Commit 0aaf2146 authored by Linus Torvalds's avatar Linus Torvalds

Merge tag 'docs-4.10-2' of git://git.lwn.net/linux

Pull more documentation updates from Jonathan Corbet:
 "This converts the crypto DocBook to Sphinx"

* tag 'docs-4.10-2' of git://git.lwn.net/linux:
  crypto: doc - optimize compilation
  crypto: doc - clarify AEAD memory structure
  crypto: doc - remove crypto_alloc_ablkcipher
  crypto: doc - add KPP documentation
  crypto: doc - fix separation of cipher / req API
  crypto: doc - fix source comments for Sphinx
  crypto: doc - remove crypto API DocBook
  crypto: doc - convert crypto API documentation to Sphinx
parents 59331c21 3fa71d0f
......@@ -13,7 +13,7 @@ DOCBOOKS := z8530book.xml \
gadget.xml libata.xml mtdnand.xml librs.xml rapidio.xml \
genericirq.xml s390-drivers.xml uio-howto.xml scsi.xml \
80211.xml sh.xml regulator.xml w1.xml \
writing_musb_glue_layer.xml crypto-API.xml iio.xml
writing_musb_glue_layer.xml iio.xml
ifeq ($(DOCBOOKS),)
......
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.1.2//EN"
"http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd" []>
<book id="KernelCryptoAPI">
<bookinfo>
<title>Linux Kernel Crypto API</title>
<authorgroup>
<author>
<firstname>Stephan</firstname>
<surname>Mueller</surname>
<affiliation>
<address>
<email>smueller@chronox.de</email>
</address>
</affiliation>
</author>
<author>
<firstname>Marek</firstname>
<surname>Vasut</surname>
<affiliation>
<address>
<email>marek@denx.de</email>
</address>
</affiliation>
</author>
</authorgroup>
<copyright>
<year>2014</year>
<holder>Stephan Mueller</holder>
</copyright>
<legalnotice>
<para>
This documentation is free software; you can redistribute
it and/or modify it under the terms of the GNU General Public
License as published by the Free Software Foundation; either
version 2 of the License, or (at your option) any later
version.
</para>
<para>
This program is distributed in the hope that it will be
useful, but WITHOUT ANY WARRANTY; without even the implied
warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
See the GNU General Public License for more details.
</para>
<para>
You should have received a copy of the GNU General Public
License along with this program; if not, write to the Free
Software Foundation, Inc., 59 Temple Place, Suite 330, Boston,
MA 02111-1307 USA
</para>
<para>
For more details see the file COPYING in the source
distribution of Linux.
</para>
</legalnotice>
</bookinfo>
<toc></toc>
<chapter id="Intro">
<title>Kernel Crypto API Interface Specification</title>
<sect1><title>Introduction</title>
<para>
The kernel crypto API offers a rich set of cryptographic ciphers as
well as other data transformation mechanisms and methods to invoke
these. This document contains a description of the API and provides
example code.
</para>
<para>
To understand and properly use the kernel crypto API a brief
explanation of its structure is given. Based on the architecture,
the API can be separated into different components. Following the
architecture specification, hints to developers of ciphers are
provided. Pointers to the API function call documentation are
given at the end.
</para>
<para>
The kernel crypto API refers to all algorithms as "transformations".
Therefore, a cipher handle variable usually has the name "tfm".
Besides cryptographic operations, the kernel crypto API also knows
compression transformations and handles them the same way as ciphers.
</para>
<para>
The kernel crypto API serves the following entity types:
<itemizedlist>
<listitem>
<para>consumers requesting cryptographic services</para>
</listitem>
<listitem>
<para>data transformation implementations (typically ciphers)
that can be called by consumers using the kernel crypto
API</para>
</listitem>
</itemizedlist>
</para>
<para>
This specification is intended for consumers of the kernel crypto
API as well as for developers implementing ciphers. This API
specification, however, does not discuss all API calls available
to data transformation implementations (i.e. implementations of
ciphers and other transformations (such as CRC or even compression
algorithms) that can register with the kernel crypto API).
</para>
<para>
Note: The terms "transformation" and cipher algorithm are used
interchangeably.
</para>
</sect1>
<sect1><title>Terminology</title>
<para>
The transformation implementation is an actual code or interface
to hardware which implements a certain transformation with precisely
defined behavior.
</para>
<para>
The transformation object (TFM) is an instance of a transformation
implementation. There can be multiple transformation objects
associated with a single transformation implementation. Each of
those transformation objects is held by a crypto API consumer or
another transformation. Transformation object is allocated when a
crypto API consumer requests a transformation implementation.
The consumer is then provided with a structure, which contains
a transformation object (TFM).
</para>
<para>
The structure that contains transformation objects may also be
referred to as a "cipher handle". Such a cipher handle is always
subject to the following phases that are reflected in the API calls
applicable to such a cipher handle:
</para>
<orderedlist>
<listitem>
<para>Initialization of a cipher handle.</para>
</listitem>
<listitem>
<para>Execution of all intended cipher operations applicable
for the handle where the cipher handle must be furnished to
every API call.</para>
</listitem>
<listitem>
<para>Destruction of a cipher handle.</para>
</listitem>
</orderedlist>
<para>
When using the initialization API calls, a cipher handle is
created and returned to the consumer. Therefore, please refer
to all initialization API calls that refer to the data
structure type a consumer is expected to receive and subsequently
to use. The initialization API calls have all the same naming
conventions of crypto_alloc_*.
</para>
<para>
The transformation context is private data associated with
the transformation object.
</para>
</sect1>
</chapter>
<chapter id="Architecture"><title>Kernel Crypto API Architecture</title>
<sect1><title>Cipher algorithm types</title>
<para>
The kernel crypto API provides different API calls for the
following cipher types:
<itemizedlist>
<listitem><para>Symmetric ciphers</para></listitem>
<listitem><para>AEAD ciphers</para></listitem>
<listitem><para>Message digest, including keyed message digest</para></listitem>
<listitem><para>Random number generation</para></listitem>
<listitem><para>User space interface</para></listitem>
</itemizedlist>
</para>
</sect1>
<sect1><title>Ciphers And Templates</title>
<para>
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.
</para>
<para>
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.
</para>
<para>
A single block cipher may even be called with multiple templates.
However, templates cannot be used without a single cipher.
</para>
<para>
See /proc/crypto and search for "name". For example:
<itemizedlist>
<listitem><para>aes</para></listitem>
<listitem><para>ecb(aes)</para></listitem>
<listitem><para>cmac(aes)</para></listitem>
<listitem><para>ccm(aes)</para></listitem>
<listitem><para>rfc4106(gcm(aes))</para></listitem>
<listitem><para>sha1</para></listitem>
<listitem><para>hmac(sha1)</para></listitem>
<listitem><para>authenc(hmac(sha1),cbc(aes))</para></listitem>
</itemizedlist>
</para>
<para>
In these examples, "aes" and "sha1" are the ciphers and all
others are the templates.
</para>
</sect1>
<sect1><title>Synchronous And Asynchronous Operation</title>
<para>
The kernel crypto API provides synchronous and asynchronous
API operations.
</para>
<para>
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.
</para>
<para>
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.
</para>
</sect1>
<sect1><title>Crypto API Cipher References And Priority</title>
<para>
A cipher is referenced by the caller with a string. That string
has the following semantics:
<programlisting>
template(single block cipher)
</programlisting>
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
<programlisting>
template1(template2(single block cipher)))
</programlisting>
</para>
<para>
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.
</para>
<para>
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.
</para>
<para>
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:
</para>
<itemizedlist>
<listitem>
<para>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)</para>
</listitem>
<listitem>
<para>driver: the unique name of the cipher -- this name can
be used by the cipher allocation API calls</para>
</listitem>
<listitem>
<para>module: the kernel module providing the cipher
implementation (or "kernel" for statically linked ciphers)</para>
</listitem>
<listitem>
<para>priority: the priority value of the cipher implementation</para>
</listitem>
<listitem>
<para>refcnt: the reference count of the respective cipher
(i.e. the number of current consumers of this cipher)</para>
</listitem>
<listitem>
<para>selftest: specification whether the self test for the
cipher passed</para>
</listitem>
<listitem>
<para>type:
<itemizedlist>
<listitem>
<para>skcipher for symmetric key ciphers</para>
</listitem>
<listitem>
<para>cipher for single block ciphers that may be used with
an additional template</para>
</listitem>
<listitem>
<para>shash for synchronous message digest</para>
</listitem>
<listitem>
<para>ahash for asynchronous message digest</para>
</listitem>
<listitem>
<para>aead for AEAD cipher type</para>
</listitem>
<listitem>
<para>compression for compression type transformations</para>
</listitem>
<listitem>
<para>rng for random number generator</para>
</listitem>
<listitem>
<para>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)</para>
</listitem>
</itemizedlist>
</para>
</listitem>
<listitem>
<para>blocksize: blocksize of cipher in bytes</para>
</listitem>
<listitem>
<para>keysize: key size in bytes</para>
</listitem>
<listitem>
<para>ivsize: IV size in bytes</para>
</listitem>
<listitem>
<para>seedsize: required size of seed data for random number
generator</para>
</listitem>
<listitem>
<para>digestsize: output size of the message digest</para>
</listitem>
<listitem>
<para>geniv: IV generation type:
<itemizedlist>
<listitem>
<para>eseqiv for encrypted sequence number based IV
generation</para>
</listitem>
<listitem>
<para>seqiv for sequence number based IV generation</para>
</listitem>
<listitem>
<para>chainiv for chain iv generation</para>
</listitem>
<listitem>
<para>&lt;builtin&gt; is a marker that the cipher implements
IV generation and handling as it is specific to the given
cipher</para>
</listitem>
</itemizedlist>
</para>
</listitem>
</itemizedlist>
</sect1>
<sect1><title>Key Sizes</title>
<para>
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.
</para>
</sect1>
<sect1><title>Cipher Allocation Type And Masks</title>
<para>
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).
</para>
<para>
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:
</para>
<itemizedlist>
<listitem>
<para>CRYPTO_ALG_TYPE_CIPHER Single block cipher</para>
</listitem>
<listitem>
<para>CRYPTO_ALG_TYPE_COMPRESS Compression</para>
</listitem>
<listitem>
<para>CRYPTO_ALG_TYPE_AEAD Authenticated Encryption with
Associated Data (MAC)</para>
</listitem>
<listitem>
<para>CRYPTO_ALG_TYPE_BLKCIPHER Synchronous multi-block cipher</para>
</listitem>
<listitem>
<para>CRYPTO_ALG_TYPE_ABLKCIPHER Asynchronous multi-block cipher</para>
</listitem>
<listitem>
<para>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)</para>
</listitem>
<listitem>
<para>CRYPTO_ALG_TYPE_DIGEST Raw message digest</para>
</listitem>
<listitem>
<para>CRYPTO_ALG_TYPE_HASH Alias for CRYPTO_ALG_TYPE_DIGEST</para>
</listitem>
<listitem>
<para>CRYPTO_ALG_TYPE_SHASH Synchronous multi-block hash</para>
</listitem>
<listitem>
<para>CRYPTO_ALG_TYPE_AHASH Asynchronous multi-block hash</para>
</listitem>
<listitem>
<para>CRYPTO_ALG_TYPE_RNG Random Number Generation</para>
</listitem>
<listitem>
<para>CRYPTO_ALG_TYPE_AKCIPHER Asymmetric cipher</para>
</listitem>
<listitem>
<para>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.</para>
</listitem>
</itemizedlist>
<para>
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.
</para>
<para>
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.
</para>
</sect1>
<sect1><title>Internal Structure of Kernel Crypto API</title>
<para>
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.
</para>
<para>
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.
</para>
<sect2><title>Generic AEAD Cipher Structure</title>
<para>
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.
</para>
<para>
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.
</para>
<para>
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.
</para>
<para>
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.
</para>
<para>
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.
</para>
<para>
The blocks below indicate the cipher type as well as the specific
logic implemented in the cipher.
</para>
<para>
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.
</para>
<programlisting>
<![CDATA[
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) |
+-----------+
]]>
</programlisting>
<para>
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:
</para>
<orderedlist>
<listitem>
<para>
esp_output() invokes crypto_aead_encrypt() to trigger an encryption
operation of the AEAD cipher with IV generator.
</para>
<para>
In case of GCM, the SEQIV implementation is registered as GIVCIPHER
in crypto_rfc4106_alloc().
</para>
<para>
The SEQIV performs its operation to generate an IV where the core
function is seqiv_geniv().
</para>
</listitem>
<listitem>
<para>
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.
</para>
<para>
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.
</para>
<para>
The GCM implementation is responsible to invoke the CTR mode AES and
the GHASH cipher in the right manner to implement the GCM
specification.
</para>
</listitem>
<listitem>
<para>
The GCM AEAD cipher type implementation now invokes the SKCIPHER API
with the instantiated CTR(AES) cipher handle.
</para>
<para>
During instantiation of the CTR(AES) cipher, the CIPHER type
implementation of AES is instantiated. The cipher handle for AES is
retained.
