This file represents rfc4880.txt with annotations and removed lines; just to see, what is missing in ruby-openpgp. You can find the original RFC document at http://ietf.org/rfc/rfc4880 Network Working Group J. Callas Request for Comments: 4880 PGP Corporation Obsoletes: 1991, 2440 L. Donnerhacke Category: Standards Track IKS GmbH H. Finney PGP Corporation D. Shaw R. Thayer November 2007 OpenPGP Message Format [...] 3.2. Multiprecision Integers [...] Unused bits of an MPI MUST be zero. [...] 3.4. Text Unless otherwise specified, the character set for text is the UTF-8 [RFC3629] encoding of Unicode [ISO10646]. Callas, et al Standards Track [Page 9] [...] 3.7. String-to-Key (S2K) Specifiers String-to-key (S2K) specifiers are used to convert passphrase strings into symmetric-key encryption/decryption keys. They are used in two places, currently: to encrypt the secret part of private keys in the private keyring, and to convert passphrases to encryption keys for symmetrically encrypted messages. 3.7.1. String-to-Key (S2K) Specifier Types There are three types of S2K specifiers currently supported, and some reserved values: ID S2K Type -- -------- 0 Simple S2K 1 Salted S2K 2 Reserved value 3 Iterated and Salted S2K 100 to 110 Private/Experimental S2K These are described in Sections 3.7.1.1 - 3.7.1.3. 3.7.1.1. Simple S2K This directly hashes the string to produce the key data. See below for how this hashing is done. Octet 0: 0x00 Octet 1: hash algorithm Simple S2K hashes the passphrase to produce the session key. The manner in which this is done depends on the size of the session key (which will depend on the cipher used) and the size of the hash algorithm's output. If the hash size is greater than the session key size, the high-order (leftmost) octets of the hash are used as the key. If the hash size is less than the key size, multiple instances of the hash context are created -- enough to produce the required key data. These instances are preloaded with 0, 1, 2, ... octets of zeros (that is to say, the first instance has no preloading, the second gets preloaded with 1 octet of zero, the third is preloaded with two octets of zeros, and so forth). As the data is hashed, it is given independently to each hash context. Since the contexts have been initialized differently, they will each produce different hash output. Once the passphrase is hashed, the output data from the multiple hashes is concatenated, first hash leftmost, to produce the key data, with any excess octets on the right discarded. 3.7.1.2. Salted S2K This includes a "salt" value in the S2K specifier -- some arbitrary data -- that gets hashed along with the passphrase string, to help prevent dictionary attacks. Octet 0: 0x01 Octet 1: hash algorithm Octets 2-9: 8-octet salt value Salted S2K is exactly like Simple S2K, except that the input to the hash function(s) consists of the 8 octets of salt from the S2K specifier, followed by the passphrase. 3.7.1.3. Iterated and Salted S2K This includes both a salt and an octet count. The salt is combined with the passphrase and the resulting value is hashed repeatedly. This further increases the amount of work an attacker must do to try dictionary attacks. Octet 0: 0x03 Octet 1: hash algorithm Octets 2-9: 8-octet salt value Octet 10: count, a one-octet, coded value The count is coded into a one-octet number using the following formula: #define EXPBIAS 6 count = ((Int32)16 + (c & 15)) << ((c >> 4) + EXPBIAS); The above formula is in C, where "Int32" is a type for a 32-bit integer, and the variable "c" is the coded count, Octet 10. Iterated-Salted S2K hashes the passphrase and salt data multiple times. The total number of octets to be hashed is specified in the encoded count in the S2K specifier. Note that the resulting count value is an octet count of how many octets will be hashed, not an iteration count. Initially, one or more hash contexts are set up as with the other S2K algorithms, depending on how many octets of key data are needed. Then the salt, followed by the passphrase data, is repeatedly hashed until the number of octets specified by the octet count has been hashed. The one exception is that if the octet count is less than the size of the salt plus passphrase, the full salt plus passphrase will be hashed even though that is greater than the octet count. After the hashing is done, the data is unloaded from the hash context(s) as with the other S2K algorithms. 3.7.2. String-to-Key Usage Implementations SHOULD use salted or iterated-and-salted S2K specifiers, as simple S2K specifiers are more vulnerable to dictionary attacks. 3.7.2.1. Secret-Key Encryption An S2K specifier can be stored in the secret keyring to specify how to convert the passphrase to a key that unlocks the secret data. Older versions of PGP just stored a cipher algorithm octet preceding the secret data or a zero to indicate that the secret data was unencrypted. The MD5 hash function was always used to convert the passphrase to a key for the specified cipher algorithm. For compatibility, when an S2K specifier is used, the special value 254 or 255 is stored in the position where the hash algorithm octet would have been in the old data structure. This is then followed immediately by a one-octet algorithm identifier, and then by the S2K specifier as encoded above. Therefore, preceding the secret data there will be one of these possibilities: 0: secret data is unencrypted (no passphrase) 255 or 254: followed by algorithm octet and S2K specifier Cipher alg: use Simple S2K algorithm using MD5 hash This last possibility, the cipher algorithm number with an implicit use of MD5 and IDEA, is provided for backward compatibility; it MAY be understood, but SHOULD NOT be generated, and is deprecated. These are followed by an Initial Vector of the same length as the block size of the cipher for the decryption of the secret values, if they are encrypted, and then the secret-key values themselves. 3.7.2.2. Symmetric-Key Message Encryption OpenPGP can create a Symmetric-key Encrypted Session Key (ESK) packet at the front of a message. This is used to allow S2K specifiers to be used for the passphrase conversion or to create messages with a mix of symmetric-key ESKs and public-key ESKs. This allows a message to be decrypted either with a passphrase or a public-key pair. PGP 2.X always used IDEA with Simple string-to-key conversion when encrypting a message with a symmetric algorithm. This is deprecated, but MAY be used for backward-compatibility. 4. Packet Syntax This section describes the packets used by OpenPGP. [...] 4.2.2.4. Partial Body Lengths A Partial Body Length header is one octet long and encodes the length of only part of the data packet. This length is a power of 2, from 1 to 1,073,741,824 (2 to the 30th power). It is recognized by its one octet value that is greater than or equal to 224, and less than 255. The Partial Body Length is equal to: partialBodyLen = 1 << (1st_octet & 0x1F); Each Partial Body Length header is followed by a portion of the packet body data. The Partial Body Length header specifies this portion's length. Another length header (one octet, two-octet, five-octet, or partial) follows that portion. The last length header in the packet MUST NOT be a Partial Body Length header. Partial Body Length headers may only be used for the non-final parts of the packet. Note also that the last Body Length header can be a zero-length header. An implementation MAY use Partial Body Lengths for data packets, be they literal, compressed, or encrypted. The first partial length MUST be at least 512 octets long. Partial Body Lengths MUST NOT be used for any other packet types. 4.2.3. Packet Length Examples [...] It might also be encoded in the following octet stream: 0xEF, first 32768 octets of data; 0xE1, next two octets of data; 0xE0, next one octet of data; 0xF0, next 65536 octets of data; 0xC5, 0xDD, last 1693 octets of data. This is just one possible encoding, and many variations are possible on the size of the Partial Body Length headers, as long as a regular Body Length header encodes the last portion of the data. [...] 5. Packet Types 5.1. Public-Key Encrypted Session Key Packets (Tag 1) A Public-Key Encrypted Session Key packet holds the session key used to encrypt a message. Zero or more Public-Key Encrypted Session Key packets and/or Symmetric-Key Encrypted Session Key packets may precede a Symmetrically Encrypted Data Packet, which holds an encrypted message. The message is encrypted with the session key, and the session key is itself encrypted and stored in the Encrypted Session Key packet(s). The Symmetrically Encrypted Data Packet is preceded by one Public-Key Encrypted Session Key packet for each OpenPGP key to which the message is encrypted. The recipient of the message finds a session key that is encrypted to their public key, decrypts the session key, and then uses the session key to decrypt the message. The body of this packet consists of: - A one-octet number giving the version number of the packet type. The currently defined value for packet version is 3. - An eight-octet number that gives the Key ID of the public key to which the session key is encrypted. If the session key is encrypted to a subkey, then the Key ID of this subkey is used here instead of the Key ID of the primary key. - A one-octet number giving the public-key algorithm used. - A string of octets that is the encrypted session key. This string takes up the remainder of the packet, and its contents are dependent on the public-key algorithm used. Algorithm Specific Fields for RSA encryption - multiprecision integer (MPI) of RSA encrypted value m**e mod n. Algorithm Specific Fields for Elgamal encryption: - MPI of Elgamal (Diffie-Hellman) value g**k mod p. - MPI of Elgamal (Diffie-Hellman) value m * y**k mod p. The value "m" in the above formulas is derived from the session key as follows. First, the session key is prefixed with a one-octet algorithm identifier that specifies the symmetric encryption algorithm used to encrypt the following Symmetrically Encrypted Data Packet. Then a two-octet checksum is appended, which is equal to the sum of the preceding session key octets, not including the algorithm identifier, modulo 65536. This value is then encoded as described in PKCS#1 block encoding EME-PKCS1-v1_5 in Section 7.2.1 of [RFC3447] to form the "m" value used in the formulas above. See Section 13.1 of this document for notes on OpenPGP's use of PKCS#1. Note that when an implementation forms several PKESKs with one session key, forming a message that can be decrypted by several keys, the implementation MUST make a new PKCS#1 encoding for each key. An implementation MAY accept or use a Key ID of zero as a "wild card" or "speculative" Key ID. In this case, the receiving implementation would try all available private keys, checking for a valid decrypted session key. This format helps reduce traffic analysis of messages. 5.2. Signature Packet (Tag 2) [...] Implementations SHOULD accept V3 signatures. Implementations SHOULD generate V4 signatures. Note that if an implementation is creating an encrypted and signed message that is encrypted to a V3 key, it is reasonable to create a V3 signature. 5.2.1. Signature Types There are a number of possible meanings for a signature, which are indicated in a signature type octet in any given signature. Please note that the vagueness of these meanings is not a flaw, but a feature of the system. Because OpenPGP places final authority for validity upon the receiver of a signature, it may be that one signer's casual act might be more rigorous than some other authority's positive act. See Section 5.2.4, "Computing Signatures", for detailed information on how to compute and verify signatures of each type. These meanings are as follows: 0x00: Signature of a binary document. This means the signer owns it, created it, or certifies that it has not been modified. 0x01: Signature of a canonical text document. This means the signer owns it, created it, or certifies that it has not been modified. The signature is calculated over the text data with its line endings converted to . 0x02: Standalone signature. This signature is a signature of only its own subpacket contents. It is calculated identically to a signature over a zero-length binary document. Note that it doesn't make sense to have a V3 standalone signature. 0x10: Generic certification of a User ID and Public-Key packet. The issuer of this certification does not make any particular assertion as to how well the certifier has checked that the owner of the key is in fact the person described by the User ID. 0x11: Persona certification of a User ID and Public-Key packet. The issuer of this certification has not done any verification of the claim that the owner of this key is the User ID specified. 0x12: Casual certification of a User ID and Public-Key packet. The issuer of this certification has done some casual verification of the claim of identity. 0x13: Positive certification of a User ID and Public-Key packet. The issuer of this certification has done substantial verification of the claim of identity. Most OpenPGP implementations make their "key signatures" as 0x10 certifications. Some implementations can issue 0x11-0x13 certifications, but few differentiate between the types. 0x18: Subkey Binding Signature This signature is a statement by the top-level signing key that indicates that it owns the subkey. This signature is calculated directly on the primary key and subkey, and not on any User ID or other packets. A signature that binds a signing subkey MUST have an Embedded Signature subpacket in this binding signature that contains a 0x19 signature made by the signing subkey on the primary key and subkey. 0x19: Primary Key Binding Signature This signature is a statement by a signing subkey, indicating that it is owned by the primary key and subkey. This signature is calculated the same way as a 0x18 signature: directly on the primary key and subkey, and not on any User ID or other packets. 0x1F: Signature directly on a key This signature is calculated directly on a key. It binds the information in the Signature subpackets to the key, and is appropriate to be used for subpackets that provide information about the key, such as the Revocation Key subpacket. It is also appropriate for statements that non-self certifiers want to make about the key itself, rather than the binding between a key and a name. 0x20: Key revocation signature The signature is calculated directly on the key being revoked. A revoked key is not to be used. Only revocation signatures by the key being revoked, or by an authorized revocation key, should be considered valid revocation signatures. 0x28: Subkey revocation signature The signature is calculated directly on the subkey being revoked. A revoked subkey is not to be used. Only revocation signatures by the top-level signature key that is bound to this subkey, or by an authorized revocation key, should be considered valid revocation signatures. 0x30: Certification revocation signature This signature revokes an earlier User ID certification signature (signature class 0x10 through 0x13) or direct-key signature (0x1F). It should be issued by the same key that issued the revoked signature or an authorized revocation key. The signature is computed over the same data as the certificate that it revokes, and should have a later creation date than that certificate. 