summaryrefslogtreecommitdiff
path: root/src/lib/libcrypto (follow)
Commit message (Collapse)AuthorAgeFilesLines
...
* err.h HISTORY up to SSLeay 0.8.1b; researched from OpenSSL gitschwarze2018-03-208-27/+45
|
* dsa.h HISTORY up to SSLeay 0.8.1b; researched from OpenSSL gitschwarze2018-03-206-18/+36
|
* dh.h HISTORY up to SSLeay 0.8.1b; researched from OpenSSL gitschwarze2018-03-206-19/+34
|
* des.h HISTORY up to SSLeay 0.8.1b; researched from OpenSSL git;schwarze2018-03-202-15/+49
| | | | while here, drop irrelevant history of Kerberos random number generator
* crypto.h HISTORY up to SSLeay 0.8.1b; researched from OpenSSL gitschwarze2018-03-204-12/+35
|
* buffer.h HISTORY up to SSLeay 0.8.1b; researched from OpenSSL gitschwarze2018-03-201-5/+7
|
* bn.h HISTORY up to SSLeay 0.8.1b; researched from OpenSSL gitschwarze2018-03-2016-65/+96
|
* bio.h HISTORY up to SSLeay 0.8.1b; researched from OpenSSL gitschwarze2018-03-2017-36/+151
|
* asn1.h HISTORY up to SSLeay 0.8.1b; researched from OpenSSL gitschwarze2018-03-209-19/+105
|
* HISTORY; from CHANGES.SSLeayschwarze2018-03-201-2/+6
|
* Two HISTORY bits from the old CHANGES.SSLeay file, which goes up toschwarze2018-03-202-5/+9
| | | | 0.9.0b. Nothing else that can be used in there.
* If X509_check_{host,email}() are called with a length of zero, use strlen()jsing2018-03-201-3/+7
| | | | | | | | | to determine the length. This is the documented behaviour and matches the OpenSSL implementation. Issue found by Michael Gmelin <freebsd at grem dot de>. ok tb@
* In x509_vfy.h rev. 1.27 2018/03/20 15:26:22, jsing@ providedschwarze2018-03-201-4/+18
| | | | X509_STORE_get_ex_new_index(3). Document it.
* bump minors after symbol additiontb2018-03-201-1/+1
|
* Provide X509_STORE_get_ex_new_index macro.jsing2018-03-201-1/+5
|
* correct funciton prototypebeck2018-03-191-3/+3
| | | | spotted by anton@
* I should not be calling ENGINE_load_builtin_engines_internalbeck2018-03-191-4/+2
| | | | | here, just the normal version. ok jsing@
* Correct mistake of loading the default openssl.conf by default during autoinit.beck2018-03-193-13/+37
| | | | | | | | This brings in the OPENSSL_INIT_LOAD_CONFIG flag with the same semantics as OpenSSL. As a result, by default the openssl.conf file is not loaded during autoinit, which makes autoinit safe for pledge(stdio). ok jsing@
* In ecdsa.h rev. 1.5 2018/03/17 15:24:44, tb@ providedschwarze2018-03-181-4/+39
| | | | | ECDSA_SIG_get0(3) and ECDSA_SIG_set0(3). Merge the documentation from OpenSSL.
* In rsa.h rev. 1.38 2018/03/17 15:12:56 and dsa.h rev. 1.30 2018/03/17schwarze2018-03-1810-15/+400
| | | | | | 15:19:12, tb@ provided a subset of the DSA_meth_*(3) and RSA_meth_*(3) functions. Merge the relevant parts of the manual pages from OpenSSL, heavily tweaked by me, in particular for conciseness.
* In x509.h rev. 1.43 2018/03/17 14:55:39, jsing@ providedschwarze2018-03-181-5/+16
| | | | X509_PUBKEY_get0(3). Merge the documentation from OpenSSL.
* In ocsp.h rev. 1.12 2018/03/17 14:44:34, jsing@ providedschwarze2018-03-181-4/+14
| | | | | OCSP_SINGLERESP_get0_id(3). OpenSSL fails to document it, so document it from scratch.
* In x509.h rev. 1.42 2018/03/17 14:33:20, jsing@ providedschwarze2018-03-181-3/+14
| | | | X509_REVOKED_dup(3). Document it.
* Fix a typo: OPENSSL_INIT_LOAD_CONFIG doesn't have double underbars.tb2018-03-181-2/+2
| | | | | | Found via port build failures reported by sthen. ok jsing
* Clean up now that autoconfiguration is available:schwarze2018-03-171-78/+55
| | | | | | | | | | - deprecation notice - say more precisely what OPENSSL_config(3) does - kill the unfounded rumour that the library might inspect environment variables; in fact, only the openssl(1) program inspects $OPENSSL_CONF - garbage collect long, irrelevant ramblings about engines - garbage collect empty RETURN VALUES section - garbage collect CAVEATS section, duplicate information only
* In crypto.h rev. 1.43 and ssl.h rev. 1.153 2018/03/17 16:20:01, beck@schwarze2018-03-174-6/+86
| | | | | | provided OPENSSL_init_crypto(3) and OPENSSL_init_ssl(3). Write the documentation from scratch because the text OpenSSL provides is full of bloat.