</para>
<para>
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.
</para>
</listitem>
<listitem>
<para>
The SKCIPHER of CTR(AES) now invokes the CIPHER API with the AES
cipher handle to encrypt one block.
</para>
</listitem>
<listitem>
<para>
The GCM AEAD implementation also invokes the GHASH cipher
implementation via the AHASH API.
</para>
</listitem>
</orderedlist>
<para>
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).
</para>
</sect2>
<sect2><title>Generic Block Cipher Structure</title>
<para>
Generic block ciphers follow the same concept as depicted with the ASCII
art picture above.
</para>
<para>
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.
</para>
</sect2>
<sect2><title>Generic Keyed Message Digest Structure</title>
<para>
Keyed message digest implementations again follow the same concept as
depicted in the ASCII art picture above.
</para>
<para>
For example, HMAC(SHA256) is implemented with hmac.c and
sha256_generic.c. The following ASCII art illustrates the
implementation:
</para>
<programlisting>
<![CDATA[
kernel crypto API | Caller
|
+-----------+ (1) |
| | <------------------ some_function
| ahash |
| (hmac) | ---+
+-----------+ |
| (2)
+-----------+ |
| | <--+
| shash |
| (sha256) |
+-----------+
]]>
</programlisting>
<para>
The following call sequence is applicable when a caller triggers
an HMAC operation:
</para>
<orderedlist>
<listitem>
<para>
The AHASH API functions are invoked by the caller. The HMAC
implementation performs its operation as needed.
</para>
<para>
During initialization of the HMAC cipher, the SHASH cipher type of
SHA256 is instantiated. The cipher handle for the SHA256 instance is
retained.
</para>
<para>
At one time, the HMAC implementation requires a SHA256 operation
where the SHA256 cipher handle is used.
</para>
</listitem>
<listitem>
<para>
The HMAC instance now invokes the SHASH API with the SHA256
cipher handle to calculate the message digest.
</para>
</listitem>
</orderedlist>
</sect2>
</sect1>
</chapter>
<chapter id="Development"><title>Developing Cipher Algorithms</title>
<sect1><title>Registering And Unregistering Transformation</title>
<para>
There are three distinct types of registration functions in
the Crypto API. One is used to register a generic cryptographic
transformation, while the other two are specific to HASH
transformations and COMPRESSion. We will discuss the latter
two in a separate chapter, here we will only look at the
generic ones.
</para>
<para>
Before discussing the register functions, the data structure
to be filled with each, struct crypto_alg, must be considered
-- see below for a description of this data structure.
</para>
<para>
The generic registration functions can be found in
include/linux/crypto.h and their definition can be seen below.
The former function registers a single transformation, while
the latter works on an array of transformation descriptions.
The latter is useful when registering transformations in bulk,
for example when a driver implements multiple transformations.
</para>
<programlisting>
int crypto_register_alg(struct crypto_alg *alg);
int crypto_register_algs(struct crypto_alg *algs, int count);
</programlisting>
<para>
The counterparts to those functions are listed below.
</para>
<programlisting>
int crypto_unregister_alg(struct crypto_alg *alg);
int crypto_unregister_algs(struct crypto_alg *algs, int count);
</programlisting>
<para>
Notice that both registration and unregistration functions
do return a value, so make sure to handle errors. A return
code of zero implies success. Any return code &lt; 0 implies
an error.
</para>
<para>
The bulk registration/unregistration functions
register/unregister each transformation in the given array of
length count. They handle errors as follows:
</para>
<itemizedlist>
<listitem>
<para>
crypto_register_algs() succeeds if and only if it
successfully registers all the given transformations. If an
error occurs partway through, then it rolls back successful
registrations before returning the error code. Note that if
a driver needs to handle registration errors for individual
transformations, then it will need to use the non-bulk
function crypto_register_alg() instead.
</para>
</listitem>
<listitem>
<para>
crypto_unregister_algs() tries to unregister all the given
transformations, continuing on error. It logs errors and
always returns zero.
</para>
</listitem>
</itemizedlist>
</sect1>
<sect1><title>Single-Block Symmetric Ciphers [CIPHER]</title>
<para>
Example of transformations: aes, arc4, ...
</para>
<para>
This section describes the simplest of all transformation
implementations, that being the CIPHER type used for symmetric
ciphers. The CIPHER type is used for transformations which
operate on exactly one block at a time and there are no
dependencies between blocks at all.
</para>
<sect2><title>Registration specifics</title>
<para>
The registration of [CIPHER] algorithm is specific in that
struct crypto_alg field .cra_type is empty. The .cra_u.cipher
has to be filled in with proper callbacks to implement this
transformation.
</para>
<para>
See struct cipher_alg below.
</para>
</sect2>
<sect2><title>Cipher Definition With struct cipher_alg</title>
<para>
Struct cipher_alg defines a single block cipher.
</para>
<para>
Here are schematics of how these functions are called when
operated from other part of the kernel. Note that the
.cia_setkey() call might happen before or after any of these
schematics happen, but must not happen during any of these
are in-flight.
</para>
<para>
<programlisting>
KEY ---. PLAINTEXT ---.
v v
.cia_setkey() -&gt; .cia_encrypt()
|
'-----&gt; CIPHERTEXT
</programlisting>
</para>
<para>
Please note that a pattern where .cia_setkey() is called
multiple times is also valid:
</para>
<para>
<programlisting>
KEY1 --. PLAINTEXT1 --. KEY2 --. PLAINTEXT2 --.
v v v v
.cia_setkey() -&gt; .cia_encrypt() -&gt; .cia_setkey() -&gt; .cia_encrypt()
| |
'---&gt; CIPHERTEXT1 '---&gt; CIPHERTEXT2
</programlisting>
</para>
</sect2>
</sect1>
<sect1><title>Multi-Block Ciphers</title>
<para>
Example of transformations: cbc(aes), ecb(arc4), ...
</para>
<para>
This section describes the multi-block cipher transformation
implementations. The multi-block ciphers are
used for transformations which operate on scatterlists of
data supplied to the transformation functions. They output
the result into a scatterlist of data as well.
</para>
<sect2><title>Registration Specifics</title>
<para>
The registration of multi-block cipher algorithms
is one of the most standard procedures throughout the crypto API.
</para>
<para>
Note, if a cipher implementation requires a proper alignment
of data, the caller should use the functions of
crypto_skcipher_alignmask() to identify a memory alignment mask.
The kernel crypto API is able to process requests that are unaligned.
This implies, however, additional overhead as the kernel
crypto API needs to perform the realignment of the data which
may imply moving of data.
</para>
</sect2>
<sect2><title>Cipher Definition With struct blkcipher_alg and ablkcipher_alg</title>
<para>
Struct blkcipher_alg defines a synchronous block cipher whereas
struct ablkcipher_alg defines an asynchronous block cipher.
</para>
<para>
Please refer to the single block cipher description for schematics
of the block cipher usage.
</para>
</sect2>
<sect2><title>Specifics Of Asynchronous Multi-Block Cipher</title>
<para>
There are a couple of specifics to the asynchronous interface.
</para>
<para>
First of all, some of the drivers will want to use the
Generic ScatterWalk in case the hardware needs to be fed
separate chunks of the scatterlist which contains the
plaintext and will contain the ciphertext. Please refer
to the ScatterWalk interface offered by the Linux kernel
scatter / gather list implementation.
</para>
</sect2>
</sect1>
<sect1><title>Hashing [HASH]</title>
<para>
Example of transformations: crc32, md5, sha1, sha256,...
</para>
<sect2><title>Registering And Unregistering The Transformation</title>
<para>
There are multiple ways to register a HASH transformation,
depending on whether the transformation is synchronous [SHASH]
or asynchronous [AHASH] and the amount of HASH transformations
we are registering. You can find the prototypes defined in
include/crypto/internal/hash.h:
</para>
<programlisting>
int crypto_register_ahash(struct ahash_alg *alg);
int crypto_register_shash(struct shash_alg *alg);
int crypto_register_shashes(struct shash_alg *algs, int count);
</programlisting>
<para>
The respective counterparts for unregistering the HASH
transformation are as follows:
</para>
<programlisting>
int crypto_unregister_ahash(struct ahash_alg *alg);
int crypto_unregister_shash(struct shash_alg *alg);
int crypto_unregister_shashes(struct shash_alg *algs, int count);
</programlisting>
</sect2>
<sect2><title>Cipher Definition With struct shash_alg and ahash_alg</title>
<para>
Here are schematics of how these functions are called when
operated from other part of the kernel. Note that the .setkey()
call might happen before or after any of these schematics happen,
but must not happen during any of these are in-flight. Please note
that calling .init() followed immediately by .finish() is also a
perfectly valid transformation.
</para>
<programlisting>
I) DATA -----------.
v
.init() -&gt; .update() -&gt; .final() ! .update() might not be called
^ | | at all in this scenario.
'----' '---&gt; HASH
II) DATA -----------.-----------.
v v
.init() -&gt; .update() -&gt; .finup() ! .update() may not be called
^ | | at all in this scenario.
'----' '---&gt; HASH
III) DATA -----------.
v
.digest() ! The entire process is handled
| by the .digest() call.
'---------------&gt; HASH
</programlisting>
<para>
Here is a schematic of how the .export()/.import() functions are
called when used from another part of the kernel.
</para>
<programlisting>
KEY--. DATA--.
v v ! .update() may not be called
.setkey() -&gt; .init() -&gt; .update() -&gt; .export() at all in this scenario.
^ | |
'-----' '--&gt; PARTIAL_HASH
----------- other transformations happen here -----------
PARTIAL_HASH--. DATA1--.
v v
.import -&gt; .update() -&gt; .final() ! .update() may not be called
^ | | at all in this scenario.
'----' '--&gt; HASH1
PARTIAL_HASH--. DATA2-.
v v
.import -&gt; .finup()
|
'---------------&gt; HASH2
</programlisting>
</sect2>
<sect2><title>Specifics Of Asynchronous HASH Transformation</title>
<para>
Some of the drivers will want to use the Generic ScatterWalk
in case the implementation needs to be fed separate chunks of the
scatterlist which contains the input data. The buffer containing
the resulting hash will always be properly aligned to
.cra_alignmask so there is no need to worry about this.
</para>
</sect2>
</sect1>
</chapter>
<chapter id="User"><title>User Space Interface</title>
<sect1><title>Introduction</title>
<para>
The concepts of the kernel crypto API visible to kernel space is fully
applicable to the user space interface as well. Therefore, the kernel
crypto API high level discussion for the in-kernel use cases applies
here as well.
</para>
<para>
The major difference, however, is that user space can only act as a
consumer and never as a provider of a transformation or cipher algorithm.
</para>
<para>
The following covers the user space interface exported by the kernel
crypto API. A working example of this description is libkcapi that
can be obtained from [1]. That library can be used by user space
applications that require cryptographic services from the kernel.
</para>
<para>
Some details of the in-kernel kernel crypto API aspects do not
apply to user space, however. This includes the difference between
synchronous and asynchronous invocations. The user space API call
is fully synchronous.
</para>
<para>
[1] <ulink url="http://www.chronox.de/libkcapi.html">http://www.chronox.de/libkcapi.html</ulink>
</para>
</sect1>
<sect1><title>User Space API General Remarks</title>
<para>
The kernel crypto API is accessible from user space. Currently,
the following ciphers are accessible:
</para>
<itemizedlist>
<listitem>
<para>Message digest including keyed message digest (HMAC, CMAC)</para>
</listitem>
<listitem>
<para>Symmetric ciphers</para>
</listitem>
<listitem>
<para>AEAD ciphers</para>
</listitem>
<listitem>
<para>Random Number Generators</para>
</listitem>
</itemizedlist>
<para>
The interface is provided via socket type using the type AF_ALG.
In addition, the setsockopt option type is SOL_ALG. In case the
user space header files do not export these flags yet, use the
following macros:
</para>
<programlisting>
#ifndef AF_ALG
#define AF_ALG 38
#endif
#ifndef SOL_ALG
#define SOL_ALG 279
#endif
</programlisting>
<para>
A cipher is accessed with the same name as done for the in-kernel
API calls. This includes the generic vs. unique naming schema for
ciphers as well as the enforcement of priorities for generic names.
</para>
<para>
To interact with the kernel crypto API, a socket must be
created by the user space application. User space invokes the cipher
operation with the send()/write() system call family. The result of the
cipher operation is obtained with the read()/recv() system call family.
</para>
<para>
The following API calls assume that the socket descriptor
is already opened by the user space application and discusses only
the kernel crypto API specific invocations.
</para>
<para>
To initialize the socket interface, the following sequence has to
be performed by the consumer:
</para>
<orderedlist>
<listitem>
<para>
Create a socket of type AF_ALG with the struct sockaddr_alg
parameter specified below for the different cipher types.
</para>
</listitem>
<listitem>
<para>
Invoke bind with the socket descriptor
</para>
</listitem>
<listitem>
<para>
Invoke accept with the socket descriptor. The accept system call
returns a new file descriptor that is to be used to interact with
the particular cipher instance. When invoking send/write or recv/read
system calls to send data to the kernel or obtain data from the
kernel, the file descriptor returned by accept must be used.