0x40: Timestamp signature. This signature is only meaningful for the timestamp contained in it. 0x50: Third-Party Confirmation signature. This signature is a signature over some other OpenPGP Signature packet(s). It is analogous to a notary seal on the signed data. A third-party signature SHOULD include Signature Target subpacket(s) to give easy identification. Note that we really do mean SHOULD. There are plausible uses for this (such as a blind party that only sees the signature, not the key or source document) that cannot include a target subpacket. 5.2.2. Version 3 Signature Packet Format [...] The concatenation of the data to be signed, the signature type, and creation time from the Signature packet (5 additional octets) is hashed. The resulting hash value is used in the signature algorithm. The high 16 bits (first two octets) of the hash are included in the Signature packet to provide a quick test to reject some invalid signatures. Algorithm-Specific Fields for RSA signatures: - multiprecision integer (MPI) of RSA signature value m**d mod n. Algorithm-Specific Fields for DSA signatures: - MPI of DSA value r. - MPI of DSA value s. The signature calculation is based on a hash of the signed data, as described above. The details of the calculation are different for DSA signatures than for RSA signatures. With RSA signatures, the hash value is encoded using PKCS#1 encoding type EMSA-PKCS1-v1_5 as described in Section 9.2 of RFC 3447. This requires inserting the hash value as an octet string into an ASN.1 structure. The object identifier for the type of hash being used is included in the structure. The hexadecimal representations for the currently defined hash algorithms are as follows: - MD5: 0x2A, 0x86, 0x48, 0x86, 0xF7, 0x0D, 0x02, 0x05 - RIPEMD-160: 0x2B, 0x24, 0x03, 0x02, 0x01 - SHA-1: 0x2B, 0x0E, 0x03, 0x02, 0x1A - SHA224: 0x60, 0x86, 0x48, 0x01, 0x65, 0x03, 0x04, 0x02, 0x04 - SHA256: 0x60, 0x86, 0x48, 0x01, 0x65, 0x03, 0x04, 0x02, 0x01 - SHA384: 0x60, 0x86, 0x48, 0x01, 0x65, 0x03, 0x04, 0x02, 0x02 - SHA512: 0x60, 0x86, 0x48, 0x01, 0x65, 0x03, 0x04, 0x02, 0x03 The ASN.1 Object Identifiers (OIDs) are as follows: - MD5: 1.2.840.113549.2.5 - RIPEMD-160: 1.3.36.3.2.1 - SHA-1: 1.3.14.3.2.26 - SHA224: 2.16.840.1.101.3.4.2.4 - SHA256: 2.16.840.1.101.3.4.2.1 - SHA384: 2.16.840.1.101.3.4.2.2 - SHA512: 2.16.840.1.101.3.4.2.3 The full hash prefixes for these are as follows: MD5: 0x30, 0x20, 0x30, 0x0C, 0x06, 0x08, 0x2A, 0x86, 0x48, 0x86, 0xF7, 0x0D, 0x02, 0x05, 0x05, 0x00, 0x04, 0x10 RIPEMD-160: 0x30, 0x21, 0x30, 0x09, 0x06, 0x05, 0x2B, 0x24, 0x03, 0x02, 0x01, 0x05, 0x00, 0x04, 0x14 SHA-1: 0x30, 0x21, 0x30, 0x09, 0x06, 0x05, 0x2b, 0x0E, 0x03, 0x02, 0x1A, 0x05, 0x00, 0x04, 0x14 SHA224: 0x30, 0x31, 0x30, 0x0d, 0x06, 0x09, 0x60, 0x86, 0x48, 0x01, 0x65, 0x03, 0x04, 0x02, 0x04, 0x05, 0x00, 0x04, 0x1C SHA256: 0x30, 0x31, 0x30, 0x0d, 0x06, 0x09, 0x60, 0x86, 0x48, 0x01, 0x65, 0x03, 0x04, 0x02, 0x01, 0x05, 0x00, 0x04, 0x20 SHA384: 0x30, 0x41, 0x30, 0x0d, 0x06, 0x09, 0x60, 0x86, 0x48, 0x01, 0x65, 0x03, 0x04, 0x02, 0x02, 0x05, 0x00, 0x04, 0x30 SHA512: 0x30, 0x51, 0x30, 0x0d, 0x06, 0x09, 0x60, 0x86, 0x48, 0x01, 0x65, 0x03, 0x04, 0x02, 0x03, 0x05, 0x00, 0x04, 0x40 DSA signatures MUST use hashes that are equal in size to the number of bits of q, the group generated by the DSA key's generator value. If the output size of the chosen hash is larger than the number of bits of q, the hash result is truncated to fit by taking the number of leftmost bits equal to the number of bits of q. This (possibly truncated) hash function result is treated as a number and used directly in the DSA signature algorithm. 5.2.3. Version 4 Signature Packet Format [...] The concatenation of the data being signed and the signature data from the version number through the hashed subpacket data (inclusive) is hashed. The resulting hash value is what is signed. The left 16 bits of the hash are included in the Signature packet to provide a quick test to reject some invalid signatures. There are two fields consisting of Signature subpackets. The first field is hashed with the rest of the signature data, while the second is unhashed. The second set of subpackets is not cryptographically protected by the signature and should include only advisory information. The algorithms for converting the hash function result to a signature are described in a section below. 5.2.3.1. Signature Subpacket Specification [...] The value of the subpacket type octet may be: 0 = Reserved 1 = Reserved 2 = Signature Creation Time 3 = Signature Expiration Time 4 = Exportable Certification 5 = Trust Signature 6 = Regular Expression 7 = Revocable 8 = Reserved 9 = Key Expiration Time 10 = Placeholder for backward compatibility 11 = Preferred Symmetric Algorithms 12 = Revocation Key 13 = Reserved 14 = Reserved 15 = Reserved 16 = Issuer 17 = Reserved 18 = Reserved 19 = Reserved 20 = Notation Data 21 = Preferred Hash Algorithms 22 = Preferred Compression Algorithms 23 = Key Server Preferences 24 = Preferred Key Server 25 = Primary User ID 26 = Policy URI 27 = Key Flags 28 = Signer's User ID 29 = Reason for Revocation 30 = Features 31 = Signature Target 32 = Embedded Signature 100 To 110 = Private or experimental An implementation SHOULD ignore any subpacket of a type that it does not recognize. Bit 7 of the subpacket type is the "critical" bit. If set, it denotes that the subpacket is one that is critical for the evaluator of the signature to recognize. If a subpacket is encountered that is marked critical but is unknown to the evaluating software, the evaluator SHOULD consider the signature to be in error. An evaluator may "recognize" a subpacket, but not implement it. The purpose of the critical bit is to allow the signer to tell an evaluator that it would prefer a new, unknown feature to generate an error than be ignored. Implementations SHOULD implement the three preferred algorithm subpackets (11, 21, and 22), as well as the "Reason for Revocation" subpacket. Note, however, that if an implementation chooses not to implement some of the preferences, it is required to behave in a polite manner to respect the wishes of those users who do implement these preferences. 5.2.3.2. Signature Subpacket Types A number of subpackets are currently defined. Some subpackets apply to the signature itself and some are attributes of the key. Subpackets that are found on a self-signature are placed on a certification made by the key itself. Note that a key may have more than one User ID, and thus may have more than one self-signature, and differing subpackets. A subpacket may be found either in the hashed or unhashed subpacket sections of a signature. If a subpacket is not hashed, then the information in it cannot be considered definitive because it is not part of the signature proper. 5.2.3.3. Notes on Self-Signatures A self-signature is a binding signature made by the key to which the signature refers. There are three types of self-signatures, the certification signatures (types 0x10-0x13), the direct-key signature (type 0x1F), and the subkey binding signature (type 0x18). For certification self-signatures, each User ID may have a self- signature, and thus different subpackets in those self-signatures. For subkey binding signatures, each subkey in fact has a self- signature. Subpackets that appear in a certification self-signature apply to the user name, and subpackets that appear in the subkey self-signature apply to the subkey. Lastly, subpackets on the direct-key signature apply to the entire key. Implementing software should interpret a self-signature's preference subpackets as narrowly as possible. For example, suppose a key has two user names, Alice and Bob. Suppose that Alice prefers the symmetric algorithm CAST5, and Bob prefers IDEA or TripleDES. If the software locates this key via Alice's name, then the preferred algorithm is CAST5; if software locates the key via Bob's name, then the preferred algorithm is IDEA. If the key is located by Key ID, the algorithm of the primary User ID of the key provides the preferred symmetric algorithm. Revoking a self-signature or allowing it to expire has a semantic meaning that varies with the signature type. Revoking the self- signature on a User ID effectively retires that user name. The self-signature is a statement, "My name X is tied to my signing key K" and is corroborated by other users' certifications. If another user revokes their certification, they are effectively saying that they no longer believe that name and that key are tied together. Similarly, if the users themselves revoke their self-signature, then the users no longer go by that name, no longer have that email address, etc. Revoking a binding signature effectively retires that subkey. Revoking a direct-key signature cancels that signature. Please see the "Reason for Revocation" subpacket (Section 5.2.3.23) for more relevant detail. Since a self-signature contains important information about the key's use, an implementation SHOULD allow the user to rewrite the self- signature, and important information in it, such as preferences and key expiration. It is good practice to verify that a self-signature imported into an implementation doesn't advertise features that the implementation doesn't support, rewriting the signature as appropriate. An implementation that encounters multiple self-signatures on the same object may resolve the ambiguity in any way it sees fit, but it is RECOMMENDED that priority be given to the most recent self- signature. [...] 5.2.3.4. Signature Creation Time The time the signature was made. MUST be present in the hashed area. [...] 5.2.3.7. Preferred Symmetric Algorithms [...] octets with the most preferred listed first. It is assumed that only algorithms listed are supported by the recipient's software. Algorithm numbers are in Section 9. This is only found on a self- signature. [...] 5.2.3.9. Preferred Compression Algorithms [...] list is ordered. Algorithm numbers are in Section 9. If this subpacket is not included, ZIP is preferred. A zero denotes that uncompressed data is preferred; the key holder's software might have no compression software in that implementation. This is only found on a self-signature. 5.2.3.11. Exportable Certification (1 octet of exportability, 0 for not, 1 for exportable) This subpacket denotes whether a certification signature is "exportable", to be used by other users than the signature's issuer. The packet body contains a Boolean flag indicating whether the signature is exportable. If this packet is not present, the certification is exportable; it is equivalent to a flag containing a 1. Non-exportable, or "local", certifications are signatures made by a user to mark a key as valid within that user's implementation only. Thus, when an implementation prepares a user's copy of a key for transport to another user (this is the process of "exporting" the key), any local certification signatures are deleted from the key. The receiver of a transported key "imports" it, and likewise trims any local certifications. In normal operation, there won't be any, assuming the import is performed on an exported key. However, there are instances where this can reasonably happen. For example, if an implementation allows keys to be imported from a key database in addition to an exported key, then this situation can arise. Some implementations do not represent the interest of a single user (for example, a key server). Such implementations always trim local certifications from any key they handle. [...] 5.2.3.20. Policy URI [...] 5.2.3.21. Key Flags (N octets of flags) This subpacket contains a list of binary flags that hold information about a key. It is a string of octets, and an implementation MUST NOT assume a fixed size. This is so it can grow over time. If a list is shorter than an implementation expects, the unstated flags are considered to be zero. The defined flags are as follows: First octet: 0x01 - This key may be used to certify other keys. 0x02 - This key may be used to sign data. 0x04 - This key may be used to encrypt communications. 0x08 - This key may be used to encrypt storage. 0x10 - The private component of this key may have been split by a secret-sharing mechanism. 0x20 - This key may be used for authentication. 0x80 - The private component of this key may be in the possession of more than one person. Usage notes: The flags in this packet may appear in self-signatures or in certification signatures. They mean different things depending on who is making the statement -- for example, a certification signature that has the "sign data" flag is stating that the certification is for that use. On the other hand, the "communications encryption" flag in a self-signature is stating a preference that a given key be used for communications. Note however, that it is a thorny issue to determine what is "communications" and what is "storage". This decision is left wholly up to the implementation; the authors of this document do not claim any special wisdom on the issue and realize that accepted opinion may change. The "split key" (0x10) and "group key" (0x80) flags are placed on a self-signature only; they are meaningless on a certification signature. They SHOULD be placed only on a direct-key signature (type 0x1F) or a subkey signature (type 0x18), one that refers to the key the flag applies to. [...] 5.2.3.23. Reason for Revocation (1 octet of revocation code, N octets of reason string) This subpacket is used only in key revocation and certification revocation signatures. It describes the reason why the key or certificate was revoked. The first octet contains a machine-readable code that denotes the reason for the revocation: 0 - No reason specified (key revocations or cert revocations) 1 - Key is superseded (key revocations) 2 - Key material has been compromised (key revocations) 3 - Key is retired and no longer used (key revocations) 32 - User ID information is no longer valid (cert revocations) 100-110 - Private Use Following the revocation code is a string of octets that gives information about the Reason for Revocation in human-readable form (UTF-8). The string may be null, that is, of zero length. The length of the subpacket is the length of the reason string plus one. An implementation SHOULD implement this subpacket, include it in all revocation signatures, and interpret revocations appropriately. There are important semantic differences between the reasons, and there are thus important reasons for revoking signatures. If a key has been revoked because of a compromise, all signatures created by that key are suspect. However, if it was merely superseded or retired, old signatures are still valid. If the revoked signature is the self-signature for certifying a User ID, a revocation denotes that that user name is no longer in use. Such a revocation SHOULD include a 0x20 code. Note that any signature may be revoked, including a certification on some other person's key. There are many good reasons for revoking a certification signature, such as the case where the keyholder leaves the employ of a business with an email address. A revoked certification is no longer a part of validity calculations. 