summaryrefslogtreecommitdiff
path: root/src/lib (follow)
Commit message (Collapse)AuthorAgeFilesLines
...
* Hide SSLerr() under #ifndef LIBRESSL_INTERNAL since we shouldn't bebeck2017-01-261-2/+4
| | | | | using it anymore ok jsing@
* Send the error function codes to rot in the depths of hell where they belongbeck2017-01-2624-798/+572
| | | | | | | We leave a single funciton code (0xFFF) to say "SSL_internal" so the public API will not break, and we replace all internal use of the two argument SSL_err() with the internal only SSL_error() that only takes a reason code. ok jsing@
* Merge the single two line function from ssl_err2.c into ssl_err.c.jsing2017-01-263-76/+12
| | | | ok beck@
* english is hard.beck2017-01-261-2/+2
|
* Limit the number of sequential empty records that we will processbeck2017-01-264-7/+30
| | | | | | before yielding, and fail if we exceed a maximum. loosely based on what boring and openssl are doing ok jsing@
* Refactor the code to generate a WANT_READ into a function, as we arebeck2017-01-261-18/+20
| | | | | using it more and more to avoid spins. ok jsing@
* Remove most of SSL3_ENC_METHOD - we can just inline the function callsjsing2017-01-2611-135/+63
| | | | | | and defines since they are the same everywhere. ok beck@
* Move relatively new version range code from ssl_lib.c into a separatejsing2017-01-263-158/+175
| | | | | | ssl_versions.c file. ok beck@
* Rename s3_{both,clnt,pkt_srvr}.c to have an ssl_ prefix since they are nojsing2017-01-265-6/+6
| | | | | | longer SSLv3 code. ok beck@
* Merge the client/server version negotiation into the existing (currentlyjsing2017-01-2616-1229/+395
| | | | | | fixed version) client/server code. ok beck@
* Document ERR_load_BN_strings(3).schwarze2017-01-261-11/+47
| | | | | | | | | | | | | | | | | jsing@ confirmed that this function is public and worth documenting. This page needs much more work, it is outrageously incomplete and unclear. For example, it remains unexplained what error strings are, what "registering" means and what the benefit for the application is, what happens if it is not done, or what happens if an error occurs after calling ERR_free_strings(3). I tried to read the code, but it is so contorted that i postponed that work. For example, it looks like there are hooks for applications to replace the functions used for registering strings by other, application-supplied functions, and, of course, there are many levels of macro and function wrappers. For now, i only documented the most obvious BUGS.
* Remove ssl3_undef_enc_method - if we have internal bugs we want to segfaultjsing2017-01-265-36/+8
| | | | | | | so that we can debug it, rather than adding a "should not be called" error to the stack. Discussed with beck@
* Remove a sess_cert reference from a comment in the public header.jsing2017-01-261-5/+2
| | | | Noted by zhuk@
* split the tls_init(3) that had grown fat to allow healthy future growth;schwarze2017-01-2515-888/+1474
| | | | suggested by jsing@; "i would just chuck it in" jmc@
* document BN_asc2bn(3);schwarze2017-01-251-3/+27
| | | | jsing@ confirmed that it is a public function worth documenting
* Limit enabled version range by the versions configured on the SSL_CTX/SSL,jsing2017-01-253-23/+84
| | | | | | | provide an ssl_supported_versions_range() function which also limits the versions to those supported by the current method. ok beck@
* Change the SSL_IS_DTLS() macro to check the version, rather than using ajsing2017-01-252-7/+4
| | | | | | | flag in the encryption methods. We can do this since there is currently only one DTLS version. This makes upcoming changes easier. ok beck@
* Construct a BN_gcd_nonct, based on BN_mod_inverse_no_branch, as suggestedbeck2017-01-256-10/+170
| | | | | | | | | | | by Alejandro Cabrera <aldaya@gmail.com> to avoid the possibility of a sidechannel timing attack during RSA private key generation. Modify BN_gcd to become not visible under LIBRESSL_INTERNAL and force the use of the _ct or _nonct versions of the function only within the library. ok jsing@
* Provide ssl3_packet_read() and ssl3_packet_extend() functions that improvejsing2017-01-253-35/+59
| | | | | | | the awkward API provided by ssl3_read_n(). Call these when we need to read or extend a packet. ok beck@
* Provide defines for SSL_CTRL_SET_CURVES/SSL_CTRL_SET_CURVES_LIST for thingsjsing2017-01-251-1/+15
| | | | | | | | that are conditioning on these. From BoringSSL. ok beck@
* BUF_MEM_free(), X509_STORE_free() and X509_VERIFY_PARAM_free() all checkjsing2017-01-242-18/+10
| | | | for NULL, as does lh_free() - do not do the same from the caller.
* sk_free() checks for NULL so do not bother doing it from the callers.jsing2017-01-244-10/+9
|
* sk_pop_free() checks for NULL so do not bother doing it from the callers.jsing2017-01-247-50/+31
|
* Within libssl a SSL_CTX * is referred to as a ctx - fix this forjsing2017-01-241-29/+29
| | | | SSL_CTX_free().