</para>
</listitem>
</orderedlist>
</sect1>
<sect1><title>In-place Cipher operation</title>
<para>
Just like the in-kernel operation of the kernel crypto API, the user
space interface allows the cipher operation in-place. That means that
the input buffer used for the send/write system call and the output
buffer used by the read/recv system call may be one and the same.
This is of particular interest for symmetric cipher operations where a
copying of the output data to its final destination can be avoided.
</para>
<para>
If a consumer on the other hand wants to maintain the plaintext and
the ciphertext in different memory locations, all a consumer needs
to do is to provide different memory pointers for the encryption and
decryption operation.
</para>
</sect1>
<sect1><title>Message Digest API</title>
<para>
The message digest type to be used for the cipher operation is
selected when invoking the bind syscall. bind requires the caller
to provide a filled struct sockaddr data structure. This data
structure must be filled as follows:
</para>
<programlisting>
struct sockaddr_alg sa = {
.salg_family = AF_ALG,
.salg_type = "hash", /* this selects the hash logic in the kernel */
.salg_name = "sha1" /* this is the cipher name */
};
</programlisting>
<para>
The salg_type value "hash" applies to message digests and keyed
message digests. Though, a keyed message digest is referenced by
the appropriate salg_name. Please see below for the setsockopt
interface that explains how the key can be set for a keyed message
digest.
</para>
<para>
Using the send() system call, the application provides the data that
should be processed with the message digest. The send system call
allows the following flags to be specified:
</para>
<itemizedlist>
<listitem>
<para>
MSG_MORE: If this flag is set, the send system call acts like a
message digest update function where the final hash is not
yet calculated. If the flag is not set, the send system call
calculates the final message digest immediately.
</para>
</listitem>
</itemizedlist>
<para>
With the recv() system call, the application can read the message
digest from the kernel crypto API. If the buffer is too small for the
message digest, the flag MSG_TRUNC is set by the kernel.
</para>
<para>
In order to set a message digest key, the calling application must use
the setsockopt() option of ALG_SET_KEY. If the key is not set the HMAC
operation is performed without the initial HMAC state change caused by
the key.
</para>
</sect1>
<sect1><title>Symmetric Cipher API</title>
<para>
The operation is very similar to the message digest discussion.
During initialization, the struct sockaddr data structure must be
filled as follows:
</para>
<programlisting>
struct sockaddr_alg sa = {
.salg_family = AF_ALG,
.salg_type = "skcipher", /* this selects the symmetric cipher */
.salg_name = "cbc(aes)" /* this is the cipher name */
};
</programlisting>
<para>
Before data can be sent to the kernel using the write/send system
call family, the consumer must set the key. The key setting is
described with the setsockopt invocation below.
</para>
<para>
Using the sendmsg() system call, the application provides the data that should be processed for encryption or decryption. In addition, the IV is
specified with the data structure provided by the sendmsg() system call.
</para>
<para>
The sendmsg system call parameter of struct msghdr is embedded into the
struct cmsghdr data structure. See recv(2) and cmsg(3) for more
information on how the cmsghdr data structure is used together with the
send/recv system call family. That cmsghdr data structure holds the
following information specified with a separate header instances:
</para>
<itemizedlist>
<listitem>
<para>
specification of the cipher operation type with one of these flags:
</para>
<itemizedlist>
<listitem>
<para>ALG_OP_ENCRYPT - encryption of data</para>
</listitem>
<listitem>
<para>ALG_OP_DECRYPT - decryption of data</para>
</listitem>
</itemizedlist>
</listitem>
<listitem>
<para>
specification of the IV information marked with the flag ALG_SET_IV
</para>
</listitem>
</itemizedlist>
<para>
The send system call family allows the following flag to be specified:
</para>
<itemizedlist>
<listitem>
<para>
MSG_MORE: If this flag is set, the send system call acts like a
cipher update function where more input data is expected
with a subsequent invocation of the send system call.
</para>
</listitem>
</itemizedlist>
<para>
Note: The kernel reports -EINVAL for any unexpected data. The caller
must make sure that all data matches the constraints given in
/proc/crypto for the selected cipher.
</para>
<para>
With the recv() system call, the application can read the result of
the cipher operation from the kernel crypto API. The output buffer
must be at least as large as to hold all blocks of the encrypted or
decrypted data. If the output data size is smaller, only as many
blocks are returned that fit into that output buffer size.
</para>
</sect1>
<sect1><title>AEAD Cipher API</title>
<para>
The operation is very similar to the symmetric cipher discussion.
During initialization, the struct sockaddr data structure must be
filled as follows:
</para>
<programlisting>
struct sockaddr_alg sa = {
.salg_family = AF_ALG,
.salg_type = "aead", /* this selects the symmetric cipher */
.salg_name = "gcm(aes)" /* this is the cipher name */
};
</programlisting>
<para>
Before data can be sent to the kernel using the write/send system
call family, the consumer must set the key. The key setting is
described with the setsockopt invocation below.
</para>
<para>
In addition, before data can be sent to the kernel using the
write/send system call family, the consumer must set the authentication
tag size. To set the authentication tag size, the caller must use the
setsockopt invocation described below.
</para>
<para>
Using the sendmsg() system call, the application provides the data that should be processed for encryption or decryption. In addition, the IV is
specified with the data structure provided by the sendmsg() system call.
</para>
<para>
The sendmsg system call parameter of struct msghdr is embedded into the
struct cmsghdr data structure. See recv(2) and cmsg(3) for more
information on how the cmsghdr data structure is used together with the
send/recv system call family. That cmsghdr data structure holds the
following information specified with a separate header instances:
</para>
<itemizedlist>
<listitem>
<para>
specification of the cipher operation type with one of these flags:
</para>
<itemizedlist>
<listitem>
<para>ALG_OP_ENCRYPT - encryption of data</para>
</listitem>
<listitem>
<para>ALG_OP_DECRYPT - decryption of data</para>
</listitem>
</itemizedlist>
</listitem>
<listitem>
<para>
specification of the IV information marked with the flag ALG_SET_IV
</para>
</listitem>
<listitem>
<para>
specification of the associated authentication data (AAD) with the
flag ALG_SET_AEAD_ASSOCLEN. The AAD is sent to the kernel together
with the plaintext / ciphertext. See below for the memory structure.
</para>
</listitem>
</itemizedlist>
<para>
The send system call family allows the following flag to be specified:
</para>
<itemizedlist>
<listitem>
<para>
MSG_MORE: If this flag is set, the send system call acts like a
cipher update function where more input data is expected
with a subsequent invocation of the send system call.
</para>
</listitem>
</itemizedlist>
<para>
Note: The kernel reports -EINVAL for any unexpected data. The caller
must make sure that all data matches the constraints given in
/proc/crypto for the selected cipher.
</para>
<para>
With the recv() system call, the application can read the result of
the cipher operation from the kernel crypto API. The output buffer
must be at least as large as defined with the memory structure below.
If the output data size is smaller, the cipher operation is not performed.
</para>
<para>
The authenticated decryption operation may indicate an integrity error.
Such breach in integrity is marked with the -EBADMSG error code.
</para>
<sect2><title>AEAD Memory Structure</title>
<para>
The AEAD cipher operates with the following information that
is communicated between user and kernel space as one data stream:
</para>
<itemizedlist>
<listitem>
<para>plaintext or ciphertext</para>
</listitem>
<listitem>
<para>associated authentication data (AAD)</para>
</listitem>
<listitem>
<para>authentication tag</para>
</listitem>
</itemizedlist>
<para>
The sizes of the AAD and the authentication tag are provided with
the sendmsg and setsockopt calls (see there). As the kernel knows
the size of the entire data stream, the kernel is now able to
calculate the right offsets of the data components in the data
stream.
</para>
<para>
The user space caller must arrange the aforementioned information
in the following order:
</para>
<itemizedlist>
<listitem>
<para>
AEAD encryption input: AAD || plaintext
</para>
</listitem>
<listitem>
<para>
AEAD decryption input: AAD || ciphertext || authentication tag
</para>
</listitem>
</itemizedlist>
<para>
The output buffer the user space caller provides must be at least as
large to hold the following data:
</para>
<itemizedlist>
<listitem>
<para>
AEAD encryption output: ciphertext || authentication tag
</para>
</listitem>
<listitem>
<para>
AEAD decryption output: plaintext
</para>
</listitem>
</itemizedlist>
</sect2>
</sect1>
<sect1><title>Random Number Generator API</title>
<para>
Again, the operation is very similar to the other APIs.
During initialization, the struct sockaddr data structure must be
filled as follows:
</para>
<programlisting>
struct sockaddr_alg sa = {
.salg_family = AF_ALG,
.salg_type = "rng", /* this selects the symmetric cipher */
.salg_name = "drbg_nopr_sha256" /* this is the cipher name */
};
</programlisting>
<para>
Depending on the RNG type, the RNG must be seeded. The seed is provided
using the setsockopt interface to set the key. For example, the
ansi_cprng requires a seed. The DRBGs do not require a seed, but
may be seeded.
</para>
<para>
Using the read()/recvmsg() system calls, random numbers can be obtained.
The kernel generates at most 128 bytes in one call. If user space
requires more data, multiple calls to read()/recvmsg() must be made.
</para>
<para>
WARNING: The user space caller may invoke the initially mentioned
accept system call multiple times. In this case, the returned file
descriptors have the same state.
</para>
</sect1>
<sect1><title>Zero-Copy Interface</title>
<para>
In addition to the send/write/read/recv system call family, the AF_ALG
interface can be accessed with the zero-copy interface of splice/vmsplice.
As the name indicates, the kernel tries to avoid a copy operation into
kernel space.
</para>
<para>
The zero-copy operation requires data to be aligned at the page boundary.
Non-aligned data can be used as well, but may require more operations of
the kernel which would defeat the speed gains obtained from the zero-copy
interface.
</para>
<para>
The system-interent limit for the size of one zero-copy operation is
16 pages. If more data is to be sent to AF_ALG, user space must slice
the input into segments with a maximum size of 16 pages.
</para>
<para>
Zero-copy can be used with the following code example (a complete working
example is provided with libkcapi):
</para>
<programlisting>
int pipes[2];
pipe(pipes);
/* input data in iov */
vmsplice(pipes[1], iov, iovlen, SPLICE_F_GIFT);
/* opfd is the file descriptor returned from accept() system call */
splice(pipes[0], NULL, opfd, NULL, ret, 0);
read(opfd, out, outlen);
</programlisting>
</sect1>
<sect1><title>Setsockopt Interface</title>
<para>
In addition to the read/recv and send/write system call handling
to send and retrieve data subject to the cipher operation, a consumer
also needs to set the additional information for the cipher operation.
This additional information is set using the setsockopt system call
that must be invoked with the file descriptor of the open cipher
(i.e. the file descriptor returned by the accept system call).
</para>
<para>
Each setsockopt invocation must use the level SOL_ALG.
</para>
<para>
The setsockopt interface allows setting the following data using
the mentioned optname:
</para>
<itemizedlist>
<listitem>
<para>
ALG_SET_KEY -- Setting the key. Key setting is applicable to:
</para>
<itemizedlist>
<listitem>
<para>the skcipher cipher type (symmetric ciphers)</para>
</listitem>
<listitem>
<para>the hash cipher type (keyed message digests)</para>
</listitem>
<listitem>
<para>the AEAD cipher type</para>
</listitem>
<listitem>
<para>the RNG cipher type to provide the seed</para>
</listitem>
</itemizedlist>
</listitem>
<listitem>
<para>
ALG_SET_AEAD_AUTHSIZE -- Setting the authentication tag size
for AEAD ciphers. For a encryption operation, the authentication
tag of the given size will be generated. For a decryption operation,
the provided ciphertext is assumed to contain an authentication tag
of the given size (see section about AEAD memory layout below).
</para>
</listitem>
</itemizedlist>
</sect1>
<sect1><title>User space API example</title>
<para>
Please see [1] for libkcapi which provides an easy-to-use wrapper
around the aforementioned Netlink kernel interface. [1] also contains
a test application that invokes all libkcapi API calls.
</para>
<para>
[1] <ulink url="http://www.chronox.de/libkcapi.html">http://www.chronox.de/libkcapi.html</ulink>
</para>
</sect1>
</chapter>
<chapter id="API"><title>Programming Interface</title>
<para>
Please note that the kernel crypto API contains the AEAD givcrypt
API (crypto_aead_giv* and aead_givcrypt_* function calls in
include/crypto/aead.h). This API is obsolete and will be removed
in the future. To obtain the functionality of an AEAD cipher with
internal IV generation, use the IV generator as a regular cipher.
For example, rfc4106(gcm(aes)) is the AEAD cipher with external
IV generation and seqniv(rfc4106(gcm(aes))) implies that the kernel
crypto API generates the IV. Different IV generators are available.
