE attribute of the key that
is wrapped and then encrypts in ECB mode the value of the CKA_VALUE attribute
of the key that is wrapped. The result is 32 bytes of the key that is wrapped
and 4 bytes of MAC.
For unwrapping (C_UnwrapKey), the mechanism first
decrypts in ECB mode the 32 bytes of the key that was wrapped and then computes
MAC from the unwrapped key. Then compared together 4 bytes MAC has computed and
4 bytes MAC of the input. If these two MACs do not match the wrapped key is
disallowed. The mechanism contributes the result as the CKA_VALUE attribute
of the unwrapped key.
It has a parameter, which is an 8-byte MAC initialization
vector. This parameter may be omitted then a zero initialization vector is
used.
Constraints on key types and the length of data are
summarized in the following table:
Table
237, GOST 28147-89 keys as KEK: Key and Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_WrapKey
|
CKK_GOST28147
|
32 bytes
|
36 bytes
|
|
C_UnwrapKey
|
CKK_GOST28147
|
32 bytes
|
36 bytes
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure are not used.
GOST R 34.11-94 is a mechanism for message digesting,
following the hash algorithm with 256-bit message digest defined in [GOST R
34.11-94].
Table
238, GOST R 34.11-94 Mechanisms vs. Functions
|
Mechanism
|
Functions
|
|
Encrypt &
Decrypt
|
Sign &
Verify
|
SR & VR
|
Digest
|
Gen. Key/ Key
Pair
|
Wrap &
Unwrap
|
Derive
|
|
CKM_GOSTR3411
|
|
|
|
ü
|
|
|
|
|
CKM_GOSTR3411_HMAC
|
|
ü
|
|
|
|
|
|
This section defines the key type “CKK_GOSTR3411” for type CK_KEY_TYPE
as used in the CKA_KEY_TYPE attribute of domain parameter objects.
Mechanisms:
CKM_GOSTR3411
CKM_GOSTR3411_HMAC
GOST R 34.11-94 domain parameter objects (object
class CKO_DOMAIN_PARAMETERS, key type CKK_GOSTR3411) hold GOST R
34.11-94 domain parameters.
The following table defines the GOST R 34.11-94 domain
parameter object attributes, in addition to the common attributes defined for
this object class:
Table 239, GOST R 34.11-94
Domain Parameter Object Attributes
|
Attribute
|
Data Type
|
Meaning
|
|
CKA_VALUE1
|
Byte array
|
DER-encoding of the domain parameters as it
was introduced in [4] section 8.2 (type GostR3411-94-ParamSetParameters)
|
|
CKA_OBJECT_ID1
|
Byte array
|
DER-encoding of the object identifier
indicating the domain parameters
|
- Refer to Table 11 for footnotes
For any particular token, there is no guarantee that a token
supports domain parameters loading up and/or fetching out. Furthermore,
applications, that make direct use of domain parameters objects, should take in
account that CKA_VALUE attribute may be inaccessible.
The following is a sample template for creating a GOST R
34.11-94 domain parameter object:
CK_OBJECT_CLASS class = CKO_DOMAIN_PARAMETERS;
CK_KEY_TYPE keyType = CKK_GOSTR3411;
CK_UTF8CHAR label[] = “A GOST R34.11-94 cryptographic parameters
object”;
CK_BYTE oid[] = {0x06, 0x07, 0x2a, 0x85, 0x03, 0x02, 0x02, 0x1e,
0x00};
CK_BYTE value[] = {
0x30,0x64,0x04,0x40,0x4e,0x57,0x64,0xd1,0xab,0x8d,0xcb,0xbf,
0x94,0x1a,0x7a,0x4d,0x2c,0xd1,0x10,0x10,0xd6,0xa0,0x57,0x35,
0x8d,0x38,0xf2,0xf7,0x0f,0x49,0xd1,0x5a,0xea,0x2f,0x8d,0x94,
0x62,0xee,0x43,0x09,0xb3,0xf4,0xa6,0xa2,0x18,0xc6,0x98,0xe3,
0xc1,0x7c,0xe5,0x7e,0x70,0x6b,0x09,0x66,0xf7,0x02,0x3c,0x8b,
0x55,0x95,0xbf,0x28,0x39,0xb3,0x2e,0xcc,0x04,0x20,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00
};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_OBJECT_ID, oid, sizeof(oid)},
{CKA_VALUE, value, sizeof(value)}
};
GOST R 34.11-94 digest, denoted CKM_GOSTR3411, is a
mechanism for message digesting based on GOST R 34.11-94 hash algorithm [GOST R
34.11-94].
As a parameter this mechanism utilizes a DER-encoding of the
object identifier. A mechanism parameter may be missed then parameters of the
object identifier id-GostR3411-94-CryptoProParamSet [RFC 4357] (section
11.2) must be used.
Constraints on the length of input and output data are
summarized in the following table. For single-part digesting, the data and the
digest may begin at the same location in memory.
Table 240, GOST R 34.11-94: Data Length
|
Function
|
Input length
|
Digest length
|
|
C_Digest
|
Any
|
32 bytes
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure are not used.
GOST R 34.11-94 HMAC mechanism, denoted CKM_GOSTR3411_HMAC,
is a mechanism for signatures and verification. It uses the HMAC construction,
based on the GOST R 34.11-94 hash function [GOST R 34.11-94] and core HMAC
algorithm [RFC 2104]. The keys it uses are of generic key type CKK_GENERIC_SECRET
or CKK_GOST28147.
To be conformed to GOST R 34.11-94 hash algorithm [GOST R
34.11-94] the block length of core HMAC algorithm is 32 bytes long (see [RFC
2104] section 2, and [RFC 4357] section 3).
As a parameter this mechanism utilizes a DER-encoding of the
object identifier. A mechanism parameter may be missed then parameters of the
object identifier id-GostR3411-94-CryptoProParamSet [RFC 4357] (section
11.2) must be used.
Signatures (MACs) produced by this mechanism are of 32 bytes
long.
Constraints on the length of input and output data are
summarized in the following table:
Table 241, GOST R 34.11-94 HMAC: Key And Data Length
|
Function
|
Key type
|
Data length
|
Signature length
|
|
C_Sign
|
CKK_GENERIC_SECRET or CKK_GOST28147
|
Any
|
32 byte
|
|
C_Verify
|
CKK_GENERIC_SECRET or CKK_GOST28147
|
Any
|
32 bytes
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure are not used.
GOST R 34.10-2001 is a mechanism for single- and
multiple-part signatures and verification, following the digital signature
algorithm defined in [GOST R 34.10-2001].
Table
242, GOST R34.10-2001 Mechanisms vs. Functions
|
Mechanism
|
Functions
|
|
Encrypt &
Decrypt
|
Sign &
Verify
|
SR & VR
|
Digest
|
Gen. Key/ Key
Pair
|
Wrap &
Unwrap
|
Derive
|
|
CKM_GOSTR3410_KEY_PAIR_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_GOSTR3410
|
|
ü1
|
|
|
|
|
|
|
CKM_GOSTR3410_WITH_GOSTR3411
|
|
ü
|
|
|
|
|
|
|
CKM_GOSTR3410_KEY_WRAP
|
|
|
|
|
|
ü
|
|
|
CKM_GOSTR3410_DERIVE
|
|
|
|
|
|
|
ü
|
1
This section defines the key type “CKK_GOSTR3410” for type
CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of key objects and domain
parameter objects.
Mechanisms:
CKM_GOSTR3410_KEY_PAIR_GEN
CKM_GOSTR3410
CKM_GOSTR3410_WITH_GOSTR3411
CKM_GOSTR3410
CKM_GOSTR3410_KEY_WRAP
CKM_GOSTR3410_DERIVE
GOST R 34.10-2001 public key objects (object class
CKO_PUBLIC_KEY, key type CKK_GOSTR3410) hold GOST R 34.10-2001
public keys.
The following table defines the GOST R 34.10-2001 public key
object attributes, in addition to the common attributes defined for this object
class:
Table 243, GOST R 34.10-2001
Public Key Object Attributes
|
Attribute
|
Data Type
|
Meaning
|
|
CKA_VALUE1,4
|
Byte array
|
64 bytes for public key; 32 bytes for each
coordinates X and Y of Elliptic Curve point P(X, Y) in little endian
order
|
|
CKA_GOSTR3410_PARAMS1,3
|
Byte array
|
DER-encoding of the object identifier indicating the data
object type of GOST R 34.10-2001.
When key is used the domain parameter object
of key type CKK_GOSTR3410 must be specified with the same attribute
CKA_OBJECT_ID
|
|
CKA_GOSTR3411_PARAMS1,3,8
|
Byte array
|
DER-encoding of the object identifier indicating the data
object type of GOST R 34.11-94.
When key is used the domain parameter object
of key type CKK_GOSTR3411 must be specified with the same attribute
CKA_OBJECT_ID
|
|
CKA_GOST28147_PARAMS8
|
Byte array
|
DER-encoding of the object identifier indicating the data
object type of GOST 28147‑89.
When key is used the domain parameter object
of key type CKK_GOST28147 must be specified with the same attribute
CKA_OBJECT_ID. The attribute value may be omitted
|
- Refer to Table 11 for footnotes
The following is a sample template for creating an GOST R
34.10-2001 public key object:
CK_OBJECT_CLASS class = CKO_PUBLIC_KEY;
CK_KEY_TYPE keyType = CKK_GOSTR3410;
CK_UTF8CHAR label[] = “A GOST R34.10-2001 public key object”;
CK_BYTE gostR3410params_oid[] =
{0x06, 0x07, 0x2a, 0x85, 0x03, 0x02, 0x02, 0x23, 0x00};
CK_BYTE gostR3411params_oid[] =
{0x06, 0x07, 0x2a, 0x85, 0x03, 0x02, 0x02, 0x1e, 0x00};
CK_BYTE gost28147params_oid[] =
{0x06, 0x07, 0x2a, 0x85, 0x03, 0x02, 0x02, 0x1f, 0x00};
CK_BYTE value[64] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_GOSTR3410_PARAMS, gostR3410params_oid,
sizeof(gostR3410params_oid)},
{CKA_GOSTR3411_PARAMS, gostR3411params_oid, sizeof(gostR3411params_oid)},
{CKA_GOST28147_PARAMS, gost28147params_oid,
sizeof(gost28147params_oid)},
{CKA_VALUE, value, sizeof(value)}
};
GOST R 34.10-2001 private key objects (object
class CKO_PRIVATE_KEY, key type CKK_GOSTR3410) hold GOST R
34.10-2001 private keys.
