path: root/Documentation/crypto/architecture.rst

Kernel Crypto API Architecture
==============================

Cipher algorithm types
----------------------

The kernel crypto API provides different API calls for the following
cipher types:

-  Symmetric ciphers

-  AEAD ciphers

-  Message digest, including keyed message digest

-  Random number generation

-  User space interface

Ciphers And Templates
---------------------

The kernel crypto API provides implementations of single block ciphers
and message digests. In addition, the kernel crypto API provides
numerous "templates" that can be used in conjunction with the single
block ciphers and message digests. Templates include all types of block
chaining mode, the HMAC mechanism, etc.

Single block ciphers and message digests can either be directly used by
a caller or invoked together with a template to form multi-block ciphers
or keyed message digests.

A single block cipher may even be called with multiple templates.
However, templates cannot be used without a single cipher.

See /proc/crypto and search for "name". For example:

-  aes

-  ecb(aes)

-  cmac(aes)

-  ccm(aes)

-  rfc4106(gcm(aes))

-  sha1

-  hmac(sha1)

-  authenc(hmac(sha1),cbc(aes))

In these examples, "aes" and "sha1" are the ciphers and all others are
the templates.

Synchronous And Asynchronous Operation
--------------------------------------

The kernel crypto API provides synchronous and asynchronous API
operations.

When using the synchronous API operation, the caller invokes a cipher
operation which is performed synchronously by the kernel crypto API.
That means, the caller waits until the cipher operation completes.
Therefore, the kernel crypto API calls work like regular function calls.
For synchronous operation, the set of API calls is small and
conceptually similar to any other crypto library.

Asynchronous operation is provided by the kernel crypto API which
implies that the invocation of a cipher operation will complete almost
instantly. That invocation triggers the cipher operation but it does not
signal its completion. Before invoking a cipher operation, the caller
must provide a callback function the kernel crypto API can invoke to
signal the completion of the cipher operation. Furthermore, the caller
must ensure it can handle such asynchronous events by applying
appropriate locking around its data. The kernel crypto API does not
perform any special serialization operation to protect the caller's data
integrity.

Crypto API Cipher References And Priority
-----------------------------------------

A cipher is referenced by the caller with a string. That string has the
following semantics:

::

        template(single block cipher)


where "template" and "single block cipher" is the aforementioned
template and single block cipher, respectively. If applicable,
additional templates may enclose other templates, such as

::

        template1(template2(single block cipher)))


The kernel crypto API may provide multiple implementations of a template
or a single block cipher. For example, AES on newer Intel hardware has
the following implementations: AES-NI, assembler implementation, or
straight C. Now, when using the string "aes" with the kernel crypto API,
which cipher implementation is used? The answer to that question is the
priority number assigned to each cipher implementation by the kernel
crypto API. When a caller uses the string to refer to a cipher during
initialization of a cipher handle, the kernel crypto API looks up all
implementations providing an implementation with that name and selects
the implementation with the highest priority.

Now, a caller may have the need to refer to a specific cipher
implementation and thus does not want to rely on the priority-based
selection. To accommodate this scenario, the kernel crypto API allows
the cipher implementation to register a unique name in addition to
common names. When using that unique name, a caller is therefore always
sure to refer to the intended cipher implementation.

The list of available ciphers is given in /proc/crypto. However, that
list does not specify all possible permutations of templates and
ciphers. Each block listed in /proc/crypto may contain the following
information -- if one of the components listed as follows are not
applicable to a cipher, it is not displayed:

-  name: the generic name of the cipher that is subject to the
   priority-based selection -- this name can be used by the cipher
   allocation API calls (all names listed above are examples for such
   generic names)

-  driver: the unique name of the cipher -- this name can be used by the
   cipher allocation API calls

-  module: the kernel module providing the cipher implementation (or
   "kernel" for statically linked ciphers)

-  priority: the priority value of the cipher implementation

-  refcnt: the reference count of the respective cipher (i.e. the number
   of current consumers of this cipher)

-  selftest: specification whether the self test for the cipher passed

-  type:

   -  skcipher for symmetric key ciphers

   -  cipher for single block ciphers that may be used with an
      additional template

   -  shash for synchronous message digest

   -  ahash for asynchronous message digest

   -  aead for AEAD cipher type

   -  compression for compression type transformations

   -  rng for random number generator

   -  givcipher for cipher with associated IV generator (see the geniv
      entry below for the specification of the IV generator type used by
      the cipher implementation)

-  blocksize: blocksize of cipher in bytes

-  keysize: key size in bytes

-  ivsize: IV size in bytes

-  seedsize: required size of seed data for random number generator

-  digestsize: output size of the message digest

-  geniv: IV generation type:

   -  eseqiv for encrypted sequence number based IV generation

   -  seqiv for sequence number based IV generation

   -  chainiv for chain iv generation

   -  <builtin> is a marker that the cipher implements IV generation and
      handling as it is specific to the given cipher

Key Sizes
---------

When allocating a cipher handle, the caller only specifies the cipher
type. Symmetric ciphers, however, typically support multiple key sizes
(e.g. AES-128 vs. AES-192 vs. AES-256). These key sizes are determined
with the length of the provided key. Thus, the kernel crypto API does
not provide a separate way to select the particular symmetric cipher key
size.

Cipher Allocation Type And Masks
--------------------------------

The different cipher handle allocation functions allow the specification
of a type and mask flag. Both parameters have the following meaning (and
are therefore not covered in the subsequent sections).

The type flag specifies the type of the cipher algorithm. The caller
usually provides a 0 when the caller wants the default handling.
Otherwise, the caller may provide the following selections which match
the aforementioned cipher types:

-  CRYPTO_ALG_TYPE_CIPHER Single block cipher

-  CRYPTO_ALG_TYPE_COMPRESS Compression

-  CRYPTO_ALG_TYPE_AEAD Authenticated Encryption with Associated Data
   (MAC)

-  CRYPTO_ALG_TYPE_BLKCIPHER Synchronous multi-block cipher

-  CRYPTO_ALG_TYPE_ABLKCIPHER Asynchronous multi-block cipher

-  CRYPTO_ALG_TYPE_GIVCIPHER Asynchronous multi-block cipher packed
   together with an IV generator (see geniv field in the /proc/crypto
   listing for the known IV generators)

-  CRYPTO_ALG_TYPE_DIGEST Raw message digest

-  CRYPTO_ALG_TYPE_HASH Alias for CRYPTO_ALG_TYPE_DIGEST

-  CRYPTO_ALG_TYPE_SHASH Synchronous multi-block hash

-  CRYPTO_ALG_TYPE_AHASH Asynchronous multi-block hash

-  CRYPTO_ALG_TYPE_RNG Random Number Generation

-  CRYPTO_ALG_TYPE_AKCIPHER Asymmetric cipher

-  CRYPTO_ALG_TYPE_PCOMPRESS Enhanced version of
   CRYPTO_ALG_TYPE_COMPRESS allowing for segmented compression /
   decompression instead of performing the operation on one segment
   only. CRYPTO_ALG_TYPE_PCOMPRESS is intended to replace
   CRYPTO_ALG_TYPE_COMPRESS once existing consumers are converted.

The mask flag restricts the type of cipher. The only allowed flag is
CRYPTO_ALG_ASYNC to restrict the cipher lookup function to
asynchronous ciphers. Usually, a caller provides a 0 for the mask flag.

When the caller provides a mask and type specification, the caller
limits the search the kernel crypto API can perform for a suitable
cipher implementation for the given cipher name. That means, even when a
caller uses a cipher name that exists during its initialization call,
the kernel crypto API may not select it due to the used type and mask
field.

Internal Structure of Kernel Crypto API
---------------------------------------

The kernel crypto API has an internal structure where a cipher
implementation may use many layers and indirections. This section shall
help to clarify how the kernel crypto API uses various components to
implement the complete cipher.

The following subsections explain the internal structure based on
existing cipher implementations. The first section addresses the most
complex scenario where all other scenarios form a logical subset.

Generic AEAD Cipher Structure
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

The following ASCII art decomposes the kernel crypto API layers when
using the AEAD cipher with the automated IV generation. The shown
example is used by the IPSEC layer.

For other use cases of AEAD ciphers, the ASCII art applies as well, but
the caller may not use the AEAD cipher with a separate IV generator. In
this case, the caller must generate the IV.

The depicted example decomposes the AEAD cipher of GCM(AES) based on the
generic C implementations (gcm.c, aes-generic.c, ctr.c, ghash-generic.c,
seqiv.c). The generic implementation serves as an example showing the
complete logic of the kernel crypto API.

It is possible that some streamlined cipher implementations (like
AES-NI) provide implementations merging aspects which in the view of the
kernel crypto API cannot be decomposed into layers any more. In case of
the AES-NI implementation, the CTR mode, the GHASH implementation and
the AES cipher are all merged into one cipher implementation registered
with the kernel crypto API. In this case, the concept described by the
following ASCII art applies too. However, the decomposition of GCM into
the individual sub-components by the kernel crypto API is not done any
more.

