path: root/lib/crypto/doc/src/crypto.xml

<?xml version="1.0" encoding="utf-8" ?>
<!DOCTYPE erlref SYSTEM "erlref.dtd">

<erlref>
  <header>
    <copyright>
      <year>1999</year><year>2018</year>
      <holder>Ericsson AB. All Rights Reserved.</holder>
    </copyright>
    <legalnotice>
      Licensed under the Apache License, Version 2.0 (the "License");
      you may not use this file except in compliance with the License.
      You may obtain a copy of the License at

          http://www.apache.org/licenses/LICENSE-2.0

      Unless required by applicable law or agreed to in writing, software
      distributed under the License is distributed on an "AS IS" BASIS,
      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
      See the License for the specific language governing permissions and
      limitations under the License.
    </legalnotice>

    <title>crypto</title>
  </header>
  <module>crypto</module>
  <modulesummary>Crypto Functions</modulesummary>
  <description>
    <p>This module provides a set of cryptographic functions.
      </p>
    <list type="bulleted">
      <item>
	<p>Hash functions -
	<url href="http://csrc.nist.gov/publications/fips/fips180-4/fips-180-4.pdf"> Secure Hash Standard</url>,
	<url href="http://www.ietf.org/rfc/rfc1321.txt"> The MD5 Message Digest Algorithm (RFC 1321)</url> and
	<url href="http://www.ietf.org/rfc/rfc1320.txt">The MD4 Message Digest Algorithm (RFC 1320)</url>
	</p>
      </item>
      <item>
        <p>Hmac functions - <url href="http://www.ietf.org/rfc/rfc2104.txt"> Keyed-Hashing for Message Authentication (RFC 2104) </url></p>
      </item>
      <item>
        <p>Cmac functions - <url href="http://www.ietf.org/rfc/rfc4493.txt">The AES-CMAC Algorithm (RFC 4493)</url></p>
      </item>
      <item>
        <p>Block ciphers - <url href="http://csrc.nist.gov/groups/ST/toolkit/block_ciphers.html"> </url> DES and AES in
        Block Cipher Modes - <url href="http://csrc.nist.gov/groups/ST/toolkit/BCM/index.html"> ECB, CBC, CFB, OFB, CTR and GCM </url></p>
      </item>
      <item>
        <p>GCM: <url href="https://csrc.nist.gov/publications/detail/sp/800-38d/final">Dworkin, M.,
	"Recommendation for Block Cipher Modes of Operation: Galois/Counter Mode (GCM) and GMAC",
        National Institute of Standards and Technology SP 800-38D, November 2007</url>.
      </p>
      </item>
      <item>
        <p><url href="http://www.ietf.org/rfc/rfc1321.txt"> RSA encryption RFC 1321 </url> </p>
      </item>
      <item>
        <p>Digital signatures <url href="http://csrc.nist.gov/publications/drafts/fips186-3/fips_186-3.pdf">Digital Signature Standard (DSS)</url> and<url href="http://csrc.nist.gov/groups/STM/cavp/documents/dss2/ecdsa2vs.pdf"> Elliptic Curve Digital
	Signature Algorithm (ECDSA) </url> </p>
      </item>
      <item>
        <p><url href="http://www.ietf.org/rfc/rfc2945.txt"> Secure Remote Password Protocol (SRP - RFC 2945) </url></p>
      </item>
    </list>

    <note>
      <p>The actual supported algorithms and features depends on their availability in the actual libcrypto used.
      See the <seealso marker="crypto:crypto_app">crypto (App)</seealso> about dependencies.
      </p>
      <p>Enabling FIPS mode will also disable algorithms and features.
      </p>
    </note>

    <p>The <seealso marker="users_guide">CRYPTO User's Guide</seealso> has more information on
    FIPS, Engines and Algorithm Details like key lengths.
    </p>
  </description>

  <datatypes>
    <datatype_title>Ciphers</datatype_title>
    <datatype>
      <name name="stream_cipher"/>
      <desc>
	<p>Stream ciphers for
	<seealso marker="#stream_encrypt-2">stream_encrypt/2</seealso> and
	<seealso marker="#stream_decrypt-2">stream_decrypt/2</seealso> .
	</p>
      </desc>
    </datatype>

    <datatype>
      <name name="block_cipher_with_iv"/>
      <name name="cbc_cipher"/>
      <name name="cfb_cipher"/>
      <desc>
	<p>Block ciphers with initialization vector for
	<seealso marker="#block_encrypt-4">block_encrypt/4</seealso> and
	<seealso marker="#block_decrypt-4">block_decrypt/4</seealso> .
	</p>
      </desc>
    </datatype>

    <datatype>
      <name name="block_cipher_without_iv"/>
      <name name="ecb_cipher"/>
      <desc>
	<p>Block ciphers without initialization vector for
	<seealso marker="#block_encrypt-3">block_encrypt/3</seealso> and
	<seealso marker="#block_decrypt-3">block_decrypt/3</seealso> .
	</p>
      </desc>
    </datatype>

    <datatype>
      <name name="aead_cipher"/>
      <desc>
	<p>Ciphers with simultaneous MAC-calculation or MAC-checking.
	<seealso marker="#block_encrypt-4">block_encrypt/4</seealso> and
	<seealso marker="#block_decrypt-4">block_decrypt/4</seealso> .
	</p>
      </desc>
    </datatype>

    <datatype_title>Digests</datatype_title>
    <datatype>
      <name name="sha1"/>
      <name name="sha2"/>
      <name name="sha3"/>
      <desc>
      </desc>
    </datatype>

    <datatype>
      <name name="compatibility_only_hash"/>
      <desc>
	<p>The <c>compatibility_only_hash()</c> algorithms are recommended only for compatibility with existing applications.</p>
      </desc>
    </datatype>

    <datatype>
      <name name="rsa_digest_type"/>
      <desc>
      </desc>
    </datatype>

    <datatype>
      <name name="dss_digest_type"/>
      <desc>
      </desc>
    </datatype>

    <datatype>
      <name name="ecdsa_digest_type"/>
      <desc>
      </desc>
    </datatype>

    <datatype_title>Elliptic Curves</datatype_title>
    <datatype>
      <name name="ec_named_curve"/>
      <name name="edwards_curve"/>
      <desc>
	<p>Note that some curves are disabled if FIPS is enabled.</p>
      </desc>
    </datatype>

    <datatype>
      <name name="ec_explicit_curve"/>
      <name name="ec_field"/>
      <name name="ec_curve"/>
      <desc>
	<p>Parametric curve definition.</p>
      </desc>
    </datatype>

    <datatype>
      <name name="ec_prime_field"/>
      <name name="ec_characteristic_two_field"/>
      <name name="ec_basis"/>
      <desc>
	<p>Curve definition details.</p>
      </desc>
    </datatype>