The following table defines the GOST R 34.10-2001 private
key object attributes, in addition to the common attributes defined for this
object class:
Table 244, GOST R 34.10-2001 Private Key Object
Attributes
|
Attribute
|
Data
Type
|
Meaning
|
|
CKA_VALUE1,4,6,7
|
Byte array
|
32
bytes for private key in little endian order
|
|
CKA_GOSTR3410_PARAMS1,4,6
|
Byte array
|
DER-encoding of the object identifier indicating the data
object type of GOST R 34.10-2001.
When
key is used the domain parameter object of key type CKK_GOSTR3410 must be
specified with the same attribute CKA_OBJECT_ID
|
|
CKA_GOSTR3411_PARAMS1,4,6,8
|
Byte array
|
DER-encoding of the object identifier indicating the data
object type of GOST R 34.11-94.
When
key is used the domain parameter object of key type CKK_GOSTR3411 must be
specified with the same attribute CKA_OBJECT_ID
|
|
CKA_GOST28147_PARAMS4,6,8
|
Byte array
|
DER-encoding of the object identifier indicating the data
object type of GOST 28147‑89.
When key is used the domain parameter object of key type
CKK_GOST28147 must be specified with the same attribute CKA_OBJECT_ID. The
attribute value may be omitted
|
- Refer to Table 11 for footnotes
Note that when generating an GOST R 34.10-2001
private key, the GOST R 34.10-2001 domain parameters are not
specified in the key’s template. This is because GOST R 34.10-2001
private keys are only generated as part of an GOST R 34.10-2001 key pair,
and the GOST R 34.10-2001 domain parameters for the pair are
specified in the template for the GOST R 34.10-2001 public key.
The following is a sample template for creating an GOST R
34.10-2001 private key object:
CK_OBJECT_CLASS class = CKO_PRIVATE_KEY;
CK_KEY_TYPE keyType = CKK_GOSTR3410;
CK_UTF8CHAR label[] = “A GOST R34.10-2001 private key object”;
CK_BYTE subject[] = {...};
CK_BYTE id[] = {123};
CK_BYTE gostR3410params_oid[] =
{0x06, 0x07, 0x2a, 0x85, 0x03, 0x02, 0x02, 0x23, 0x00};
CK_BYTE gostR3411params_oid[] =
{0x06, 0x07, 0x2a, 0x85, 0x03, 0x02, 0x02, 0x1e, 0x00};
CK_BYTE gost28147params_oid[] =
{0x06, 0x07, 0x2a, 0x85, 0x03, 0x02, 0x02, 0x1f, 0x00};
CK_BYTE value[32] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_SUBJECT, subject, sizeof(subject)},
{CKA_ID, id, sizeof(id)},
{CKA_SENSITIVE, &true, sizeof(true)},
{CKA_SIGN, &true, sizeof(true)},
{CKA_GOSTR3410_PARAMS, gostR3410params_oid,
sizeof(gostR3410params_oid)},
{CKA_GOSTR3411_PARAMS, gostR3411params_oid,
sizeof(gostR3411params_oid)},
{CKA_GOST28147_PARAMS, gost28147params_oid,
sizeof(gost28147params_oid)},
{CKA_VALUE, value, sizeof(value)}
};
GOST R 34.10-2001 domain parameter objects (object
class CKO_DOMAIN_PARAMETERS, key type CKK_GOSTR3410) hold
GOST R 34.10‑2001 domain parameters.
The following table defines the GOST R 34.10-2001 domain
parameter object attributes, in addition to the common attributes defined for
this object class:
Table 245, GOST R 34.10-2001
Domain Parameter Object Attributes
|
Attribute
|
Data Type
|
Meaning
|
|
CKA_VALUE1
|
Byte array
|
DER-encoding of the domain parameters as it
was introduced in [4] section 8.4 (type GostR3410-2001-ParamSetParameters)
|
|
CKA_OBJECT_ID1
|
Byte array
|
DER-encoding of the object identifier
indicating the domain parameters
|
- Refer to Table 11 for footnotes
For any particular token, there is no guarantee that a token
supports domain parameters loading up and/or fetching out. Furthermore,
applications, that make direct use of domain parameters objects, should take in
account that CKA_VALUE attribute may be inaccessible.
The following is a sample template for creating a GOST R
34.10-2001 domain parameter object:
CK_OBJECT_CLASS class = CKO_DOMAIN_PARAMETERS;
CK_KEY_TYPE keyType = CKK_GOSTR3410;
CK_UTF8CHAR label[] = “A GOST R34.10-2001 cryptographic
parameters object”;
CK_BYTE oid[] =
{0x06, 0x07, 0x2a, 0x85, 0x03, 0x02, 0x02, 0x23, 0x00};
CK_BYTE value[] = {
0x30,0x81,0x90,0x02,0x01,0x07,0x02,0x20,0x5f,0xbf,0xf4,0x98,
0xaa,0x93,0x8c,0xe7,0x39,0xb8,0xe0,0x22,0xfb,0xaf,0xef,0x40,
0x56,0x3f,0x6e,0x6a,0x34,0x72,0xfc,0x2a,0x51,0x4c,0x0c,0xe9,
0xda,0xe2,0x3b,0x7e,0x02,0x21,0x00,0x80,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
0x00,0x04,0x31,0x02,0x21,0x00,0x80,0x00,0x00,0x00,0x00,0x00,
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x01,0x50,0xfe,
0x8a,0x18,0x92,0x97,0x61,0x54,0xc5,0x9c,0xfc,0x19,0x3a,0xcc,
0xf5,0xb3,0x02,0x01,0x02,0x02,0x20,0x08,0xe2,0xa8,0xa0,0xe6,
0x51,0x47,0xd4,0xbd,0x63,0x16,0x03,0x0e,0x16,0xd1,0x9c,0x85,
0xc9,0x7f,0x0a,0x9c,0xa2,0x67,0x12,0x2b,0x96,0xab,0xbc,0xea,
0x7e,0x8f,0xc8
};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_OBJECT_ID, oid, sizeof(oid)},
{CKA_VALUE, value, sizeof(value)}
};
♦ CK_GOSTR3410_KEY_WRAP_PARAMS
CK_GOSTR3410_KEY_WRAP_PARAMS is a structure that
provides the parameters to the CKM_GOSTR3410_KEY_WRAP mechanism. It is
defined as follows:
typedef struct CK_GOSTR3410_KEY_WRAP_PARAMS {
CK_BYTE_PTR pWrapOID;
CK_ULONG ulWrapOIDLen;
CK_BYTE_PTR pUKM;
CK_ULONG ulUKMLen;
CK_OBJECT_HANDLE hKey;
} CK_GOSTR3410_KEY_WRAP_PARAMS;
The fields of the structure have the following meanings:
|
pWrapOID
|
|
pointer to a data with DER-encoding of the object
identifier indicating the data object type of GOST 28147‑89. If
pointer takes NULL_PTR value in C_WrapKey operation then parameters are specified
in object identifier of attribute CKA_GOSTR3411_PARAMS must be used. For
C_UnwrapKey operation the pointer is not used and must take NULL_PTR value
anytime
|
|
ulWrapOIDLen
|
|
length of data with DER-encoding of the object identifier indicating
the data object type of GOST 28147‑89
|
|
pUKM
|
|
pointer to a data with UKM. If pointer takes NULL_PTR
value in C_WrapKey operation then random value of UKM will be used. If
pointer takes non-NULL_PTR value in C_UnwrapKey operation then the pointer
value will be compared with UKM value of wrapped key. If these two values do
not match the wrapped key will be rejected
|
|
ulUKMLen
|
|
length of UKM data. If pUKM-pointer is different
from NULL_PTR then equal to 8
|
|
hKey
|
|
key handle. Key handle of a sender for C_WrapKey
operation. Key handle of a receiver for C_UnwrapKey operation. When key
handle takes CK_INVALID_HANDLE value then an ephemeral (one time) key pair of
a sender will be used
|
CK_GOSTR3410_KEY_WRAP_PARAMS_PTR is a pointer to a
CK_GOSTR3410_KEY_WRAP_PARAMS.
♦ CK_GOSTR3410_DERIVE_PARAMS
CK_GOSTR3410_DERIVE_PARAMS is a structure that
provides the parameters to the CKM_GOSTR3410_DERIVE mechanism. It is
defined as follows:
typedef struct CK_GOSTR3410_DERIVE_PARAMS {
CK_EC_KDF_TYPE kdf;
CK_BYTE_PTR pPublicData;
CK_ULONG ulPublicDataLen;
CK_BYTE_PTR pUKM;
CK_ULONG ulUKMLen;
} CK_GOSTR3410_DERIVE_PARAMS;
The fields of the structure have the following meanings:
|
kdf
|
|
additional
key diversification algorithm identifier. Possible values are CKD_NULL and
CKD_CPDIVERSIFY_KDF. In case of CKD_NULL, result of the key derivation
function
described in
[RFC 4357], section 5.2 is used directly; In case of CKD_CPDIVERSIFY_KDF, the
resulting key value is additionally processed with algorithm from [RFC 4357],
section 6.5.
|
|
pPublicData1
|
|
pointer to data with public key of a receiver
|
|
ulPublicDataLen
|
|
length of data with public key of a receiver (must be 64)
|
|
pUKM
|
|
pointer to a UKM data
|
|
ulUKMLen
|
|
length of UKM data in bytes (must be 8)
|
1
CK_GOSTR3410_DERIVE_PARAMS_PTR is a pointer to a
CK_GOSTR3410_DERIVE_PARAMS.