5.2.3.24. Features (N octets of flags) The Features subpacket denotes which advanced OpenPGP features a user's implementation supports. This is so that as features are added to OpenPGP that cannot be backwards-compatible, a user can state that they can use that feature. The flags are single bits that indicate that a given feature is supported. This subpacket is similar to a preferences subpacket, and only appears in a self-signature. An implementation SHOULD NOT use a feature listed when sending to a user who does not state that they can use it. Defined features are as follows: First octet: 0x01 - Modification Detection (packets 18 and 19) If an implementation implements any of the defined features, it SHOULD implement the Features subpacket, too. An implementation may freely infer features from other suitable implementation-dependent mechanisms. 5.2.3.25. Signature Target (1 octet public-key algorithm, 1 octet hash algorithm, N octets hash) This subpacket identifies a specific target signature to which a signature refers. For revocation signatures, this subpacket provides explicit designation of which signature is being revoked. For a third-party or timestamp signature, this designates what signature is signed. All arguments are an identifier of that target signature. The N octets of hash data MUST be the size of the hash of the signature. For example, a target signature with a SHA-1 hash MUST have 20 octets of hash data. [...] 5.2.4. Computing Signatures All signatures are formed by producing a hash over the signature data, and then using the resulting hash in the signature algorithm. For binary document signatures (type 0x00), the document data is hashed directly. For text document signatures (type 0x01), the document is canonicalized by converting line endings to , and the resulting data is hashed. When a signature is made over a key, the hash data starts with the octet 0x99, followed by a two-octet length of the key, and then body of the key packet. (Note that this is an old-style packet header for a key packet with two-octet length.) A subkey binding signature (type 0x18) or primary key binding signature (type 0x19) then hashes the subkey using the same format as the main key (also using 0x99 as the first octet). Key revocation signatures (types 0x20 and 0x28) hash only the key being revoked. A certification signature (type 0x10 through 0x13) hashes the User ID being bound to the key into the hash context after the above data. A V3 certification hashes the contents of the User ID or attribute packet packet, without any header. A V4 certification hashes the constant 0xB4 for User ID certifications or the constant 0xD1 for User Attribute certifications, followed by a four-octet number giving the length of the User ID or User Attribute data, and then the User ID or User Attribute data. When a signature is made over a Signature packet (type 0x50), the hash data starts with the octet 0x88, followed by the four-octet length of the signature, and then the body of the Signature packet. (Note that this is an old-style packet header for a Signature packet with the length-of-length set to zero.) The unhashed subpacket data of the Signature packet being hashed is not included in the hash, and the unhashed subpacket data length value is set to zero. Once the data body is hashed, then a trailer is hashed. A V3 signature hashes five octets of the packet body, starting from the signature type field. This data is the signature type, followed by the four-octet signature time. A V4 signature hashes the packet body starting from its first field, the version number, through the end of the hashed subpacket data. Thus, the fields hashed are the signature version, the signature type, the public-key algorithm, the hash algorithm, the hashed subpacket length, and the hashed subpacket body. V4 signatures also hash in a final trailer of six octets: the version of the Signature packet, i.e., 0x04; 0xFF; and a four-octet, big-endian number that is the length of the hashed data from the Signature packet (note that this number does not include these final six octets). After all this has been hashed in a single hash context, the resulting hash field is used in the signature algorithm and placed at the end of the Signature packet. 5.2.4.1. Subpacket Hints It is certainly possible for a signature to contain conflicting information in subpackets. For example, a signature may contain multiple copies of a preference or multiple expiration times. In most cases, an implementation SHOULD use the last subpacket in the signature, but MAY use any conflict resolution scheme that makes more sense. Please note that we are intentionally leaving conflict resolution to the implementer; most conflicts are simply syntax errors, and the wishy-washy language here allows a receiver to be generous in what they accept, while putting pressure on a creator to be stingy in what they generate. Some apparent conflicts may actually make sense -- for example, suppose a keyholder has a V3 key and a V4 key that share the same RSA key material. Either of these keys can verify a signature created by the other, and it may be reasonable for a signature to contain an issuer subpacket for each key, as a way of explicitly tying those keys to the signature. 5.3. Symmetric-Key Encrypted Session Key Packets (Tag 3) The Symmetric-Key Encrypted Session Key packet holds the symmetric-key encryption of a session key used to encrypt a message. Zero or more Public-Key Encrypted Session Key packets and/or Symmetric-Key Encrypted Session Key packets may precede a Symmetrically Encrypted Data packet that holds an encrypted message. The message is encrypted with a session key, and the session key is itself encrypted and stored in the Encrypted Session Key packet or the Symmetric-Key Encrypted Session Key packet. If the Symmetrically Encrypted Data packet is preceded by one or more Symmetric-Key Encrypted Session Key packets, each specifies a passphrase that may be used to decrypt the message. This allows a message to be encrypted to a number of public keys, and also to one or more passphrases. This packet type is new and is not generated by PGP 2.x or PGP 5.0. The body of this packet consists of: - A one-octet version number. The only currently defined version is 4. - A one-octet number describing the symmetric algorithm used. - A string-to-key (S2K) specifier, length as defined above. - Optionally, the encrypted session key itself, which is decrypted with the string-to-key object. If the encrypted session key is not present (which can be detected on the basis of packet length and S2K specifier size), then the S2K algorithm applied to the passphrase produces the session key for decrypting the file, using the symmetric cipher algorithm from the Symmetric-Key Encrypted Session Key packet. If the encrypted session key is present, the result of applying the S2K algorithm to the passphrase is used to decrypt just that encrypted session key field, using CFB mode with an IV of all zeros. The decryption result consists of a one-octet algorithm identifier that specifies the symmetric-key encryption algorithm used to encrypt the following Symmetrically Encrypted Data packet, followed by the session key octets themselves. Note: because an all-zero IV is used for this decryption, the S2K specifier MUST use a salt value, either a Salted S2K or an Iterated-Salted S2K. The salt value will ensure that the decryption key is not repeated even if the passphrase is reused. 5.4. One-Pass Signature Packets (Tag 4) The One-Pass Signature packet precedes the signed data and contains enough information to allow the receiver to begin calculating any hashes needed to verify the signature. It allows the Signature packet to be placed at the end of the message, so that the signer can compute the entire signed message in one pass. A One-Pass Signature does not interoperate with PGP 2.6.x or earlier. The body of this packet consists of: - A one-octet version number. The current version is 3. - A one-octet signature type. Signature types are described in Section 5.2.1. - A one-octet number describing the hash algorithm used. - A one-octet number describing the public-key algorithm used. - An eight-octet number holding the Key ID of the signing key. - A one-octet number holding a flag showing whether the signature is nested. A zero value indicates that the next packet is another One-Pass Signature packet that describes another signature to be applied to the same message data. Note that if a message contains more than one one-pass signature, then the Signature packets bracket the message; that is, the first Signature packet after the message corresponds to the last one-pass packet and the final Signature packet corresponds to the first one-pass packet. 5.5. Key Material Packet A key material packet contains all the information about a public or private key. There are four variants of this packet type, and two major versions. Consequently, this section is complex. 5.5.1. Key Packet Variants 5.5.1.1. Public-Key Packet (Tag 6) A Public-Key packet starts a series of packets that forms an OpenPGP key (sometimes called an OpenPGP certificate). DONE 5.5.1.2. Public-Subkey Packet (Tag 14) A Public-Subkey packet (tag 14) has exactly the same format as a Public-Key packet, but denotes a subkey. One or more subkeys may be associated with a top-level key. By convention, the top-level key provides signature services, and the subkeys provide encryption services. Note: in PGP 2.6.x, tag 14 was intended to indicate a comment packet. This tag was selected for reuse because no previous version of PGP ever emitted comment packets but they did properly ignore them. Public-Subkey packets are ignored by PGP 2.6.x and do not cause it to fail, providing a limited degree of backward compatibility. 5.5.1.3. Secret-Key Packet (Tag 5) A Secret-Key packet contains all the information that is found in a Public-Key packet, including the public-key material, but also includes the secret-key material after all the public-key fields. 5.5.1.4. Secret-Subkey Packet (Tag 7) A Secret-Subkey packet (tag 7) is the subkey analog of the Secret Key packet and has exactly the same format. DONE 5.5.2. Public-Key Packet Formats There are two versions of key-material packets. Version 3 packets were first generated by PGP 2.6. Version 4 keys first appeared in PGP 5.0 and are the preferred key version for OpenPGP. OpenPGP implementations MUST create keys with version 4 format. V3 keys are deprecated; an implementation MUST NOT generate a V3 key, but MAY accept it. A version 3 public key or public-subkey packet contains: - A one-octet version number (3). - A four-octet number denoting the time that the key was created. - A two-octet number denoting the time in days that this key is valid. If this number is zero, then it does not expire. - A one-octet number denoting the public-key algorithm of this key. - A series of multiprecision integers comprising the key material: - a multiprecision integer (MPI) of RSA public modulus n; - an MPI of RSA public encryption exponent e. V3 keys are deprecated. They contain three weaknesses. First, it is relatively easy to construct a V3 key that has the same Key ID as any other key because the Key ID is simply the low 64 bits of the public modulus. Secondly, because the fingerprint of a V3 key hashes the key material, but not its length, there is an increased opportunity for fingerprint collisions. Third, there are weaknesses in the MD5 hash algorithm that make developers prefer other algorithms. See below for a fuller discussion of Key IDs and fingerprints. V2 keys are identical to the deprecated V3 keys except for the version number. An implementation MUST NOT generate them and MAY accept or reject them as it sees fit. The version 4 format is similar to the version 3 format except for the absence of a validity period. This has been moved to the Signature packet. In addition, fingerprints of version 4 keys are calculated differently from version 3 keys, as described in the section "Enhanced Key Formats". A version 4 packet contains: - A one-octet version number (4). - A four-octet number denoting the time that the key was created. - A one-octet number denoting the public-key algorithm of this key. - A series of multiprecision integers comprising the key material. This algorithm-specific portion is: Algorithm-Specific Fields for RSA public keys: - multiprecision integer (MPI) of RSA public modulus n; - MPI of RSA public encryption exponent e. Algorithm-Specific Fields for DSA public keys: - MPI of DSA prime p; - MPI of DSA group order q (q is a prime divisor of p-1); - MPI of DSA group generator g; - MPI of DSA public-key value y (= g**x mod p where x is secret). Algorithm-Specific Fields for Elgamal public keys: - MPI of Elgamal prime p; - MPI of Elgamal group generator g; - MPI of Elgamal public key value y (= g**x mod p where x is secret). 5.5.3. Secret-Key Packet Formats The Secret-Key and Secret-Subkey packets contain all the data of the Public-Key and Public-Subkey packets, with additional algorithm- specific secret-key data appended, usually in encrypted form. The packet contains: - A Public-Key or Public-Subkey packet, as described above. - One octet indicating string-to-key usage conventions. Zero indicates that the secret-key data is not encrypted. 