    <datatype_title>Keys</datatype_title>
    <datatype>
      <name name="key"/>
      <name name="des3_key"/>
      <desc>
	<p>For keylengths, iv-sizes and blocksizes see the
	<seealso marker="crypto:algorithm_details#ciphers">User's Guide</seealso>.
	</p>
	<p>A key for des3 is a list of three iolists</p>
      </desc>
    </datatype>

    <datatype>
      <name name="key_integer"/>
      <desc>
	<p>Always <c>binary()</c> when used as return value</p>
      </desc>
    </datatype>

    <datatype_title>Public/Private Keys</datatype_title>
    <datatype>
      <name name="rsa_public"/>
      <name name="rsa_private"/>
      <name name="rsa_params"/>
      <desc>
	<code>rsa_public() = [E, N]</code>
	<code>rsa_private() = [E, N, D] | [E, N, D, P1, P2, E1, E2, C]</code>
	<p>Where E is the public exponent, N is public modulus and D is
	the private exponent. The longer key format contains redundant
	information that will make the calculation faster. P1,P2 are first
	and second prime factors. E1,E2 are first and second exponents. C
	is the CRT coefficient. Terminology is taken from <url href="http://www.ietf.org/rfc/rfc3477.txt"> RFC 3447</url>.</p>
      </desc>
    </datatype>

    <datatype>
      <name name="dss_public"/>
      <name name="dss_private"/>
      <desc>
	<code>dss_public() = [P, Q, G, Y] </code>
	<p>Where P, Q and G are the dss parameters and Y is the public key.</p>

	<code>dss_private() = [P, Q, G, X] </code>
	<p>Where P, Q and G are the dss parameters and X is the private key.</p>
      </desc>
    </datatype>

    <datatype>
      <name name="ecdsa_public"/>
      <name name="ecdsa_private"/>
      <name name="ecdsa_params"/>
      <desc>
      </desc>
    </datatype>

    <datatype>
      <name name="srp_public"/>
      <name name="srp_private"/>
      <desc>
	<code>srp_public() = key_integer() </code>
	<p>Where is <c>A</c> or <c>B</c> from <url href="http://srp.stanford.edu/design.html">SRP design</url></p>
	
	<code>srp_private() = key_integer() </code>
	<p>Where is  <c>a</c> or <c>b</c> from <url href="http://srp.stanford.edu/design.html">SRP design</url></p>
      </desc>
    </datatype>

    <datatype>
      <name name="srp_gen_params"/>
      <name name="srp_comp_params"/>
      <desc>
	<marker id="type-srp_user_gen_params"/>
	<code>srp_user_gen_params() = [DerivedKey::binary(), Prime::binary(), Generator::binary(), Version::atom()]</code>
	<marker id="type-srp_host_gen_params"/>
	<code>srp_host_gen_params() = [Verifier::binary(), Prime::binary(), Version::atom() ]</code>
	<marker id="type-srp_user_comp_params"/>
	<code>srp_user_comp_params() = [DerivedKey::binary(), Prime::binary(), Generator::binary(), Version::atom() | ScramblerArg::list()]</code>
	<marker id="type-srp_host_comp_params"/>
	<code>srp_host_comp_params() = [Verifier::binary(), Prime::binary(), Version::atom() | ScramblerArg::list()]</code>
	<p>Where Verifier is <c>v</c>, Generator is <c>g</c> and Prime is<c> N</c>, DerivedKey is <c>X</c>, and Scrambler is
	<c>u</c> (optional will be generated if not provided) from <url href="http://srp.stanford.edu/design.html">SRP design</url>
	Version = '3' |  '6' |  '6a'
	</p>
      </desc>
    </datatype>

    <datatype_title>Public Key Ciphers</datatype_title>

    <datatype>
      <name name="pk_encrypt_decrypt_algs"/>
      <desc>
	<p>Algorithms for public key encrypt/decrypt. Only RSA is supported.</p>
      </desc>
    </datatype>

    <datatype>
      <name name="pk_encrypt_decrypt_opts"/>
      <name name="rsa_opt"/>
      <name name="rsa_padding"/>
      <desc>
	<p>Options for public key encrypt/decrypt. Only RSA is supported.</p>
      </desc>
    </datatype>

    <datatype>
      <name name="rsa_compat_opts"/>
      <desc>
	<p>Those option forms are kept only for compatibility and should not be used in new code.</p>
      </desc>
    </datatype>

    <datatype_title>Public Key Sign and Verify</datatype_title>

    <datatype>
      <name name="pk_sign_verify_algs"/>
      <desc>
	<p>Algorithms for sign and verify.</p>
      </desc>
    </datatype>

    <datatype>
      <name name="pk_sign_verify_opts"/>
      <name name="rsa_sign_verify_opt"/>
      <name name="rsa_sign_verify_padding"/>
      <desc>
	<p>Options for sign and verify.</p>
      </desc>
    </datatype>

    <datatype_title>Diffie-Hellman Keys and parameters</datatype_title>
    <datatype>
      <name name="dh_public"/>
      <name name="dh_private"/>
      <desc>
      </desc>
    </datatype>

    <datatype>
      <name name="dh_params"/>
      <desc>
	<code>dh_params() = [P, G] | [P, G, PrivateKeyBitLength]</code>
      </desc>
    </datatype>

    <datatype>
      <name name="ecdh_public"/>
      <name name="ecdh_private"/>
      <name name="ecdh_params"/>
      <desc>
      </desc>
    </datatype>

    <datatype_title>Types for Engines</datatype_title>

    <datatype>
      <name name="engine_key_ref"/>
      <name name="engine_ref"/>
      <desc>
	<p>The result of a call to <seealso marker="#engine_load-3">engine_load/3</seealso>.
	</p>
      </desc>
    </datatype>

    <datatype>
      <name name="key_id"/>
      <desc>
	<p>Identifies the key to be used. The format depends on the loaded engine. It is passed to
	the <c>ENGINE_load_(private|public)_key</c> functions in libcrypto.
	</p>
      </desc>
    </datatype>

    <datatype>
      <name name="password"/>
      <desc>
	<p>The password of the key stored in an engine.
	</p>
      </desc>
    </datatype>

    <datatype>
      <name name="engine_method_type"/>
    </datatype>

    <datatype>
      <name name="engine_cmnd"/>
      <desc>
	<p>Pre and Post commands for <seealso marker="#engine_load-3">engine_load/3 and /4</seealso>.
	</p>
      </desc>
    </datatype>

    <datatype_title>Internal data types</datatype_title>

    <datatype>
      <name name="stream_state"/>
      <name name="hmac_state"/>
      <name name="hash_state"/>
      <desc>
	<p>Contexts with an internal state that should not be manipulated but passed between function calls.
	</p>
      </desc>
    </datatype>