The GOST R 34.10‑2001 key pair generation
mechanism, denoted CKM_GOSTR3410_KEY_PAIR_GEN, is a key pair generation
mechanism for GOST R 34.10‑2001.
This mechanism does not have a parameter.
The mechanism generates GOST R 34.10‑2001
public/private key pairs with particular GOST R 34.10‑2001
domain parameters, as specified in the CKA_GOSTR3410_PARAMS, CKA_GOSTR3411_PARAMS,
and CKA_GOST28147_PARAMS attributes of the template for the public key.
Note that CKA_GOST28147_PARAMS attribute may not be present in the
template.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new public key and the CKA_CLASS,
CKA_KEY_TYPE, CKA_VALUE, and CKA_GOSTR3410_PARAMS, CKA_GOSTR3411_PARAMS,
CKA_GOST28147_PARAMS attributes to the new private key.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure are not used.
The GOST R 34.10‑2001 without hashing
mechanism, denoted CKM_GOSTR3410, is a mechanism for single-part
signatures and verification for GOST R 34.10‑2001. (This
mechanism corresponds only to the part of GOST R 34.10‑2001
that processes the 32-bytes hash value; it does not compute the hash value.)
This mechanism does not have a parameter.
For the purposes of these mechanisms, a
GOST R 34.10‑2001 signature is an octet string of 64 bytes
long. The signature octets correspond to the concatenation of the
GOST R 34.10‑2001 values s and r’, both
represented as a 32 bytes octet string in big endian order with the most
significant byte first [RFC 4490] section 3.2, and [RFC 4491] section 2.2.2.
The input for the mechanism is an octet string of 32 bytes
long with digest has computed by means of GOST R 34.11‑94 hash
algorithm in the context of signed or should be signed message.
Table 246, GOST R 34.10-2001 without hashing: Key and Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Sign1
|
CKK_GOSTR3410
|
32 bytes
|
64 bytes
|
|
C_Verify1
|
CKK_GOSTR3410
|
32 bytes
|
64 bytes
|
1
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure are not used.
The GOST R 34.10‑2001 with
GOST R 34.11‑94, denoted CKM_GOSTR3410_WITH_GOSTR3411,
is a mechanism for signatures and verification for GOST R 34.10‑2001.
This mechanism computes the entire GOST R 34.10‑2001
specification, including the hashing with GOST R 34.11‑94 hash
algorithm.
As a parameter this mechanism utilizes a DER-encoding of the
object identifier indicating GOST R 34.11‑94 data object type.
A mechanism parameter may be missed then parameters are specified in object
identifier of attribute CKA_GOSTR3411_PARAMS must be used.
For the purposes of these mechanisms, a
GOST R 34.10‑2001 signature is an octet string of 64 bytes
long. The signature octets correspond to the concatenation of the
GOST R 34.10‑2001 values s and r’, both
represented as a 32 bytes octet string in big endian order with the most
significant byte first [RFC 4490] section 3.2, and [RFC 4491] section 2.2.2.
The input for the mechanism is signed or should be signed
message of any length. Single- and multiple-part signature operations are
available.
Table 247, GOST R 34.10-2001
with GOST R 34.11-94: Key and Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
|
C_Sign
|
CKK_GOSTR3410
|
Any
|
64 bytes
|
|
C_Verify
|
CKK_GOSTR3410
|
Any
|
64 bytes
|
For this mechanism, the ulMinKeySize and ulMaxKeySize fields
of the CK_MECHANISM_INFO structure are not used.
GOST R 34.10-2001 keys as a KEK (key encryption keys) for
encryption GOST 28147 keys, denoted by CKM_GOSTR3410_KEY_WRAP, is a
mechanism for key wrapping; and key unwrapping, based on GOST R 34.10-2001. Its
purpose is to encrypt and decrypt keys have been generated by key generation
mechanism for GOST 28147‑89. An encryption algorithm from [RFC 4490]
(section 5.2) must be used. Encrypted key is a DER-encoded structure of ASN.1 GostR3410-KeyTransport
type [RFC 4490] section 4.2.
It has a parameter, a CK_GOSTR3410_KEY_WRAP_PARAMS structure
defined in section 6.57.5.
For unwrapping (C_UnwrapKey), the mechanism decrypts
the wrapped key, and contributes the result as the CKA_VALUE attribute
of the new key.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure are not used.
Common key derivation, denoted CKM_GOSTR3410_DERIVE, is
a mechanism for key derivation with assistance of GOST R 34.10‑2001
private and public keys. The key of the mechanism must be of
object class CKO_DOMAIN_PARAMETERS and key type CKK_GOSTR3410.
An algorithm for key derivation from [RFC 4357] (section 5.2) must be used.
The mechanism contributes the result as the CKA_VALUE attribute
of the new private key. All other attributes must be specified in a template
for creating private key object.
ChaCha20 is a secret-key stream cipher described in [CHACHA].
Table
248,
ChaCha20 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_CHACHA20_KEY_GEN
|
|
|
|
|
✓
|
|
|
|
CKM_CHACHA20
|
✓
|
|
|
|
|
✓
|
|
This section defines the key type “CKK_CHACHA20” for type
CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of key objects.
Mechanisms:
CKM_CHACHA20_KEY_GEN
CKM_CHACHA20
ChaCha20 secret key objects
(object class CKO_SECRET_KEY, key type CKK_CHACHA20) hold ChaCha20 keys. The
following table defines the ChaCha20 secret key object attributes, in addition
to the common attributes defined for this object class:
Table 249,
ChaCha20 Secret Key Object
|
Attribute
|
Data type
|
Meaning
|
|
CKA_VALUE1,4,6,7
|
Byte array
|
Key length is fixed at 256 bits. Bit length
restricted to a byte array.
|
|
CKA_VALUE_LEN2,3
|
CK_ULONG
|
Length in bytes of key value
|
The following is a sample template for creating a ChaCha20
secret key object:
CK_OBJECT_CLASS class = CKO_SECRET_KEY;
CK_KEY_TYPE keyType = CKK_CHACHA20;
CK_UTF8CHAR label[] = “A ChaCha20 secret key object”;
CK_BYTE value[32] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_ENCRYPT, &true, sizeof(true)},
{CKA_VALUE, value, sizeof(value)}
};
CKA_CHECK_VALUE: The value of this
attribute is derived from the key object by taking the first three bytes of the
SHA-1 hash of the ChaCha20 secret key object’s CKA_VALUE attribute.
¨
CK_CHACHA20_PARAMS;
CK_CHACHA20_PARAMS_PTR
CK_CHACHA20_PARAMS provides the parameters to the CKM_CHACHA20
mechanism. It is defined as follows:
typedef struct CK_CHACHA20_PARAMS {
CK_BYTE_PTR pBlockCounter;
CK_ULONG blockCounterBits;
CK_BYTE_PTR pNonce;
CK_ULONG ulNonceBits;
} CK_CHACHA20_PARAMS;
The fields of the structure have the following meanings:
pBlockCounter pointer
to block counter
ulblockCounterBits length
of block counter in bits (can be either 32 or 64)
pNonce nonce
(This should be never re-used with the same key.)
ulNonceBits length
of nonce in bits (is 64 for original, 96 for IETF and 192 for xchacha20
variant)
The block counter is used to address 512 bit blocks in the
stream. In certain settings (e.g. disk encryption) it is necessary to address
these blocks in random order, thus this counter is exposed here.
CK_CHACHA20_PARAMS_PTR is a pointer to CK_CHACHA20_PARAMS.
The ChaCha20 key generation mechanism, denoted CKM_CHACHA20_KEY_GEN,
is a key generation mechanism for ChaCha20.
It does not have a parameter.
The mechanism generates ChaCha20 keys of 256 bits.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key. Other attributes supported by
the key type (specifically, the flags indicating which functions the key
supports) may be specified in the template for the key, or else are assigned
default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
key sizes in bytes. As a practical matter, the key size for ChaCha20 is fixed
at 256 bits.
ChaCha20, denoted CKM_CHACHA20, is a mechanism for
single and multiple-part encryption and decryption based on the ChaCha20 stream
cipher. It comes in 3 variants, which only differ in the size and handling of
their nonces, affecting the safety of using random nonces and the maximum size
that can be encrypted safely.
Chacha20 has a parameter, CK_CHACHA20_PARAMS, which
indicates the nonce and initial block counter value.
Constraints on key types and the length of input and output
data are summarized in the following table:
Table 250,
ChaCha20: Key and Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
Comments
|
|
C_Encrypt
|
ChaCha20
|
Any / only up to 256 GB in case of IETF variant
|
Same as input length
|
No final part
|
|
C_Decrypt
|
ChaCha20
|
Any / only up to 256 GB in case of IETF variant
|
Same as input length
|
No final part
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
ChaCha20 key sizes, in bits.
Table 251,
ChaCha20: Nonce and block counter lengths
|
Variant
|
Nonce
|
Block counter
|
Maximum message
|
Nonce generation
|
|
original
|
64 bit
|
64 bit
|
Virtually unlimited
|
1st msg: nonce0=random
nth msg: noncen-1++
|
|
IETF
|
96 bit
|
32 bit
|
Max ~256 GB
|
1st msg: nonce0=random
nth msg: noncen-1++
|
|
XChaCha20
|
192 bit
|
64 bit
|
Virtually unlimited
|
Each nonce can be randomly generated.
|
Nonces must not ever be reused with the same key. However
due to the birthday paradox the first two variants cannot guarantee that
randomly generated nonces are never repeating. Thus the recommended way to
handle this is to generate the first nonce randomly, then increase this for
follow-up messages. Only the last (XChaCha20) has large enough nonces so that
it is virtually impossible to trigger with randomly generated nonces the
birthday paradox.