* In x509_cmp.c rev. 1.30 2018/03/17 14:57:23, jsing@ adjustedschwarze2018-03-171-3/+14
| | | | | X509_get0_pubkey(3) to the same semantics as in OpenSSL. Merge the documentation.
* crank majorstb2018-03-171-2/+2
| | | | req by deraadt
* Bump minors after symbol additiontb2018-03-171-1/+1
|
* Bring in compatibility for OpenSSL 1.1 style init functions.beck2018-03-1714-33/+241
| | | | | | | | | This adds OPENSSL_init_crypto and OPENSSL_init_ssl, as well thread safety modifications for the existing LibreSSL init functions. The initialization routines are called automatically by the normal entry points into the library, as in newer OpenSSL ok jsing@, nits by tb@ and deraadt@
* Provide X509_STORE_get0_param()tb2018-03-173-2/+10
| | | | ok jsing
* Provide X509_OBJECT_get_type(). Instead of the X509_LOOKUP_TYPE enumtb2018-03-173-2/+10
| | | | | | (which we don't have) it returns a plain int. ok jsing
* Provide X509_NAME_ENTRY_set()tb2018-03-173-2/+10
| | | | ok jsing
* Provide ECDSA_SIG_{g,s}et0().tb2018-03-173-2/+40
| | | | ok jsing
* Add DSA_meth_{dup,free,new,set_{finish,sign}}()tb2018-03-174-2/+93
| | | | | | | As in RSA_meth_*, note that these functions return NULL in out-of-memory situations, but they do not set an error explicitly. ok jsing
* Provide RSA_meth_{dup,free,new,set_{finish,priv_{dec,enc}}}()tb2018-03-174-3/+104
| | | | | | | Note that these functions return NULL in out-of-memory situations, but contrary to OpenSSL's versions they do not set an error. ok jsing
* Make BIO_meth_get_write() public. Omission spotted by schwarze.tb2018-03-172-1/+3
| | | | ok jsing
* Fix X509_get0_pubkey() - X509_get_pubkey() is a misnamed "get1" function,jsing2018-03-171-2/+4
| | | | | | so call X509_PUBKEY_get0() instead. Spotted by schwarze@ while documenting.
* Provide X509_PUBKEY_get0() by splitting X509_PUBKEY_get() and turning itjsing2018-03-173-7/+19
| | | | into a wrapper that calls X509_PUBKEY_get0() and up refs.
* Provide OCSP_SINGLERESP_get0_id().jsing2018-03-173-2/+10
|
* Provide object identifier for chacha20-poly1305.jsing2018-03-172-0/+3
|
* Provide object identifiers for TLS cipher suite key exchange andjsing2018-03-172-0/+20
| | | | authentication.
* Provide X509_REVOKED_dup().jsing2018-03-173-2/+10
|
* Consistently spell "IPsec" in comments and debug outputs.mpi2018-03-161-2/+2
| | | | From Raf Czlonka, ok sthen@
* ensure SYS_getrandom and GRND_NONBLOCK are both defined before using ↵bcook2018-03-131-4/+4
| | | | | | | getrandom(2) Based on discussion here https://github.com/libressl-portable/openbsd/pull/82 Suggested fix from jsing@
* Fix for processing of EC public keyinoguchi2018-03-122-15/+27
| | | | | | | | | | | | | | Prevents segmentation fault while reading EC private key without public key. Generates missing EC public key when reading EC private key. Refer to these OpenSSL commits: 1f2b943254ce590867717375e4f364860a9b7154 2083f7c465d07867dd9867b8742bb71c03d1f203 Reported on GitHub https://github.com/libressl-portable/portable/issues/395 by Anton Bukov (@k06a) . ok beck@
* Two new manual pages OPENSSL_sk_new(3) and STACK_OF(3), written fromschwarze2018-03-013-1/+749
| | | | scratch, tweaks and OK jsing@ and jmc@.
* Garbage collect RC5 documentation.schwarze2018-02-272-25/+6
| | | | | | We have no code implementing it, tedu@ took care of the rc5/ directory in 2014, the related files are commented out in arch/*/Makefile.inc, and opensslfeatures.h contains an explicit #define OPENSSL_NO_RC5.
* Garbage collect MD2 documentation.schwarze2018-02-273-73/+21
| | | | We have no code implementing those functions.
* In x509.h rev. 1.41 2018/02/22 17:22:02, jsing@ providedschwarze2018-02-261-6/+31
| | | | | X509_chain_up_ref(3). Merge the documentation from OpenSSL, tweaked by me.