  </datatypes>

  <!--================ FUNCTIONS ================-->
  <funcs>
    <func>
      <name name="block_encrypt" arity="3"/>
      <fsummary>Encrypt <c>PlainText</c> according to <c>Type</c> block cipher</fsummary>
      <desc>
	<p>Encrypt <c>PlainText</c> according to <c>Type</c> block cipher.</p>
	<p>May raise exception <c>error:notsup</c> in case the chosen <c>Type</c>
	is not supported by the underlying libcrypto implementation.</p>
	<p>For keylengths and blocksizes see the
	<seealso marker="crypto:algorithm_details#ciphers">User's Guide</seealso>.
	</p>
      </desc>
    </func>

    <func>
      <name name="block_decrypt" arity="3"/>
      <fsummary>Decrypt <c>CipherText</c> according to <c>Type</c> block cipher</fsummary>
      <desc>
        <p>Decrypt <c>CipherText</c> according to <c>Type</c> block cipher.</p>
	<p>May raise exception <c>error:notsup</c> in case the chosen <c>Type</c>
	is not supported by the underlying libcrypto implementation.</p>
	<p>For keylengths and blocksizes see the
	<seealso marker="crypto:algorithm_details#ciphers">User's Guide</seealso>.
	</p>
      </desc>
    </func>

    <func>
      <name>block_encrypt(Type, Key, Ivec, PlainText) -> CipherText</name>
      <name>block_encrypt(AeadType, Key, Ivec, {AAD, PlainText}) -> {CipherText, CipherTag}</name>
      <name>block_encrypt(aes_gcm, Key, Ivec, {AAD, PlainText, TagLength}) -> {CipherText, CipherTag}</name>
      <fsummary>Encrypt <c>PlainText</c> according to <c>Type</c> block cipher</fsummary>
      <type>
	<v>Type = <seealso marker="#type-block_cipher_with_iv">block_cipher_with_iv()</seealso></v>
	<v>AeadType = <seealso marker="#type-aead_cipher">aead_cipher()</seealso></v>
	<v>Key = <seealso marker="#type-key">key()</seealso> | <seealso marker="#type-des3_key">des3_key()</seealso></v>
        <v>PlainText = iodata()</v>
        <v>AAD = IVec = CipherText = CipherTag = binary()</v>
        <v>TagLength = 1..16</v>
      </type>
      <desc>
        <p>Encrypt <c>PlainText</c> according to <c>Type</c> block cipher.
	<c>IVec</c> is an arbitrary initializing vector.</p>
	<p>In AEAD (Authenticated Encryption with Associated Data) mode, encrypt
	<c>PlainText</c>according to <c>Type</c> block cipher and calculate
	<c>CipherTag</c> that also authenticates the <c>AAD</c> (Associated Authenticated Data).</p>
	<p>May raise exception <c>error:notsup</c> in case the chosen <c>Type</c>
	is not supported by the underlying libcrypto implementation.</p>
	<p>For keylengths, iv-sizes and blocksizes see the
	<seealso marker="crypto:algorithm_details#ciphers">User's Guide</seealso>.
	</p>
      </desc>
    </func>

    <func>
      <name>block_decrypt(Type, Key, Ivec, CipherText) -> PlainText</name>
      <name>block_decrypt(AeadType, Key, Ivec, {AAD, CipherText, CipherTag}) -> PlainText | error</name>
      <fsummary>Decrypt <c>CipherText</c> according to <c>Type</c> block cipher</fsummary>
      <type>
	<v>Type = <seealso marker="#type-block_cipher_with_iv">block_cipher_with_iv()</seealso></v>
	<v>AeadType = <seealso marker="#type-aead_cipher">aead_cipher()</seealso></v>
	<v>Key = <seealso marker="#type-key">key()</seealso> | <seealso marker="#type-des3_key">des3_key()</seealso></v>
        <v>PlainText = iodata()</v>
        <v>AAD = IVec = CipherText = CipherTag = binary()</v>
      </type>
      <desc>
        <p>Decrypt <c>CipherText</c> according to <c>Type</c> block cipher.
	<c>IVec</c> is an arbitrary initializing vector.</p>
        <p>In AEAD (Authenticated Encryption with Associated Data) mode, decrypt
	<c>CipherText</c>according to <c>Type</c> block cipher and check the authenticity
	the <c>PlainText</c> and <c>AAD</c> (Associated Authenticated Data) using the
	<c>CipherTag</c>. May return <c>error</c> if the decryption or validation fail's</p>
	<p>May raise exception <c>error:notsup</c> in case the chosen <c>Type</c>
	is not supported by the underlying libcrypto implementation.</p>
	<p>For keylengths, iv-sizes and blocksizes see the
	<seealso marker="crypto:algorithm_details#ciphers">User's Guide</seealso>.
	</p>
      </desc>
    </func>

     <func>
      <name name="bytes_to_integer" arity="1"/>
      <fsummary>Convert binary representation, of an integer, to an Erlang integer.</fsummary>
      <desc>
        <p>Convert binary representation, of an integer, to an Erlang integer.
         </p>
      </desc>
    </func>

    <func>
      <name name="compute_key" arity="4"/>
      <fsummary>Computes the shared secret</fsummary>
      <desc>
	<p>Computes the shared secret from the private key and the other party's public key.
	 See also <seealso marker="public_key:public_key#compute_key-2">public_key:compute_key/2</seealso>
	</p>
      </desc>
    </func>

    <func>
      <name name="exor" arity="2"/>
      <fsummary>XOR data</fsummary>
      <desc>
	<p>Performs bit-wise XOR (exclusive or) on the data supplied.</p>
      </desc>
    </func>


    <func>
      <name name="generate_key" arity="2"/>
      <name name="generate_key" arity="3"/>
      <fsummary>Generates a public key of type <c>Type</c></fsummary>
      <desc>
	<p>Generates a public key of type <c>Type</c>.
	See also <seealso marker="public_key:public_key#generate_key-1">public_key:generate_key/1</seealso>.
	May raise exception:
        </p>
        <list type="bulleted">
          <item><c>error:badarg</c>: an argument is of wrong type or has an illegal value,</item>
	  <item><c>error:low_entropy</c>: the random generator failed due to lack of secure "randomness",</item>
          <item><c>error:computation_failed</c>: the computation fails of another reason than <c>low_entropy</c>.</item>
        </list>
	<note>
	  <p>RSA key generation is only available if the runtime was
	  built with dirty scheduler support. Otherwise, attempting to
	  generate an RSA key will raise exception <c>error:notsup</c>.</p>
	</note>
      </desc>
    </func>