Salsa20 is a secret-key stream cipher described in [SALSA].
Table
252,
Salsa20 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_SALSA20_KEY_GEN
|
|
|
|
|
✓
|
|
|
|
CKM_SALSA20
|
✓
|
|
|
|
|
✓
|
|
This section defines the key type “CKK_SALSA20” and
“CKK_SALSA20” for type CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of key
objects.
Mechanisms:
CKM_SALSA20_KEY_GEN
CKM_SALSA20
Salsa20 secret key objects
(object class CKO_SECRET_KEY, key type CKK_SALSA20) hold Salsa20 keys. The
following table defines the Salsa20 secret key object attributes, in addition
to the common attributes defined for this object class:
Table 253,
ChaCha20 Secret Key Object
|
Attribute
|
Data type
|
Meaning
|
|
CKA_VALUE1,4,6,7
|
Byte array
|
Key length is fixed at 256 bits. Bit length
restricted to a byte array.
|
|
CKA_VALUE_LEN2,3
|
CK_ULONG
|
Length in bytes of key value
|
The following is a sample template for creating a Salsa20
secret key object:
CK_OBJECT_CLASS class = CKO_SECRET_KEY;
CK_KEY_TYPE keyType = CKK_SALSA20;
CK_UTF8CHAR label[] = “A Salsa20 secret key object”;
CK_BYTE value[32] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_ENCRYPT, &true, sizeof(true)},
{CKA_VALUE, value, sizeof(value)}
};
CKA_CHECK_VALUE: The value of this attribute is derived
from the key object by taking the first three bytes of the SHA-1 hash of the
ChaCha20 secret key object’s CKA_VALUE attribute.
¨
CK_SALSA20_PARAMS;
CK_SALSA20_PARAMS_PTR
CK_SALSA20_PARAMS provides the parameters to the CKM_SALSA20
mechanism. It is defined as follows:
typedef struct CK_SALSA20_PARAMS {
CK_BYTE_PTR pBlockCounter;
CK_BYTE_PTR pNonce;
CK_ULONG ulNonceBits;
} CK_SALSA20_PARAMS;
The fields of the structure have the following meanings:
pBlockCounter pointer
to block counter (64 bits)
pNonce nonce
ulNonceBits size
of the nonce in bits (64 for classic and 192 for XSalsa20)
The block counter is used to address 512 bit blocks in the
stream. In certain settings (e.g. disk encryption) it is necessary to address
these blocks in random order, thus this counter is exposed here.
CK_SALSA20_PARAMS_PTR is a pointer to CK_SALSA20_PARAMS.
The Salsa20 key generation mechanism, denoted CKM_SALSA20_KEY_GEN,
is a key generation mechanism for Salsa20.
It does not have a parameter.
The mechanism generates Salsa20 keys of 256 bits.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
and CKA_VALUE attributes to the new key. Other attributes supported by
the key type (specifically, the flags indicating which functions the key
supports) may be specified in the template for the key, or else are assigned
default initial values.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
key sizes in bytes. As a practical matter, the key size for Salsa20 is fixed
at 256 bits.
Salsa20, denoted CKM_SALSA20, is a mechanism for
single and multiple-part encryption and decryption based on the Salsa20 stream
cipher. Salsa20 comes in two variants which only differ in the size and
handling of their nonces, affecting the safety of using random nonces.
Salsa20 has a parameter, CK_SALSA20_PARAMS, which
indicates the nonce and initial block counter value.
Constraints on key types and the length of input and output
data are summarized in the following table:
Table 254,
Salsa20: Key and Data Length
|
Function
|
Key type
|
Input length
|
Output length
|
Comments
|
|
C_Encrypt
|
Salsa20
|
Any
|
Same as input length
|
No final part
|
|
C_Decrypt
|
Salsa20
|
Any
|
Same as input length
|
No final part
|
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure specify the supported range of
ChaCha20 key sizes, in bits.
Table 255,
Salsa20: Nonce sizes
|
Variant
|
Nonce
|
Maximum message
|
Nonce generation
|
|
original
|
64 bit
|
Virtually unlimited
|
1st msg: nonce0=random
nth msg: noncen-1++
|
|
XSalsa20
|
192 bit
|
Virtually unlimited
|
Each nonce can be randomly generated.
|
Nonces must not ever be reused with the same key. However
due to the birthday paradox the original variant cannot guarantee that randomly
generated nonces are never repeating. Thus the recommended way to handle this
is to generate the first nonce randomly, then increase this for follow-up messages.
Only the XSalsa20 has large enough nonces so that it is virtually impossible to
trigger with randomly generated nonces the birthday paradox.
Poly1305 is a message authentication code designed by D.J
Bernsterin [POLY1305]. Poly1305 takes a 256 bit key and a message and
produces a 128 bit tag that is used to verify the message.
Table
256,
Poly1305 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_POLY1305_KEY_GEN
|
|
|
|
|
✓
|
|
|
|
CKM_POLY1305
|
|
✓
|
|
|
|
|
|
This section defines the key type “CKK_POLY1305” for type
CK_KEY_TYPE as used in the CKA_KEY_TYPE attribute of key objects.
Mechanisms:
CKM_POLY1305_KEY_GEN
CKM_POLY1305
Poly1305 secret key objects
(object class CKO_SECRET_KEY, key type CKK_POLY1305) hold Poly1305 keys. The
following table defines the Poly1305 secret key object attributes, in addition
to the common attributes defined for this object class:
Table 257,
Poly1305 Secret Key Object
|
Attribute
|
Data type
|
Meaning
|
|
CKA_VALUE1,4,6,7
|
Byte array
|
Key length is fixed at 256 bits. Bit length
restricted to a byte array.
|
|
CKA_VALUE_LEN2,3
|
CK_ULONG
|
Length in bytes of key value
|
The following is a sample template for creating a Poly1305
secret key object:
CK_OBJECT_CLASS class = CKO_SECRET_KEY;
CK_KEY_TYPE keyType = CKK_POLY1305;
CK_UTF8CHAR label[] = “A Poly1305 secret key object”;
CK_BYTE value[32] = {...};
CK_BBOOL true = CK_TRUE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &class, sizeof(class)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_SIGN, &true, sizeof(true)},
{CKA_VALUE, value, sizeof(value)}
};
Poly1305, denoted CKM_POLY1305, is a mechanism for
producing an output tag based on a 256 bit key and arbitrary length input.
It has no parameters.
Signatures (MACs) produced by this mechanism will be fixed
at 128 bits in size.
Table 258,
Poly1305: Key and Data Length
|
Function
|
Key type
|
Data length
|
Signature Length
|
|
C_Sign
|
Poly1305
|
Any
|
128 bits
|
|
C_Verify
|
Poly1305
|
Any
|
128 bits
|
The stream ciphers Salsa20 and ChaCha20 are normally used in
conjunction with the Poly1305 authenticator, in such a construction they also
provide Authenticated Encryption with Associated Data (AEAD). This section
defines the combined mechanisms and their usage in an AEAD setting.
Table 259,
Poly1305 Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_CHACHA20_POLY1305
|
✓
|
|
|
|
|
|
|
|
CKM_SALSA20_POLY1305
|
✓
|
|
|
|
|
|
|
Mechanisms:
CKM_CHACHA20_POLY1305
CKM_SALSA20_POLY1305
Generic ChaCha20, Salsa20, Poly1305 modes are described in
[CHACHA], [SALSA] and [POLY1305]. To set up for ChaCha20/Poly1305 or
Salsa20/Poly1305 use the following process. ChaCha20/Poly1305 and
Salsa20/Poly1305 both use CK_SALSA20_CHACHA20_POLY1305_PARAMS for Encrypt,
Decrypt and CK_SALSA20_CHACHA20_POLY1305_MSG_PARAMS for MessageEncrypt, and
MessageDecrypt.
Encrypt:
- Set
the Nonce length ulNonceLen in the parameter block. (this affects
which variant of Chacha20 will be used: 64 bits → original, 96 bits
→ IETF, 192 bits → XChaCha20)
- Set
the Nonce data pNonce in the parameter block.
- Set
the AAD data pAAD and size ulAADLen in the parameter block. pAAD
may be NULL if ulAADLen is 0.
- Call
C_EncryptInit() for CKM_CHACHA20_POLY1305 or CKM_SALSA20_POLY1305
mechanism with parameters and key K.
- Call
C_Encrypt(), or C_EncryptUpdate()*
C_EncryptFinal(), for the plaintext obtaining ciphertext and
authentication tag output.
Decrypt:
- Set
the Nonce length ulNonceLen in the parameter block. (this affects
which variant of Chacha20 will be used: 64 bits → original, 96 bits
→ IETF, 192 bits → XChaCha20)
- Set
the Nonce data pNonce in the parameter block.
- Set
the AAD data pAAD and size ulAADLen in the parameter block. pAAD
may be NULL if ulAADLen is 0.
- Call
C_DecryptInit() for CKM_CHACHA20_POLY1305 or CKM_SALSA20_POLY1305
mechanism with parameters and key K.
- Call
C_Decrypt(), or C_DecryptUpdate()*1 C_DecryptFinal(), for the
ciphertext, including the appended tag, obtaining plaintext output. Note:
since CKM_CHACHA20_POLY1305 and CKM_SALSA20_POLY1305 are
AEAD ciphers, no data should be returned until C_Decrypt() or
C_DecryptFinal().
MessageEncrypt::
- Set
the Nonce length ulNonceLen in the parameter block. (this affects
which variant of Chacha20 will be used: 64 bits → original, 96 bits
→ IETF, 192 bits → XChaCha20)
- Set
the Nonce data pNonce in the parameter block.