255 or 254 indicates that a string-to-key specifier is being given. Any other value is a symmetric-key encryption algorithm identifier. - [Optional] If string-to-key usage octet was 255 or 254, a one- octet symmetric encryption algorithm. - [Optional] If string-to-key usage octet was 255 or 254, a string-to-key specifier. The length of the string-to-key specifier is implied by its type, as described above. - [Optional] If secret data is encrypted (string-to-key usage octet not zero), an Initial Vector (IV) of the same length as the cipher's block size. - Plain or encrypted multiprecision integers comprising the secret key data. These algorithm-specific fields are as described below. - If the string-to-key usage octet is zero or 255, then a two-octet checksum of the plaintext of the algorithm-specific portion (sum of all octets, mod 65536). If the string-to-key usage octet was 254, then a 20-octet SHA-1 hash of the plaintext of the algorithm-specific portion. This checksum or hash is encrypted together with the algorithm-specific fields (if string-to-key usage octet is not zero). Note that for all other values, a two-octet checksum is required. Algorithm-Specific Fields for RSA secret keys: - multiprecision integer (MPI) of RSA secret exponent d. - MPI of RSA secret prime value p. - MPI of RSA secret prime value q (p < q). - MPI of u, the multiplicative inverse of p, mod q. Algorithm-Specific Fields for DSA secret keys: - MPI of DSA secret exponent x. Algorithm-Specific Fields for Elgamal secret keys: - MPI of Elgamal secret exponent x. Secret MPI values can be encrypted using a passphrase. If a string- to-key specifier is given, that describes the algorithm for converting the passphrase to a key, else a simple MD5 hash of the passphrase is used. Implementations MUST use a string-to-key specifier; the simple hash is for backward compatibility and is deprecated, though implementations MAY continue to use existing private keys in the old format. The cipher for encrypting the MPIs is specified in the Secret-Key packet. Encryption/decryption of the secret data is done in CFB mode using the key created from the passphrase and the Initial Vector from the packet. A different mode is used with V3 keys (which are only RSA) than with other key formats. With V3 keys, the MPI bit count prefix (i.e., the first two octets) is not encrypted. Only the MPI non- prefix data is encrypted. Furthermore, the CFB state is resynchronized at the beginning of each new MPI value, so that the CFB block boundary is aligned with the start of the MPI data. With V4 keys, a simpler method is used. All secret MPI values are encrypted in CFB mode, including the MPI bitcount prefix. The two-octet checksum that follows the algorithm-specific portion is the algebraic sum, mod 65536, of the plaintext of all the algorithm- specific octets (including MPI prefix and data). With V3 keys, the checksum is stored in the clear. With V4 keys, the checksum is encrypted like the algorithm-specific data. This value is used to check that the passphrase was correct. However, this checksum is deprecated; an implementation SHOULD NOT use it, but should rather use the SHA-1 hash denoted with a usage octet of 254. The reason for this is that there are some attacks that involve undetectably modifying the secret key. 5.6. Compressed Data Packet (Tag 8) The Compressed Data packet contains compressed data. Typically, this packet is found as the contents of an encrypted packet, or following a Signature or One-Pass Signature packet, and contains a literal data packet. The body of this packet consists of: - One octet that gives the algorithm used to compress the packet. - Compressed data, which makes up the remainder of the packet. A Compressed Data Packet's body contains an block that compresses some set of packets. See section "Packet Composition" for details on how messages are formed. ZIP-compressed packets are compressed with raw RFC 1951 [RFC1951] DEFLATE blocks. Note that PGP V2.6 uses 13 bits of compression. If an implementation uses more bits of compression, PGP V2.6 cannot decompress it. ZLIB-compressed packets are compressed with RFC 1950 [RFC1950] ZLIB- style blocks. BZip2-compressed packets are compressed using the BZip2 [BZ2] algorithm. 5.7. Symmetrically Encrypted Data Packet (Tag 9) The Symmetrically Encrypted Data packet contains data encrypted with a symmetric-key algorithm. When it has been decrypted, it contains other packets (usually a literal data packet or compressed data packet, but in theory other Symmetrically Encrypted Data packets or sequences of packets that form whole OpenPGP messages). The body of this packet consists of: - Encrypted data, the output of the selected symmetric-key cipher operating in OpenPGP's variant of Cipher Feedback (CFB) mode. The symmetric cipher used may be specified in a Public-Key or Symmetric-Key Encrypted Session Key packet that precedes the Symmetrically Encrypted Data packet. In that case, the cipher algorithm octet is prefixed to the session key before it is encrypted. If no packets of these types precede the encrypted data, the IDEA algorithm is used with the session key calculated as the MD5 hash of the passphrase, though this use is deprecated. The data is encrypted in CFB mode, with a CFB shift size equal to the cipher's block size. The Initial Vector (IV) is specified as all zeros. Instead of using an IV, OpenPGP prefixes a string of length equal to the block size of the cipher plus two to the data before it is encrypted. The first block-size octets (for example, 8 octets for a 64-bit block length) are random, and the following two octets are copies of the last two octets of the IV. For example, in an 8-octet block, octet 9 is a repeat of octet 7, and octet 10 is a repeat of octet 8. In a cipher of length 16, octet 17 is a repeat of octet 15 and octet 18 is a repeat of octet 16. As a pedantic clarification, in both these examples, we consider the first octet to be numbered 1. After encrypting the first block-size-plus-two octets, the CFB state is resynchronized. The last block-size octets of ciphertext are passed through the cipher and the block boundary is reset. The repetition of 16 bits in the random data prefixed to the message allows the receiver to immediately check whether the session key is incorrect. See the "Security Considerations" section for hints on the proper use of this "quick check". DONE 5.8. Marker Packet (Obsolete Literal Packet) (Tag 10) An experimental version of PGP used this packet as the Literal packet, but no released version of PGP generated Literal packets with this tag. With PGP 5.x, this packet has been reassigned and is reserved for use as the Marker packet. The body of this packet consists of: - The three octets 0x50, 0x47, 0x50 (which spell "PGP" in UTF-8). Such a packet MUST be ignored when received. It may be placed at the beginning of a message that uses features not available in PGP 2.6.x in order to cause that version to report that newer software is necessary to process the message. 5.9. Literal Data Packet (Tag 11) A Literal Data packet contains the body of a message; data that is not to be further interpreted. The body of this packet consists of: - A one-octet field that describes how the data is formatted. If it is a 'b' (0x62), then the Literal packet contains binary data. If it is a 't' (0x74), then it contains text data, and thus may need line ends converted to local form, or other text-mode changes. The tag 'u' (0x75) means the same as 't', but also indicates that implementation believes that the literal data contains UTF-8 text. Early versions of PGP also defined a value of 'l' as a 'local' mode for machine-local conversions. RFC 1991 [RFC1991] incorrectly stated this local mode flag as '1' (ASCII numeral one). Both of these local modes are deprecated. - File name as a string (one-octet length, followed by a file name). This may be a zero-length string. Commonly, if the source of the encrypted data is a file, this will be the name of the encrypted file. An implementation MAY consider the file name in the Literal packet to be a more authoritative name than the actual file name. If the special name "_CONSOLE" is used, the message is considered to be "for your eyes only". This advises that the message data is unusually sensitive, and the receiving program should process it more carefully, perhaps avoiding storing the received data to disk, for example. - A four-octet number that indicates a date associated with the literal data. Commonly, the date might be the modification date of a file, or the time the packet was created, or a zero that indicates no specific time. - The remainder of the packet is literal data. Text data is stored with text endings (i.e., network- normal line endings). These should be converted to native line endings by the receiving software. 5.10. Trust Packet (Tag 12) [...] Trust packets SHOULD NOT be emitted to output streams that are transferred to other users, and they SHOULD be ignored on any input other than local keyring files. 5.12. User Attribute Packet (Tag 17) [...] The only currently defined subpacket type is 1, signifying an image. An implementation SHOULD ignore any subpacket of a type that it does not recognize. Subpacket types 100 through 110 are reserved for private or experimental use. [...] 5.13. Sym. Encrypted Integrity Protected Data Packet (Tag 18) The Symmetrically Encrypted Integrity Protected Data packet is a variant of the Symmetrically Encrypted Data packet. It is a new feature created for OpenPGP that addresses the problem of detecting a modification to encrypted data. It is used in combination with a Modification Detection Code packet. There is a corresponding feature in the features Signature subpacket that denotes that an implementation can properly use this packet type. An implementation MUST support decrypting these packets and SHOULD prefer generating them to the older Symmetrically Encrypted Data packet when possible. Since this data packet protects against modification attacks, this standard encourages its proliferation. While blanket adoption of this data packet would create interoperability problems, rapid adoption is nevertheless important. An implementation SHOULD specifically denote support for this packet, but it MAY infer it from other mechanisms. For example, an implementation might infer from the use of a cipher such as Advanced Encryption Standard (AES) or Twofish that a user supports this feature. It might place in the unhashed portion of another user's key signature a Features subpacket. It might also present a user with an opportunity to regenerate their own self- signature with a Features subpacket. This packet contains data encrypted with a symmetric-key algorithm and protected against modification by the SHA-1 hash algorithm. When it has been decrypted, it will typically contain other packets (often a Literal Data packet or Compressed Data packet). The last decrypted packet in this packet's payload MUST be a Modification Detection Code packet. The body of this packet consists of: - A one-octet version number. The only currently defined value is 1. - Encrypted data, the output of the selected symmetric-key cipher operating in Cipher Feedback mode with shift amount equal to the block size of the cipher (CFB-n where n is the block size). The symmetric cipher used MUST be specified in a Public-Key or Symmetric-Key Encrypted Session Key packet that precedes the Symmetrically Encrypted Data packet. In either case, the cipher algorithm octet is prefixed to the session key before it is encrypted. The data is encrypted in CFB mode, with a CFB shift size equal to the cipher's block size. The Initial Vector (IV) is specified as all zeros. Instead of using an IV, OpenPGP prefixes an octet string to the data before it is encrypted. The length of the octet string equals the block size of the cipher in octets, plus two. The first octets in the group, of length equal to the block size of the cipher, are random; the last two octets are each copies of their 2nd preceding octet. For example, with a cipher whose block size is 128 bits or 16 octets, the prefix data will contain 16 random octets, then two more octets, which are copies of the 15th and 16th octets, respectively. Unlike the Symmetrically Encrypted Data Packet, no special CFB resynchronization is done after encrypting this prefix data. See "OpenPGP CFB Mode" below for more details. The repetition of 16 bits in the random data prefixed to the message allows the receiver to immediately check whether the session key is incorrect. The plaintext of the data to be encrypted is passed through the SHA-1 hash function, and the result of the hash is appended to the plaintext in a Modification Detection Code packet. The input to the hash function includes the prefix data described above; it includes all of the plaintext, and then also includes two octets of values 0xD3, 0x14. These represent the encoding of a Modification Detection Code packet tag and length field of 20 octets. The resulting hash value is stored in a Modification Detection Code (MDC) packet, which MUST use the two octet encoding just given to represent its tag and length field. The body of the MDC packet is the 20-octet output of the SHA-1 hash. The Modification Detection Code packet is appended to the plaintext and encrypted along with the plaintext using the same CFB context. During decryption, the plaintext data should be hashed with SHA-1, including the prefix data as well as the packet tag and length field of the Modification Detection Code packet. The body of the MDC packet, upon decryption, is compared with the result of the SHA-1 hash. Any failure of the MDC indicates that the message has been modified and MUST be treated as a security problem. Failures include a difference in the hash values, but also the absence of an MDC packet, or an MDC packet in any position other than the end of the plaintext. Any failure SHOULD be reported to the user. Note: future designs of new versions of this packet should consider rollback attacks since it will be possible for an attacker to change the version back to 1. NON-NORMATIVE EXPLANATION The MDC system, as packets 18 and 19 are called, were created to provide an integrity mechanism that is less strong than a signature, yet stronger than bare CFB encryption. It is a limitation of CFB encryption that damage to the ciphertext will corrupt the affected cipher blocks and the block following. Additionally, if data is removed from the end of a CFB-encrypted block, that removal is undetectable. (Note also that CBC mode has a similar limitation, but data removed from the front of the block is undetectable.) The obvious way to protect or authenticate an encrypted block is to digitally sign it. However, many people do not wish to habitually sign data, for a large number of reasons beyond the scope of this document. Suffice it to say that many people consider properties such as deniability to be as valuable as integrity. OpenPGP addresses this desire to have more security than raw encryption and yet preserve deniability with the MDC system. An MDC is intentionally not a MAC. Its name was not selected by accident. It is analogous to a checksum. Despite the fact that it is a relatively modest system, it has proved itself in the real world. It is an effective defense to several attacks that have surfaced since it has been created. It has met its modest goals admirably. Consequently, because it is a modest security system, it has modest requirements on the hash function(s) it employs. It does not rely on a hash function being collision-free, it relies on a hash function being one-way. If a forger, Frank, wishes to send Alice a (digitally) unsigned message that says, "I've always secretly loved you, signed Bob", it is far easier for him to construct a new message than it is to modify anything intercepted from Bob. (Note also that if Bob wishes to communicate secretly with Alice, but without authentication or identification and with a threat model that includes forgers, he has a problem that transcends mere cryptography.) Note also that unlike nearly every other OpenPGP subsystem, there are no parameters in the MDC system. It hard-defines SHA-1 as its hash function. This is not an accident. It is an intentional choice to avoid downgrade and cross-grade attacks while making a simple, fast system. (A downgrade attack would be an attack that replaced SHA-256 with SHA-1, for example. A cross-grade attack would replace SHA-1 with another 160-bit hash, such as RIPE- MD/160, for example.) However, given the present state of hash function cryptanalysis and cryptography, it may be desirable to upgrade the MDC system to a new hash function. See Section 13.11 in the "IANA Considerations" for guidance. 