     <func>
      <name name="hash" arity="2"/>
      <fsummary></fsummary>
      <desc>
        <p>Computes a message digest of type <c>Type</c> from <c>Data</c>.</p>
	<p>May raise exception <c>error:notsup</c> in case the chosen <c>Type</c>
	is not supported by the underlying libcrypto implementation.</p>
      </desc>
    </func>

    <func>
      <name name="hash_init" arity="1"/>
      <fsummary></fsummary>
      <desc>
        <p>Initializes the context for streaming hash operations. <c>Type</c> determines
        which digest to use. The returned context should be used as argument
	to <seealso marker="#hash_update-2">hash_update</seealso>.</p>
	<p>May raise exception <c>error:notsup</c> in case the chosen <c>Type</c>
	is not supported by the underlying libcrypto implementation.</p>
      </desc>
    </func>

    <func>
      <name name="hash_update" arity="2"/>
      <fsummary></fsummary>
      <desc>
        <p>Updates the digest represented by <c>Context</c> using the given <c>Data</c>. <c>Context</c>
        must have been generated using <seealso marker="#hash_init-1">hash_init</seealso>
	or a previous call to this function. <c>Data</c> can be any length. <c>NewContext</c>
        must be passed into the next call to <c>hash_update</c>
	or <seealso marker="#hash_final-1">hash_final</seealso>.</p>
      </desc>
    </func>

    <func>
      <name name="hash_final" arity="1"/>
      <fsummary></fsummary>
      <desc>
        <p>Finalizes the hash operation referenced by <c>Context</c> returned
	from a previous call to <seealso marker="#hash_update-2">hash_update</seealso>.
	The size of <c>Digest</c> is determined by the type of hash
	function used to generate it.</p>
      </desc>
    </func>

    <func>
      <name name="hmac" arity="3"/>
      <name name="hmac" arity="4"/>
      <fsummary></fsummary>
      <desc>
        <p>Computes a HMAC of type <c>Type</c> from <c>Data</c> using
        <c>Key</c> as the authentication key.</p> <p><c>MacLength</c>
        will limit the size of the resultant <c>Mac</c>.</p>
      </desc>
    </func>

    <func>
      <name name="hmac_init" arity="2"/>
      <fsummary></fsummary>
      <desc>
        <p>Initializes the context for streaming HMAC operations. <c>Type</c> determines
        which hash function to use in the HMAC operation. <c>Key</c> is the authentication
        key. The key can be any length.</p>
      </desc>
    </func>

    <func>
      <name name="hmac_update" arity="2"/>
      <fsummary></fsummary>
      <desc>
        <p>Updates the HMAC represented by <c>Context</c> using the given <c>Data</c>. <c>Context</c>
        must have been generated using an HMAC init function (such as
        <seealso marker="#hmac_init-2">hmac_init</seealso>). <c>Data</c> can be any length. <c>NewContext</c>
        must be passed into the next call to <c>hmac_update</c>
	or to one of the functions <seealso marker="#hmac_final-1">hmac_final</seealso> and
	<seealso marker="#hmac_final_n-2">hmac_final_n</seealso>
	</p>
	<warning><p>Do not use a <c>Context</c> as argument in more than one
	call to	hmac_update or hmac_final. The semantics of reusing old contexts
	in any way is undefined and could even crash the VM in earlier releases.
	The reason for this limitation is a lack of support in the underlying
	libcrypto	API.</p></warning>
      </desc>
    </func>

    <func>
      <name name="hmac_final" arity="1"/>
      <fsummary></fsummary>
      <desc>
        <p>Finalizes the HMAC operation referenced by <c>Context</c>. The size of the resultant MAC is
        determined by the type of hash function used to generate it.</p>
      </desc>
    </func>

    <func>
      <name name="hmac_final_n" arity="2"/>
      <fsummary></fsummary>
      <desc>
        <p>Finalizes the HMAC operation referenced by <c>Context</c>. <c>HashLen</c> must be greater than
        zero. <c>Mac</c> will be a binary with at most <c>HashLen</c> bytes. Note that if HashLen is greater than the actual number of bytes returned from the underlying hash, the returned hash will have fewer than <c>HashLen</c> bytes.</p>
      </desc>
    </func>

    <func>
      <name name="cmac" arity="3"/>
      <name name="cmac" arity="4"/>
      <fsummary>Calculates the Cipher-based Message Authentication Code.</fsummary>
      <desc>
        <p>Computes a CMAC of type <c>Type</c> from <c>Data</c> using
        <c>Key</c> as the authentication key.</p> <p><c>MacLength</c>
        will limit the size of the resultant <c>Mac</c>.</p>
      </desc>
    </func>

    <func>
      <name name="info_fips" arity="0"/>
      <fsummary>Provides information about the FIPS operating status.</fsummary>
      <desc>
        <p>Provides information about the FIPS operating status of
        crypto and the underlying libcrypto library. If crypto was built
        with FIPS support this can be either <c>enabled</c> (when
        running in FIPS mode) or <c>not_enabled</c>. For other builds
        this value is always <c>not_supported</c>.
	</p>
	<p>See <seealso marker="#enable_fips_mode-1">enable_fips_mode/1</seealso> about how to enable
	FIPS mode.
	</p>
        <warning>
          <p>In FIPS mode all non-FIPS compliant algorithms are
          disabled and raise exception <c>error:notsup</c>. Check
          <seealso marker="#supports-0">supports</seealso> that in
          FIPS mode returns the restricted list of available
          algorithms.</p>
        </warning>
      </desc>
    </func>

    <func>
      <name name="enable_fips_mode" arity="1"/>
      <fsummary>Change FIPS mode.</fsummary>
      <desc>
        <p>Enables (<c>Enable = true</c>) or disables (<c>Enable = false</c>) FIPS mode. Returns <c>true</c> if
	the operation was successful or <c>false</c> otherwise.
	</p>
	<p>Note that to enable FIPS mode succesfully, OTP must be built with the configure option <c>--enable-fips</c>,
	and the underlying libcrypto must also support FIPS.
	</p>
	<p>See also <seealso marker="#info_fips-0">info_fips/0</seealso>.
	</p>
      </desc>
    </func>

    <func>
      <name name="info_lib" arity="0"/>
      <fsummary>Provides information about the libraries used by crypto.</fsummary>
      <desc>
        <p>Provides the name and version of the libraries used by crypto.</p>
        <p><c>Name</c> is the name of the library. <c>VerNum</c> is
        the numeric version according to the library's own versioning
        scheme. <c>VerStr</c> contains a text variant of the version.</p>
        <pre>
> <input>info_lib().</input>
[{&lt;&lt;"OpenSSL"&gt;&gt;,269484095,&lt;&lt;"OpenSSL 1.1.0c  10 Nov 2016""&gt;&gt;}]
        </pre>
	<note><p>
	  From OTP R16 the <em>numeric version</em> represents the version of the OpenSSL
	  <em>header files</em> (<c>openssl/opensslv.h</c>) used when crypto was compiled.
	  The text variant represents the libcrypto library used at runtime.
	  In earlier OTP versions both numeric and text was taken from the library.
	</p></note>
      </desc>
    </func>