- Set
pTag to hold the tag data returned from C_EncryptMessage() or the final
C_EncryptMessageNext().
- Call
C_MessageEncryptInit() for CKM_CHACHA20_POLY1305 or CKM_SALSA20_POLY1305
mechanism with key K.
- Call
C_EncryptMessage(), or C_EncryptMessageBegin followed by
C_EncryptMessageNext()*.
The mechanism parameter is passed to all three of these functions.
- Call
C_MessageEncryptFinal() to close the message decryption.
MessageDecrypt:
- Set
the Nonce length ulNonceLen in the parameter block. (this affects
which variant of Chacha20 will be used: 64 bits → original, 96 bits
→ IETF, 192 bits → XChaCha20)
- Set
the Nonce data pNonce in the parameter block.
- Set
the tag data pTag in the parameter block before C_DecryptMessage or the final
C_DecryptMessageNext()
- Call
C_MessageDecryptInit() for CKM_CHACHA20_POLY1305 or CKM_SALSA20_POLY1305
mechanism with key K.
- Call
C_DecryptMessage(), or C_DecryptMessageBegin followed by
C_DecryptMessageNext()*.
The mechanism parameter is passed to all three of these functions.
- Call
C_MessageDecryptFinal() to close the message decryption
ulNonceLen is the length of the nonce in bits.
In Encrypt and Decrypt the tag is appended to the cipher
text. In MessageEncrypt the tag is returned in the pTag filed of
CK_SALSA20_CHACHA20_POLY1305_MSG_PARAMS. In MesssageDecrypt the tag is provided
by the pTag field of CK_SALSA20_CHACHA20_POLY1305_MSG_PARAMS. The application
must provide 16 bytes of space for the tag.
The key type for K must be compatible with CKM_CHACHA20
or CKM_SALSA20 respectively and the C_EncryptInit/C_DecryptInit calls
shall behave, with respect to K, as if they were called directly with CKM_CHACHA20
or CKM_SALSA20, K and NULL parameters.
Unlike the atomic Salsa20/ChaCha20 mechanism the AEAD
mechanism based on them does not expose the block counter, as the AEAD
construction is based on a message metaphor in which random access is not
needed.
¨
CK_SALSA20_CHACHA20_POLY1305_PARAMS;
CK_SALSA20_CHACHA20_POLY1305_PARAMS_PTR
CK_SALSA20_CHACHA20_POLY1305_PARAMS is a structure
that provides the parameters to the CKM_CHACHA20_POLY1305 and CKM_SALSA20_POLY1305
mechanisms. It is defined as follows:
typedef struct CK_SALSA20_CHACHA20_POLY1305_PARAMS {
CK_BYTE_PTR pNonce;
CK_ULONG ulNonceLen;
CK_BYTE_PTR pAAD;
CK_ULONG ulAADLen;
} CK_SALSA20_CHACHA20_POLY1305_PARAMS;
The fields of the structure have the following meanings:
pNonce nonce
(This should be never re-used with the same key.)
ulNonceLen length
of nonce in bits (is 64 for original, 96 for IETF (only for chacha20) and 192
for xchacha20/xsalsa20 variant)
pAAD pointer
to additional authentication data. This data is authenticated but not
encrypted.
ulAADLen length
of pAAD in bytes.
CK_SALSA20_CHACHA20_POLY1305_PARAMS_PTR is a pointer
to a CK_SALSA20_CHACHA20_POLY1305_PARAMS.
¨
CK_SALSA20_CHACHA20_POLY1305_MSG_PARAMS;
CK_SALSA20_CHACHA20_POLY1305_MSG_PARAMS_PTR
CK_CHACHA20POLY1305_PARAMS is a structure that provides the
parameters to the CKM_ CHACHA20_POLY1305 mechanism. It is defined as follows:
typedef struct CK_SALSA20_CHACHA20_POLY1305_MSG_PARAMS {
CK_BYTE_PTR pNonce;
CK_ULONG ulNonceLen;
CK_BYTE_PTR pTag;
} CK_SALSA20_CHACHA20_POLY1305_MSG_PARAMS;
The fields of the structure have the following meanings:
pNonce pointer
to nonce
ulNonceLen length
of nonce in bits. The length of the influences which variant of the ChaCha20
will be used (64 original, 96 IETF(only for ChaCha20), 192 XChaCha20/XSalsa20)
pTag location
of the authentication tag which is returned on MessageEncrypt, and provided on
MessageDecrypt.
CK_SALSA20_CHACHA20_POLY1305_MSG_PARAMS_PTR is a
pointer to a CK_SALSA20_CHACHA20_POLY1305_MSG_PARAMS.
Details for HKDF key derivation mechanisms can be found in
[RFC 5869].
Table
260, HKDF
Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_HKDF_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_HKDF_DATA
|
|
|
|
|
|
|
ü
|
|
CKM_HKDF_KEY_GEN
|
|
|
|
|
ü
|
|
|
Mechanisms:
CKM_HKDF_DERIVE
CKM_HKDF_DATA
CKM_HKDF_KEY_GEN
Key Types:
CKK_HKDF
6.62.2 HKDF
mechanism parameters
¨
CK_HKDF_PARAMS;
CK_HKDF_PARAMS_PTR
CK_HKDF_PARAMS is a structure that provides the
parameters to the CKM_HKDF_DERIVE and CKM_HKDF_DATA mechanisms.
It is defined as follows:
typedef struct CK_HKDF_PARAMS {
CK_BBOOL bExtract;
CK_BBOOL bExpand;
CK_MECHANISM_TYPE prfHashMechanism;
CK_ULONG ulSaltType;
CK_BYTE_PTR pSalt;
CK_ULONG ulSaltLen;
CK_OBJECT_HANDLE hSaltKey;
CK_BYTE_PTR pInfo;
CK_ULONG ulInfoLen;
} CK_HKDF_PARAMS;
The fields of the structure have the following meanings:
bExtract execute
the extract portion of HKDF.
bExpand execute
the expand portion of HKDF.
prfHashMechanism base hash used
for the HMAC in the underlying HKDF operation.
ulSaltType specifies
how the salt for the extract portion of the KDF is supplied.
CKF_HKDF_SALT_NULL
no salt is supplied.
CKF_HKDF_SALT_DATA
salt is supplied as a data in pSalt with length ulSaltLen.
CKF_HKDF_SALT_KEY
salt is supplied as a key in hSaltKey.
pSalt pointer
to the salt.
ulSaltLen length
of the salt pointed to in pSalt.
hSaltKey object
handle to the salt key.
pInfo info
string for the expand stage.
ulInfoLen length
of the info string for the expand stage.
CK_HKDF_PARAMS_PTR is a pointer to a CK_HKDF_PARAMS.
HKDF derivation implements the HKDF as specified in [RFC
5869]. The two booleans bExtract and bExpand control whether the extract
section of the HKDF or the expand section of the HKDF is in use.
It has a parameter, a CK_HKDF_PARAMS structure, which
allows for the passing of the salt and or the expansion info. The structure
contains the bools bExtract and bExpand which control whether the
extract or expand portions of the HKDF is to be used. This structure is defined
in Section 6.62.2.
The input key must be of type CKK_HKDF or CKK_GENERIC_SECRET
and the length must be the size of the underlying hash function specified in prfHashMechanism.
The exception is a data object which has the same size as the underlying hash
function, and which may be supplied as an input key. In this case bExtract
should be true and non-null salt should be supplied.
Either bExtract or bExpand must be set to
true. If they are both set to true, input key is first extracted then expanded.
The salt is used in the extraction stage. If bExtract is set to true and no
salt is given, a ‘zero’ salt (salt whose length is the same as the underlying
hash and values all set to zero) is used as specified by the RFC. If bExpand is
set to true, CKA_VALUE_LEN should be set to the desired key length. If
it is false CKA_VALUE_LEN may be set to the length of the hash, but that is not
necessary as the mechanism will supply this value. The salt should be ignored
if bExtract is false. The pInfo should be ignored if bExpand
is set to false.
The mechanism also contributes the CKA_CLASS, and CKA_VALUE
attributes to the new key. Other attributes may be specified in the template,
or else are assigned default values.
The template sent along with this mechanism during a C_DeriveKey
call may indicate that the object class is CKO_SECRET_KEY. However,
since these facts are all implicit in the mechanism, there is no need to
specify any of them.
This mechanism has the following rules about key sensitivity
and extractability:
·
The CKA_SENSITIVE and CKA_EXTRACTABLE attributes in
the template for the new key can both be specified to be either CK_TRUE or
CK_FALSE. If omitted, these attributes each take on some default value.
·
If the base key has its CKA_ALWAYS_SENSITIVE attribute set
to CK_FALSE, then the derived key will as well. If the base key has its CKA_ALWAYS_SENSITIVE
attribute set to CK_TRUE, then the derived key has its CKA_ALWAYS_SENSITIVE
attribute set to the same value as its CKA_SENSITIVE attribute.
·
Similarly, if the base key has its CKA_NEVER_EXTRACTABLE
attribute set to CK_FALSE, then the derived key will, too. If the base key has
its CKA_NEVER_EXTRACTABLE attribute set to CK_TRUE, then the derived key
has its CKA_NEVER_EXTRACTABLE attribute set to the opposite value
from its CKA_EXTRACTABLE attribute.
HKDF Data derive mechanism, denoted CKM_HKDF_DATA, is
identical to HKDF Derive except the output is a CKO_DATA object whose
value is the result to the derive operation. Some tokens may restrict what data
may be successfully derived based on the pInfo portion of the CK_HKDF_PARAMS.
Tokens may reject requests based on the pInfo values. Allowed pInfo
values are specified in the profile document and applications could then query
the appropriate profile before depending on the mechanism.
HKDF key gen, denoted CKM_HKDF_KEY_GEN generates a new
random HKDF key. CKA_VALUE_LEN must be set in the template.