5.14. Modification Detection Code Packet (Tag 19) The Modification Detection Code packet contains a SHA-1 hash of plaintext data, which is used to detect message modification. It is only used with a Symmetrically Encrypted Integrity Protected Data packet. The Modification Detection Code packet MUST be the last packet in the plaintext data that is encrypted in the Symmetrically Encrypted Integrity Protected Data packet, and MUST appear in no other place. A Modification Detection Code packet MUST have a length of 20 octets. The body of this packet consists of: - A 20-octet SHA-1 hash of the preceding plaintext data of the Symmetrically Encrypted Integrity Protected Data packet, including prefix data, the tag octet, and length octet of the Modification Detection Code packet. Note that the Modification Detection Code packet MUST always use a new format encoding of the packet tag, and a one-octet encoding of the packet length. The reason for this is that the hashing rules for modification detection include a one-octet tag and one-octet length in the data hash. While this is a bit restrictive, it reduces complexity. DONE 6. Radix-64 Conversions As stated in the introduction, OpenPGP's underlying native representation for objects is a stream of arbitrary octets, and some systems desire these objects to be immune to damage caused by character set translation, data conversions, etc. In principle, any printable encoding scheme that met the requirements of the unsafe channel would suffice, since it would not change the underlying binary bit streams of the native OpenPGP data structures. The OpenPGP standard specifies one such printable encoding scheme to ensure interoperability. OpenPGP's Radix-64 encoding is composed of two parts: a base64 encoding of the binary data and a checksum. The base64 encoding is identical to the MIME base64 content-transfer-encoding [RFC2045]. The checksum is a 24-bit Cyclic Redundancy Check (CRC) converted to four characters of radix-64 encoding by the same MIME base64 transformation, preceded by an equal sign (=). The CRC is computed by using the generator 0x864CFB and an initialization of 0xB704CE. The accumulation is done on the data before it is converted to radix-64, rather than on the converted data. A sample implementation of this algorithm is in the next section. The checksum with its leading equal sign MAY appear on the first line after the base64 encoded data. Rationale for CRC-24: The size of 24 bits fits evenly into printable base64. The nonzero initialization can detect more errors than a zero initialization. 6.1. An Implementation of the CRC-24 in "C" #define CRC24_INIT 0xB704CEL #define CRC24_POLY 0x1864CFBL typedef long crc24; crc24 crc_octets(unsigned char *octets, size_t len) { crc24 crc = CRC24_INIT; int i; while (len--) { crc ^= (*octets++) << 16; for (i = 0; i < 8; i++) { crc <<= 1; if (crc & 0x1000000) crc ^= CRC24_POLY; } } return crc & 0xFFFFFFL; } 6.2. Forming ASCII Armor When OpenPGP encodes data into ASCII Armor, it puts specific headers around the Radix-64 encoded data, so OpenPGP can reconstruct the data later. An OpenPGP implementation MAY use ASCII armor to protect raw binary data. OpenPGP informs the user what kind of data is encoded in the ASCII armor through the use of the headers. Concatenating the following data creates ASCII Armor: - An Armor Header Line, appropriate for the type of data - Armor Headers - A blank (zero-length, or containing only whitespace) line - The ASCII-Armored data - An Armor Checksum - The Armor Tail, which depends on the Armor Header Line An Armor Header Line consists of the appropriate header line text surrounded by five (5) dashes ('-', 0x2D) on either side of the header line text. The header line text is chosen based upon the type of data that is being encoded in Armor, and how it is being encoded. Header line texts include the following strings: BEGIN PGP MESSAGE Used for signed, encrypted, or compressed files. BEGIN PGP PUBLIC KEY BLOCK Used for armoring public keys. BEGIN PGP PRIVATE KEY BLOCK Used for armoring private keys. BEGIN PGP MESSAGE, PART X/Y Used for multi-part messages, where the armor is split amongst Y parts, and this is the Xth part out of Y. BEGIN PGP MESSAGE, PART X Used for multi-part messages, where this is the Xth part of an unspecified number of parts. Requires the MESSAGE-ID Armor Header to be used. BEGIN PGP SIGNATURE Used for detached signatures, OpenPGP/MIME signatures, and cleartext signatures. Note that PGP 2.x uses BEGIN PGP MESSAGE for detached signatures. Note that all these Armor Header Lines are to consist of a complete line. That is to say, there is always a line ending preceding the starting five dashes, and following the ending five dashes. The header lines, therefore, MUST start at the beginning of a line, and MUST NOT have text other than whitespace following them on the same line. These line endings are considered a part of the Armor Header Line for the purposes of determining the content they delimit. This is particularly important when computing a cleartext signature (see below). The Armor Headers are pairs of strings that can give the user or the receiving OpenPGP implementation some information about how to decode or use the message. The Armor Headers are a part of the armor, not a part of the message, and hence are not protected by any signatures applied to the message. The format of an Armor Header is that of a key-value pair. A colon (':' 0x38) and a single space (0x20) separate the key and value. OpenPGP should consider improperly formatted Armor Headers to be corruption of the ASCII Armor. Unknown keys should be reported to the user, but OpenPGP should continue to process the message. Note that some transport methods are sensitive to line length. While there is a limit of 76 characters for the Radix-64 data (Section 6.3), there is no limit to the length of Armor Headers. Care should be taken that the Armor Headers are short enough to survive transport. One way to do this is to repeat an Armor Header key multiple times with different values for each so that no one line is overly long. Currently defined Armor Header Keys are as follows: - "Version", which states the OpenPGP implementation and version used to encode the message. - "Comment", a user-defined comment. OpenPGP defines all text to be in UTF-8. A comment may be any UTF-8 string. However, the whole point of armoring is to provide seven-bit-clean data. Consequently, if a comment has characters that are outside the US-ASCII range of UTF, they may very well not survive transport. - "MessageID", a 32-character string of printable characters. The string must be the same for all parts of a multi-part message that uses the "PART X" Armor Header. MessageID strings should be unique enough that the recipient of the mail can associate all the parts of a message with each other. A good checksum or cryptographic hash function is sufficient. The MessageID SHOULD NOT appear unless it is in a multi-part message. If it appears at all, it MUST be computed from the finished (encrypted, signed, etc.) message in a deterministic fashion, rather than contain a purely random value. This is to allow the legitimate recipient to determine that the MessageID cannot serve as a covert means of leaking cryptographic key information. - "Hash", a comma-separated list of hash algorithms used in this message. This is used only in cleartext signed messages. - "Charset", a description of the character set that the plaintext is in. Please note that OpenPGP defines text to be in UTF-8. An implementation will get best results by translating into and out of UTF-8. However, there are many instances where this is easier said than done. Also, there are communities of users who have no need for UTF-8 because they are all happy with a character set like ISO Latin-5 or a Japanese character set. In such instances, an implementation MAY override the UTF-8 default by using this header key. An implementation MAY implement this key and any translations it cares to; an implementation MAY ignore it and assume all text is UTF-8. The Armor Tail Line is composed in the same manner as the Armor Header Line, except the string "BEGIN" is replaced by the string "END". 6.3. Encoding Binary in Radix-64 The encoding process represents 24-bit groups of input bits as output strings of 4 encoded characters. Proceeding from left to right, a 24-bit input group is formed by concatenating three 8-bit input groups. These 24 bits are then treated as four concatenated 6-bit groups, each of which is translated into a single digit in the Radix-64 alphabet. When encoding a bit stream with the Radix-64 encoding, the bit stream must be presumed to be ordered with the most significant bit first. That is, the first bit in the stream will be the high-order bit in the first 8-bit octet, and the eighth bit will be the low-order bit in the first 8-bit octet, and so on. +--first octet--+-second octet--+--third octet--+ |7 6 5 4 3 2 1 0|7 6 5 4 3 2 1 0|7 6 5 4 3 2 1 0| +-----------+---+-------+-------+---+-----------+ |5 4 3 2 1 0|5 4 3 2 1 0|5 4 3 2 1 0|5 4 3 2 1 0| +--1.index--+--2.index--+--3.index--+--4.index--+ Each 6-bit group is used as an index into an array of 64 printable characters from the table below. The character referenced by the index is placed in the output string. Value Encoding Value Encoding Value Encoding Value Encoding 0 A 17 R 34 i 51 z 1 B 18 S 35 j 52 0 2 C 19 T 36 k 53 1 3 D 20 U 37 l 54 2 4 E 21 V 38 m 55 3 5 F 22 W 39 n 56 4 6 G 23 X 40 o 57 5 7 H 24 Y 41 p 58 6 8 I 25 Z 42 q 59 7 9 J 26 a 43 r 60 8 10 K 27 b 44 s 61 9 11 L 28 c 45 t 62 + 12 M 29 d 46 u 63 / 13 N 30 e 47 v 14 O 31 f 48 w (pad) = 15 P 32 g 49 x 16 Q 33 h 50 y The encoded output stream must be represented in lines of no more than 76 characters each. Special processing is performed if fewer than 24 bits are available at the end of the data being encoded. There are three possibilities: 1. The last data group has 24 bits (3 octets). No special processing is needed. 2. The last data group has 16 bits (2 octets). The first two 6-bit groups are processed as above. The third (incomplete) data group has two zero-value bits added to it, and is processed as above. A pad character (=) is added to the output. 3. The last data group has 8 bits (1 octet). The first 6-bit group is processed as above. The second (incomplete) data group has four zero-value bits added to it, and is processed as above. Two pad characters (=) are added to the output. 6.4. Decoding Radix-64 In Radix-64 data, characters other than those in the table, line breaks, and other white space probably indicate a transmission error, about which a warning message or even a message rejection might be appropriate under some circumstances. Decoding software must ignore all white space. Because it is used only for padding at the end of the data, the occurrence of any "=" characters may be taken as evidence that the end of the data has been reached (without truncation in transit). No such assurance is possible, however, when the number of octets transmitted was a multiple of three and no "=" characters are present. 6.5. Examples of Radix-64 Input data: 0x14FB9C03D97E Hex: 1 4 F B 9 C | 0 3 D 9 7 E 8-bit: 00010100 11111011 10011100 | 00000011 11011001 11111110 6-bit: 000101 001111 101110 011100 | 000000 111101 100111 111110 Decimal: 5 15 46 28 0 61 37 62 Output: F P u c A 9 l + Input data: 0x14FB9C03D9 Hex: 1 4 F B 9 C | 0 3 D 9 8-bit: 00010100 11111011 10011100 | 00000011 11011001 pad with 00 6-bit: 000101 001111 101110 011100 | 000000 111101 100100 Decimal: 5 15 46 28 0 61 36 pad with = Output: F P u c A 9 k = Input data: 0x14FB9C03 Hex: 1 4 F B 9 C | 0 3 8-bit: 00010100 11111011 10011100 | 00000011 pad with 0000 6-bit: 000101 001111 101110 011100 | 000000 110000 Decimal: 5 15 46 28 0 48 pad with = = Output: F P u c A w = = 6.6. Example of an ASCII Armored Message -----BEGIN PGP MESSAGE----- Version: OpenPrivacy 0.99 yDgBO22WxBHv7O8X7O/jygAEzol56iUKiXmV+XmpCtmpqQUKiQrFqclFqUDBovzS vBSFjNSiVHsuAA== =njUN -----END PGP MESSAGE----- Note that this example has extra indenting; an actual armored message would have no leading whitespace. 7. Cleartext Signature Framework It is desirable to be able to sign a textual octet stream without ASCII armoring the stream itself, so the signed text is still readable without special software. In order to bind a signature to such a cleartext, this framework is used. (Note that this framework is not intended to be reversible. RFC 3156 [RFC3156] defines another way to sign cleartext messages for environments that support MIME.) The cleartext signed message consists of: - The cleartext header '-----BEGIN PGP SIGNED MESSAGE-----' on a single line, - One or more "Hash" Armor Headers, - Exactly one empty line not included into the message digest, - The dash-escaped cleartext that is included into the message digest, - The ASCII armored signature(s) including the '-----BEGIN PGP SIGNATURE-----' Armor Header and Armor Tail Lines. If the "Hash" Armor Header is given, the specified message digest algorithm(s) are used for the signature. If there are no such headers, MD5 is used. If MD5 is the only hash used, then an implementation MAY omit this header for improved V2.x compatibility. If more than one message digest is used in the signature, the "Hash" armor header contains a comma-delimited list of used message digests. Current message digest names are described below with the algorithm IDs. An implementation SHOULD add a line break after the cleartext, but MAY omit it if the cleartext ends with a line break. This is for visual clarity. 7.1. Dash-Escaped Text The cleartext content of the message must also be dash-escaped. Dash-escaped cleartext is the ordinary cleartext where every line starting with a dash '-' (0x2D) is prefixed by the sequence dash '-' (0x2D) and space ' ' (0x20). This prevents the parser from recognizing armor headers of the cleartext itself. An implementation MAY dash-escape any line, SHOULD dash-escape lines commencing "From" followed by a space, and MUST dash-escape any line commencing in a dash. The message digest is computed using the cleartext itself, not the dash-escaped form. As with binary signatures on text documents, a cleartext signature is calculated on the text using canonical line endings. The line ending (i.e., the ) before the '-----BEGIN PGP SIGNATURE-----' line that terminates the signed text is not considered part of the signed text. When reversing dash-escaping, an implementation MUST strip the string "- " if it occurs at the beginning of a line, and SHOULD warn on "-" and any character other than a space at the beginning of a line. Also, any trailing whitespace -- spaces (0x20) and tabs (0x09) -- at the end of any line is removed when the cleartext signature is generated. 8. Regular Expressions A regular expression is zero or more branches, separated by '|'. It matches anything that matches one of the branches. A branch is zero or more pieces, concatenated. It matches a match for the first, followed by a match for the second, etc. A piece is an atom possibly followed by '*', '+', or '?'