    <func>
      <name name="mod_pow" arity="3"/>
      <fsummary>Computes the function: N^P mod M</fsummary>
      <desc>
	<p>Computes the function <c>N^P mod M</c>.</p>
      </desc>
    </func>

    <func>
      <name name="next_iv" arity="2"/>
      <name name="next_iv" arity="3"/>
      <fsummary></fsummary>
      <desc>
	<p>Returns the initialization vector to be used in the next
	iteration of encrypt/decrypt of type <c>Type</c>. <c>Data</c> is the
	encrypted data from the previous iteration step. The <c>IVec</c>
	argument is only needed for <c>des_cfb</c> as the vector used
	in the previous iteration step.</p>
      </desc>
    </func>

    <func>
      <name name="poly1305" arity="2"/>
      <fsummary></fsummary>
      <desc>
        <p>Computes a POLY1305 message authentication code (<c>Mac</c>) from <c>Data</c> using
        <c>Key</c> as the authentication key.</p>
      </desc>
    </func>

    <func>
      <name name="private_decrypt" arity="4"/>
      <fsummary>Decrypts CipherText using the private Key.</fsummary>
      <desc>
	<p>Decrypts the <c>CipherText</c>, encrypted with
	<seealso marker="#public_encrypt-4">public_encrypt/4</seealso> (or equivalent function)
	  using the <c>PrivateKey</c>, and returns the
	  plaintext (message digest). This is a low level signature  verification operation
	used for instance by older versions of the SSL protocol.
	  See also <seealso marker="public_key:public_key#decrypt_private-2">public_key:decrypt_private/[2,3]</seealso>
	</p>
      </desc>
    </func>

    <func>
      <name name="private_encrypt" arity="4"/>
      <fsummary>Encrypts PlainText using the private Key.</fsummary>
      <desc>
	<p>Encrypts the <c>PlainText</c> using the <c>PrivateKey</c>
	and returns the ciphertext. This is a low level signature operation
	used for instance by older versions of the SSL protocol. See
	also <seealso
	marker="public_key:public_key#encrypt_private-2">public_key:encrypt_private/[2,3]</seealso>
	</p>
      </desc>
    </func>

    <func>
      <name name="public_decrypt" arity="4"/>
      <fsummary>Decrypts CipherText using the public Key.</fsummary>
      <desc>
	<p>Decrypts the <c>CipherText</c>, encrypted with
	<seealso marker="#private_encrypt-4">private_encrypt/4</seealso>(or equivalent function)
	  using the <c>PrivateKey</c>, and returns the
	  plaintext (message digest). This is a low level signature verification operation
	  used for instance by older versions of the SSL protocol.
	  See also <seealso marker="public_key:public_key#decrypt_public-2">public_key:decrypt_public/[2,3]</seealso>
	</p>
      </desc>
    </func>

    <func>
      <name name="public_encrypt" arity="4"/>
      <fsummary>Encrypts PlainText using the public Key.</fsummary>
      <desc>
	<p>Encrypts the <c>PlainText</c> (message digest) using the <c>PublicKey</c>
	and returns the <c>CipherText</c>. This is a low level signature operation
	used for instance by older versions of the SSL protocol. See also <seealso
	marker="public_key:public_key#encrypt_public-2">public_key:encrypt_public/[2,3]</seealso>
	</p>
      </desc>
    </func>

    <func>
      <name name="rand_seed" arity="1"/>
      <fsummary>Set the seed for random bytes generation</fsummary>
      <desc>
        <p>Set the seed for PRNG to the given binary. This calls the
        RAND_seed function from openssl. Only use this if the system
        you are running on does not have enough "randomness" built in.
        Normally this is when
        <seealso marker="#strong_rand_bytes/1">strong_rand_bytes/1</seealso>
        raises <c>error:low_entropy</c></p>
      </desc>
    </func>

     <func>
      <name>rand_uniform(Lo, Hi) -> N</name>
      <fsummary>Generate a random number</fsummary>
      <type>
        <v>Lo, Hi, N = integer()</v>
      </type>
      <desc>
        <p>Generate a random number <c><![CDATA[N, Lo =< N < Hi.]]></c> Uses the
	<c>crypto</c> library pseudo-random number generator.
	<c>Hi</c> must be larger than <c>Lo</c>.</p>
      </desc>
    </func>

    <func>
      <name name="start" arity="0"/>
      <fsummary> Equivalent to application:start(crypto). </fsummary>
      <desc>
        <p> Equivalent to application:start(crypto).</p>
      </desc>
    </func>

    <func>
      <name name="stop" arity="0"/>
      <fsummary> Equivalent to application:stop(crypto).</fsummary>
      <desc>
        <p> Equivalent to application:stop(crypto).</p>
      </desc>
    </func>

    <func>
      <name name="strong_rand_bytes" arity="1"/>
      <fsummary>Generate a binary of random bytes</fsummary>
      <desc>
        <p>Generates N bytes randomly uniform 0..255, and returns the
        result in a binary. Uses a cryptographically secure prng seeded and
        periodically mixed with operating system provided entropy. By default
        this is the <c>RAND_bytes</c> method from OpenSSL.</p>
	<p>May raise exception <c>error:low_entropy</c> in case the random generator
	failed due to lack of secure "randomness".</p>
      </desc>
    </func>

    <func>
      <name name="rand_seed" arity="0"/>
      <fsummary>Strong random number generation plugin state</fsummary>
      <desc>
	<p>
          Creates state object for
	  <seealso marker="stdlib:rand">random number generation</seealso>,
          in order to generate cryptographically strong random numbers
          (based on OpenSSL's <c>BN_rand_range</c>),
          and saves it in the process dictionary before returning it as well.
          See also
	  <seealso marker="stdlib:rand#seed-1">rand:seed/1</seealso> and
	  <seealso marker="#rand_seed_s-0">rand_seed_s/0</seealso>.
	</p>
	<p>
	  When using the state object from this function the 
	  <seealso marker="stdlib:rand">rand</seealso> functions using it
	  may raise exception <c>error:low_entropy</c> in case the random generator
	  failed due to lack of secure "randomness".
	</p>
        <p><em>Example</em></p>
        <pre>
_ = crypto:rand_seed(),
_IntegerValue = rand:uniform(42), % [1; 42]
_FloatValue = rand:uniform().     % [0.0; 1.0[</pre>
      </desc>
    </func>