</para>
<sect1><title>Block Cipher Context Data Structures</title>
!Pinclude/linux/crypto.h Block Cipher Context Data Structures
!Finclude/crypto/aead.h aead_request
</sect1>
<sect1><title>Block Cipher Algorithm Definitions</title>
!Pinclude/linux/crypto.h Block Cipher Algorithm Definitions
!Finclude/linux/crypto.h crypto_alg
!Finclude/linux/crypto.h ablkcipher_alg
!Finclude/crypto/aead.h aead_alg
!Finclude/linux/crypto.h blkcipher_alg
!Finclude/linux/crypto.h cipher_alg
!Finclude/crypto/rng.h rng_alg
</sect1>
<sect1><title>Symmetric Key Cipher API</title>
!Pinclude/crypto/skcipher.h Symmetric Key Cipher API
!Finclude/crypto/skcipher.h crypto_alloc_skcipher
!Finclude/crypto/skcipher.h crypto_free_skcipher
!Finclude/crypto/skcipher.h crypto_has_skcipher
!Finclude/crypto/skcipher.h crypto_skcipher_ivsize
!Finclude/crypto/skcipher.h crypto_skcipher_blocksize
!Finclude/crypto/skcipher.h crypto_skcipher_setkey
!Finclude/crypto/skcipher.h crypto_skcipher_reqtfm
!Finclude/crypto/skcipher.h crypto_skcipher_encrypt
!Finclude/crypto/skcipher.h crypto_skcipher_decrypt
</sect1>
<sect1><title>Symmetric Key Cipher Request Handle</title>
!Pinclude/crypto/skcipher.h Symmetric Key Cipher Request Handle
!Finclude/crypto/skcipher.h crypto_skcipher_reqsize
!Finclude/crypto/skcipher.h skcipher_request_set_tfm
!Finclude/crypto/skcipher.h skcipher_request_alloc
!Finclude/crypto/skcipher.h skcipher_request_free
!Finclude/crypto/skcipher.h skcipher_request_set_callback
!Finclude/crypto/skcipher.h skcipher_request_set_crypt
</sect1>
<sect1><title>Asynchronous Block Cipher API - Deprecated</title>
!Pinclude/linux/crypto.h Asynchronous Block Cipher API
!Finclude/linux/crypto.h crypto_alloc_ablkcipher
!Finclude/linux/crypto.h crypto_free_ablkcipher
!Finclude/linux/crypto.h crypto_has_ablkcipher
!Finclude/linux/crypto.h crypto_ablkcipher_ivsize
!Finclude/linux/crypto.h crypto_ablkcipher_blocksize
!Finclude/linux/crypto.h crypto_ablkcipher_setkey
!Finclude/linux/crypto.h crypto_ablkcipher_reqtfm
!Finclude/linux/crypto.h crypto_ablkcipher_encrypt
!Finclude/linux/crypto.h crypto_ablkcipher_decrypt
</sect1>
<sect1><title>Asynchronous Cipher Request Handle - Deprecated</title>
!Pinclude/linux/crypto.h Asynchronous Cipher Request Handle
!Finclude/linux/crypto.h crypto_ablkcipher_reqsize
!Finclude/linux/crypto.h ablkcipher_request_set_tfm
!Finclude/linux/crypto.h ablkcipher_request_alloc
!Finclude/linux/crypto.h ablkcipher_request_free
!Finclude/linux/crypto.h ablkcipher_request_set_callback
!Finclude/linux/crypto.h ablkcipher_request_set_crypt
</sect1>
<sect1><title>Authenticated Encryption With Associated Data (AEAD) Cipher API</title>
!Pinclude/crypto/aead.h Authenticated Encryption With Associated Data (AEAD) Cipher API
!Finclude/crypto/aead.h crypto_alloc_aead
!Finclude/crypto/aead.h crypto_free_aead
!Finclude/crypto/aead.h crypto_aead_ivsize
!Finclude/crypto/aead.h crypto_aead_authsize
!Finclude/crypto/aead.h crypto_aead_blocksize
!Finclude/crypto/aead.h crypto_aead_setkey
!Finclude/crypto/aead.h crypto_aead_setauthsize
!Finclude/crypto/aead.h crypto_aead_encrypt
!Finclude/crypto/aead.h crypto_aead_decrypt
</sect1>
<sect1><title>Asynchronous AEAD Request Handle</title>
!Pinclude/crypto/aead.h Asynchronous AEAD Request Handle
!Finclude/crypto/aead.h crypto_aead_reqsize
!Finclude/crypto/aead.h aead_request_set_tfm
!Finclude/crypto/aead.h aead_request_alloc
!Finclude/crypto/aead.h aead_request_free
!Finclude/crypto/aead.h aead_request_set_callback
!Finclude/crypto/aead.h aead_request_set_crypt
!Finclude/crypto/aead.h aead_request_set_ad
</sect1>
<sect1><title>Synchronous Block Cipher API - Deprecated</title>
!Pinclude/linux/crypto.h Synchronous Block Cipher API
!Finclude/linux/crypto.h crypto_alloc_blkcipher
!Finclude/linux/crypto.h crypto_free_blkcipher
!Finclude/linux/crypto.h crypto_has_blkcipher
!Finclude/linux/crypto.h crypto_blkcipher_name
!Finclude/linux/crypto.h crypto_blkcipher_ivsize
!Finclude/linux/crypto.h crypto_blkcipher_blocksize
!Finclude/linux/crypto.h crypto_blkcipher_setkey
!Finclude/linux/crypto.h crypto_blkcipher_encrypt
!Finclude/linux/crypto.h crypto_blkcipher_encrypt_iv
!Finclude/linux/crypto.h crypto_blkcipher_decrypt
!Finclude/linux/crypto.h crypto_blkcipher_decrypt_iv
!Finclude/linux/crypto.h crypto_blkcipher_set_iv
!Finclude/linux/crypto.h crypto_blkcipher_get_iv
</sect1>
<sect1><title>Single Block Cipher API</title>
!Pinclude/linux/crypto.h Single Block Cipher API
!Finclude/linux/crypto.h crypto_alloc_cipher
!Finclude/linux/crypto.h crypto_free_cipher
!Finclude/linux/crypto.h crypto_has_cipher
!Finclude/linux/crypto.h crypto_cipher_blocksize
!Finclude/linux/crypto.h crypto_cipher_setkey
!Finclude/linux/crypto.h crypto_cipher_encrypt_one
!Finclude/linux/crypto.h crypto_cipher_decrypt_one
</sect1>
<sect1><title>Message Digest Algorithm Definitions</title>
!Pinclude/crypto/hash.h Message Digest Algorithm Definitions
!Finclude/crypto/hash.h hash_alg_common
!Finclude/crypto/hash.h ahash_alg
!Finclude/crypto/hash.h shash_alg
</sect1>
<sect1><title>Asynchronous Message Digest API</title>
!Pinclude/crypto/hash.h Asynchronous Message Digest API
!Finclude/crypto/hash.h crypto_alloc_ahash
!Finclude/crypto/hash.h crypto_free_ahash
!Finclude/crypto/hash.h crypto_ahash_init
!Finclude/crypto/hash.h crypto_ahash_digestsize
!Finclude/crypto/hash.h crypto_ahash_reqtfm
!Finclude/crypto/hash.h crypto_ahash_reqsize
!Finclude/crypto/hash.h crypto_ahash_setkey
!Finclude/crypto/hash.h crypto_ahash_finup
!Finclude/crypto/hash.h crypto_ahash_final
!Finclude/crypto/hash.h crypto_ahash_digest
!Finclude/crypto/hash.h crypto_ahash_export
!Finclude/crypto/hash.h crypto_ahash_import
</sect1>
<sect1><title>Asynchronous Hash Request Handle</title>
!Pinclude/crypto/hash.h Asynchronous Hash Request Handle
!Finclude/crypto/hash.h ahash_request_set_tfm
!Finclude/crypto/hash.h ahash_request_alloc
!Finclude/crypto/hash.h ahash_request_free
!Finclude/crypto/hash.h ahash_request_set_callback
!Finclude/crypto/hash.h ahash_request_set_crypt
</sect1>
<sect1><title>Synchronous Message Digest API</title>
!Pinclude/crypto/hash.h Synchronous Message Digest API
!Finclude/crypto/hash.h crypto_alloc_shash
!Finclude/crypto/hash.h crypto_free_shash
!Finclude/crypto/hash.h crypto_shash_blocksize
!Finclude/crypto/hash.h crypto_shash_digestsize
!Finclude/crypto/hash.h crypto_shash_descsize
!Finclude/crypto/hash.h crypto_shash_setkey
!Finclude/crypto/hash.h crypto_shash_digest
!Finclude/crypto/hash.h crypto_shash_export
!Finclude/crypto/hash.h crypto_shash_import
!Finclude/crypto/hash.h crypto_shash_init
!Finclude/crypto/hash.h crypto_shash_update
!Finclude/crypto/hash.h crypto_shash_final
!Finclude/crypto/hash.h crypto_shash_finup
</sect1>
<sect1><title>Crypto API Random Number API</title>
!Pinclude/crypto/rng.h Random number generator API
!Finclude/crypto/rng.h crypto_alloc_rng
!Finclude/crypto/rng.h crypto_rng_alg
!Finclude/crypto/rng.h crypto_free_rng
!Finclude/crypto/rng.h crypto_rng_generate
!Finclude/crypto/rng.h crypto_rng_get_bytes
!Finclude/crypto/rng.h crypto_rng_reset
!Finclude/crypto/rng.h crypto_rng_seedsize
!Cinclude/crypto/rng.h
</sect1>
<sect1><title>Asymmetric Cipher API</title>
!Pinclude/crypto/akcipher.h Generic Public Key API
!Finclude/crypto/akcipher.h akcipher_alg
!Finclude/crypto/akcipher.h akcipher_request
!Finclude/crypto/akcipher.h crypto_alloc_akcipher
!Finclude/crypto/akcipher.h crypto_free_akcipher
!Finclude/crypto/akcipher.h crypto_akcipher_set_pub_key
!Finclude/crypto/akcipher.h crypto_akcipher_set_priv_key
</sect1>
<sect1><title>Asymmetric Cipher Request Handle</title>
!Finclude/crypto/akcipher.h akcipher_request_alloc
!Finclude/crypto/akcipher.h akcipher_request_free
!Finclude/crypto/akcipher.h akcipher_request_set_callback
!Finclude/crypto/akcipher.h akcipher_request_set_crypt
!Finclude/crypto/akcipher.h crypto_akcipher_maxsize
!Finclude/crypto/akcipher.h crypto_akcipher_encrypt
!Finclude/crypto/akcipher.h crypto_akcipher_decrypt
!Finclude/crypto/akcipher.h crypto_akcipher_sign
!Finclude/crypto/akcipher.h crypto_akcipher_verify
</sect1>
</chapter>
<chapter id="Code"><title>Code Examples</title>
<sect1><title>Code Example For Symmetric Key Cipher Operation</title>
<programlisting>
struct tcrypt_result {
struct completion completion;
int err;
};
/* tie all data structures together */
struct skcipher_def {
struct scatterlist sg;
struct crypto_skcipher *tfm;
struct skcipher_request *req;
struct tcrypt_result result;
};
/* Callback function */
static void test_skcipher_cb(struct crypto_async_request *req, int error)
{
struct tcrypt_result *result = req-&gt;data;
if (error == -EINPROGRESS)
return;
result-&gt;err = error;
complete(&amp;result-&gt;completion);
pr_info("Encryption finished successfully\n");
}
/* Perform cipher operation */
static unsigned int test_skcipher_encdec(struct skcipher_def *sk,
int enc)
{
int rc = 0;
if (enc)
rc = crypto_skcipher_encrypt(sk-&gt;req);
else
rc = crypto_skcipher_decrypt(sk-&gt;req);
switch (rc) {
case 0:
break;
case -EINPROGRESS:
case -EBUSY:
rc = wait_for_completion_interruptible(
&amp;sk-&gt;result.completion);
if (!rc &amp;&amp; !sk-&gt;result.err) {
reinit_completion(&amp;sk-&gt;result.completion);
break;
}
default:
pr_info("skcipher encrypt returned with %d result %d\n",
rc, sk-&gt;result.err);
break;
}
init_completion(&amp;sk-&gt;result.completion);
return rc;
}
/* Initialize and trigger cipher operation */
static int test_skcipher(void)
{
struct skcipher_def sk;
struct crypto_skcipher *skcipher = NULL;
struct skcipher_request *req = NULL;
char *scratchpad = NULL;
char *ivdata = NULL;
unsigned char key[32];
int ret = -EFAULT;
skcipher = crypto_alloc_skcipher("cbc-aes-aesni", 0, 0);
if (IS_ERR(skcipher)) {
pr_info("could not allocate skcipher handle\n");
return PTR_ERR(skcipher);
}
req = skcipher_request_alloc(skcipher, GFP_KERNEL);
if (!req) {
pr_info("could not allocate skcipher request\n");
ret = -ENOMEM;
goto out;
}
skcipher_request_set_callback(req, CRYPTO_TFM_REQ_MAY_BACKLOG,
test_skcipher_cb,
&amp;sk.result);
/* AES 256 with random key */
get_random_bytes(&amp;key, 32);
if (crypto_skcipher_setkey(skcipher, key, 32)) {
pr_info("key could not be set\n");
ret = -EAGAIN;
goto out;
}
/* IV will be random */
ivdata = kmalloc(16, GFP_KERNEL);
if (!ivdata) {
pr_info("could not allocate ivdata\n");
goto out;
}
get_random_bytes(ivdata, 16);
/* Input data will be random */
scratchpad = kmalloc(16, GFP_KERNEL);
if (!scratchpad) {
pr_info("could not allocate scratchpad\n");
goto out;
}
get_random_bytes(scratchpad, 16);
sk.tfm = skcipher;
sk.req = req;
/* We encrypt one block */
sg_init_one(&amp;sk.sg, scratchpad, 16);
skcipher_request_set_crypt(req, &amp;sk.sg, &amp;sk.sg, 16, ivdata);
init_completion(&amp;sk.result.completion);
/* encrypt data */
ret = test_skcipher_encdec(&amp;sk, 1);
if (ret)
goto out;
pr_info("Encryption triggered successfully\n");
out:
if (skcipher)
crypto_free_skcipher(skcipher);
if (req)
skcipher_request_free(req);
if (ivdata)
kfree(ivdata);
if (scratchpad)
kfree(scratchpad);
return ret;
}
</programlisting>
</sect1>
<sect1><title>Code Example For Use of Operational State Memory With SHASH</title>
<programlisting>
struct sdesc {
struct shash_desc shash;
char ctx[];
};
static struct sdescinit_sdesc(struct crypto_shash *alg)
{
struct sdescsdesc;
int size;
size = sizeof(struct shash_desc) + crypto_shash_descsize(alg);
sdesc = kmalloc(size, GFP_KERNEL);
if (!sdesc)
return ERR_PTR(-ENOMEM);
sdesc-&gt;shash.tfm = alg;
sdesc-&gt;shash.flags = 0x0;
return sdesc;