Each block in the following ASCII art is an independent cipher instance
obtained from the kernel crypto API. Each block is accessed by the
caller or by other blocks using the API functions defined by the kernel
crypto API for the cipher implementation type.

The blocks below indicate the cipher type as well as the specific logic
implemented in the cipher.

The ASCII art picture also indicates the call structure, i.e. who calls
which component. The arrows point to the invoked block where the caller
uses the API applicable to the cipher type specified for the block.

::


    kernel crypto API                                |   IPSEC Layer
                                                     |
    +-----------+                                    |
    |           |            (1)
    |   aead    | <-----------------------------------  esp_output
    |  (seqiv)  | ---+
    +-----------+    |
                     | (2)
    +-----------+    |
    |           | <--+                (2)
    |   aead    | <-----------------------------------  esp_input
    |   (gcm)   | ------------+
    +-----------+             |
          | (3)               | (5)
          v                   v
    +-----------+       +-----------+
    |           |       |           |
    |  skcipher |       |   ahash   |
    |   (ctr)   | ---+  |  (ghash)  |
    +-----------+    |  +-----------+
                     |
    +-----------+    | (4)
    |           | <--+
    |   cipher  |
    |   (aes)   |
    +-----------+



The following call sequence is applicable when the IPSEC layer triggers
an encryption operation with the esp_output function. During
configuration, the administrator set up the use of rfc4106(gcm(aes)) as
the cipher for ESP. The following call sequence is now depicted in the
ASCII art above:

1. esp_output() invokes crypto_aead_encrypt() to trigger an
   encryption operation of the AEAD cipher with IV generator.

   In case of GCM, the SEQIV implementation is registered as GIVCIPHER
   in crypto_rfc4106_alloc().

   The SEQIV performs its operation to generate an IV where the core
   function is seqiv_geniv().

2. Now, SEQIV uses the AEAD API function calls to invoke the associated
   AEAD cipher. In our case, during the instantiation of SEQIV, the
   cipher handle for GCM is provided to SEQIV. This means that SEQIV
   invokes AEAD cipher operations with the GCM cipher handle.

   During instantiation of the GCM handle, the CTR(AES) and GHASH
   ciphers are instantiated. The cipher handles for CTR(AES) and GHASH
   are retained for later use.

   The GCM implementation is responsible to invoke the CTR mode AES and
   the GHASH cipher in the right manner to implement the GCM
   specification.

3. The GCM AEAD cipher type implementation now invokes the SKCIPHER API
   with the instantiated CTR(AES) cipher handle.

   During instantiation of the CTR(AES) cipher, the CIPHER type
   implementation of AES is instantiated. The cipher handle for AES is
   retained.

   That means that the SKCIPHER implementation of CTR(AES) only
   implements the CTR block chaining mode. After performing the block
   chaining operation, the CIPHER implementation of AES is invoked.

4. The SKCIPHER of CTR(AES) now invokes the CIPHER API with the AES
   cipher handle to encrypt one block.

5. The GCM AEAD implementation also invokes the GHASH cipher
   implementation via the AHASH API.

When the IPSEC layer triggers the esp_input() function, the same call
sequence is followed with the only difference that the operation starts
with step (2).

Generic Block Cipher Structure
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Generic block ciphers follow the same concept as depicted with the ASCII
art picture above.

For example, CBC(AES) is implemented with cbc.c, and aes-generic.c. The
ASCII art picture above applies as well with the difference that only
step (4) is used and the SKCIPHER block chaining mode is CBC.

Generic Keyed Message Digest Structure
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Keyed message digest implementations again follow the same concept as
depicted in the ASCII art picture above.

For example, HMAC(SHA256) is implemented with hmac.c and
sha256_generic.c. The following ASCII art illustrates the
implementation:

::


    kernel crypto API            |       Caller
                                 |
    +-----------+         (1)    |
    |           | <------------------  some_function
    |   ahash   |
    |   (hmac)  | ---+
    +-----------+    |
                     | (2)
    +-----------+    |
    |           | <--+
    |   shash   |
    |  (sha256) |
    +-----------+



The following call sequence is applicable when a caller triggers an HMAC
operation:

1. The AHASH API functions are invoked by the caller. The HMAC
   implementation performs its operation as needed.

   During initialization of the HMAC cipher, the SHASH cipher type of
   SHA256 is instantiated. The cipher handle for the SHA256 instance is
   retained.

   At one time, the HMAC implementation requires a SHA256 operation
   where the SHA256 cipher handle is used.

2. The HMAC instance now invokes the SHASH API with the SHA256 cipher
   handle to calculate the message digest.