    <func>
      <name name="rand_seed_s" arity="0"/>
      <fsummary>Strong random number generation plugin state</fsummary>
      <desc>
	<p>
          Creates state object for
	  <seealso marker="stdlib:rand">random number generation</seealso>,
          in order to generate cryptographically strongly random numbers
          (based on OpenSSL's <c>BN_rand_range</c>).
          See also
	  <seealso marker="stdlib:rand#seed_s-1">rand:seed_s/1</seealso>.
	</p>
	<p>
	  When using the state object from this function the 
	  <seealso marker="stdlib:rand">rand</seealso> functions using it
	  may raise exception <c>error:low_entropy</c> in case the random generator
	  failed due to lack of secure "randomness".
	</p>
	<note>
	  <p>
	    The state returned from this function can not be used
	    to get a reproducable random sequence as from
	    the other 
	    <seealso marker="stdlib:rand">rand</seealso>
	    functions,
	    since reproducability does not match cryptographically safe.
	  </p>
	  <p>
	    The only supported usage is to generate one distinct
	    random sequence from this start state.
	  </p>
	</note>
      </desc>
    </func>

    <func>
      <name>rand_seed_alg(Alg) -> rand:state()</name>
      <fsummary>Strong random number generation plugin state</fsummary>
      <type>
        <v>Alg = crypto | crypto_cache</v>
      </type>
      <desc>
	<marker id="rand_seed_alg-1" />
	<p>
          Creates state object for
	  <seealso marker="stdlib:rand">random number generation</seealso>,
          in order to generate cryptographically strong random numbers.
          See also
	  <seealso marker="stdlib:rand#seed-1">rand:seed/1</seealso> and
	  <seealso marker="#rand_seed_alg_s-1">rand_seed_alg_s/1</seealso>.
	</p>
	<p>
	  When using the state object from this function the 
	  <seealso marker="stdlib:rand">rand</seealso> functions using it
	  may raise exception <c>error:low_entropy</c> in case the random generator
	  failed due to lack of secure "randomness".
	</p>
	<p>
	  The cache size can be changed from its default value using the
	  <seealso marker="crypto_app">
	    crypto app's
	  </seealso> configuration parameter <c>rand_cache_size</c>.
	</p>
        <p><em>Example</em></p>
        <pre>
_ = crypto:rand_seed_alg(crypto_cache),
_IntegerValue = rand:uniform(42), % [1; 42]
_FloatValue = rand:uniform().     % [0.0; 1.0[</pre>
      </desc>
    </func>

    <func>
      <name>rand_seed_alg_s(Alg) -> rand:state()</name>
      <fsummary>Strong random number generation plugin state</fsummary>
      <type>
        <v>Alg = crypto | crypto_cache</v>
      </type>
      <desc>
	<marker id="rand_seed_alg_s-1" />
	<p>
          Creates state object for
	  <seealso marker="stdlib:rand">random number generation</seealso>,
          in order to generate cryptographically strongly random numbers.
          See also
	  <seealso marker="stdlib:rand#seed_s-1">rand:seed_s/1</seealso>.
	</p>
	<p>
	  If <c>Alg</c> is <c>crypto</c> this function behaves exactly like
	  <seealso marker="#rand_seed_s-0">rand_seed_s/0</seealso>.
	</p>
	<p>
	  If <c>Alg</c> is <c>crypto_cache</c> this function
	  fetches random data with OpenSSL's <c>RAND_bytes</c>
	  and caches it for speed using an internal word size
	  of 56 bits that makes calculations fast on 64 bit machines.
	</p>
	<p>
	  When using the state object from this function the 
	  <seealso marker="stdlib:rand">rand</seealso> functions using it
	  may raise exception <c>error:low_entropy</c> in case the random generator
	  failed due to lack of secure "randomness".
	</p>
	<p>
	  The cache size can be changed from its default value using the
	  <seealso marker="crypto_app">
	    crypto app's
	  </seealso> configuration parameter <c>rand_cache_size</c>.
	</p>
	<note>
	  <p>
	    The state returned from this function can not be used
	    to get a reproducable random sequence as from
	    the other 
	    <seealso marker="stdlib:rand">rand</seealso>
	    functions,
	    since reproducability does not match cryptographically safe.
	  </p>
	  <p>
	    In fact since random data is cached some numbers may
	    get reproduced if you try, but this is unpredictable.
	  </p>
	  <p>
	    The only supported usage is to generate one distinct
	    random sequence from this start state.
	  </p>
	</note>
      </desc>
    </func>

    <func>
      <name name="stream_init" arity="2"/>
      <fsummary></fsummary>
      <desc>
        <p>Initializes the state for use in RC4 stream encryption
        <seealso marker="#stream_encrypt-2">stream_encrypt</seealso> and
        <seealso marker="#stream_decrypt-2">stream_decrypt</seealso></p>
	<p>For keylengths see the
	<seealso marker="crypto:algorithm_details#stream-ciphers">User's Guide</seealso>.
	</p>
      </desc>
    </func>

   <func>
      <name name="stream_init" arity="3"/>
      <fsummary></fsummary>
      <desc>
        <p>Initializes the state for use in streaming AES encryption using Counter mode (CTR).
        <c>Key</c> is the AES key and must be either 128, 192, or 256 bits long. <c>IVec</c> is
        an arbitrary initializing vector of 128 bits (16 bytes). This state is for use with
        <seealso marker="#stream_encrypt-2">stream_encrypt</seealso> and
        <seealso marker="#stream_decrypt-2">stream_decrypt</seealso>.</p>
	<p>For keylengths and iv-sizes see the
	<seealso marker="crypto:algorithm_details#stream-ciphers">User's Guide</seealso>.
	</p>
      </desc>
    </func>

    <func>
      <name name="stream_encrypt" arity="2"/>
      <fsummary></fsummary>
      <desc>
	<p>Encrypts <c>PlainText</c> according to the stream cipher <c>Type</c> specified in stream_init/3.
	<c>Text</c> can be any number of bytes. The initial <c>State</c> is created using
        <seealso marker="#stream_init-2">stream_init</seealso>.
	<c>NewState</c> must be passed into the next call to <c>stream_encrypt</c>.</p>
      </desc>
    </func>

    <func>
      <name name="stream_decrypt" arity="2"/>
      <fsummary></fsummary>
      <desc>
	<p>Decrypts <c>CipherText</c> according to the stream cipher <c>Type</c> specified in stream_init/3.
	<c>PlainText</c> can be any number of bytes. The initial <c>State</c> is created using
        <seealso marker="#stream_init-2">stream_init</seealso>.
	<c>NewState</c> must be passed into the next call to <c>stream_decrypt</c>.</p>
      </desc>
    </func>