CKM_NULL is a mechanism used to implement the trivial
pass-through function.
Table
261, CKM_NULL Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_NULL
|
ü
|
ü
|
ü
|
ü
|
|
ü
|
ü
|
|
1SR =
SignRecover, VR = VerifyRecover
|
Mechanisms:
CKM_NULL
CKM_NULL does not have a parameter.
When used for encrypting / decrypting data, the input data
is copied unchanged to the output data.
When used for signing, the input data is copied to the
signature. When used for signature verification, it compares the input data and
the signature, and returns CKR_OK (indicating that both are identical) or
CKR_SIGNATURE_INVALID.
When used for digesting data, the input data is copied to
the message digest.
When used for wrapping a private or secret key object, the wrapped
key will be identical to the key to be wrapped. When used for unwrapping, a new
object with the same value as the wrapped key will be created.
When used for deriving a key, the derived key has the same
value as the base key.
Table 262, IKE
Mechanisms vs. Functions
|
|
Functions
|
|
Mechanism
|
Encrypt
&
Decrypt
|
Sign
&
Verify
|
SR
&
VR1
|
Digest
|
Gen.
Key/
Key
Pair
|
Wrap
&
Unwrap
|
Derive
|
|
CKM_IKE2_PRF_PLUS_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_IKE_PRF_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_IKE1_PRF_DERIVE
|
|
|
|
|
|
|
ü
|
|
CKM_IKE1_EXTENDED_DERIVE
|
|
|
|
|
|
|
ü
|
Mechanisms:
CKM_IKE2_PRF_PLUS_DERIVE
CKM_IKE_PRF_DERIVE
CKM_IKE1_PRF_DERIVE
CKM_IKE1_EXTENDED_DERIVE
¨
CK_IKE2_PRF_PLUS_DERIVE_PARAMS;
CK_IKE2_PRF_PLUS_DERIVE_PARAMS_PTR
CK_IKE2_PRF_PLUS_DERIVE_PARAMS is a structure that
provides the parameters to the CKM_IKE2_PRF_PLUS_DERIVE mechanism. It is
defined as follows:
typedef struct CK_IKE2_PRF_PLUS_DERIVE_PARAMS {
CK_MECHANISM_TYPE prfMechanism;
CK_BBOOL bHasSeedKey;
CK_OBJECT_HANDLE hSeedKey;
CK_BYTE_PTR pSeedData;
CK_ULONG ulSeedDataLen;
} CK_IKE2_PRF_PLUS_DERIVE_PARAMS;
The fields of the structure have the following meanings:
prfMechanism underlying
MAC mechanism used to generate the prf
bHasSeedKey hSeed
key is present
hSeedKey optional
seed from key
pSeedData optional
seed from data
ulSeedDataLen length
of optional seed data. If no seed data is present this value is 0
CK_IKE2_PRF_PLUS_DERIVE_PARAMS_PTR is a pointer to a
CK_IKE2_PRF_PLUS_DERIVE_PARAMS.
¨
CK_IKE_PRF_DERIVE_PARAMS;
CK_IKE_PRF_DERIVE_PARAMS_PTR
CK_IKE_PRF_DERIVE_PARAMS is a structure that provides
the parameters to the CKM_IKE_PRF_DERIVE mechanism. It is defined as
follows:
typedef struct CK_IKE_PRF_DERIVE_PARAMS {
CK_MECHANISM_TYPE prfMechanism;
CK_BBOOL bDataAsKey;
CK_BBOOL bRekey;
CK_BYTE_PTR pNi;
CK_ULONG ulNiLen;
CK_BYTE_PTR pNr;
CK_ULONG ulNrLen;
CK_OBJECT_HANDLE hNewKey;
} CK_IKE_PRF_DERIVE_PARAMS;
The fields of the structure have the following meanings:
prfMechanism underlying
MAC mechanism used to generate the prf
bDataAsKey Ni||Nr
is used as the key for the prf rather than baseKey
bRekey rekey
operation. hNewKey must be present
pNi Ni
value
ulNiLen length
of Ni
pNr Nr
value
ulNrLen length
of Nr
hNewKey New
key value to drive the rekey.
CK_IKE_PRF_DERIVE_PARAMS_PTR is a pointer to a CK_IKE_PRF_DERIVE_PARAMS.
¨
CK_IKE1_PRF_DERIVE_PARAMS;
CK_IKE1_PRF_DERIVE_PARAMS_PTR
CK_IKE1_PRF_DERIVE_PARAMS is a structure that
provides the parameters to the CKM_IKE1_PRF_DERIVE mechanism. It is
defined as follows:
typedef struct CK_IKE1_PRF_DERIVE_PARAMS {
CK_MECHANISM_TYPE prfMechanism;
CK_BBOOL bHasPrevKey;
CK_OBJECT_HANDLE hKeygxy;
CK_OBJECT_HANDLE hPrevKey;
CK_BYTE_PTR pCKYi;
CK_ULONG ulCKYiLen;
CK_BYTE_PTR pCKYr;
CK_ULONG ulCKYrLen;
CK_BYTE keyNumber;
} CK_IKE1_PRF_DERIVE_PARAMS;
The fields of the structure have the following meanings:
prfMechanism underlying
MAC mechanism used to generate the prf
bHasPrevkey hPrevKey
is present
hKeygxy handle
to the exchanged g^xy key
hPrevKey handle
to the previously derived key
pCKYi CKYi
value
ulCKYiLen length
of CKYi
pCKYr CKYr
value
ulCKYrLen length
of CKYr
keyNumber unique
number for this key derivation
CK_IKE1_PRF_DERIVE_PARAMS_PTR is a pointer to a CK_IKE1_PRF_DERIVE_PARAMS.
¨
CK_IKE1_EXTENDED_DERIVE_PARAMS;
CK_IKE1_EXTENDED_DERIVE_PARAMS_PTR
CK_IKE1_EXTENDED_DERIVE_PARAMS is a structure that
provides the parameters to the CKM_IKE1_EXTENDED_DERIVE mechanism. It
is defined as follows:
typedef struct CK_IKE1_EXTENDED_DERIVE_PARAMS {
CK_MECHANISM_TYPE prfMechanism;
CK_BBOOL bHasKeygxy;
CK_OBJECT_HANDLE hKeygxy;
CK_BYTE_PTR pExtraData;
CK_ULONG ulExtraDataLen;
} CK_IKE1_EXTENDED_DERIVE_PARAMS;
The fields of the structure have the following meanings:
prfMechanism underlying
MAC mechanism used to generate the prf
bHasKeygxy hKeygxy
key is present
hKeygxy optional
key g^xy
pExtraData optional
extra data
ulExtraDataLen length
of optional extra data. If no extra data is present this value is 0
CK_IKE2_PRF_PLUS_DERIVE_PARAMS_PTR is a pointer to a CK_IKE2_PRF_PLUS_DERIVE_PARAMS.
The IKE PRF Derive mechanism denoted CKM_IKE_PRF_DERIVE
is used in IPSEC both IKEv1 and IKEv2 to generate an initial key that is used
to generate additional keys. It takes a CK_IKE_PRF_DERIVE_PARAMS as a
mechanism parameter. baseKey is the base key passed into C_DeriveKey.
baseKey must be of type CKK_GENERIC_SECRET if bDataAsKey
is TRUE and the key type of the underlying prf if bDataAsKey is FALSE. hNewKey
must be of type CKK_GENERIC_SECRET. Depending on the parameter settings,
it generates keys with a CKA_VALUE of:
1. prf(pNi|pNr, baseKey); (bDataAsKey=TRUE,
bRekey=FALSE)
2. prf(baseKey, pNi|pNr); (bDataAsKkey=FALSE,
bRekey=FALSE)
3. prf(baseKey, ValueOf(hNewKey)| pNi | pNr);
(bDataAsKey=FALSE, bRekey=TRUE)
The resulting output key is always the length of the
underlying prf. The combination of bDataAsKey=TRUE and bRekey=TRUE
is not allowed. If both are set, CKR_ARGUMENTS_BAD is returned.
Case 1 is used in
a. ikev2 (RFC 5996) baseKey is called g^ir, the
output is called SKEYSEED
b. ikev1 (RFC 2409) baseKey is called g^ir, the
output is called SKEYID
Case 2 is used in ikev1 (RFC 2409) inkey is called
pre-shared-key, output is called SKEYID
Case 3 is used in ikev2 (RFC 5996) rekey case, baseKey
is SK_d, hNewKey is g^ir (new), the output is called SKEYSEED. The derived key
will have a length of the length of the underlying prf. If CKA_VALUE_LEN
is specified, it must equal the underlying prf or CKR_KEY_SIZE_RANGE is
returned. If CKA_KEY_TYPE is not specified in the template, it will be
the underlying key type of the prf.
The IKEv1 PRF Derive mechanism denoted CKM_IKE1_PRF_DERIVE
is used in IPSEC IKEv1 to generate various additional keys from the initial
SKEYID. It takes a CK_IKE1_PRF_DERIVE_PARAMS as a mechanism parameter.
SKEYID is the base key passed into C_DeriveKey.
This mechanism derives a key with CKA_VALUE set to
either:
prf(baseKey, ValueOf(hKeygxy) || pCKYi || pCKYr ||
key_number)
or
prf(baseKey, ValueOf(hPrevKey) || ValueOf(hKeygxy) ||
pCKYi || pCKYr || key_number)
depending on the state of bHasPrevKey.
The key type of baseKey must be the key type of the
prf, and the key type of hKeygxy must be CKK_GENERIC_SECRET. The
key type of hPrevKey can be any key type.
