| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
|
| |
unravelling the maze of function pointers and callbacks by directly
calling ssl3_{get,put}_cipher_by_char() and removing the
ssl_{get,put}_cipher_by_char macros.
Prompted by similar changes in boringssl.
ok guenther.
|
| |
|
|
| |
to only apply to s23_srvr.c.
|
| |
|
|
|
|
|
|
|
|
|
| |
saying that you expect it to return that value and compare it against zero
because it is supposedly faster, for this leads to bugs (especially given the
high rate of sloppy cut'n'paste within ssl3 and dtls1 routines in this
library).
Instead, compare for the exact value it ought to return upon success.
ok deraadt@
|
| |
|
|
|
|
| |
Based on changes to OpenSSL trunk.
ok beck@ miod@
|
| |
|
|
| |
ok beck@ miod@
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
bounds check, after reading the 2-, 3- or 4-byte size of the next chunk to
process. But the size fields themselves are not checked for being entirely
contained in the buffer.
Since reading past your bounds is bad practice, and may not possible if you
are using a secure memory allocator, we need to add the necessary bounds check,
at the expense of some readability.
As a bonus, a wrong size GOST session key will now trigger an error instead of
a printf to stderr and it being handled as if it had the correct size.
Creating this diff made my eyes bleed (in the real sense); reviewing it
made guenther@'s and beck@'s eyes bleed too (in the literal sense).
ok guenther@ beck@
|
| |
|
|
|
| |
baggage.
ok miod@ jsing@
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
c30718b5e7480add42598158
Don't know the full story, but it looks like a "can't do random
perfectly, so do it god awful" problem was found in 2013, and
replaced with "only do it badly if a flag is set". New flags
(SSL_MODE_SEND_SERVERHELLO_TIME and SSL_MODE_SEND_SERVERHELLO_TIME)
were added [Ben Laurie?] to support the old scheme of "use time_t
for first 4 bytes of the random buffer".
Nothing uses these flags [ecosystem scan by sthen]
Fully discourage use of these flags in the future by removing
support & definition of them. The buflen < 4 check is also interesting,
because no entropy would be returned. No callers passed such small
buffers.
ok miod sthen
|
| |
|
|
|
| |
implicit NULL checks, so there is no point ensuring that the pointer is
non-NULL before calling them.
|
| | |
|
| |
|
|
|
|
| |
OPENSSL_NO_TLSEXT.
ok tedu@
|
| |
|
|
| |
a not quite appropriate data structure. ok jsing
|
| |
|
|
| |
ok deraadt jsing
|
| |
|
|
|
|
|
|
| |
since free already does this for us. Also remove some pointless NULL
assignments, where the result from malloc(3) is immediately assigned to the
same variable.
ok miod@
|
| |
|
|
|
|
|
|
| |
readable and one less layer of abstraction. Use C99 initialisers for
clarity, grepability and to protect from future field reordering/removal.
ok miod@ (tedu@ also thought it was a wonderful idea, beck@ also agreed,
but ran away squealing since it reminded him of the VOP layer...)
|
| |
|
|
|
| |
all on their own and we can't effectively maintain them without using them,
which we don't. If the need arises, the code can be resurrected.
|
| |
|
|
|
|
|
|
| |
``debug'' code from a 15+ years old bugfix and the SSL_OP_PKCS1_CHECK_*
constants have had a value of zero since ages. No production code should use
them.
ok beck@
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
| |
OPENSSL_foo wrappers. This changes:
OPENSSL_malloc->malloc
OPENSSL_free->free
OPENSSL_relloc->realloc
OPENSSL_freeFunc->free
|
| | |
|
| |
|
|
|
|
|
| |
in a bunch of places inside the TLS engine, to try to keep entropy high.
I wonder if their moto is "If you can't solve a problem, at least try
to do it badly".
ok miod
|
| |
|
|
|
| |
readable. This pass is whitespace only and can readily be verified using
tr and md5.
|
| |
|
|
| |
ok miod@, deraadt@
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