   <func>
      <name name="supports" arity="0"/>
      <fsummary>Provide a list of available crypto algorithms.</fsummary>
      <desc>
        <p> Can be used to determine which crypto algorithms that are supported
	by the underlying libcrypto library</p>
      </desc>
    </func>

    <func>
      <name name="ec_curves" arity="0"/>
      <fsummary>Provide a list of available named elliptic curves.</fsummary>
      <desc>
        <p>Can be used to determine which named elliptic curves are supported.</p>
      </desc>
    </func>

    <func>
      <name name="ec_curve" arity="1"/>
      <fsummary>Get the defining parameters of a elliptic curve.</fsummary>
      <desc>
        <p>Return the defining parameters of a elliptic curve.</p>
      </desc>
    </func>

    <func>
      <name name="sign" arity="4"/>
      <name name="sign" arity="5"/>
      <fsummary> Create digital signature.</fsummary>
      <desc>
	<p>Creates a digital signature.</p>
	<p>The msg is either the binary "cleartext" data to be
	signed or it is the hashed value of "cleartext" i.e. the
	digest (plaintext).</p>
	<p>Algorithm <c>dss</c> can only be used together with digest type
	<c>sha</c>.</p>
	<p>See also <seealso marker="public_key:public_key#sign-3">public_key:sign/3</seealso>.</p>
      </desc>
    </func>

    <func>
      <name name="verify" arity="5"/>
      <name name="verify" arity="6"/>
      <fsummary>Verifies a digital signature.</fsummary>
      <desc>
	<p>Verifies a digital signature</p>
	<p>The msg is either the binary "cleartext" data to be
	signed or it is the hashed value of "cleartext" i.e. the
	digest (plaintext).</p>
	<p>Algorithm <c>dss</c> can only be used together with digest type
	<c>sha</c>.</p>

	<p>See also <seealso marker="public_key:public_key#verify-4">public_key:verify/4</seealso>.</p>
      </desc>
    </func>

    <!-- Engine functions -->
    <func>
      <name name="privkey_to_pubkey" arity="2"/>
      <fsummary>Fetches a public key from an Engine stored private key.</fsummary>
      <desc>
	<p>Fetches the corresponding public key from a private key stored in an Engine.
	The key must be of the type indicated by the Type parameter.
	</p>	
      </desc>
    </func>

    <func>
      <name name="engine_get_all_methods" arity="0"/>
      <fsummary>Return list of all possible engine methods</fsummary>
      <desc>
	<p>
	  Returns a list of all possible engine methods.
	</p>
	<p>
	  May raise exception <c>error:notsup</c> in case there is
	  no engine support in the underlying OpenSSL implementation.
	</p>
	<p>
	  See also the chapter <seealso marker="crypto:engine_load#engine_load">Engine Load</seealso>
	  in the User's Guide.
	</p>
      </desc>
    </func>

    <func>
      <name name="engine_load" arity="3"/>
      <fsummary>Dynamical load an encryption engine</fsummary>
      <desc>
	<p>
	  Loads the OpenSSL engine given by <c>EngineId</c> if it is available and then returns ok and
	  an engine handle. This function is the same as calling <c>engine_load/4</c> with
	  <c>EngineMethods</c> set to a list of all the possible methods.  An error tuple is
	  returned if the engine can't be loaded.
	</p>
	<p>
	  The function raises a <c>error:badarg</c> if the parameters are in wrong format.
	  It may also raise the exception <c>error:notsup</c> in case there is
	  no engine support in the underlying OpenSSL implementation.
	</p>
	<p>
	  See also the chapter <seealso marker="crypto:engine_load#engine_load">Engine Load</seealso>
	  in the User's Guide.
	</p>
      </desc>
    </func>

    <func>
      <name name="engine_load" arity="4"/>
      <fsummary>Dynamical load an encryption engine</fsummary>
      <desc>
	<p>
	  Loads the OpenSSL engine given by <c>EngineId</c> if it is available and then returns ok and
	  an engine handle. An error tuple is returned if the engine can't be loaded.
	</p>
	<p>
	  The function raises a <c>error:badarg</c> if the parameters are in wrong format.
	  It may also raise the exception <c>error:notsup</c> in case there is
	  no engine support in the underlying OpenSSL implementation.
	</p>
	<p>
	  See also the chapter <seealso marker="crypto:engine_load#engine_load">Engine Load</seealso>
	  in the User's Guide.
	</p>
      </desc>
    </func>

    <func>
      <name name="engine_unload" arity="1"/>
      <fsummary>Dynamical load an encryption engine</fsummary>
      <desc>
	<p>
	  Unloads the OpenSSL engine given by <c>Engine</c>.
	  An error tuple is returned if the engine can't be unloaded.
	</p>
	<p>
	  The function raises a <c>error:badarg</c> if the parameter is in wrong format.
	  It may also raise the exception <c>error:notsup</c> in case there is
	  no engine support in the underlying OpenSSL implementation.
	</p>
	<p>
	  See also the chapter <seealso marker="crypto:engine_load#engine_load">Engine Load</seealso>
	  in the User's Guide.
	</p>
      </desc>
    </func>

    <func>
      <name name="engine_by_id" arity="1"/>
      <fsummary>Get a reference to an already loaded engine</fsummary>
      <desc>
	<p>
	  Get a reference to an already loaded engine with <c>EngineId</c>.
	  An error tuple is returned if the engine can't be unloaded.
	</p>
	<p>
	  The function raises a <c>error:badarg</c> if the parameter is in wrong format.
	  It may also raise the exception <c>error:notsup</c> in case there is
	  no engine support in the underlying OpenSSL implementation.
	</p>
	<p>
	  See also the chapter <seealso marker="crypto:engine_load#engine_load">Engine Load</seealso>
	  in the User's Guide.
	</p>
      </desc>
    </func>

    <func>
      <name name="engine_ctrl_cmd_string" arity="3"/>
      <fsummary>Sends ctrl commands to an OpenSSL engine</fsummary>
      <desc>
	<p>
	  Sends ctrl commands to the OpenSSL engine given by <c>Engine</c>.
	  This function is the same as calling <c>engine_ctrl_cmd_string/4</c> with
	  <c>Optional</c> set to <c>false</c>.
	</p>
	<p>
	  The function raises a <c>error:badarg</c> if the parameters are in wrong format.
	  It may also raise the exception <c>error:notsup</c> in case there is
	  no engine support in the underlying OpenSSL implementation.
	</p>
      </desc>
    </func>

