openbsd/src/lib/libssl/ssl_asn1.c, branch OPENBSD_6_8_BASEA mirror of https://github.com/libressl/openbsd.git
https://git.lua4.win/openbsd/atom?h=OPENBSD_6_8_BASE2018-08-27T16:42:48+00:00Simplify new session ticket encoding/generation.2018-08-27T16:42:48+00:00jsing2018-08-27T16:42:48+00:00urn:sha1:3214e561f5d16cffff1ac02e8f9fcf7cd78bafc8
The original code did a crazy encode/malloc/encode/decode/modify/encode
dance, in order to encode a session in the form needed to encrypt then add
to a session ticket. By modifying the encoding functions slightly, we can
do this entire dance as a single encode.
Inspired by similar changes in BoringSSL.
ok inoguchi@ tb@
In i2d_SSL_SESSION(), on error call CBB_cleanup() with the correct CBB.2018-03-20T16:10:57+00:00jsing2018-03-20T16:10:57+00:00urn:sha1:61074e318aa46e48a32d930ca8c714491dbab771
Spotted by Coverity, although reported as a different issue.
Add missing $OpenBSD$ tags.2017-05-06T16:18:36+00:00jsing2017-05-06T16:18:36+00:00urn:sha1:0de4d35571b79288296f8df357dedc38c201ad7eSwitch back to freezero() and explicitly initialise data_len to zero. The2017-04-30T04:21:53+00:00jsing2017-04-30T04:21:53+00:00urn:sha1:f79e820651388e3e43865d862d25f125435d633d
previous code was safe since data would always be NULL if data_len was
uninitialised, however compilers cannot know this.
backout previous, data_len is not always initialized2017-04-16T06:26:55+00:00otto2017-04-16T06:26:55+00:00urn:sha1:cf60f6b359a0d738aa963c901f0f283688efaeb5Switch i2d_SSL_SESSION() back to freezero(3) now that the size constraints2017-04-14T15:20:16+00:00jsing2017-04-14T15:20:16+00:00urn:sha1:bb3a2224371df56a095bf21607f36458081bda63
have been relaxed.
Revert ssl_asn1.c r1.50 - CBB and freezero(3) do not play nicely together.2017-04-11T13:14:08+00:00jsing2017-04-11T13:14:08+00:00urn:sha1:cd55c3b0c4a69244140d116d0a9458ea69ab5bd1
Back this out while we investigate and implement a solution.
Found the hard way by sthen@
Use freezero() for i2d_SSL_SESSION() - one line of code instead of three.2017-04-10T16:47:08+00:00jsing2017-04-10T16:47:08+00:00urn:sha1:8d4a6ac759c1f9f5fafef26397422f7241dd61e9
In this case the memory allocated can also be significant, in which case
freezero() will have less overhead than explicit_bzero() (munmap instead
of touching all of the memory to write zeros).
Change SSLerror() back to taking two args, with the first one being an SSL *.2017-02-07T02:08:38+00:00beck2017-02-07T02:08:38+00:00urn:sha1:9a5920738bea15430db1fdd138e67d9bbc3a95d3
Make a table of "function codes" which maps the internal state of the SSL *
to something like a useful name so in a typical error in the connection you
know in what sort of place in the handshake things happened. (instead of
by arcane function name).
Add SSLerrorx() for when we don't have an SSL *
ok jsing@ after us both being prodded by bluhm@ to make it not terrible
Send the error function codes to rot in the depths of hell where they belong2017-01-26T10:40:21+00:00beck2017-01-26T10:40:21+00:00urn:sha1:a4abf558fd44464a5a48bfeb5393b01002f66c5e
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@