}
static int calc_hash(struct crypto_shashalg,
const unsigned chardata, unsigned int datalen,
unsigned chardigest) {
struct sdescsdesc;
int ret;
sdesc = init_sdesc(alg);
if (IS_ERR(sdesc)) {
pr_info("trusted_key: can't alloc %s\n", hash_alg);
return PTR_ERR(sdesc);
}
ret = crypto_shash_digest(&amp;sdesc-&gt;shash, data, datalen, digest);
kfree(sdesc);
return ret;
}
</programlisting>
</sect1>
<sect1><title>Code Example For Random Number Generator Usage</title>
<programlisting>
static int get_random_numbers(u8 *buf, unsigned int len)
{
struct crypto_rngrng = NULL;
chardrbg = "drbg_nopr_sha256"; /* Hash DRBG with SHA-256, no PR */
int ret;
if (!buf || !len) {
pr_debug("No output buffer provided\n");
return -EINVAL;
}
rng = crypto_alloc_rng(drbg, 0, 0);
if (IS_ERR(rng)) {
pr_debug("could not allocate RNG handle for %s\n", drbg);
return -PTR_ERR(rng);
}
ret = crypto_rng_get_bytes(rng, buf, len);
if (ret &lt; 0)
pr_debug("generation of random numbers failed\n");
else if (ret == 0)
pr_debug("RNG returned no data");
else
pr_debug("RNG returned %d bytes of data\n", ret);
out:
crypto_free_rng(rng);
return ret;
}
</programlisting>
</sect1>
</chapter>
</book>
Authenticated Encryption With Associated Data (AEAD) Algorithm Definitions
--------------------------------------------------------------------------
.. kernel-doc:: include/crypto/aead.h
:doc: Authenticated Encryption With Associated Data (AEAD) Cipher API
.. kernel-doc:: include/crypto/aead.h
:functions: aead_request aead_alg
Authenticated Encryption With Associated Data (AEAD) Cipher API
---------------------------------------------------------------
.. kernel-doc:: include/crypto/aead.h
:functions: crypto_alloc_aead crypto_free_aead crypto_aead_ivsize crypto_aead_authsize crypto_aead_blocksize crypto_aead_setkey crypto_aead_setauthsize crypto_aead_encrypt crypto_aead_decrypt
Asynchronous AEAD Request Handle
--------------------------------
.. kernel-doc:: include/crypto/aead.h
:doc: Asynchronous AEAD Request Handle
.. kernel-doc:: include/crypto/aead.h
:functions: crypto_aead_reqsize aead_request_set_tfm aead_request_alloc aead_request_free aead_request_set_callback aead_request_set_crypt aead_request_set_ad
Asymmetric Cipher Algorithm Definitions
---------------------------------------
.. kernel-doc:: include/crypto/akcipher.h
:functions: akcipher_alg akcipher_request
Asymmetric Cipher API
---------------------
.. kernel-doc:: include/crypto/akcipher.h
:doc: Generic Public Key API
.. kernel-doc:: include/crypto/akcipher.h
:functions: crypto_alloc_akcipher crypto_free_akcipher crypto_akcipher_set_pub_key crypto_akcipher_set_priv_key crypto_akcipher_maxsize crypto_akcipher_encrypt crypto_akcipher_decrypt crypto_akcipher_sign crypto_akcipher_verify
Asymmetric Cipher Request Handle
--------------------------------
.. kernel-doc:: include/crypto/akcipher.h
:functions: akcipher_request_alloc akcipher_request_free akcipher_request_set_callback akcipher_request_set_crypt
Message Digest Algorithm Definitions
------------------------------------
.. kernel-doc:: include/crypto/hash.h
:doc: Message Digest Algorithm Definitions
.. kernel-doc:: include/crypto/hash.h
:functions: hash_alg_common ahash_alg shash_alg
Asynchronous Message Digest API
-------------------------------
.. kernel-doc:: include/crypto/hash.h
:doc: Asynchronous Message Digest API
.. kernel-doc:: include/crypto/hash.h
:functions: crypto_alloc_ahash crypto_free_ahash crypto_ahash_init crypto_ahash_digestsize crypto_ahash_reqtfm crypto_ahash_reqsize crypto_ahash_setkey crypto_ahash_finup crypto_ahash_final crypto_ahash_digest crypto_ahash_export crypto_ahash_import
Asynchronous Hash Request Handle
--------------------------------
.. kernel-doc:: include/crypto/hash.h
:doc: Asynchronous Hash Request Handle
.. kernel-doc:: include/crypto/hash.h
:functions: ahash_request_set_tfm ahash_request_alloc ahash_request_free ahash_request_set_callback ahash_request_set_crypt
Synchronous Message Digest API
------------------------------
.. kernel-doc:: include/crypto/hash.h
:doc: Synchronous Message Digest API
.. kernel-doc:: include/crypto/hash.h
:functions: crypto_alloc_shash crypto_free_shash crypto_shash_blocksize crypto_shash_digestsize crypto_shash_descsize crypto_shash_setkey crypto_shash_digest crypto_shash_export crypto_shash_import crypto_shash_init crypto_shash_update crypto_shash_final crypto_shash_finup
Key-agreement Protocol Primitives (KPP) Cipher Algorithm Definitions
--------------------------------------------------------------------
.. kernel-doc:: include/crypto/kpp.h
:functions: kpp_request crypto_kpp kpp_alg kpp_secret
Key-agreement Protocol Primitives (KPP) Cipher API
--------------------------------------------------
.. kernel-doc:: include/crypto/kpp.h
:doc: Generic Key-agreement Protocol Primitives API
.. kernel-doc:: include/crypto/kpp.h
:functions: crypto_alloc_kpp crypto_free_kpp crypto_kpp_set_secret crypto_kpp_generate_public_key crypto_kpp_compute_shared_secret crypto_kpp_maxsize
Key-agreement Protocol Primitives (KPP) Cipher Request Handle
-------------------------------------------------------------
.. kernel-doc:: include/crypto/kpp.h
:functions: kpp_request_alloc kpp_request_free kpp_request_set_callback kpp_request_set_input kpp_request_set_output
ECDH Helper Functions
---------------------
.. kernel-doc:: include/crypto/ecdh.h
:doc: ECDH Helper Functions
.. kernel-doc:: include/crypto/ecdh.h
:functions: ecdh crypto_ecdh_key_len crypto_ecdh_encode_key crypto_ecdh_decode_key
DH Helper Functions
-------------------
.. kernel-doc:: include/crypto/dh.h
:doc: DH Helper Functions
.. kernel-doc:: include/crypto/dh.h
:functions: dh crypto_dh_key_len crypto_dh_encode_key crypto_dh_decode_key
Random Number Algorithm Definitions
-----------------------------------
.. kernel-doc:: include/crypto/rng.h
:functions: rng_alg
Crypto API Random Number API
----------------------------
.. kernel-doc:: include/crypto/rng.h
:doc: Random number generator API
.. kernel-doc:: include/crypto/rng.h
:functions: crypto_alloc_rng crypto_rng_alg crypto_free_rng crypto_rng_generate crypto_rng_get_bytes crypto_rng_reset crypto_rng_seedsize
Code Examples
=============
Code Example For Symmetric Key Cipher Operation
-----------------------------------------------
::
struct tcrypt_result {
struct completion completion;
int err;
};
/* tie all data structures together */
struct skcipher_def {
struct scatterlist sg;
struct crypto_skcipher *tfm;
struct skcipher_request *req;
struct tcrypt_result result;
};
/* Callback function */
static void test_skcipher_cb(struct crypto_async_request *req, int error)
{
struct tcrypt_result *result = req->data;
if (error == -EINPROGRESS)
return;
result->err = error;
complete(&result->completion);
pr_info("Encryption finished successfully\n");
}
/* Perform cipher operation */
static unsigned int test_skcipher_encdec(struct skcipher_def *sk,
int enc)
{
int rc = 0;
if (enc)
rc = crypto_skcipher_encrypt(sk->req);
else
rc = crypto_skcipher_decrypt(sk->req);
switch (rc) {
case 0:
break;
case -EINPROGRESS:
case -EBUSY:
rc = wait_for_completion_interruptible(
&sk->result.completion);
if (!rc && !sk->result.err) {
reinit_completion(&sk->result.completion);
break;
}
default:
pr_info("skcipher encrypt returned with %d result %d\n",
rc, sk->result.err);
break;
}
init_completion(&sk->result.completion);
return rc;
}
/* Initialize and trigger cipher operation */
static int test_skcipher(void)
{
struct skcipher_def sk;
struct crypto_skcipher *skcipher = NULL;
struct skcipher_request *req = NULL;
char *scratchpad = NULL;
char *ivdata = NULL;
unsigned char key[32];
int ret = -EFAULT;
skcipher = crypto_alloc_skcipher("cbc-aes-aesni", 0, 0);
if (IS_ERR(skcipher)) {
pr_info("could not allocate skcipher handle\n");
return PTR_ERR(skcipher);
}
req = skcipher_request_alloc(skcipher, GFP_KERNEL);
if (!req) {
pr_info("could not allocate skcipher request\n");
ret = -ENOMEM;
goto out;
}
skcipher_request_set_callback(req, CRYPTO_TFM_REQ_MAY_BACKLOG,
test_skcipher_cb,
&sk.result);
/* AES 256 with random key */
get_random_bytes(&key, 32);
if (crypto_skcipher_setkey(skcipher, key, 32)) {
pr_info("key could not be set\n");
ret = -EAGAIN;
goto out;
}
/* IV will be random */
ivdata = kmalloc(16, GFP_KERNEL);
if (!ivdata) {
pr_info("could not allocate ivdata\n");
goto out;
}
get_random_bytes(ivdata, 16);
/* Input data will be random */
scratchpad = kmalloc(16, GFP_KERNEL);
if (!scratchpad) {
pr_info("could not allocate scratchpad\n");
goto out;
}
get_random_bytes(scratchpad, 16);
sk.tfm = skcipher;
sk.req = req;
/* We encrypt one block */
sg_init_one(&sk.sg, scratchpad, 16);
skcipher_request_set_crypt(req, &sk.sg, &sk.sg, 16, ivdata);
init_completion(&sk.result.completion);
/* encrypt data */
ret = test_skcipher_encdec(&sk, 1);
if (ret)
goto out;
pr_info("Encryption triggered successfully\n");
out:
if (skcipher)
crypto_free_skcipher(skcipher);
if (req)
skcipher_request_free(req);
if (ivdata)
kfree(ivdata);
if (scratchpad)
kfree(scratchpad);
return ret;
}
Code Example For Use of Operational State Memory With SHASH
-----------------------------------------------------------
::
struct sdesc {
struct shash_desc shash;
char ctx[];
};
static struct sdescinit_sdesc(struct crypto_shash *alg)
{
struct sdescsdesc;
int size;
size = sizeof(struct shash_desc) + crypto_shash_descsize(alg);
sdesc = kmalloc(size, GFP_KERNEL);
if (!sdesc)
return ERR_PTR(-ENOMEM);
sdesc->shash.tfm = alg;
sdesc->shash.flags = 0x0;
return sdesc;
}
static int calc_hash(struct crypto_shashalg,
const unsigned chardata, unsigned int datalen,
unsigned chardigest) {
struct sdescsdesc;
int ret;
sdesc = init_sdesc(alg);
if (IS_ERR(sdesc)) {
pr_info("trusted_key: can't alloc %s\n", hash_alg);
return PTR_ERR(sdesc);
}
ret = crypto_shash_digest(&sdesc->shash, data, datalen, digest);
kfree(sdesc);
return ret;
}
Code Example For Random Number Generator Usage
----------------------------------------------
::
static int get_random_numbers(u8 *buf, unsigned int len)
{
struct crypto_rngrng = NULL;
chardrbg = "drbg_nopr_sha256"; /* Hash DRBG with SHA-256, no PR */
int ret;
if (!buf || !len) {
pr_debug("No output buffer provided\n");
return -EINVAL;
}
rng = crypto_alloc_rng(drbg, 0, 0);
if (IS_ERR(rng)) {
pr_debug("could not allocate RNG handle for %s\n", drbg);
return -PTR_ERR(rng);
}
ret = crypto_rng_get_bytes(rng, buf, len);
if (ret < 0)
pr_debug("generation of random numbers failed\n");
else if (ret == 0)
pr_debug("RNG returned no data");
else
pr_debug("RNG returned %d bytes of data\n", ret);
out:
crypto_free_rng(rng);
return ret;
}
Block Cipher Algorithm Definitions
----------------------------------
.. kernel-doc:: include/linux/crypto.h
:doc: Block Cipher Algorithm Definitions
.. kernel-doc:: include/linux/crypto.h
:functions: crypto_alg ablkcipher_alg blkcipher_alg cipher_alg
Symmetric Key Cipher API
------------------------
.. kernel-doc:: include/crypto/skcipher.h
:doc: Symmetric Key Cipher API
.. kernel-doc:: include/crypto/skcipher.h
:functions: crypto_alloc_skcipher crypto_free_skcipher crypto_has_skcipher crypto_skcipher_ivsize crypto_skcipher_blocksize crypto_skcipher_setkey crypto_skcipher_reqtfm crypto_skcipher_encrypt crypto_skcipher_decrypt
Symmetric Key Cipher Request Handle
-----------------------------------
.. kernel-doc:: include/crypto/skcipher.h
:doc: Symmetric Key Cipher Request Handle
.. kernel-doc:: include/crypto/skcipher.h
:functions: crypto_skcipher_reqsize skcipher_request_set_tfm skcipher_request_alloc skcipher_request_free skcipher_request_set_callback skcipher_request_set_crypt