This is defined in RFC 2409. For each of the following keys.
baseKey is SKEYID, hKeygxy is g^xy
for outKey = SKEYID_d, bHasPrevKey = false,
key_number = 0
for outKey = SKEYID_a, hPrevKey= SKEYID_d,
key_number = 1
for outKey = SKEYID_e, hPrevKey= SKEYID_a,
key_number = 2
If CKA_VALUE_LEN is not specified, the resulting key
will be the length of the prf. If CKA_VALUE_LEN is greater then the
prf, CKR_KEY_SIZE_RANGE is returned. If it is less the key is truncated
taking the left most bytes. The value CKA_KEY_TYPE must be specified in
the template or CKR_TEMPLATE_INCOMPLETE is returned.
The IKEv2 PRF PLUS Derive mechanism denoted CKM_IKE2_PRF_PLUS_DERIVE
is used in IPSEC IKEv2 to derive various additional keys from the initial
SKEYSEED. It takes a CK_IKE2_PRF_PLUS_DERIVE_PARAMS as a mechanism
parameter. SKEYSEED is the base key passed into C_DeriveKey. The key
type of baseKey must be the key type of the underlying prf. This
mechanism uses the base key and a feedback version of the prf to generate a
single key with sufficient bytes to cover all additional keys. The application
will then use CKM_EXTRACT_KEY_FROM_KEY several times to pull out the
various keys. CKA_VALUE_LEN must be set in the template and its value
must not be bigger than 255 times the size of the prf function output or CKR_KEY_SIZE_RANGE
will be returned. If CKA_KEY_TYPE is not specified, the output key type
will be CKK_GENERIC_SECRET.
This mechanism derives a key with a CKA_VALUE of
(from RFC 5996):
prfplus = T1 | T2 | T3 | T4 |... Tn
where:
T1 = prf(K, S | 0x01)
T2 = prf(K, T1 | S | 0x02)
T3 = prf(K, T3 | S | 0x03)
T4 = prf(K, T4 | S | 0x04)
.
Tn = prf(K, T(n-1) | n)
K = baseKey, S = valueOf(hSeedKey)
| pSeedData
The IKE Extended Derive mechanism denoted CKM_IKE1_EXTENDED_DERIVE
is used in IPSEC IKEv1 to derive longer keys than CKM_IKE1_EXTENDED_DERIVE
can from the initial SKEYID. It is used to support RFC 2409 appendix B and RFC
2409 section 5.5 (Quick Mode). It takes a CK_IKE1_EXTENDED_DERIVE_PARAMS
as a mechanism parameter. SKEYID is the base key passed into C_DeriveKey.
CKA_VALUE_LEN must be set in the template and its value must not be
bigger than 255 times the size of the prf function output or CKR_KEY_SIZE_RANGE
will be returned. If CKA_KEY_TYPE is not specified, the output key type
will be CKK_GENERIC_SECRET. The key type of SKEYID must be the key type
of the prf, and the key type of hKeygxy (if present) must be CKK_GENERIC_SECRET.
This mechanism derives a key with CKA_VALUE (from RFC
2409 appendix B and section 5.5):
Ka = K1 | K2 | K3 | K4 |... Kn
where:
K1 = prf(K, valueOf(hKeygxy)|pExtraData)
or prf(K,0x00) if bHashKeygxy is FALSE and ulExtraData is 0
K2 = prf(K, K1|valueOf(hKeygxy)|pExtraData)
K3 = prf(K, K2|valueOf(hKeygxy)|pExtraData)
K4 = prf(K, K3|valueOf(hKeygxy)|pExtraData)
.
Kn = prf(K, K(n-1)|valueOf(hKeygxy)|pExtraData)
K = baseKey
If CKA_VALUE_LEN is less then or equal to the prf
length and bHasKeygxy is CK_FALSE, then the new key is simply the base key
truncated to CKA_VALUE_LEN (specified in RFC 2409 appendix B). Otherwise
the prf is executed and the derived keys value is CKA_VALUE_LEN bytes of
the resulting prf.
HSS is a mechanism for single-part signatures and
verification, following the digital signature algorithm defined in [RFC 8554]
and [NIST 802-208].
Table 263, HSS
Mechanisms vs. Functions
|
Mechanism
|
Functions
|
|
Encrypt & Decrypt
|
Sign & Verify
|
SR & VR
|
Digest
|
Gen. Key/ Key Pair
|
Wrap & Unwrap
|
Derive
|
|
CKM_HSS_KEY_PAIR_GEN
|
|
|
|
|
ü
|
|
|
|
CKM_HSS
|
|
ü1
|
|
|
|
|
|
1
This section defines the key type CKK_HSS for type CK_KEY_TYPE
as used in the CKA_KEY_TYPE attribute of key objects and domain
parameter objects.
Mechanisms:
CKM_HSS_KEY_PAIR_GEN
CKM_HSS
HSS public key objects (object class CKO_PUBLIC_KEY, key
type CKK_HSS) hold HSS public keys.
The following table defines the HSS public key object
attributes, in addition to the common attributes defined for this object class:
Table 264, HSS Public Key Object Attributes
|
Attribute
|
Data Type
|
Meaning
|
|
CKA_HSS_LEVELS2,4
|
CK_ULONG
|
The number of levels in the HSS scheme.
|
|
CKA_HSS_LMS_TYPE2,4
|
CK_ULONG
|
The encoding for the Merkle tree
heights of the top level LMS tree in the hierarchy.
|
|
CKA_HSS_LMOTS_TYPE2,4
|
CK_ULONG
|
The encoding for the Winternitz
parameter of the one-time-signature scheme of the top level LMS tree.
|
|
CKA_VALUE1,4
|
Byte array
|
XDR-encoded public key as defined in
[RFC8554].
|
- Refer to Table 11 for footnotes
The following is a sample template for creating an HSS
public key object:
CK_OBJECT_CLASS keyClass = CKO_PUBLIC_KEY;
CK_KEY_TYPE keyType = CKK_HSS;
CK_UTF8CHAR label[] = “An HSS public key object”;
CK_BYTE value[] = {...};
CK_BBOOL true = CK_TRUE;
CK_BBOOL false = CK_FALSE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &keyClass, sizeof(keyClass)},
{CKA_KEY_TYPE, &keyType, sizeof(keyType)},
{CKA_TOKEN, &false, sizeof(false)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_VALUE, value, sizeof(value)},
{CKA_VERIFY, &true, sizeof(true)}
};
HSS private key objects (object class CKO_PRIVATE_KEY, key
type CKK_HSS) hold HSS private keys.
The following table defines the HSS private key object
attributes, in addition to the common attributes defined for this object class:
Table 265,
HSS Private Key Object Attributes
|
Attribute
|
Data
Type
|
Meaning
|
|
CKA_HSS_LEVELS1,3
|
CK_ULONG
|
The number of levels in the HSS scheme.
|
|
CKA_HSS_LMS_TYPES1,3
|
CK_ULONG_PTR
|
A list of encodings for the Merkle tree
heights of the LMS trees in the hierarchy from top to bottom. The number of
encodings in the array is the ulValueLen component of the attribute divided
by the size of CK_ULONG. This number must match the CKA_HSS_LEVELS attribute
value.
|
|
CKA_HSS_LMOTS_TYPES1,3
|
CK_ULONG_PTR
|
A list of encodings for the Winternitz
parameter of the one-time-signature scheme of
the LMS trees in the hierarchy from top to bottom. The number of encodings in
the array is the ulValueLen component of the attribute divided by the size of
CK_ULONG. This number must match the CKA_HSS_LEVELS attribute value.
|
|
CKA_VALUE1,4,6,7
|
Byte
array
|
Vendor
defined, must include state information.
Note
that exporting this value is dangerous as it would allow key reuse.
|
|
CKA_HSS_KEYS_REMAINING2,4
|
CK_ULONG
|
The minimum
of the following two values: 1) The number of one-time private keys
remaining; 2) 2^32-1
|
- Refer to Table 11 for footnotes
The encodings for CKA_HSS_LMOTS_TYPES and CKA_HSS_LMS_TYPES
are defined in [RFC 8554] and [NIST 802-208].
The following is a sample template for creating an LMS
private key object:
CK_OBJECT_CLASS keyClass = CKO_PRIVATE_KEY;
CK_KEY_TYPE keyType = CKK_HSS;
CK_UTF8CHAR label[] = “An HSS private key object”;
CK_ULONG hssLevels = 123;
CK_ULONG lmsTypes[] = {123,...};
CK_ULONG lmotsTypes[] = {123,...};
CK_BYTE value[] = {...};
CK_BBOOL
true = CK_TRUE;
CK_BBOOL false = CK_FALSE;
CK_ATTRIBUTE template[] = {
{CKA_CLASS, &keyClass, sizeof(keyClass)},
{CKA_KEY_TYPE, &keyType,
sizeof(keyType)},
{CKA_TOKEN, &true, sizeof(true)},
{CKA_LABEL, label, sizeof(label)-1},
{CKA_SENSITIVE, &true, sizeof(true)},
{CKA_EXTRACTABLE, &false, sizeof(true)},
{CKA_HSS_LEVELS, &hssLevels,
sizeof(hssLevels)},
{CKA_HSS_LMS_TYPES, lmsTypes,
sizeof(lmsTypes)},
{CKA_HSS_LMOTS_TYPES, lmotsTypes, sizeof(lmotsTypes)},
{CKA_VALUE, value, sizeof(value)},
{CKA_SIGN, &true, sizeof(true)}
};
CKA_SENSITIVE MUST be true, CKA_EXTRACTABLE
MUST be false, and CKA_COPYABLE MUST be false for this key.
The HSS key pair generation mechanism, denoted CKM_HSS_KEY_PAIR_GEN,
is a key pair generation mechanism for HSS.
This mechanism does not have a parameter.
The mechanism generates HSS public/private key pairs for the
scheme specified by the CKA_HSS_LEVELS, CKA_HSS_LMS_TYPES, and CKA_HSS_LMOTS_TYPES
attributes of the template for the private key.