. An atom followed by '*' matches a sequence of 0 or more matches of the atom. An atom followed by '+' matches a sequence of 1 or more matches of the atom. An atom followed by '?' matches a match of the atom, or the null string. An atom is a regular expression in parentheses (matching a match for the regular expression), a range (see below), '.' (matching any single character), '^' (matching the null string at the beginning of the input string), '$' (matching the null string at the end of the input string), a '\' followed by a single character (matching that character), or a single character with no other significance (matching that character). A range is a sequence of characters enclosed in '[]'. It normally matches any single character from the sequence. If the sequence begins with '^', it matches any single character not from the rest of the sequence. If two characters in the sequence are separated by '-', this is shorthand for the full list of ASCII characters between them (e.g., '[0-9]' matches any decimal digit). To include a literal ']' in the sequence, make it the first character (following a possible '^'). To include a literal '-', make it the first or last character. [...] 9.1. Public-Key Algorithms [...] Implementations MUST implement DSA for signatures, and Elgamal for encryption. Implementations SHOULD implement RSA keys (1). RSA Encrypt-Only (2) and RSA Sign-Only are deprecated and SHOULD NOT be generated, but may be interpreted. See Section 13.5. See Section 13.8 for notes on Elliptic Curve (18), ECDSA (19), Elgamal Encrypt or Sign (20), and X9.42 (21). Implementations MAY implement any other algorithm. 9.2. Symmetric-Key Algorithms [...] Implementations MUST implement TripleDES. Implementations SHOULD implement AES-128 and CAST5. Implementations that interoperate with PGP 2.6 or earlier need to support IDEA, as that is the only symmetric cipher those versions use. Implementations MAY implement any other algorithm. 9.3. Compression Algorithms [...] Implementations MUST implement uncompressed data. Implementations SHOULD implement ZIP. Implementations MAY implement any other algorithm. 9.4. Hash Algorithms ID Algorithm Text Name -- --------- --------- 1 - MD5 [HAC] "MD5" 2 - SHA-1 [FIPS180] "SHA1" 3 - RIPE-MD/160 [HAC] "RIPEMD160" 4 - Reserved 5 - Reserved 6 - Reserved 7 - Reserved 8 - SHA256 [FIPS180] "SHA256" 9 - SHA384 [FIPS180] "SHA384" 10 - SHA512 [FIPS180] "SHA512" 11 - SHA224 [FIPS180] "SHA224" 100 to 110 - Private/Experimental algorithm Implementations MUST implement SHA-1. Implementations MAY implement other algorithms. MD5 is deprecated. 10. IANA Considerations OpenPGP is highly parameterized, and consequently there are a number of considerations for allocating parameters for extensions. This section describes how IANA should look at extensions to the protocol as described in this document. 10.1. New String-to-Key Specifier Types OpenPGP S2K specifiers contain a mechanism for new algorithms to turn a string into a key. This specification creates a registry of S2K specifier types. The registry includes the S2K type, the name of the S2K, and a reference to the defining specification. The initial values for this registry can be found in Section 3.7.1. Adding a new S2K specifier MUST be done through the IETF CONSENSUS method, as described in [RFC2434]. 10.2. New Packets Major new features of OpenPGP are defined through new packet types. This specification creates a registry of packet types. The registry includes the packet type, the name of the packet, and a reference to the defining specification. The initial values for this registry can be found in Section 4.3. Adding a new packet type MUST be done through the IETF CONSENSUS method, as described in [RFC2434]. 10.2.1. User Attribute Types The User Attribute packet permits an extensible mechanism for other types of certificate identification. This specification creates a registry of User Attribute types. The registry includes the User Attribute type, the name of the User Attribute, and a reference to the defining specification. The initial values for this registry can be found in Section 5.12. Adding a new User Attribute type MUST be done through the IETF CONSENSUS method, as described in [RFC2434]. 10.2.1.1. Image Format Subpacket Types Within User Attribute packets, there is an extensible mechanism for other types of image-based user attributes. This specification creates a registry of Image Attribute subpacket types. The registry includes the Image Attribute subpacket type, the name of the Image Attribute subpacket, and a reference to the defining specification. The initial values for this registry can be found in Section 5.12.1. Adding a new Image Attribute subpacket type MUST be done through the IETF CONSENSUS method, as described in [RFC2434]. 10.2.2. New Signature Subpackets OpenPGP signatures contain a mechanism for signed (or unsigned) data to be added to them for a variety of purposes in the Signature subpackets as discussed in Section 5.2.3.1. This specification creates a registry of Signature subpacket types. The registry includes the Signature subpacket type, the name of the subpacket, and a reference to the defining specification. The initial values for this registry can be found in Section 5.2.3.1. Adding a new Signature subpacket MUST be done through the IETF CONSENSUS method, as described in [RFC2434]. 10.2.2.1. Signature Notation Data Subpackets OpenPGP signatures further contain a mechanism for extensions in signatures. These are the Notation Data subpackets, which contain a key/value pair. Notations contain a user space that is completely unmanaged and an IETF space. This specification creates a registry of Signature Notation Data types. The registry includes the Signature Notation Data type, the name of the Signature Notation Data, its allowed values, and a reference to the defining specification. The initial values for this registry can be found in Section 5.2.3.16. Adding a new Signature Notation Data subpacket MUST be done through the EXPERT REVIEW method, as described in [RFC2434]. 10.2.2.2. Key Server Preference Extensions OpenPGP signatures contain a mechanism for preferences to be specified about key servers. This specification creates a registry of key server preferences. The registry includes the key server preference, the name of the preference, and a reference to the defining specification. The initial values for this registry can be found in Section 5.2.3.17. Adding a new key server preference MUST be done through the IETF CONSENSUS method, as described in [RFC2434]. 10.2.2.3. Key Flags Extensions OpenPGP signatures contain a mechanism for flags to be specified about key usage. This specification creates a registry of key usage flags. The registry includes the key flags value, the name of the flag, and a reference to the defining specification. The initial values for this registry can be found in Section 5.2.3.21. Adding a new key usage flag MUST be done through the IETF CONSENSUS method, as described in [RFC2434]. 10.2.2.4. Reason for Revocation Extensions OpenPGP signatures contain a mechanism for flags to be specified about why a key was revoked. This specification creates a registry of "Reason for Revocation" flags. The registry includes the "Reason for Revocation" flags value, the name of the flag, and a reference to the defining specification. The initial values for this registry can be found in Section 5.2.3.23. Adding a new feature flag MUST be done through the IETF CONSENSUS method, as described in [RFC2434]. 10.2.2.5. Implementation Features OpenPGP signatures contain a mechanism for flags to be specified stating which optional features an implementation supports. This specification creates a registry of feature-implementation flags. The registry includes the feature-implementation flags value, the name of the flag, and a reference to the defining specification. The initial values for this registry can be found in Section 5.2.3.24. Adding a new feature-implementation flag MUST be done through the IETF CONSENSUS method, as described in [RFC2434]. Also see Section 13.12 for more information about when feature flags are needed. 10.2.3. New Packet Versions The core OpenPGP packets all have version numbers, and can be revised by introducing a new version of an existing packet. This specification creates a registry of packet types. The registry includes the packet type, the number of the version, and a reference to the defining specification. The initial values for this registry can be found in Section 5. Adding a new packet version MUST be done through the IETF CONSENSUS method, as described in [RFC2434]. 10.3. New Algorithms Section 9 lists the core algorithms that OpenPGP uses. Adding in a new algorithm is usually simple. For example, adding in a new symmetric cipher usually would not need anything more than allocating a constant for that cipher. If that cipher had other than a 64-bit or 128-bit block size, there might need to be additional documentation describing how OpenPGP-CFB mode would be adjusted. Similarly, when DSA was expanded from a maximum of 1024-bit public keys to 3072-bit public keys, the revision of FIPS 186 contained enough information itself to allow implementation. Changes to this document were made mainly for emphasis. 10.3.1. Public-Key Algorithms OpenPGP specifies a number of public-key algorithms. This specification creates a registry of public-key algorithm identifiers. The registry includes the algorithm name, its key sizes and parameters, and a reference to the defining specification. The initial values for this registry can be found in Section 9. Adding a new public-key algorithm MUST be done through the IETF CONSENSUS method, as described in [RFC2434]. 10.3.2. Symmetric-Key Algorithms OpenPGP specifies a number of symmetric-key algorithms. This specification creates a registry of symmetric-key algorithm identifiers. The registry includes the algorithm name, its key sizes and block size, and a reference to the defining specification. The initial values for this registry can be found in Section 9. Adding a new symmetric-key algorithm MUST be done through the IETF CONSENSUS method, as described in [RFC2434]. 10.3.3. Hash Algorithms OpenPGP specifies a number of hash algorithms. This specification creates a registry of hash algorithm identifiers. The registry includes the algorithm name, a text representation of that name, its block size, an OID hash prefix, and a reference to the defining specification. The initial values for this registry can be found in Section 9 for the algorithm identifiers and text names, and Section 5.2.2 for the OIDs and expanded signature prefixes. Adding a new hash algorithm MUST be done through the IETF CONSENSUS method, as described in [RFC2434]. 10.3.4. Compression Algorithms OpenPGP specifies a number of compression algorithms. This specification creates a registry of compression algorithm identifiers. The registry includes the algorithm name and a reference to the defining specification. The initial values for this registry can be found in Section 9.3. Adding a new compression key algorithm MUST be done through the IETF CONSENSUS method, as described in [RFC2434]. 11. Packet Composition OpenPGP packets are assembled into sequences in order to create messages and to transfer keys. Not all possible packet sequences are meaningful and correct. This section describes the rules for how packets should be placed into sequences. 11.1. Transferable Public Keys OpenPGP users may transfer public keys. The essential elements of a transferable public key are as follows: - One Public-Key packet - Zero or more revocation signatures - One or more User ID packets - After each User ID packet, zero or more Signature packets (certifications) - Zero or more User Attribute packets - After each User Attribute packet, zero or more Signature packets (certifications) - Zero or more Subkey packets - After each Subkey packet, one Signature packet, plus optionally a revocation The Public-Key packet occurs first. Each of the following User ID packets provides the identity of the owner of this public key. If there are multiple User ID packets, this corresponds to multiple means of identifying the same unique individual user; for example, a user may have more than one email address, and construct a User ID for each one. Immediately following each User ID packet, there are zero or more Signature packets. Each Signature packet is calculated on the immediately preceding User ID packet and the initial Public-Key packet. The signature serves to certify the corresponding public key and User ID. In effect, the signer is testifying to his or her belief that this public key belongs to the user identified by this User ID. Within the same section as the User ID packets, there are zero or more User Attribute packets. Like the User ID packets, a User Attribute packet is followed by zero or more Signature packets calculated on the immediately preceding User Attribute packet and the initial Public-Key packet. User Attribute packets and User ID packets may be freely intermixed in this section, so long as the signatures that follow them are maintained on the proper User Attribute or User ID packet. After the User ID packet or Attribute packet, there may be zero or more Subkey packets. In general, subkeys are provided in cases where the top-level public key is a signature-only key. However, any V4 key may have subkeys, and the subkeys may be encryption-only keys, signature-only keys, or general-purpose keys. V3 keys MUST NOT have subkeys. Each Subkey packet MUST be followed by one Signature packet, which should be a subkey binding signature issued by the top-level key. For subkeys that can issue signatures, the subkey binding signature MUST contain an Embedded Signature subpacket with a primary key binding signature (0x19) issued by the subkey on the top-level key. Subkey and Key packets may each be followed by a revocation Signature packet to indicate that the key is revoked. Revocation signatures are only accepted if they are issued by the key itself, or by a key that is authorized to issue revocations via a Revocation Key subpacket in a self-signature by the top-level key. Transferable public-key packet sequences may be concatenated to allow transferring multiple public keys in one operation. 11.2. Transferable Secret Keys OpenPGP users may transfer secret keys. The format of a transferable secret key is the same as a transferable public key except that secret-key and secret-subkey packets are used instead of the public key and public-subkey packets. Implementations SHOULD include self- signatures on any user IDs and subkeys, as this allows for a complete public key to be automatically extracted from the transferable secret key. Implementations MAY choose to omit the self-signatures, especially if a transferable public key accompanies the transferable secret key. 11.3. OpenPGP Messages An OpenPGP message is a packet or sequence of packets that corresponds to the following grammatical rules (comma represents sequential composition, and vertical bar separates alternatives): OpenPGP Message :- Encrypted Message | Signed Message | Compressed Message | Literal Message. Compressed Message :- Compressed Data Packet. Literal Message :- Literal Data Packet. ESK :- Public-Key Encrypted Session Key Packet | Symmetric-Key Encrypted Session Key Packet. ESK Sequence :- ESK | ESK Sequence, ESK. Encrypted Data :- Symmetrically Encrypted Data Packet | Symmetrically Encrypted Integrity Protected Data Packet Encrypted Message :- Encrypted Data | ESK Sequence, Encrypted Data. One-Pass Signed Message :- One-Pass Signature Packet, OpenPGP Message, Corresponding Signature Packet. Signed Message :- Signature Packet, OpenPGP Message | One-Pass Signed Message. In addition, decrypting a Symmetrically Encrypted Data packet or a Symmetrically Encrypted Integrity Protected Data packet as well as decompressing a Compressed Data packet must yield a valid OpenPGP Message. 