Single Block Cipher API
-----------------------
.. kernel-doc:: include/linux/crypto.h
:doc: Single Block Cipher API
.. kernel-doc:: include/linux/crypto.h
:functions: crypto_alloc_cipher crypto_free_cipher crypto_has_cipher crypto_cipher_blocksize crypto_cipher_setkey crypto_cipher_encrypt_one crypto_cipher_decrypt_one
Asynchronous Block Cipher API - Deprecated
------------------------------------------
.. kernel-doc:: include/linux/crypto.h
:doc: Asynchronous Block Cipher API
.. kernel-doc:: include/linux/crypto.h
:functions: crypto_free_ablkcipher crypto_has_ablkcipher crypto_ablkcipher_ivsize crypto_ablkcipher_blocksize crypto_ablkcipher_setkey crypto_ablkcipher_reqtfm crypto_ablkcipher_encrypt crypto_ablkcipher_decrypt
Asynchronous Cipher Request Handle - Deprecated
-----------------------------------------------
.. kernel-doc:: include/linux/crypto.h
:doc: Asynchronous Cipher Request Handle
.. kernel-doc:: include/linux/crypto.h
:functions: crypto_ablkcipher_reqsize ablkcipher_request_set_tfm ablkcipher_request_alloc ablkcipher_request_free ablkcipher_request_set_callback ablkcipher_request_set_crypt
Synchronous Block Cipher API - Deprecated
-----------------------------------------
.. kernel-doc:: include/linux/crypto.h
:doc: Synchronous Block Cipher API
.. kernel-doc:: include/linux/crypto.h
:functions: crypto_alloc_blkcipher rypto_free_blkcipher crypto_has_blkcipher crypto_blkcipher_name crypto_blkcipher_ivsize crypto_blkcipher_blocksize crypto_blkcipher_setkey crypto_blkcipher_encrypt crypto_blkcipher_encrypt_iv crypto_blkcipher_decrypt crypto_blkcipher_decrypt_iv crypto_blkcipher_set_iv crypto_blkcipher_get_iv
Programming Interface
=====================
Please note that the kernel crypto API contains the AEAD givcrypt API
(crypto_aead_giv\* and aead_givcrypt\* function calls in
include/crypto/aead.h). This API is obsolete and will be removed in the
future. To obtain the functionality of an AEAD cipher with internal IV
generation, use the IV generator as a regular cipher. For example,
rfc4106(gcm(aes)) is the AEAD cipher with external IV generation and
seqniv(rfc4106(gcm(aes))) implies that the kernel crypto API generates
the IV. Different IV generators are available.
.. class:: toc-title
Table of contents
.. toctree::
:maxdepth: 2
api-skcipher
api-aead
api-digest
api-rng
api-akcipher
api-kpp
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)
- kpp for a Key-agreement Protocol Primitive (KPP) cipher such as
an ECDH or DH 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_KPP Key-agreement Protocol Primitive (KPP) such as
an ECDH or DH implementation
- 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.
Developing Cipher Algorithms
============================
Registering And Unregistering Transformation
--------------------------------------------
There are three distinct types of registration functions in the Crypto
API. One is used to register a generic cryptographic transformation,
while the other two are specific to HASH transformations and
COMPRESSion. We will discuss the latter two in a separate chapter, here
we will only look at the generic ones.
Before discussing the register functions, the data structure to be
filled with each, struct crypto_alg, must be considered -- see below
for a description of this data structure.
The generic registration functions can be found in
include/linux/crypto.h and their definition can be seen below. The
former function registers a single transformation, while the latter
works on an array of transformation descriptions. The latter is useful
when registering transformations in bulk, for example when a driver
implements multiple transformations.
::
int crypto_register_alg(struct crypto_alg *alg);
int crypto_register_algs(struct crypto_alg *algs, int count);
The counterparts to those functions are listed below.
::
int crypto_unregister_alg(struct crypto_alg *alg);
int crypto_unregister_algs(struct crypto_alg *algs, int count);
Notice that both registration and unregistration functions do return a
value, so make sure to handle errors. A return code of zero implies
success. Any return code < 0 implies an error.
The bulk registration/unregistration functions register/unregister each
transformation in the given array of length count. They handle errors as
follows:
- crypto_register_algs() succeeds if and only if it successfully
registers all the given transformations. If an error occurs partway
through, then it rolls back successful registrations before returning
the error code. Note that if a driver needs to handle registration
errors for individual transformations, then it will need to use the
non-bulk function crypto_register_alg() instead.
- crypto_unregister_algs() tries to unregister all the given
transformations, continuing on error. It logs errors and always
returns zero.
Single-Block Symmetric Ciphers [CIPHER]
---------------------------------------
Example of transformations: aes, arc4, ...
This section describes the simplest of all transformation
implementations, that being the CIPHER type used for symmetric ciphers.
The CIPHER type is used for transformations which operate on exactly one
block at a time and there are no dependencies between blocks at all.
Registration specifics
~~~~~~~~~~~~~~~~~~~~~~
The registration of [CIPHER] algorithm is specific in that struct
crypto_alg field .cra_type is empty. The .cra_u.cipher has to be
filled in with proper callbacks to implement this transformation.
See struct cipher_alg below.
Cipher Definition With struct cipher_alg
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Struct cipher_alg defines a single block cipher.
Here are schematics of how these functions are called when operated from
other part of the kernel. Note that the .cia_setkey() call might happen
before or after any of these schematics happen, but must not happen
during any of these are in-flight.
::
KEY ---. PLAINTEXT ---.
v v
.cia_setkey() -> .cia_encrypt()
|
'-----> CIPHERTEXT
Please note that a pattern where .cia_setkey() is called multiple times
is also valid:
::
KEY1 --. PLAINTEXT1 --. KEY2 --. PLAINTEXT2 --.
v v v v
.cia_setkey() -> .cia_encrypt() -> .cia_setkey() -> .cia_encrypt()
| |
'---> CIPHERTEXT1 '---> CIPHERTEXT2
Multi-Block Ciphers
-------------------
Example of transformations: cbc(aes), ecb(arc4), ...
This section describes the multi-block cipher transformation
implementations. The multi-block ciphers are used for transformations
which operate on scatterlists of data supplied to the transformation
functions. They output the result into a scatterlist of data as well.
Registration Specifics
~~~~~~~~~~~~~~~~~~~~~~
The registration of multi-block cipher algorithms is one of the most
standard procedures throughout the crypto API.
Note, if a cipher implementation requires a proper alignment of data,
the caller should use the functions of crypto_skcipher_alignmask() to
identify a memory alignment mask. The kernel crypto API is able to
process requests that are unaligned. This implies, however, additional
overhead as the kernel crypto API needs to perform the realignment of
the data which may imply moving of data.
Cipher Definition With struct blkcipher_alg and ablkcipher_alg
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Struct blkcipher_alg defines a synchronous block cipher whereas struct
ablkcipher_alg defines an asynchronous block cipher.
Please refer to the single block cipher description for schematics of
the block cipher usage.
Specifics Of Asynchronous Multi-Block Cipher
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
There are a couple of specifics to the asynchronous interface.
First of all, some of the drivers will want to use the Generic
ScatterWalk in case the hardware needs to be fed separate chunks of the
scatterlist which contains the plaintext and will contain the
ciphertext. Please refer to the ScatterWalk interface offered by the
Linux kernel scatter / gather list implementation.
Hashing [HASH]
--------------
Example of transformations: crc32, md5, sha1, sha256,...
Registering And Unregistering The Transformation
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
There are multiple ways to register a HASH transformation, depending on
whether the transformation is synchronous [SHASH] or asynchronous
[AHASH] and the amount of HASH transformations we are registering. You
can find the prototypes defined in include/crypto/internal/hash.h:
::
int crypto_register_ahash(struct ahash_alg *alg);
int crypto_register_shash(struct shash_alg *alg);
int crypto_register_shashes(struct shash_alg *algs, int count);
The respective counterparts for unregistering the HASH transformation
are as follows:
::
int crypto_unregister_ahash(struct ahash_alg *alg);
int crypto_unregister_shash(struct shash_alg *alg);
int crypto_unregister_shashes(struct shash_alg *algs, int count);
Cipher Definition With struct shash_alg and ahash_alg
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Here are schematics of how these functions are called when operated from
other part of the kernel. Note that the .setkey() call might happen
before or after any of these schematics happen, but must not happen
during any of these are in-flight. Please note that calling .init()
followed immediately by .finish() is also a perfectly valid
transformation.
::
I) DATA -----------.
v
.init() -> .update() -> .final() ! .update() might not be called
^ | | at all in this scenario.
'----' '---> HASH
II) DATA -----------.-----------.
v v
.init() -> .update() -> .finup() ! .update() may not be called
^ | | at all in this scenario.
'----' '---> HASH
III) DATA -----------.
v
.digest() ! The entire process is handled
| by the .digest() call.
'---------------> HASH
Here is a schematic of how the .export()/.import() functions are called
when used from another part of the kernel.
::
KEY--. DATA--.
v v ! .update() may not be called
.setkey() -> .init() -> .update() -> .export() at all in this scenario.
^ | |
'-----' '--> PARTIAL_HASH
----------- other transformations happen here -----------
PARTIAL_HASH--. DATA1--.
v v
.import -> .update() -> .final() ! .update() may not be called
^ | | at all in this scenario.
'----' '--> HASH1
PARTIAL_HASH--. DATA2-.
v v
.import -> .finup()
|
'---------------> HASH2
Specifics Of Asynchronous HASH Transformation
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Some of the drivers will want to use the Generic ScatterWalk in case the
implementation needs to be fed separate chunks of the scatterlist which
contains the input data. The buffer containing the resulting hash will
always be properly aligned to .cra_alignmask so there is no need to
worry about this.
=======================
Linux Kernel Crypto API
=======================
:Author: Stephan Mueller
:Author: Marek Vasut
This documentation outlines the Linux kernel crypto API with its
concepts, details about developing cipher implementations, employment of the API
for cryptographic use cases, as well as programming examples.
.. class:: toc-title
Table of contents
.. toctree::
:maxdepth: 2
intro
architecture
devel-algos
userspace-if
api
api-samples
Kernel Crypto API Interface Specification
=========================================
Introduction
------------
The kernel crypto API offers a rich set of cryptographic ciphers as well
as other data transformation mechanisms and methods to invoke these.
This document contains a description of the API and provides example
code.