    <func>
      <name name="engine_ctrl_cmd_string" arity="4"/>
      <fsummary>Sends ctrl commands to an OpenSSL engine</fsummary>
      <desc>
	<p>
	  Sends ctrl commands to the OpenSSL engine given by <c>Engine</c>.
	  <c>Optional</c> is a boolean argument that can relax the semantics of the function.
	  If set to <c>true</c> it will only return failure if the ENGINE supported the given
	  command name but failed while executing it, if the ENGINE doesn't support the command
	  name it will simply return success without doing anything. In this case we assume
	  the user is only supplying commands specific to the given ENGINE so we set this to
	  <c>false</c>.
	</p>
	<p>
	  The function raises a <c>error:badarg</c> if the parameters are in wrong format.
	  It may also raise the exception <c>error:notsup</c> in case there is
	  no engine support in the underlying OpenSSL implementation.
	</p>
      </desc>
    </func>

    <func>
      <name name="engine_add" arity="1"/>
      <fsummary>Add engine to OpenSSL internal list</fsummary>
      <desc>
	<p>Add the engine to OpenSSL's internal list.</p>
	<p>
	  The function raises a <c>error:badarg</c> if the parameters are in wrong format.
	  It may also raise the exception <c>error:notsup</c> in case there is
	  no engine support in the underlying OpenSSL implementation.
	</p>
      </desc>
    </func>

    <func>
      <name name="engine_remove" arity="1"/>
      <fsummary>Remove engine to OpenSSL internal list</fsummary>
      <desc>
	<p>Remove the engine from OpenSSL's internal list.</p>
	<p>
	  The function raises a <c>error:badarg</c> if the parameters are in wrong format.
	  It may also raise the exception <c>error:notsup</c> in case there is
	  no engine support in the underlying OpenSSL implementation.
	</p>
      </desc>
    </func>

    <func>
      <name name="engine_get_id" arity="1"/>
      <fsummary>Fetch engine ID</fsummary>
      <desc>
	<p>Return the ID for the engine, or an empty binary if there is no id set.</p>
	<p>
	  The function raises a <c>error:badarg</c> if the parameters are in wrong format.
	  It may also raise the exception <c>error:notsup</c> in case there is
	  no engine support in the underlying OpenSSL implementation.
	</p>
      </desc>
    </func>

    <func>
      <name name="engine_get_name" arity="1"/>
      <fsummary>Fetch engine name</fsummary>
      <desc>
	<p>Return the name (eg a description) for the engine, or an empty binary if there is no name set.</p>
	<p>
	  The function raises a <c>error:badarg</c> if the parameters are in wrong format.
	  It may also raise the exception <c>error:notsup</c> in case there is
	  no engine support in the underlying OpenSSL implementation.
	</p>
      </desc>
    </func>

    <func>
      <name name="engine_list" arity="0"/>
      <fsummary>List the known engine ids</fsummary>
      <desc>
	<p>List the id's of all engines in OpenSSL's internal list.</p>
	<p>
	  It may also raise the exception <c>error:notsup</c> in case there is
	  no engine support in the underlying OpenSSL implementation.
	</p>
	<p>
	  See also the chapter <seealso marker="crypto:engine_load#engine_load">Engine Load</seealso>
	  in the User's Guide.
	</p>
	<p>
	  May raise exception <c>error:notsup</c> in case engine functionality is not supported by the underlying
	  OpenSSL implementation.
	</p>
      </desc>
    </func>

    <func>
      <name name="ensure_engine_loaded" arity="2"/>
      <fsummary>Ensure encryption engine just loaded once</fsummary>
      <desc>
	<p>
	  Loads the OpenSSL engine given by <c>EngineId</c> and the path to the dynamic library
	  implementing the engine. This function is the same as calling <c>ensure_engine_loaded/3</c> with
	  <c>EngineMethods</c> set to a list of all the possible methods. An error tuple is
	  returned if the engine can't be loaded.
	</p>
	<p>
	  The function raises a <c>error:badarg</c> if the parameters are in wrong format.
	  It may also raise the exception <c>error:notsup</c> in case there is
	  no engine support in the underlying OpenSSL implementation.
	</p>
	<p>
	  See also the chapter <seealso marker="crypto:engine_load#engine_load">Engine Load</seealso>
	  in the User's Guide.
	</p>
      </desc>
    </func>

    <func>
      <name name="ensure_engine_loaded" arity="3"/>
      <fsummary>Ensure encryption engine just loaded once</fsummary>
      <desc>
	<p>
	  Loads the OpenSSL engine given by <c>EngineId</c> and the path to the dynamic library
	  implementing the engine. This function differs from the normal engine_load in that sense it
	  also add the engine id to the internal list in OpenSSL. Then in the following calls to the function
	  it just fetch the reference to the engine instead of loading it again.
	  An error tuple is returned if the engine can't be loaded.
	</p>
	<p>
	  The function raises a <c>error:badarg</c> if the parameters are in wrong format.
	  It may also raise the exception <c>error:notsup</c> in case there is
	  no engine support in the underlying OpenSSL implementation.
	</p>
	<p>
	  See also the chapter <seealso marker="crypto:engine_load#engine_load">Engine Load</seealso>
	  in the User's Guide.
	</p>
      </desc>
    </func>

    <func>
      <name name="ensure_engine_unloaded" arity="1"/>
      <fsummary>Unload an engine loaded with the ensure function</fsummary>
      <desc>
	<p>
	  Unloads an engine loaded with the <c>ensure_engine_loaded</c> function.
	  It both removes the label from the OpenSSL internal engine list and unloads the engine.
	  This function is the same as calling <c>ensure_engine_unloaded/2</c> with
	  <c>EngineMethods</c> set to a list of all the possible methods. An error tuple is
	  returned if the engine can't be unloaded.
	</p>
	<p>
	  The function raises a <c>error:badarg</c> if the parameters are in wrong format.
	  It may also raise the exception <c>error:notsup</c> in case there is
	  no engine support in the underlying OpenSSL implementation.
	</p>
	<p>
	  See also the chapter <seealso marker="crypto:engine_load#engine_load">Engine Load</seealso>
	  in the User's Guide.
	</p>
      </desc>
    </func>

    <func>
      <name name="ensure_engine_unloaded" arity="2"/>
      <fsummary>Unload an engine loaded with the ensure function</fsummary>
      <desc>
	<p>
	  Unloads an engine loaded with the <c>ensure_engine_loaded</c> function.
	  It both removes the label from the OpenSSL internal engine list and unloads the engine.
	  An error tuple is returned if the engine can't be unloaded.
	</p>
	<p>
	  The function raises a <c>error:badarg</c> if the parameters are in wrong format.
	  It may also raise the exception <c>error:notsup</c> in case there is
	  no engine support in the underlying OpenSSL implementation.
	</p>
	<p>
	  See also the chapter <seealso marker="crypto:engine_load#engine_load">Engine Load</seealso>
	  in the User's Guide.
	</p>
      </desc>
    </func>

 </funcs>


</erlref>