11.4. Detached Signatures Some OpenPGP applications use so-called "detached signatures". For example, a program bundle may contain a file, and with it a second file that is a detached signature of the first file. These detached signatures are simply a Signature packet stored separately from the data for which they are a signature. 12. Enhanced Key Formats 12.1. Key Structures The format of an OpenPGP V3 key is as follows. Entries in square brackets are optional and ellipses indicate repetition. RSA Public Key [Revocation Self Signature] User ID [Signature ...] [User ID [Signature ...] ...] Each signature certifies the RSA public key and the preceding User ID. The RSA public key can have many User IDs and each User ID can have many signatures. V3 keys are deprecated. Implementations MUST NOT generate new V3 keys, but MAY continue to use existing ones. The format of an OpenPGP V4 key that uses multiple public keys is similar except that the other keys are added to the end as "subkeys" of the primary key. Primary-Key [Revocation Self Signature] [Direct Key Signature...] User ID [Signature ...] [User ID [Signature ...] ...] [User Attribute [Signature ...] ...] [[Subkey [Binding-Signature-Revocation] Primary-Key-Binding-Signature] ...] A subkey always has a single signature after it that is issued using the primary key to tie the two keys together. This binding signature may be in either V3 or V4 format, but SHOULD be V4. Subkeys that can issue signatures MUST have a V4 binding signature due to the REQUIRED embedded primary key binding signature. In the above diagram, if the binding signature of a subkey has been revoked, the revoked key may be removed, leaving only one key. In a V4 key, the primary key MUST be a key capable of certification. The subkeys may be keys of any other type. There may be other constructions of V4 keys, too. For example, there may be a single- key RSA key in V4 format, a DSA primary key with an RSA encryption key, or RSA primary key with an Elgamal subkey, etc. It is also possible to have a signature-only subkey. This permits a primary key that collects certifications (key signatures), but is used only for certifying subkeys that are used for encryption and signatures. 12.2. Key IDs and Fingerprints For a V3 key, the eight-octet Key ID consists of the low 64 bits of the public modulus of the RSA key. The fingerprint of a V3 key is formed by hashing the body (but not the two-octet length) of the MPIs that form the key material (public modulus n, followed by exponent e) with MD5. Note that both V3 keys and MD5 are deprecated. A V4 fingerprint is the 160-bit SHA-1 hash of the octet 0x99, followed by the two-octet packet length, followed by the entire Public-Key packet starting with the version field. The Key ID is the low-order 64 bits of the fingerprint. Here are the fields of the hash material, with the example of a DSA key: a.1) 0x99 (1 octet) a.2) high-order length octet of (b)-(e) (1 octet) a.3) low-order length octet of (b)-(e) (1 octet) b) version number = 4 (1 octet); c) timestamp of key creation (4 octets); d) algorithm (1 octet): 17 = DSA (example); e) Algorithm-specific fields. Algorithm-Specific Fields for DSA keys (example): e.1) MPI of DSA prime p; e.2) MPI of DSA group order q (q is a prime divisor of p-1); e.3) MPI of DSA group generator g; e.4) MPI of DSA public-key value y (= g**x mod p where x is secret). Note that it is possible for there to be collisions of Key IDs -- two different keys with the same Key ID. Note that there is a much smaller, but still non-zero, probability that two different keys have the same fingerprint. Also note that if V3 and V4 format keys share the same RSA key material, they will have different Key IDs as well as different fingerprints. Finally, the Key ID and fingerprint of a subkey are calculated in the same way as for a primary key, including the 0x99 as the first octet (even though this is not a valid packet ID for a public subkey). 13. Notes on Algorithms 13.1. PKCS#1 Encoding in OpenPGP This standard makes use of the PKCS#1 functions EME-PKCS1-v1_5 and EMSA-PKCS1-v1_5. However, the calling conventions of these functions has changed in the past. To avoid potential confusion and interoperability problems, we are including local copies in this document, adapted from those in PKCS#1 v2.1 [RFC3447]. RFC 3447 should be treated as the ultimate authority on PKCS#1 for OpenPGP. Nonetheless, we believe that there is value in having a self- contained document that avoids problems in the future with needed changes in the conventions. 13.1.1. EME-PKCS1-v1_5-ENCODE Input: k = the length in octets of the key modulus M = message to be encoded, an octet string of length mLen, where mLen <= k - 11 Output: EM = encoded message, an octet string of length k Error: "message too long" 1. Length checking: If mLen > k - 11, output "message too long" and stop. 2. Generate an octet string PS of length k - mLen - 3 consisting of pseudo-randomly generated nonzero octets. The length of PS will be at least eight octets. 3. Concatenate PS, the message M, and other padding to form an encoded message EM of length k octets as EM = 0x00 || 0x02 || PS || 0x00 || M. 4. Output EM. 13.1.2. EME-PKCS1-v1_5-DECODE Input: EM = encoded message, an octet string Output: M = message, an octet string Error: "decryption error" To decode an EME-PKCS1_v1_5 message, separate the encoded message EM into an octet string PS consisting of nonzero octets and a message M as follows EM = 0x00 || 0x02 || PS || 0x00 || M. If the first octet of EM does not have hexadecimal value 0x00, if the second octet of EM does not have hexadecimal value 0x02, if there is no octet with hexadecimal value 0x00 to separate PS from M, or if the length of PS is less than 8 octets, output "decryption error" and stop. See also the security note in Section 14 regarding differences in reporting between a decryption error and a padding error. 13.1.3. EMSA-PKCS1-v1_5 This encoding method is deterministic and only has an encoding operation. Option: Hash - a hash function in which hLen denotes the length in octets of the hash function output Input: M = message to be encoded mL = intended length in octets of the encoded message, at least tLen + 11, where tLen is the octet length of the DER encoding T of a certain value computed during the encoding operation Output: EM = encoded message, an octet string of length emLen Errors: "message too long"; "intended encoded message length too short" Steps: 1. Apply the hash function to the message M to produce a hash value H: H = Hash(M). If the hash function outputs "message too long," output "message too long" and stop. 2. Using the list in Section 5.2.2, produce an ASN.1 DER value for the hash function used. Let T be the full hash prefix from Section 5.2.2, and let tLen be the length in octets of T. 3. If emLen < tLen + 11, output "intended encoded message length too short" and stop. 4. Generate an octet string PS consisting of emLen - tLen - 3 octets with hexadecimal value 0xFF. The length of PS will be at least 8 octets. 5. Concatenate PS, the hash prefix T, and other padding to form the encoded message EM as EM = 0x00 || 0x01 || PS || 0x00 || T. 6. Output EM. 13.2. Symmetric Algorithm Preferences The symmetric algorithm preference is an ordered list of algorithms that the keyholder accepts. Since it is found on a self-signature, it is possible that a keyholder may have multiple, different preferences. For example, Alice may have TripleDES only specified for "alice@work.com" but CAST5, Blowfish, and TripleDES specified for "alice@home.org". Note that it is also possible for preferences to be in a subkey's binding signature. Since TripleDES is the MUST-implement algorithm, if it is not explicitly in the list, it is tacitly at the end. However, it is good form to place it there explicitly. Note also that if an implementation does not implement the preference, then it is implicitly a TripleDES-only implementation. An implementation MUST NOT use a symmetric algorithm that is not in the recipient's preference list. When encrypting to more than one recipient, the implementation finds a suitable algorithm by taking the intersection of the preferences of the recipients. Note that the MUST-implement algorithm, TripleDES, ensures that the intersection is not null. The implementation may use any mechanism to pick an algorithm in the intersection. If an implementation can decrypt a message that a keyholder doesn't have in their preferences, the implementation SHOULD decrypt the message anyway, but MUST warn the keyholder that the protocol has been violated. For example, suppose that Alice, above, has software that implements all algorithms in this specification. Nonetheless, she prefers subsets for work or home. If she is sent a message encrypted with IDEA, which is not in her preferences, the software warns her that someone sent her an IDEA-encrypted message, but it would ideally decrypt it anyway. 13.3. Other Algorithm Preferences Other algorithm preferences work similarly to the symmetric algorithm preference, in that they specify which algorithms the keyholder accepts. There are two interesting cases that other comments need to be made about, though, the compression preferences and the hash preferences. 13.3.1. Compression Preferences Compression has been an integral part of PGP since its first days. OpenPGP and all previous versions of PGP have offered compression. In this specification, the default is for messages to be compressed, although an implementation is not required to do so. Consequently, the compression preference gives a way for a keyholder to request that messages not be compressed, presumably because they are using a minimal implementation that does not include compression. Additionally, this gives a keyholder a way to state that it can support alternate algorithms. Like the algorithm preferences, an implementation MUST NOT use an algorithm that is not in the preference vector. If the preferences are not present, then they are assumed to be [ZIP(1), Uncompressed(0)]. Additionally, an implementation MUST implement this preference to the degree of recognizing when to send an uncompressed message. A robust implementation would satisfy this requirement by looking at the recipient's preference and acting accordingly. A minimal implementation can satisfy this requirement by never generating a compressed message, since all implementations can handle messages that have not been compressed. 13.3.2. Hash Algorithm Preferences Typically, the choice of a hash algorithm is something the signer does, rather than the verifier, because a signer rarely knows who is going to be verifying the signature. This preference, though, allows a protocol based upon digital signatures ease in negotiation. Thus, if Alice is authenticating herself to Bob with a signature, it makes sense for her to use a hash algorithm that Bob's software uses. This preference allows Bob to state in his key which algorithms Alice may use. Since SHA1 is the MUST-implement hash algorithm, if it is not explicitly in the list, it is tacitly at the end. However, it is good form to place it there explicitly. 13.4. Plaintext Algorithm 0, "plaintext", may only be used to denote secret keys that are stored in the clear. Implementations MUST NOT use plaintext in Symmetrically Encrypted Data packets; they must use Literal Data packets to encode unencrypted or literal data. 13.5. RSA There are algorithm types for RSA Sign-Only, and RSA Encrypt-Only keys. These types are deprecated. The "key flags" subpacket in a signature is a much better way to express the same idea, and generalizes it to all algorithms. An implementation SHOULD NOT create such a key, but MAY interpret it. An implementation SHOULD NOT implement RSA keys of size less than 1024 bits. 13.6. DSA An implementation SHOULD NOT implement DSA keys of size less than 1024 bits. It MUST NOT implement a DSA key with a q size of less than 160 bits. DSA keys MUST also be a multiple of 64 bits, and the q size MUST be a multiple of 8 bits. The Digital Signature Standard (DSS) [FIPS186] specifies that DSA be used in one of the following ways: * 1024-bit key, 160-bit q, SHA-1, SHA-224, SHA-256, SHA-384, or SHA-512 hash * 2048-bit key, 224-bit q, SHA-224, SHA-256, SHA-384, or SHA-512 hash * 2048-bit key, 256-bit q, SHA-256, SHA-384, or SHA-512 hash * 3072-bit key, 256-bit q, SHA-256, SHA-384, or SHA-512 hash The above key and q size pairs were chosen to best balance the strength of the key with the strength of the hash. Implementations SHOULD use one of the above key and q size pairs when generating DSA keys. If DSS compliance is desired, one of the specified SHA hashes must be used as well. [FIPS186] is the ultimate authority on DSS, and should be consulted for all questions of DSS compliance. Note that earlier versions of this standard only allowed a 160-bit q with no truncation allowed, so earlier implementations may not be able to handle signatures with a different q size or a truncated hash. 13.7. Elgamal An implementation SHOULD NOT implement Elgamal keys of size less than 1024 bits. 13.8. Reserved Algorithm Numbers A number of algorithm IDs have been reserved for algorithms that would be useful to use in an OpenPGP implementation, yet there are issues that prevent an implementer from actually implementing the algorithm. These are marked in Section 9.1, "Public-Key Algorithms", as "reserved for". The reserved public-key algorithms, Elliptic Curve (18), ECDSA (19), and X9.42 (21), do not have the necessary parameters, parameter order, or semantics defined. Previous versions of OpenPGP permitted Elgamal [ELGAMAL] signatures with a public-key identifier of 20. These are no longer permitted. An implementation MUST NOT generate such keys. An implementation MUST NOT generate Elgamal signatures. See [BLEICHENBACHER]. 13.9. OpenPGP CFB Mode OpenPGP does symmetric encryption using a variant of Cipher Feedback mode (CFB mode). This section describes the procedure it uses in detail. This mode is what is used for Symmetrically Encrypted Data Packets; the mechanism used for encrypting secret-key material is similar, and is described in the sections above. In the description below, the value BS is the block size in octets of the cipher. Most ciphers have a block size of 8 octets. The AES and Twofish have a block size of 16 octets. Also note that the description below assumes that the IV and CFB arrays start with an index of 1 (unlike the C language, which assumes arrays start with a zero index). OpenPGP CFB mode uses an initialization vector (IV) of all zeros, and prefixes the plaintext with BS+2 octets of random data, such that octets BS+1 and BS+2 match octets BS-1 and BS. It does a CFB resynchronization after encrypting those BS+2 octets. Thus, for an algorithm that has a block size of 8 octets (64 bits), the IV is 10 octets long and octets 7 and 8 of the IV are the same as octets 9 and 10. For an algorithm with a block size of 16 octets (128 bits), the IV is 18 octets long, and octets 17 and 18 replicate octets 15 and 16. Those extra two octets are an easy check for a correct key. Step by step, here is the procedure: 1. The feedback register (FR) is set to the IV, which is all zeros. 