To understand and properly use the kernel crypto API a brief explanation
of its structure is given. Based on the architecture, the API can be
separated into different components. Following the architecture
specification, hints to developers of ciphers are provided. Pointers to
the API function call documentation are given at the end.
The kernel crypto API refers to all algorithms as "transformations".
Therefore, a cipher handle variable usually has the name "tfm". Besides
cryptographic operations, the kernel crypto API also knows compression
transformations and handles them the same way as ciphers.
The kernel crypto API serves the following entity types:
- consumers requesting cryptographic services
- data transformation implementations (typically ciphers) that can be
called by consumers using the kernel crypto API
This specification is intended for consumers of the kernel crypto API as
well as for developers implementing ciphers. This API specification,
however, does not discuss all API calls available to data transformation
implementations (i.e. implementations of ciphers and other
transformations (such as CRC or even compression algorithms) that can
register with the kernel crypto API).
Note: The terms "transformation" and cipher algorithm are used
interchangeably.
Terminology
-----------
The transformation implementation is an actual code or interface to
hardware which implements a certain transformation with precisely
defined behavior.
The transformation object (TFM) is an instance of a transformation
implementation. There can be multiple transformation objects associated
with a single transformation implementation. Each of those
transformation objects is held by a crypto API consumer or another
transformation. Transformation object is allocated when a crypto API
consumer requests a transformation implementation. The consumer is then
provided with a structure, which contains a transformation object (TFM).
The structure that contains transformation objects may also be referred
to as a "cipher handle". Such a cipher handle is always subject to the
following phases that are reflected in the API calls applicable to such
a cipher handle:
1. Initialization of a cipher handle.
2. Execution of all intended cipher operations applicable for the handle
where the cipher handle must be furnished to every API call.
3. Destruction of a cipher handle.
When using the initialization API calls, a cipher handle is created and
returned to the consumer. Therefore, please refer to all initialization
API calls that refer to the data structure type a consumer is expected
to receive and subsequently to use. The initialization API calls have
all the same naming conventions of crypto_alloc\*.
The transformation context is private data associated with the
transformation object.
User Space Interface
====================
Introduction
------------
The concepts of the kernel crypto API visible to kernel space is fully
applicable to the user space interface as well. Therefore, the kernel
crypto API high level discussion for the in-kernel use cases applies
here as well.
The major difference, however, is that user space can only act as a
consumer and never as a provider of a transformation or cipher
algorithm.
The following covers the user space interface exported by the kernel
crypto API. A working example of this description is libkcapi that can
be obtained from [1]. That library can be used by user space
applications that require cryptographic services from the kernel.
Some details of the in-kernel kernel crypto API aspects do not apply to
user space, however. This includes the difference between synchronous
and asynchronous invocations. The user space API call is fully
synchronous.
[1] http://www.chronox.de/libkcapi.html
User Space API General Remarks
------------------------------
The kernel crypto API is accessible from user space. Currently, the
following ciphers are accessible:
- Message digest including keyed message digest (HMAC, CMAC)
- Symmetric ciphers
- AEAD ciphers
- Random Number Generators
The interface is provided via socket type using the type AF_ALG. In
addition, the setsockopt option type is SOL_ALG. In case the user space
header files do not export these flags yet, use the following macros:
::
#ifndef AF_ALG
#define AF_ALG 38
#endif
#ifndef SOL_ALG
#define SOL_ALG 279
#endif
A cipher is accessed with the same name as done for the in-kernel API
calls. This includes the generic vs. unique naming schema for ciphers as
well as the enforcement of priorities for generic names.
To interact with the kernel crypto API, a socket must be created by the
user space application. User space invokes the cipher operation with the
send()/write() system call family. The result of the cipher operation is
obtained with the read()/recv() system call family.
The following API calls assume that the socket descriptor is already
opened by the user space application and discusses only the kernel
crypto API specific invocations.
To initialize the socket interface, the following sequence has to be
performed by the consumer:
1. Create a socket of type AF_ALG with the struct sockaddr_alg
parameter specified below for the different cipher types.
2. Invoke bind with the socket descriptor
3. Invoke accept with the socket descriptor. The accept system call
returns a new file descriptor that is to be used to interact with the
particular cipher instance. When invoking send/write or recv/read
system calls to send data to the kernel or obtain data from the
kernel, the file descriptor returned by accept must be used.
In-place Cipher operation
-------------------------
Just like the in-kernel operation of the kernel crypto API, the user
space interface allows the cipher operation in-place. That means that
the input buffer used for the send/write system call and the output
buffer used by the read/recv system call may be one and the same. This
is of particular interest for symmetric cipher operations where a
copying of the output data to its final destination can be avoided.
If a consumer on the other hand wants to maintain the plaintext and the
ciphertext in different memory locations, all a consumer needs to do is
to provide different memory pointers for the encryption and decryption
operation.
Message Digest API
------------------
The message digest type to be used for the cipher operation is selected
when invoking the bind syscall. bind requires the caller to provide a
filled struct sockaddr data structure. This data structure must be
filled as follows:
::
struct sockaddr_alg sa = {
.salg_family = AF_ALG,
.salg_type = "hash", /* this selects the hash logic in the kernel */
.salg_name = "sha1" /* this is the cipher name */
};
The salg_type value "hash" applies to message digests and keyed message
digests. Though, a keyed message digest is referenced by the appropriate
salg_name. Please see below for the setsockopt interface that explains
how the key can be set for a keyed message digest.
Using the send() system call, the application provides the data that
should be processed with the message digest. The send system call allows
the following flags to be specified:
- MSG_MORE: If this flag is set, the send system call acts like a
message digest update function where the final hash is not yet
calculated. If the flag is not set, the send system call calculates
the final message digest immediately.
With the recv() system call, the application can read the message digest
from the kernel crypto API. If the buffer is too small for the message
digest, the flag MSG_TRUNC is set by the kernel.
In order to set a message digest key, the calling application must use
the setsockopt() option of ALG_SET_KEY. If the key is not set the HMAC
operation is performed without the initial HMAC state change caused by
the key.
Symmetric Cipher API
--------------------
The operation is very similar to the message digest discussion. During
initialization, the struct sockaddr data structure must be filled as
follows:
::
struct sockaddr_alg sa = {
.salg_family = AF_ALG,
.salg_type = "skcipher", /* this selects the symmetric cipher */
.salg_name = "cbc(aes)" /* this is the cipher name */
};
Before data can be sent to the kernel using the write/send system call
family, the consumer must set the key. The key setting is described with
the setsockopt invocation below.
Using the sendmsg() system call, the application provides the data that
should be processed for encryption or decryption. In addition, the IV is
specified with the data structure provided by the sendmsg() system call.
The sendmsg system call parameter of struct msghdr is embedded into the
struct cmsghdr data structure. See recv(2) and cmsg(3) for more
information on how the cmsghdr data structure is used together with the
send/recv system call family. That cmsghdr data structure holds the
following information specified with a separate header instances:
- specification of the cipher operation type with one of these flags:
- ALG_OP_ENCRYPT - encryption of data
- ALG_OP_DECRYPT - decryption of data
- specification of the IV information marked with the flag ALG_SET_IV
The send system call family allows the following flag to be specified:
- MSG_MORE: If this flag is set, the send system call acts like a
cipher update function where more input data is expected with a
subsequent invocation of the send system call.
Note: The kernel reports -EINVAL for any unexpected data. The caller
must make sure that all data matches the constraints given in
/proc/crypto for the selected cipher.
With the recv() system call, the application can read the result of the
cipher operation from the kernel crypto API. The output buffer must be
at least as large as to hold all blocks of the encrypted or decrypted
data. If the output data size is smaller, only as many blocks are
returned that fit into that output buffer size.
AEAD Cipher API
---------------
The operation is very similar to the symmetric cipher discussion. During
initialization, the struct sockaddr data structure must be filled as
follows:
::
struct sockaddr_alg sa = {
.salg_family = AF_ALG,
.salg_type = "aead", /* this selects the symmetric cipher */
.salg_name = "gcm(aes)" /* this is the cipher name */
};
Before data can be sent to the kernel using the write/send system call
family, the consumer must set the key. The key setting is described with
the setsockopt invocation below.
In addition, before data can be sent to the kernel using the write/send
system call family, the consumer must set the authentication tag size.
To set the authentication tag size, the caller must use the setsockopt
invocation described below.
Using the sendmsg() system call, the application provides the data that
should be processed for encryption or decryption. In addition, the IV is
specified with the data structure provided by the sendmsg() system call.
The sendmsg system call parameter of struct msghdr is embedded into the
struct cmsghdr data structure. See recv(2) and cmsg(3) for more
information on how the cmsghdr data structure is used together with the
send/recv system call family. That cmsghdr data structure holds the
following information specified with a separate header instances:
- specification of the cipher operation type with one of these flags:
- ALG_OP_ENCRYPT - encryption of data
- ALG_OP_DECRYPT - decryption of data
- specification of the IV information marked with the flag ALG_SET_IV
- specification of the associated authentication data (AAD) with the
flag ALG_SET_AEAD_ASSOCLEN. The AAD is sent to the kernel together
with the plaintext / ciphertext. See below for the memory structure.
The send system call family allows the following flag to be specified:
- MSG_MORE: If this flag is set, the send system call acts like a
cipher update function where more input data is expected with a
subsequent invocation of the send system call.
Note: The kernel reports -EINVAL for any unexpected data. The caller
must make sure that all data matches the constraints given in
/proc/crypto for the selected cipher.
With the recv() system call, the application can read the result of the
cipher operation from the kernel crypto API. The output buffer must be
at least as large as defined with the memory structure below. If the
output data size is smaller, the cipher operation is not performed.
The authenticated decryption operation may indicate an integrity error.
Such breach in integrity is marked with the -EBADMSG error code.
AEAD Memory Structure
~~~~~~~~~~~~~~~~~~~~~
The AEAD cipher operates with the following information that is
communicated between user and kernel space as one data stream:
- plaintext or ciphertext
- associated authentication data (AAD)
- authentication tag
The sizes of the AAD and the authentication tag are provided with the
sendmsg and setsockopt calls (see there). As the kernel knows the size
of the entire data stream, the kernel is now able to calculate the right
offsets of the data components in the data stream.
The user space caller must arrange the aforementioned information in the
following order:
- AEAD encryption input: AAD \|\| plaintext
- AEAD decryption input: AAD \|\| ciphertext \|\| authentication tag
The output buffer the user space caller provides must be at least as
large to hold the following data:
- AEAD encryption output: ciphertext \|\| authentication tag
- AEAD decryption output: plaintext
Random Number Generator API
---------------------------
Again, the operation is very similar to the other APIs. During
initialization, the struct sockaddr data structure must be filled as
follows:
::
struct sockaddr_alg sa = {
.salg_family = AF_ALG,
.salg_type = "rng", /* this selects the symmetric cipher */
.salg_name = "drbg_nopr_sha256" /* this is the cipher name */
};
Depending on the RNG type, the RNG must be seeded. The seed is provided
using the setsockopt interface to set the key. For example, the
ansi_cprng requires a seed. The DRBGs do not require a seed, but may be
seeded.
Using the read()/recvmsg() system calls, random numbers can be obtained.
The kernel generates at most 128 bytes in one call. If user space
requires more data, multiple calls to read()/recvmsg() must be made.
WARNING: The user space caller may invoke the initially mentioned accept
system call multiple times. In this case, the returned file descriptors
have the same state.
Zero-Copy Interface
-------------------
In addition to the send/write/read/recv system call family, the AF_ALG
interface can be accessed with the zero-copy interface of
splice/vmsplice. As the name indicates, the kernel tries to avoid a copy
operation into kernel space.
The zero-copy operation requires data to be aligned at the page
boundary. Non-aligned data can be used as well, but may require more
operations of the kernel which would defeat the speed gains obtained
from the zero-copy interface.
The system-interent limit for the size of one zero-copy operation is 16
pages. If more data is to be sent to AF_ALG, user space must slice the
input into segments with a maximum size of 16 pages.
Zero-copy can be used with the following code example (a complete
working example is provided with libkcapi):
::
int pipes[2];
pipe(pipes);
/* input data in iov */
vmsplice(pipes[1], iov, iovlen, SPLICE_F_GIFT);
/* opfd is the file descriptor returned from accept() system call */
splice(pipes[0], NULL, opfd, NULL, ret, 0);
read(opfd, out, outlen);
Setsockopt Interface
--------------------
In addition to the read/recv and send/write system call handling to send
and retrieve data subject to the cipher operation, a consumer also needs
to set the additional information for the cipher operation. This
additional information is set using the setsockopt system call that must
be invoked with the file descriptor of the open cipher (i.e. the file
descriptor returned by the accept system call).
Each setsockopt invocation must use the level SOL_ALG.
