Start ensuring all OpenSSL "free" routines allow NULL, and remove
any if check before calling them.
This gets DH_free, DSA_free, RSA_free
Reviewed-by: NMatt Caswell <matt@openssl.org>

Reviewed-by: NMatt Caswell <matt@openssl.org>

It was saying that it was an illegal parameter / unsupported cipher
Reviewed-by: NMatt Caswell <matt@openssl.org>

The certificate already contains the DH parameters in that case.
ssl3_send_server_key_exchange() would fail in that case anyway.
Reviewed-by: NMatt Caswell <matt@openssl.org>

Also removes for it being NULL
Reviewed-by: NMatt Caswell <matt@openssl.org>

Fix missing return value check in dtls1_listen when calling SSL_clear().
Reviewed-by: NRichard Levitte <levitte@openssl.org>

Change ssl_set_handshake_header from return void to returning int, and
handle error return code appropriately.
Reviewed-by: NRichard Levitte <levitte@openssl.org>

Ensure that all functions have their return values checked where
appropriate. This covers all functions defined and called from within
libssl.
Reviewed-by: NRichard Levitte <levitte@openssl.org>

Mark most functions returning a result defined in any libssl header file
with __owur to warn if they are used without checking the return value.
Use -DUNUSED_RETURN compiler flag with gcc to activate these warnings.
Some functions returning a result are skipped if it is common and valid to
use these functions without checking the return value.
Reviewed-by: NRichard Levitte <levitte@openssl.org>

Don't check that the curve appears in the list of acceptable curves for the
peer, if they didn't send us such a list (RFC 4492 does not require that the
extension be sent).
Reviewed-by: NEmilia Käsper <emilia@openssl.org>

Reviewed-by: NEmilia Käsper <emilia@openssl.org>

Reviewed-by: NEmilia Käsper <emilia@openssl.org>

If client auth is used then a server can seg fault in the event of a DHE
cipher being used and a zero length ClientKeyExchange message being sent
by the client. This could be exploited in a DoS attack.

CVE-2015-1787
Reviewed-by: NRichard Levitte <levitte@openssl.org>

If a client renegotiates using an invalid signature algorithms extension
it will crash a server with a NULL pointer dereference.

Thanks to David Ramos of Stanford University for reporting this bug.

CVE-2015-0291
Reviewed-by: NTim Hudson <tjh@openssl.org>

The DTLSv1_listen function is intended to be stateless and processes
the initial ClientHello from many peers. It is common for user code to
loop over the call to DTLSv1_listen until a valid ClientHello is received
with an associated cookie. A defect in the implementation of DTLSv1_listen
means that state is preserved in the SSL object from one invokation to the
next that can lead to a segmentation fault. Erorrs processing the initial
ClientHello can trigger this scenario. An example of such an error could
be that a DTLS1.0 only client is attempting to connect to a DTLS1.2 only
server.

CVE-2015-0207
Reviewed-by: NRichard Levitte <levitte@openssl.org>

OpenSSL 1.0.2 introduced the "multiblock" performance improvement. This
feature only applies on 64 bit x86 architecture platforms that support AES
NI instructions. A defect in the implementation of "multiblock" can cause
OpenSSL's internal write buffer to become incorrectly set to NULL when
using non-blocking IO. Typically, when the user application is using a
socket BIO for writing, this will only result in a failed connection.
However if some other BIO is used then it is likely that a segmentation
fault will be triggered, thus enabling a potential DoS attack.

CVE-2015-0290
Reviewed-by: NRichard Levitte <levitte@openssl.org>
Reviewed-by: NAndy Polyakov <appro@openssl.org>

The function tls1_PRF counts the number of digests in use and partitions
security evenly between them. There always needs to be at least one digest
in use, otherwise this is an internal error. Add a sanity check for this.
Reviewed-by: NRichard Levitte <levitte@openssl.org>

Missed one file in the #ifdef merge; thanks Kurt.
Reviewed-by: NKurt Roeckx <kurt@openssl.org>

It created the cert structure in SSL_CTX or SSL if it was NULL, but they can
never be NULL as the comments already said.
Reviewed-by: NDr. Stephen Henson <steve@openssl.org>

If SSL_check_chain is called with a NULL X509 object or a NULL EVP_PKEY
or the type of the public key is unrecognised then the local variable
|cpk| in tls1_check_chain does not get initialised. Subsequently an
attempt is made to deref it (after the "end" label), and a seg fault will
result.
Reviewed-by: NDr. Stephen Henson <steve@openssl.org>

[and respect error return value in ssltest.c]
Reviewed-by: NMatt Caswell <matt@openssl.org>

Suggested by John Foley <foleyj@cisco.com>.
Reviewed-by: NMatt Caswell <matt@openssl.org>

Cleanse various intermediate buffers used by the PRF.
Reviewed-by: NRichard Levitte <levitte@openssl.org>

Ensure all malloc failures return -1.