The mechanism contributes the CKA_CLASS, CKA_KEY_TYPE,
CKA_HSS_LEVELS, CKA_HSS_LMS_TYPE, CKA_HSS_LMOTS_TYPE, and CKA_VALUE
attributes to the new public key and the CKA_CLASS, CKA_KEY_TYPE,
CKA_VALUE, and CKA_HSS_KEYS_REMAINING attributes to the new
private key.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure are not used and must be set
to 0.
The HSS without hashing mechanism, denoted CKM_HSS,
is a mechanism for single-part signatures and verification for HSS. (This
mechanism corresponds only to the part of LMS that processes the hash value, which may be of any length; it does not
compute the hash value.)
This mechanism does not have a parameter.
For the purposes of these mechanisms, an
HSS signature is a byte string with length depending on CKA_HSS_LEVELS,
CKA_HSS_LMS_TYPES, CKA_HSS_LMOTS_TYPES as described
in the following table.
Table 266, HSS without hashing: Key and Data Length
|
Function
|
Key
type
|
Input length
|
Output length
|
|
C_Sign1
|
HSS
Private Key
|
any
|
1296-749882
|
|
C_Verify1
|
HSS
Public Key
|
any, 1296-749882
|
N/A
|
1 Single-part operations only.
2 4+(levels-1)*56+levels*(8+(36+32*p)+h*32)
where p has values (265, 133, 67, 34) for lmots type (W1, W2, W4,
W8) and h is the number of levels in the LMS Merkle trees.
For this mechanism, the ulMinKeySize and ulMaxKeySize
fields of the CK_MECHANISM_INFO structure are not used and must be set
to 0.
If the number of signatures is exhausted, CKR_KEY_EXHAUSTED
will be returned.
An implementation is a conforming PKCS#11 Consumer if the
implementation meets the conditions specified in one or more consumer profiles
specified in [PKCS11-Prof].
A PKCS#11 consumer implementation SHALL be a conforming
PKCS#11 Consumer.
If a PKCS#11 consumer implementation claims support for a
particular consumer profile, then the implementation SHALL conform to all
normative statements within the clauses specified for that profile and for any
subclauses to each of those clauses.
An implementation is a conforming PKCS#11 Provider if the
implementation meets the conditions specified in one or more provider profiles
specified in [PKCS11-Prof].
A PKCS#11 provider implementation SHALL be a conforming
PKCS#11 Provider.
The following individuals have participated in the creation
of this specification and are gratefully acknowledged:
Participants:
|
Salutation
|
First Name
|
Last Name
|
Company
|
|
Dr.
|
Warren
|
Armstrong
|
QuintessenceLabs Pty Ltd.
|
|
|
Anthony
|
Berglas
|
Cryptsoft Pty Ltd.
|
|
Mr.
|
Dieter
|
Bong
|
Utimaco IS GmbH
|
|
Mr.
|
Roland
|
Bramm
|
PrimeKey Solutions AB
|
|
|
Andrew
|
Byrne
|
Dell
|
|
|
Hamish
|
Cameron
|
nCipher
|
|
|
Kenli
|
Chong
|
QuintessenceLabs Pty Ltd.
|
|
Mr.
|
Justin
|
Corlett
|
Cryptsoft Pty Ltd.
|
|
|
Xuelei
|
Fan
|
Oracle
|
|
Mr.
|
Jan
|
Friedel
|
Oracle
|
|
Ms.
|
Susan
|
Gleeson
|
Oracle
|
|
Mr.
|
Thomas
|
Hardjono
|
M.I.T.
|
|
Mrs.
|
Jane
|
Harnad
|
OASIS
|
|
|
David
|
Horton
|
Dell
|
|
|
Tim
|
Hudson
|
Cryptsoft Pty Ltd.
|
|
Mr.
|
Gershon
|
Janssen
|
Individual
|
|
Mr.
|
Jakub
|
Jelen
|
Red Hat
|
|
Dr.
|
Mark
|
Joseph
|
P6R, Inc
|
|
Mr.
|
Paul
|
King
|
nCipher
|
|
Ms.
|
Dina
|
Kurktchi-Nimeh
|
Oracle
|
|
|
John
|
Leiseboer
|
QuintessenceLabs Pty Ltd.
|
|
Mr.
|
John
|
Leser
|
Oracle
|
|
|
Chris
|
Malafis
|
Red Hat
|
|
Dr.
|
Michael
|
Markowitz
|
Information Security Corporation
|
|
Mr.
|
Scott
|
Marshall
|
Cryptsoft Pty Ltd.
|
|
Mr.
|
Chris
|
Meyer
|
Utimaco IS GmbH
|
|
Mr.
|
Darren
|
Moffat
|
Oracle
|
|
Dr.
|
Florian
|
Poppa
|
QuintessenceLabs Pty Ltd.
|
|
|
Roland
|
Reichenberg
|
Utimaco IS GmbH
|
|
Mr.
|
Robert
|
Relyea
|
Red Hat
|
|
Mr.
|
Jonathan
|
Schulze-Hewett
|
Information Security Corporation
|
|
Mr.
|
Greg
|
Scott
|
Cryptsoft Pty Ltd.
|
|
Mr.
|
Martin
|
Shannon
|
QuintessenceLabs Pty Ltd.
|
|
Mr.
|
Oscar
|
So
|
Individual
|
|
|
Patrick
|
Steuer
|
IBM
|
|
Mr.
|
Gerald
|
Stueve
|
Fornetix
|
|
|
Jim
|
Susoy
|
P6R, Inc
|
|
Mr.
|
Sander
|
Temme
|
nCipher
|
|
Mr.
|
Manish
|
Upasani
|
Utimaco IS GmbH
|
|
Mr.
|
Charles
|
White
|
Fornetix
|
|
Ms.
|
Magda
|
Zdunkiewicz
|
Cryptsoft Pty Ltd.
|
The definitions for manifest
constants specified in this document can be found in the Additional artifacts section.
|
Revision
|
Date
|
Editor
|
Changes Made
|
|
CSD02 WD01
|
12 May 2022
|
Dieter Bong
|
Changes made compared to Committee Specification CSD01, as
working draft of Committee Specification CSD02
- Editorial changes resolving comments by Paul Knight,
OASIS, https://lists.oasis-open.org/archives/pkcs11-comment/202203/msg00001.html
- Reference [PKCS11-curr] replaced by reference within
document
- Correction of typos
|
|
|
|
|
|
Copyright © OASIS Open 2023. All Rights Reserved.
All capitalized terms in the following text have the
meanings assigned to them in the OASIS Intellectual Property Rights Policy (the
"OASIS IPR Policy"). The full Policy may be
found at the OASIS website: [https://www.oasis-open.org/policies-guidelines/ipr/].
This document and translations of it may be copied and
furnished to others, and derivative works that comment on or otherwise explain
it or assist in its implementation may be prepared, copied, published, and
distributed, in whole or in part, without restriction of any kind, provided
that the above copyright notice and this section are included on all such
copies and derivative works. However, this document itself may not be modified
in any way, including by removing the copyright notice or references to OASIS,
except as needed for the purpose of developing any document or deliverable
produced by an OASIS Technical Committee (in which case the rules applicable to
copyrights, as set forth in the OASIS IPR Policy, must be followed) or as
required to translate it into languages other than English.
The limited permissions granted above are perpetual and will
not be revoked by OASIS or its successors or assigns.
This document and the information contained herein is
provided on an "AS IS" basis and OASIS DISCLAIMS ALL WARRANTIES,
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF
THE INFORMATION HEREIN WILL NOT INFRINGE ANY OWNERSHIP RIGHTS OR ANY IMPLIED
WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. OASIS AND
ITS MEMBERS WILL NOT BE LIABLE FOR ANY DIRECT, INDIRECT, SPECIAL OR
CONSEQUENTIAL DAMAGES ARISING OUT OF ANY USE OF THIS DOCUMENT OR ANY PART
THEREOF.
As stated in the OASIS IPR Policy, the following three
paragraphs in brackets apply to OASIS Standards Final Deliverable documents (Committee
Specifications, OASIS Standards, or Approved Errata).
[OASIS requests that any OASIS Party or any other party that
believes it has patent claims that would necessarily be infringed by
implementations of this OASIS Standards Final Deliverable, to notify OASIS TC
Administrator and provide an indication of its willingness to grant patent
licenses to such patent claims in a manner consistent with the IPR Mode of the
OASIS Technical Committee that produced this deliverable.]
[OASIS invites any party to contact the OASIS TC Administrator
if it is aware of a claim of ownership of any patent claims that would
necessarily be infringed by implementations of this OASIS Standards Final
Deliverable by a patent holder that is not willing to provide a license to such
patent claims in a manner consistent with the IPR Mode of the OASIS Technical
Committee that produced this OASIS Standards Final Deliverable. OASIS may
include such claims on its website, but disclaims any obligation to do so.]
[OASIS takes no position regarding the validity or scope of
any intellectual property or other rights that might be claimed to pertain to
the implementation or use of the technology described in this OASIS Standards
Final Deliverable or the extent to which any license under such rights might or
might not be available; neither does it represent that it has made any effort
to identify any such rights. Information on OASIS' procedures with respect to
rights in any document or deliverable produced by an OASIS Technical Committee
can be found on the OASIS website. Copies of claims of rights made available
for publication and any assurances of licenses to be made available, or the
result of an attempt made to obtain a general license or permission for the use
of such proprietary rights by implementers or users of this OASIS Standards
Final Deliverable, can be obtained from the OASIS TC Administrator. OASIS makes
no representation that any information or list of intellectual property rights
will at any time be complete, or that any claims in such list are, in fact, Essential
Claims.]
The name "OASIS" is a trademark of OASIS, the owner and developer of this document,
and should be used only to refer to the organization and its official outputs.
OASIS welcomes reference to, and implementation and use of, documents, while
reserving the right to enforce its marks against misleading uses. Please see https://www.oasis-open.org/policies-guidelines/trademark/
for above guidance.