* in resolver(3), document that _EDNS0 and _DNSSEC are no ops;jmc2017-01-241-6/+17
| | | | | | | diff from kirill miazine while here, bump all the no op texts to one standard blurb; help/ok jca
* #if 0 the ecformats_list and eccurves_list - these are currently unused butjsing2017-01-241-2/+5
| | | | will be revisited at some point in the near future.
* Remove unused cert variable.jsing2017-01-241-3/+1
| | | | Found by bcook@
* Bump libssl and libtls minors due to symbol additions.jsing2017-01-242-2/+2
|
* Add support for setting the supported EC curves viajsing2017-01-247-26/+197
| | | | | | | | | | | | | SSL{_CTX}_set1_groups{_list}() - also provide defines for the previous SSL{_CTX}_set1_curves{_list} names. This also changes the default list of EC curves to be X25519, P-256 and P-384. If you want others (such a brainpool) you need to configure this yourself. Inspired by parts of BoringSSL and OpenSSL. ok beck@
* Correct bounds checks used when generating the EC curves extension.jsing2017-01-241-3/+3
| | | | ok beck@
* accross -> across;jmc2017-01-241-2/+2
|
* Fix typo in brainpool curve name within a comment.jsing2017-01-241-2/+2
|
* There is no point returning then breaking...jsing2017-01-241-2/+1
|
* unifdef OPENSSL_NO_BIO - we do not support this in any form.jsing2017-01-241-15/+1
| | | | ok beck@
* Introduce ticket support. To enable them it is enough to set a positiveclaudio2017-01-246-14/+251
| | | | | | | | | | | | | | | | | | | | lifetime with tls_config_set_session_lifetime(). This enables tickets and uses an internal automatic rekeying mode for the ticket keys. If multiple processes are involved the following functions can be used to make tickets work accross all instances: - tls_config_set_session_id() sets the session identifier - tls_config_add_ticket_key() adds an encryption and authentication key For now only the last 4 keys added will be used (unless they are too old). If tls_config_add_ticket_key() is used the caller must ensure to add new keys regularly. It is best to do this 4 times per session lifetime (which is also the ticket key lifetime). Since tickets break PFS it is best to minimize the session lifetime according to needs. With a lot of help, input and OK beck@, jsing@
* ssl_cert_free() checks for NULL itself.jsing2017-01-241-10/+5
|
* Remove a "free up if allocated" comment that exists before code that freesjsing2017-01-241-2/+1
| | | | | | things if they are allocated. ok captainobvious@
* sk_SSL_CIPHER_free() checks for NULL so do not bother doing the same fromjsing2017-01-244-27/+16
| | | | the callers.
* ssl_sess_cert_free() checks for NULL, so do not bother doing it at thejsing2017-01-242-6/+8
| | | | call sites.
* There is no point in zeroing fields that exist within a struct that isjsing2017-01-241-3/+1
| | | | about to be explicit_bzero'd and freed.
* Add missing documentation for tls_config_set_verify_depthclaudio2017-01-241-1/+5
| | | | Done together with jsing@
* Shuffle the deck chairs to bring them back in order.claudio2017-01-241-9/+9
|
* tls_config_add_keypair_mem is the function to add additional keypairs andclaudio2017-01-241-5/+5
| | | | | | ocsp_staple functions set the OCSP response they don't add them (which implies you can call them multiple times). Discussed with jsing@ beck@
* move default_passwd_cb and default_passwd_cb_userdata back intobeck2017-01-234-30/+35
| | | | | | | the ssl_ctx from internal - these are used directly by python and openvpn and a few other things - we have the set accessors but the get accessors were added in 1.1 and these roll their own caveat OPENSSL_VERSION chickenpluckery
* Move options and mode from SSL_CTX and SSL to internal, since these can bejsing2017-01-2312-66/+70
| | | | set and cleared via existing functions.
* Split most of SSL_METHOD out into an internal variant, which is opaque.jsing2017-01-2323-303/+378
| | | | Discussed with beck@
* Remove ssl_ctrl, ssl_ctx_ctrl, ssl_callback_ctrl and ssl_ctx_callback_ctrljsing2017-01-238-76/+14
| | | | | | | from SSL_METHOD, replacing usage with direct calls to the appropriate functions. ok beck@
* send state and rstate from ssl_st into internal. There are accessorsbeck2017-01-2316-295/+296
| | | | | so these should not be diddled with directly ok jsing@
* move back read_hash and enc_read_ctx into ssl_st. wpa_supplicant andbeck2017-01-238-52/+57
| | | | other perversions touches them sickly and unnaturally.
* Move a large part of ssl_st into internal, so we can see what squeals.beck2017-01-2322-888/+892
| | | | ok jsing@
generated by cgit v1.2.3-55-g6feb (git 2.39.1) at 2025-04-28 16:52:16 +0000