2. FR is encrypted to produce FRE (FR Encrypted). This is the encryption of an all-zero value. 3. FRE is xored with the first BS octets of random data prefixed to the plaintext to produce C[1] through C[BS], the first BS octets of ciphertext. 4. FR is loaded with C[1] through C[BS]. 5. FR is encrypted to produce FRE, the encryption of the first BS octets of ciphertext. 6. The left two octets of FRE get xored with the next two octets of data that were prefixed to the plaintext. This produces C[BS+1] and C[BS+2], the next two octets of ciphertext. 7. (The resynchronization step) FR is loaded with C[3] through C[BS+2]. 8. FR is encrypted to produce FRE. 9. FRE is xored with the first BS octets of the given plaintext, now that we have finished encrypting the BS+2 octets of prefixed data. This produces C[BS+3] through C[BS+(BS+2)], the next BS octets of ciphertext. 10. FR is loaded with C[BS+3] to C[BS + (BS+2)] (which is C11-C18 for an 8-octet block). 11. FR is encrypted to produce FRE. 12. FRE is xored with the next BS octets of plaintext, to produce the next BS octets of ciphertext. These are loaded into FR, and the process is repeated until the plaintext is used up. 13.10. Private or Experimental Parameters S2K specifiers, Signature subpacket types, user attribute types, image format types, and algorithms described in Section 9 all reserve the range 100 to 110 for private and experimental use. Packet types reserve the range 60 to 63 for private and experimental use. These are intentionally managed with the PRIVATE USE method, as described in [RFC2434]. However, implementations need to be careful with these and promote them to full IANA-managed parameters when they grow beyond the original, limited system. 13.11. Extension of the MDC System As described in the non-normative explanation in Section 5.13, the MDC system is uniquely unparameterized in OpenPGP. This was an intentional decision to avoid cross-grade attacks. If the MDC system is extended to a stronger hash function, care must be taken to avoid downgrade and cross-grade attacks. One simple way to do this is to create new packets for a new MDC. For example, instead of the MDC system using packets 18 and 19, a new MDC could use 20 and 21. This has obvious drawbacks (it uses two packet numbers for each new hash function in a space that is limited to a maximum of 60). Another simple way to extend the MDC system is to create new versions of packet 18, and reflect this in packet 19. For example, suppose that V2 of packet 18 implicitly used SHA-256. This would require packet 19 to have a length of 32 octets. The change in the version in packet 18 and the size of packet 19 prevent a downgrade attack. There are two drawbacks to this latter approach. The first is that using the version number of a packet to carry algorithm information is not tidy from a protocol-design standpoint. It is possible that there might be several versions of the MDC system in common use, but this untidiness would reflect untidiness in cryptographic consensus about hash function security. The second is that different versions of packet 19 would have to have unique sizes. If there were two versions each with 256-bit hashes, they could not both have 32-octet packet 19s without admitting the chance of a cross-grade attack. Yet another, complex approach to extend the MDC system would be a hybrid of the two above -- create a new pair of MDC packets that are fully parameterized, and yet protected from downgrade and cross- grade. Any change to the MDC system MUST be done through the IETF CONSENSUS method, as described in [RFC2434]. 13.12. Meta-Considerations for Expansion If OpenPGP is extended in a way that is not backwards-compatible, meaning that old implementations will not gracefully handle their absence of a new feature, the extension proposal can be declared in the key holder's self-signature as part of the Features signature subpacket. We cannot state definitively what extensions will not be upwards- compatible, but typically new algorithms are upwards-compatible, whereas new packets are not. If an extension proposal does not update the Features system, it SHOULD include an explanation of why this is unnecessary. If the proposal contains neither an extension to the Features system nor an explanation of why such an extension is unnecessary, the proposal SHOULD be rejected. 14. Security Considerations * As with any technology involving cryptography, you should check the current literature to determine if any algorithms used here have been found to be vulnerable to attack. * This specification uses Public-Key Cryptography technologies. It is assumed that the private key portion of a public-private key pair is controlled and secured by the proper party or parties. * Certain operations in this specification involve the use of random numbers. An appropriate entropy source should be used to generate these numbers (see [RFC4086]). * The MD5 hash algorithm has been found to have weaknesses, with collisions found in a number of cases. MD5 is deprecated for use in OpenPGP. Implementations MUST NOT generate new signatures using MD5 as a hash function. They MAY continue to consider old signatures that used MD5 as valid. * SHA-224 and SHA-384 require the same work as SHA-256 and SHA-512, respectively. In general, there are few reasons to use them outside of DSS compatibility. You need a situation where one needs more security than smaller hashes, but does not want to have the full 256-bit or 512-bit data length. * Many security protocol designers think that it is a bad idea to use a single key for both privacy (encryption) and integrity (signatures). In fact, this was one of the motivating forces behind the V4 key format with separate signature and encryption keys. If you as an implementer promote dual-use keys, you should at least be aware of this controversy. * The DSA algorithm will work with any hash, but is sensitive to the quality of the hash algorithm. Verifiers should be aware that even if the signer used a strong hash, an attacker could have modified the signature to use a weak one. Only signatures using acceptably strong hash algorithms should be accepted as valid. * As OpenPGP combines many different asymmetric, symmetric, and hash algorithms, each with different measures of strength, care should be taken that the weakest element of an OpenPGP message is still sufficiently strong for the purpose at hand. While consensus about the strength of a given algorithm may evolve, NIST Special Publication 800-57 [SP800-57] recommends the following list of equivalent strengths: Asymmetric | Hash | Symmetric key size | size | key size ------------+--------+----------- 1024 160 80 2048 224 112 3072 256 128 7680 384 192 15360 512 256 * There is a somewhat-related potential security problem in signatures. If an attacker can find a message that hashes to the same hash with a different algorithm, a bogus signature structure can be constructed that evaluates correctly. For example, suppose Alice DSA signs message M using hash algorithm H. Suppose that Mallet finds a message M' that has the same hash value as M with H'. Mallet can then construct a signature block that verifies as Alice's signature of M' with H'. However, this would also constitute a weakness in either H or H' or both. Should this ever occur, a revision will have to be made to this document to revise the allowed hash algorithms. * If you are building an authentication system, the recipient may specify a preferred signing algorithm. However, the signer would be foolish to use a weak algorithm simply because the recipient requests it. * Some of the encryption algorithms mentioned in this document have been analyzed less than others. For example, although CAST5 is presently considered strong, it has been analyzed less than TripleDES. Other algorithms may have other controversies surrounding them. * In late summer 2002, Jallad, Katz, and Schneier published an interesting attack on the OpenPGP protocol and some of its implementations [JKS02]. In this attack, the attacker modifies a message and sends it to a user who then returns the erroneously decrypted message to the attacker. The attacker is thus using the user as a random oracle, and can often decrypt the message. Compressing data can ameliorate this attack. The incorrectly decrypted data nearly always decompresses in ways that defeat the attack. However, this is not a rigorous fix, and leaves open some small vulnerabilities. For example, if an implementation does not compress a message before encryption (perhaps because it knows it was already compressed), then that message is vulnerable. Because of this happenstance -- that modification attacks can be thwarted by decompression errors -- an implementation SHOULD treat a decompression error as a security problem, not merely a data problem. This attack can be defeated by the use of Modification Detection, provided that the implementation does not let the user naively return the data to the attacker. An implementation MUST treat an MDC failure as a security problem, not merely a data problem. In either case, the implementation MAY allow the user access to the erroneous data, but MUST warn the user as to potential security problems should that data be returned to the sender. While this attack is somewhat obscure, requiring a special set of circumstances to create it, it is nonetheless quite serious as it permits someone to trick a user to decrypt a message. Consequently, it is important that: 1. Implementers treat MDC errors and decompression failures as security problems. 2. Implementers implement Modification Detection with all due speed and encourage its spread. 3. Users migrate to implementations that support Modification Detection with all due speed. * PKCS#1 has been found to be vulnerable to attacks in which a system that reports errors in padding differently from errors in decryption becomes a random oracle that can leak the private key in mere millions of queries. Implementations must be aware of this attack and prevent it from happening. The simplest solution is to report a single error code for all variants of decryption errors so as not to leak information to an attacker. * Some technologies mentioned here may be subject to government control in some countries. * In winter 2005, Serge Mister and Robert Zuccherato from Entrust released a paper describing a way that the "quick check" in OpenPGP CFB mode can be used with a random oracle to decrypt two octets of every cipher block [MZ05]. They recommend as prevention not using the quick check at all. Many implementers have taken this advice to heart for any data that is symmetrically encrypted and for which the session key is public-key encrypted. In this case, the quick check is not needed as the public-key encryption of the session key should guarantee that it is the right session key. In other cases, the implementation should use the quick check with care. On the one hand, there is a danger to using it if there is a random oracle that can leak information to an attacker. In plainer language, there is a danger to using the quick check if timing information about the check can be exposed to an attacker, particularly via an automated service that allows rapidly repeated queries. On the other hand, it is inconvenient to the user to be informed that they typed in the wrong passphrase only after a petabyte of data is decrypted. There are many cases in cryptographic engineering where the implementer must use care and wisdom, and this is one. 15. Implementation Nits This section is a collection of comments to help an implementer, particularly with an eye to backward compatibility. Previous implementations of PGP are not OpenPGP compliant. Often the differences are small, but small differences are frequently more vexing than large differences. Thus, this is a non-comprehensive list of potential problems and gotchas for a developer who is trying to be backward-compatible. * The IDEA algorithm is patented, and yet it is required for PGP 2.x interoperability. It is also the de-facto preferred algorithm for a V3 key with a V3 self-signature (or no self- signature). * When exporting a private key, PGP 2.x generates the header "BEGIN PGP SECRET KEY BLOCK" instead of "BEGIN PGP PRIVATE KEY BLOCK". All previous versions ignore the implied data type, and look directly at the packet data type. DONE * PGP 2.0 through 2.5 generated V2 Public-Key packets. These are identical to the deprecated V3 keys except for the version number. An implementation MUST NOT generate them and may accept or reject them as it sees fit. Some older PGP versions generated V2 PKESK packets (Tag 1) as well. An implementation may accept or reject V2 PKESK packets as it sees fit, and MUST NOT generate them. * PGP 2.6.x will not accept key-material packets with versions greater than 3. * There are many ways possible for two keys to have the same key material, but different fingerprints (and thus Key IDs). Perhaps the most interesting is an RSA key that has been "upgraded" to V4 format, but since a V4 fingerprint is constructed by hashing the key creation time along with other things, two V4 keys created at different times, yet with the same key material will have different fingerprints. * If an implementation is using zlib to interoperate with PGP 2.x, then the "windowBits" parameter should be set to -13. * The 0x19 back signatures were not required for signing subkeys until relatively recently. Consequently, there may be keys in the wild that do not have these back signatures. Implementing software may handle these keys as it sees fit. * OpenPGP does not put limits on the size of public keys. However, larger keys are not necessarily better keys. Larger keys take more computation time to use, and this can quickly become impractical. Different OpenPGP implementations may also use different upper bounds for public key sizes, and so care should be taken when choosing sizes to maintain interoperability. As of 2007 most implementations have an upper bound of 4096 bits. * ASCII armor is an optional feature of OpenPGP. The OpenPGP working group strives for a minimal set of mandatory-to-implement features, and since there could be useful implementations that only use binary object formats, this is not a "MUST" feature for an implementation. For example, an implementation that is using OpenPGP as a mechanism for file signatures may find ASCII armor unnecessary. OpenPGP permits an implementation to declare what features it does and does not support, but ASCII armor is not one of these. Since most implementations allow binary and armored objects to be used indiscriminately, an implementation that does not implement ASCII armor may find itself with compatibility issues with general-purpose implementations. Moreover, implementations of OpenPGP-MIME [RFC3156] already have a requirement for ASCII armor so those implementations will necessarily have support. [...]