Reported by Adam Langley (Google).
Reviewed-by: NMatt Caswell <matt@openssl.org>

Fix security issue where under certain conditions a client can complete a
handshake with an unseeded PRNG. The conditions are:
- Client is on a platform where the PRNG has not been seeded, and the
user has not seeded manually
- A protocol specific client method version has been used (i.e. not
SSL_client_methodv23)
- A ciphersuite is used that does not require additional random data
from the PRNG beyond the initial ClientHello client random
(e.g. PSK-RC4-SHA)

If the handshake succeeds then the client random that has been used will
have been generated from a PRNG with insufficient entropy and therefore
the output may be predictable.

For example using the following command with an unseeded openssl will
succeed on an unpatched platform:

openssl s_client -psk 1a2b3c4d -tls1_2 -cipher PSK-RC4-SHA

CVE-2015-0285
Reviewed-by: NRichard Levitte <levitte@openssl.org>

Since commit 741c9959 ("DTLS revision."), we put the wrong protocol
version into our ClientHello for DTLS1_BAD_VER. The old DTLS
code which used ssl->version was replaced by the more generic SSL3 code
which uses ssl->client_version. The Cisco ASA no longer likes our
ClientHello.

RT#3711
Reviewed-by: NRich Salz <rsalz@openssl.org>

Commit 9cf0f187 in HEAD, and 68039af3 in 1.0.2, removed a version check
from dtls1_buffer_message() which was needed to distinguish between DTLS
1.x and Cisco's pre-standard version of DTLS (DTLS1_BAD_VER).

Based on an original patch by David Woodhouse <dwmw2@infradead.org>
RT#3703
Reviewed-by: NTim Hudson <tjh@openssl.org>

This patch uses warning/fatal constants instead of numbers with comments for
warning/alerts in d1_pkt.c and s3_pkt.c

RT#3725
Reviewed-by: NRich Salz <rsalz@openssl.org>

Miscellaneous unchecked malloc fixes. Also fixed some mem leaks on error
paths as I spotted them along the way.
Reviewed-by: NTim Hudson <tjh@openssl.org>

These ciphers are removed:
    TLS1_CK_RSA_EXPORT1024_WITH_RC4_56_MD5
    TLS1_CK_RSA_EXPORT1024_WITH_RC2_CBC_56_MD5
    TLS1_CK_RSA_EXPORT1024_WITH_DES_CBC_SHA
    TLS1_CK_DHE_DSS_EXPORT1024_WITH_DES_CBC_SHA
    TLS1_CK_RSA_EXPORT1024_WITH_RC4_56_SHA
    TLS1_CK_DHE_DSS_EXPORT1024_WITH_RC4_56_SHA
    TLS1_CK_DHE_DSS_WITH_RC4_128_SHA
They were defined in a long-expired IETF internet-draft:
draft-ietf-tls-56-bit-ciphersuites-01.txt
Reviewed-by: NRichard Levitte <levitte@openssl.org>

Some Cisco appliances use a pre-standard version number for DTLS. We support
this as DTLS1_BAD_VER within the code.

This change fixes d2i_SSL_SESSION for that DTLS version.

Based on an original patch by David Woodhouse <dwmw2@infradead.org>

RT#3704
Reviewed-by: NTim Hudson <tjh@openssl.org>

Added various missing return value checks in tls1_change_cipher_state.
Reviewed-by: NRichard Levitte <levitte@openssl.org>

Fixed various missing return value checks in ssl3_send_newsession_ticket.
Also a mem leak on error.
Reviewed-by: NRichard Levitte <levitte@openssl.org>

NETSCAPE_HANG_BUG is a workaround for a browser bug from many years ago
(2000).
It predates DTLS, so certainly has no place in d1_srvr.c.
In s3_srvr.c it forces the ServerDone to appear in the same record as the
CertificateRequest when doing client auth.

BoringSSL have already made the same commit:
79ae85e4f777f94d91b7be19e8a62016cb55b3c5
Reviewed-by: NTim Hudson <tjh@openssl.org>

the "-hack" option from s_server that set this option.
Reviewed-by: NTim Hudson <tjh@openssl.org>

Reviewed-by: NTim Hudson <tjh@openssl.org>

Reviewed-by: NAndy Polyakov <appro@openssl.org>
Reviewed-by: NRichard Levitte <levitte@openssl.org>

Reviewed-by: NTim Hudson <tjh@openssl.org>

When writing out the hint, if the hint > 0, then we write it out otherwise
we skip it.

Previously when reading the hint back in, if were expecting to see one
(because the ticket length > 0), but it wasn't present then we set the hint
to -1, otherwise we set it to 0. This fails to set the hint to the same as
when it was written out.

The hint should never be negative because the RFC states the hint is
unsigned. It is valid for a server to set the hint to 0 (this means the
lifetime is unspecified according to the RFC). If the server set it to 0, it
should still be 0 when we read it back in.
Reviewed-by: NTim Hudson <tjh@openssl.org>

SSL_SESSION_get_ticket_lifetime_hint. The latter has been reported as
required to fix Qt for OpenSSL 1.1.0. I have also added the former in order
to determine whether a ticket is present or not - otherwise it is difficult
to know whether a zero lifetime hint is because the server set it to 0, or
because there is no ticket.
Reviewed-by: NTim Hudson <tjh@openssl.org>