The setsockopt interface allows setting the following data using the
mentioned optname:
- ALG_SET_KEY -- Setting the key. Key setting is applicable to:
- the skcipher cipher type (symmetric ciphers)
- the hash cipher type (keyed message digests)
- the AEAD cipher type
- the RNG cipher type to provide the seed
- ALG_SET_AEAD_AUTHSIZE -- Setting the authentication tag size for
AEAD ciphers. For a encryption operation, the authentication tag of
the given size will be generated. For a decryption operation, the
provided ciphertext is assumed to contain an authentication tag of
the given size (see section about AEAD memory layout below).
User space API example
----------------------
Please see [1] for libkcapi which provides an easy-to-use wrapper around
the aforementioned Netlink kernel interface. [1] also contains a test
application that invokes all libkcapi API calls.
[1] http://www.chronox.de/libkcapi.html
......@@ -58,6 +58,7 @@ needed).
gpu/index
security/index
sound/index
crypto/index
Korean translations
-------------------
......
......@@ -556,18 +556,8 @@ static int aead_recvmsg_sync(struct socket *sock, struct msghdr *msg, int flags)
lock_sock(sk);
/*
* AEAD memory structure: For encryption, the tag is appended to the
* ciphertext which implies that the memory allocated for the ciphertext
* must be increased by the tag length. For decryption, the tag
* is expected to be concatenated to the ciphertext. The plaintext
* therefore has a memory size of the ciphertext minus the tag length.
*
* The memory structure for cipher operation has the following
* structure:
* AEAD encryption input: assoc data || plaintext
* AEAD encryption output: cipherntext || auth tag
* AEAD decryption input: assoc data || ciphertext || auth tag
* AEAD decryption output: plaintext
* Please see documentation of aead_request_set_crypt for the
* description of the AEAD memory structure expected from the caller.
*/
if (ctx->more) {
......
......@@ -55,14 +55,14 @@
* The scatter list pointing to the input data must contain:
*
* * for RFC4106 ciphers, the concatenation of
* associated authentication data || IV || plaintext or ciphertext. Note, the
* same IV (buffer) is also set with the aead_request_set_crypt call. Note,
* the API call of aead_request_set_ad must provide the length of the AAD and
* the IV. The API call of aead_request_set_crypt only points to the size of
* the input plaintext or ciphertext.
* associated authentication data || IV || plaintext or ciphertext. Note, the
* same IV (buffer) is also set with the aead_request_set_crypt call. Note,
* the API call of aead_request_set_ad must provide the length of the AAD and
* the IV. The API call of aead_request_set_crypt only points to the size of
* the input plaintext or ciphertext.
*
* * for "normal" AEAD ciphers, the concatenation of
* associated authentication data || plaintext or ciphertext.
* associated authentication data || plaintext or ciphertext.
*
* It is important to note that if multiple scatter gather list entries form
* the input data mentioned above, the first entry must not point to a NULL
......@@ -452,7 +452,7 @@ static inline void aead_request_free(struct aead_request *req)
* completes
*
* The callback function is registered with the aead_request handle and
* must comply with the following template
* must comply with the following template::
*
* void callback_function(struct crypto_async_request *req, int error)
*/
......@@ -483,30 +483,18 @@ static inline void aead_request_set_callback(struct aead_request *req,
* destination is the ciphertext. For a decryption operation, the use is
* reversed - the source is the ciphertext and the destination is the plaintext.
*
* For both src/dst the layout is associated data, plain/cipher text,
* authentication tag.
*
* The content of the AD in the destination buffer after processing
* will either be untouched, or it will contain a copy of the AD
* from the source buffer. In order to ensure that it always has
* a copy of the AD, the user must copy the AD over either before
* or after processing. Of course this is not relevant if the user
* is doing in-place processing where src == dst.
*
* IMPORTANT NOTE AEAD requires an authentication tag (MAC). For decryption,
* the caller must concatenate the ciphertext followed by the
* authentication tag and provide the entire data stream to the
* decryption operation (i.e. the data length used for the
* initialization of the scatterlist and the data length for the
* decryption operation is identical). For encryption, however,
* the authentication tag is created while encrypting the data.
* The destination buffer must hold sufficient space for the
* ciphertext and the authentication tag while the encryption
* invocation must only point to the plaintext data size. The
* following code snippet illustrates the memory usage
* buffer = kmalloc(ptbuflen + (enc ? authsize : 0));
* sg_init_one(&sg, buffer, ptbuflen + (enc ? authsize : 0));
* aead_request_set_crypt(req, &sg, &sg, ptbuflen, iv);
* The memory structure for cipher operation has the following structure:
*
* - AEAD encryption input: assoc data || plaintext
* - AEAD encryption output: assoc data || cipherntext || auth tag
* - AEAD decryption input: assoc data || ciphertext || auth tag
* - AEAD decryption output: assoc data || plaintext
*
* Albeit the kernel requires the presence of the AAD buffer, however,
* the kernel does not fill the AAD buffer in the output case. If the
* caller wants to have that data buffer filled, the caller must either
* use an in-place cipher operation (i.e. same memory location for
* input/output memory location).
*/
static inline void aead_request_set_crypt(struct aead_request *req,
struct scatterlist *src,
......
......@@ -13,6 +13,27 @@
#ifndef _CRYPTO_DH_
#define _CRYPTO_DH_
/**
* DOC: DH Helper Functions
*
* To use DH with the KPP cipher API, the following data structure and
* functions should be used.
*
* To use DH with KPP, the following functions should be used to operate on
* a DH private key. The packet private key that can be set with
* the KPP API function call of crypto_kpp_set_secret.
*/
/**
* struct dh - define a DH private key
*
* @key: Private DH key
* @p: Diffie-Hellman parameter P
* @g: Diffie-Hellman generator G
* @key_size: Size of the private DH key
* @p_size: Size of DH parameter P
* @g_size: Size of DH generator G
*/
struct dh {
void *key;
void *p;
......@@ -22,8 +43,45 @@ struct dh {
unsigned int g_size;
};
/**
* crypto_dh_key_len() - Obtain the size of the private DH key
* @params: private DH key
*
* This function returns the packet DH key size. A caller can use that
* with the provided DH private key reference to obtain the required
* memory size to hold a packet key.
*
* Return: size of the key in bytes
*/
int crypto_dh_key_len(const struct dh *params);
/**
* crypto_dh_encode_key() - encode the private key
* @buf: Buffer allocated by the caller to hold the packet DH
* private key. The buffer should be at least crypto_dh_key_len
* bytes in size.
* @len: Length of the packet private key buffer
* @params: Buffer with the caller-specified private key
*
* The DH implementations operate on a packet representation of the private
* key.
*
* Return: -EINVAL if buffer has insufficient size, 0 on success
*/
int crypto_dh_encode_key(char *buf, unsigned int len, const struct dh *params);
/**
* crypto_dh_decode_key() - decode a private key
* @buf: Buffer holding a packet key that should be decoded
* @len: Lenth of the packet private key buffer
* @params: Buffer allocated by the caller that is filled with the
* unpacket DH private key.
*
* The unpacking obtains the private key by pointing @p to the correct location
* in @buf. Thus, both pointers refer to the same memory.
*
* Return: -EINVAL if buffer has insufficient size, 0 on success
*/
int crypto_dh_decode_key(const char *buf, unsigned int len, struct dh *params);
#endif
......@@ -13,18 +13,76 @@
#ifndef _CRYPTO_ECDH_
#define _CRYPTO_ECDH_
/**
* DOC: ECDH Helper Functions
*
* To use ECDH with the KPP cipher API, the following data structure and
* functions should be used.
*
* The ECC curves known to the ECDH implementation are specified in this
* header file.
*
* To use ECDH with KPP, the following functions should be used to operate on
* an ECDH private key. The packet private key that can be set with
* the KPP API function call of crypto_kpp_set_secret.
*/
/* Curves IDs */
#define ECC_CURVE_NIST_P192 0x0001
#define ECC_CURVE_NIST_P256 0x0002
/**
* struct ecdh - define an ECDH private key
*
* @curve_id: ECC curve the key is based on.
* @key: Private ECDH key
* @key_size: Size of the private ECDH key
*/
struct ecdh {
unsigned short curve_id;
char *key;
unsigned short key_size;
};
/**
* crypto_ecdh_key_len() - Obtain the size of the private ECDH key
* @params: private ECDH key
*
* This function returns the packet ECDH key size. A caller can use that
* with the provided ECDH private key reference to obtain the required
* memory size to hold a packet key.
*
* Return: size of the key in bytes
*/
int crypto_ecdh_key_len(const struct ecdh *params);
/**
* crypto_ecdh_encode_key() - encode the private key
* @buf: Buffer allocated by the caller to hold the packet ECDH
* private key. The buffer should be at least crypto_ecdh_key_len
* bytes in size.
* @len: Length of the packet private key buffer
* @p: Buffer with the caller-specified private key
*
* The ECDH implementations operate on a packet representation of the private
* key.
*
* Return: -EINVAL if buffer has insufficient size, 0 on success
*/
int crypto_ecdh_encode_key(char *buf, unsigned int len, const struct ecdh *p);
/**
* crypto_ecdh_decode_key() - decode a private key
* @buf: Buffer holding a packet key that should be decoded
* @len: Lenth of the packet private key buffer
* @p: Buffer allocated by the caller that is filled with the
* unpacket ECDH private key.
*
* The unpacking obtains the private key by pointing @p to the correct location
* in @buf. Thus, both pointers refer to the same memory.
*
* Return: -EINVAL if buffer has insufficient size, 0 on success
*/
int crypto_ecdh_decode_key(const char *buf, unsigned int len, struct ecdh *p);
#endif
......@@ -605,7 +605,7 @@ static inline struct ahash_request *ahash_request_cast(
* the cipher operation completes.
*
* The callback function is registered with the &ahash_request handle and
* must comply with the following template
* must comply with the following template::
*
* void callback_function(struct crypto_async_request *req, int error)
*/
......
......@@ -71,7 +71,7 @@ struct crypto_kpp {
*
* @reqsize: Request context size required by algorithm
* implementation
* @base Common crypto API algorithm data structure
* @base: Common crypto API algorithm data structure
*/
struct kpp_alg {
int (*set_secret)(struct crypto_kpp *tfm, void *buffer,
......@@ -89,7 +89,7 @@ struct kpp_alg {
};
/**
* DOC: Generic Key-agreement Protocol Primitevs API
* DOC: Generic Key-agreement Protocol Primitives API
*
* The KPP API is used with the algorithm type
* CRYPTO_ALG_TYPE_KPP (listed as type "kpp" in /proc/crypto)
......@@ -264,6 +264,12 @@ struct kpp_secret {
* Function invokes the specific kpp operation for a given alg.
*
* @tfm: tfm handle
* @buffer: Buffer holding the packet representation of the private
* key. The structure of the packet key depends on the particular
* KPP implementation. Packing and unpacking helpers are provided
* for ECDH and DH (see the respective header files for those
* implementations).
* @len: Length of the packet private key buffer.
*
* Return: zero on success; error code in case of error
*/
......@@ -279,7 +285,10 @@ static inline int crypto_kpp_set_secret(struct crypto_kpp *tfm, void *buffer,
* crypto_kpp_generate_public_key() - Invoke kpp operation
*
* Function invokes the specific kpp operation for generating the public part
* for a given kpp algorithm
* for a given kpp algorithm.
*
* To generate a private key, the caller should use a random number generator.
* The output of the requested length serves as the private key.
*
* @req: kpp key request
*
......
......@@ -516,7 +516,7 @@ static inline void skcipher_request_zero(struct skcipher_request *req)
* skcipher_request_set_callback() - set asynchronous callback function
* @req: request handle
* @flags: specify zero or an ORing of the flags
* CRYPTO_TFM_REQ_MAY_BACKLOG the request queue may back log and
* CRYPTO_TFM_REQ_MAY_BACKLOG the request queue may back log and
* increase the wait queue beyond the initial maximum size;
* CRYPTO_TFM_REQ_MAY_SLEEP the request processing may sleep
* @compl: callback function pointer to be registered with the request handle
......@@ -533,7 +533,7 @@ static inline void skcipher_request_zero(struct skcipher_request *req)
* cipher operation completes.
*
* The callback function is registered with the skcipher_request handle and
* must comply with the following template
* must comply with the following template::
*
* void callback_function(struct crypto_async_request *req, int error)
*/
......
......@@ -963,7 +963,7 @@ static inline void ablkcipher_request_free(struct ablkcipher_request *req)
* ablkcipher_request_set_callback() - set asynchronous callback function
* @req: request handle
* @flags: specify zero or an ORing of the flags
* CRYPTO_TFM_REQ_MAY_BACKLOG the request queue may back log and
* CRYPTO_TFM_REQ_MAY_BACKLOG the request queue may back log and
* increase the wait queue beyond the initial maximum size;
* CRYPTO_TFM_REQ_MAY_SLEEP the request processing may sleep
* @compl: callback function pointer to be registered with the request handle
......@@ -980,7 +980,7 @@ static inline void ablkcipher_request_free(struct ablkcipher_request *req)
* cipher operation completes.
*
* The callback function is registered with the ablkcipher_request handle and
* must comply with the following template
* must comply with the following template::
*
* void callback_function(struct crypto_async_request *req, int error)
*/
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment