RT468: SSL_CTX_sess_set_cache_size wrong

The documentation is wrong about what happens when the
session cache fills up.

Reviewed-by: Tim Hudson <tjh@openssl.org>
(cherry picked from commit e9edfc419674f20b482a9beff9c246519f9c503e)

RT3301: Discard too-long heartbeat requests

Reviewed-by: Tim Hudson <tjh@openssl.org>
(cherry picked from commit af4c6e348e4bad6303e7d214cdcf2536487aabe4)

RT2518: fix pod2man errors

pod2man now complains when item tags are not sequential.
Also complains about missing =back and other tags.
Silence the warnings; most were already done.

Reviewed-by: Tim Hudson <tjh@openssl.org>
(cherry picked from commit fe7573042fa7f406fedb78d959659b39a7a1dcfb)

RT3108: OPENSSL_NO_SOCK should imply OPENSSL_NO_DGRAM

Reviewed-by: Dr. Stephen Henson <steve@openssl.org>
(cherry picked from commit be0bd11d698677bb7dde14cde73af098da94da18)

RT3031: Need to #undef some names for win32

Copy the ifdef/undef stanza from x509.h to x509v3.h

Reviewed-by: Dr. Stephen Henson <steve@openssl.org>
(cherry picked from commit 83e4e03eeb22d2fbaec516a466330f2ccab22864)

RT2843: Remove another spurious close-comment token

Reviewed-by: Dr. Stephen Henson <steve@openssl.org>
(cherry picked from commit 683cd7c9485009efcd5b522357519f0c7e1d4c47)

RT2842: Remove spurious close-comment marker.

Also, I (rsalz) changed "#ifdef undef" to "#if 0"

Reviewed-by: Dr. Stephen Henson <steve@openssl.org>
(cherry picked from commit 6b0dc6eff1a59274730802db923d55802378d011)

Merge branch 'OpenSSL_1_0_2-stable' of git.openssl.org:openssl into OpenSSL_1_0_2-stable

another empty merge???

Reviewed-by: Dr. Stephen Henson <steve@openssl.org>

Empty merge

Merge branch 'OpenSSL_1_0_2-stable' of git.openssl.org:openssl into OpenSSL_1_0_2-stable

Reviewed-by: Dr. Stephen Henson <steve@openssl.org>

Empty merge

Merge branch 'OpenSSL_1_0_2-stable' of git.openssl.org:openssl into OpenSSL_1_0_2-stable

Reviewed-by: Dr. Stephen Henson <steve@openssl.org>

RT1834: Fix PKCS7_verify return value

The function returns 0 or 1, only.

Reviewed-by: Dr. Stephen Henson <steve@openssl.org>
(cherry picked from commit b0e659cfaca9ff4a481cc63b7f6b6e97303ad8fe)

RT1832: Fix PKCS7_verify return value

The function returns 0 or 1, only.

Reviewed-by: Dr. Stephen Henson <steve@openssl.org>
(cherry picked from commit b0e659cfaca9ff4a481cc63b7f6b6e97303ad8fe)

RT1771: Add string.h include.

Reviewed-by: Dr. Stephen Henson <steve@openssl.org>
(cherry picked from commit 8842987e5a76535597b9795b0408565baabf18d1)

RT1325,2973: Add more extensions to c_rehash

Regexp was bracketed wrong.

Reviewed-by: Tim Hudson <tjh@openssl.org>
(cherry picked from commit 5a8addc432503d99dba39474892b07345a619641)

Add CHANGES entry for SCT viewer code.

Reviewed-by: Emilia Käsper <emilia@openssl.org>
(cherry picked from commit b2774f6e17d4204f19a4c009cef9db58821c4456)

psk_client_callback, 128-byte id bug.

Fix a bug in handling of 128 byte long PSK identity in
psk_client_callback.

OpenSSL supports PSK identities of up to (and including) 128 bytes in
length. PSK identity is obtained via the psk_client_callback,
implementors of which are expected to provide a NULL-terminated
identity. However, the callback is invoked with only 128 bytes of
storage thus making it impossible to return a 128 byte long identity and
the required additional NULL byte.

This CL fixes the issue by passing in a 129 byte long buffer into the
psk_client_callback. As a safety precaution, this CL also zeroes out the
buffer before passing it into the callback, uses strnlen for obtaining
the length of the identity returned by the callback, and aborts the
handshake if the identity (without the NULL terminator) is longer than
128 bytes.

(Original patch amended to achieve strnlen in a different way.)

Reviewed-by: Rich Salz <rsalz@openssl.org>
(cherry picked from commit be0d851732bad7370640702bc9c4a33189ede287)

Ensure that x**0 mod 1 = 0.

(cherry picked from commit 2b0180c37fa6ffc48ee40caa831ca398b828e680)

Reviewed-by: Ben Laurie <ben@openssl.org>

Followup on RT3334 fix: make sure that a directory that's the empty
string returns 0 with errno = ENOENT.

Reviewed-by: Andy Polyakov <appro@openssl.org>
(cherry picked from commit 360928b7d0f16dde70e26841bbf9e1af727e8b8f)

RT3334: Fix crypto/LPdir_win.c

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Andy Polyakov <appro@openssl.org>
(cherry picked from commit 6a14fe7576e7a14a46ba14df8be8fe478536b4fb)

RT3140: Possibly-unit variable in pem_lib.c

Can't really happen, but the flow of control isn't obvious.
Add an initializer.

Reviewed-by: Matt Caswell <matt@openssl.org>
(cherry picked from commit 0ff3687eab8b0915198ad84d83da5998860b11b3)

Make the inline const-time functions static.

"inline" without static is not correct as the compiler may choose to ignore it
and will then either emit an external definition, or expect one.

Reviewed-by: Geoff Thorpe <geoff@openssl.org>
(cherry picked from commit 86f50b36e63275a916b147f9d8764e3c0c060fdb)

RT3511: doc fix; req default serial is random

RT842, closed back in 2004, changed the default serial number
to be a random number rather than zero.  Finally time to update
the doc

Reviewed-by: Tim Hudson <tjh@openssl.org>
(cherry picked from commit 3aba132d61baeecffb9a6f8da7d0809352cbfb2d)

Add t1_ext and ssl_utst to the VMS build as well.

Reviewed-by: Dr Stephen Henson <steve@openssl.org>

RT1325,2973: Add more extensions to c_rehash

Add .crt/.cer/.crl to the filenames parsed.

I also updated the podpage (since it didn't exist when
this ticket was first created, nor when it was re-created
seven years later).

Reviewed-by: Tim Hudson <tjh@openssl.org>
(cherry picked from commit 80ec8d4e3ee212786dc3092b1c97305b871827f0)

md5-x86_64.pl: work around warning.

Reviewed-by: Rich Salz <rsalz@openssl.org>
(cherry picked from commit 4d86e8df6be69ed13abb73fd564f1f894eea0a98)

x86[_64] assembly pack: add Silvermont performance data.

Reviewed-by: Rich Salz <rsalz@openssl.org>
(cherry picked from commit b59f92e75d334c9281082a02faa6c68afb614fd2)

RT2820: case-insensitive filenames on Darwin

Andy pointed out there is also darwin64, so tweak the pattern.

Reviewed-by: Andy Polyakov <appro@openssl.org>
(cherry picked from commit 457f7b14ecce885dce0b1a16bcd0723f1d7a2792)

Add tags/TAGS

Reviewed-by: Tim Hudson <tjh@openssl.org>
(cherry picked from commit 9d6253cfd3395dfe7147dae134579c6b16544c53)

RT2119,3407: Updated to dgst.pod

Re-order algorithm list.
Be consistent in command synopsis.
Add content about signing.
Add EXAMPLE section
Add some missing options: -r, -fips-fingerprint -non-fips-allow
Various other fixes.

Reviewed-by: Andy Polyakov <appro@openssl.org>

RT1941: c_rehash.pod is missing

Add the file written by James Westby, graciously contributed
under the terms of the OpenSSL license.

Reviewed-by: Andy Polyakov <appro@openssl.org>
(cherry picked from commit cf2239b3b397174a8a6b1cc84ff68aba34ed5941)

RT2379: Bug in BIO_set_accept_port.pod

The doc says that port can be "*" to mean any port.
That's wrong.

Reviewed-by: Dr. Stephen Henson <steve@openssl.org>
(cherry picked from commit 07e3b31fae98b985d3d2aad7066144b11833f688)

RT2880: HFS is case-insensitive filenames

Add Darwin to list of case-insensitive filenames when
installing manapges.  When doing this, I noticed that
we weren't setting "filecase" for the HTML doc install.

Reviewed-by: Dr. Stephen Henson <steve@openssl.org>
(cherry picked from commit 82d9185ae53f6ba93953ae0e484179be89c8508a)

update ordinals

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

Fix comments, add new test.

Fix comments in ssltest.c: return value of 0 now means extension is
omitted and add_cb is not called for servers if the corresponding
extension is absent in ClientHello.

Test add_cb is not called if extension is not received.
Reviewed-by: Emilia Käsper <emilia@openssl.org>
(cherry picked from commit f47e203975133ddbae3cde20c8c3c0516f62066c)

Custom extension documentation.

Reviewed-by: Emilia Käsper <emilia@openssl.org>
(cherry picked from commit f3f56c2a87951e115a7f82d06826e72c9e13987f)

Rename some callbacks, fix alignment.

Reviewed-by: Emilia Käsper <emilia@openssl.org>
(cherry picked from commit 0cfefe4b6dcc6947c236b0f10a7f9e2f02273075)

Use consistent function naming.

Instead of SSL_CTX_set_custom_cli_ext and SSL_CTX_set_custom_srv_ext
use SSL_CTX_add_client_custom_ext and SSL_CTX_add_server_custom_ext.
Reviewed-by: Emilia Käsper <emilia@openssl.org>
(cherry picked from commit 8cafe9e8bfcc99d12adf083c61411955995668c4)

New function SSL_extension_supported().

Reviewed-by: Emilia Käsper <emilia@openssl.org>
(cherry picked from commit c846a5f5678a7149bc6cbd37dbdae886a5108364)

New extension callback features.

Support separate parse and add callback arguments.
Add new callback so an application can free extension data.
Change return value for send functions so < 0 is an error 0
omits extension and > 0 includes it. This is more consistent
with the behaviour of other functions in OpenSSL.

Modify parse_cb handling so <= 0 is an error.

Make SSL_CTX_set_custom_cli_ext and SSL_CTX_set_custom_cli_ext argument
order consistent.

NOTE: these changes WILL break existing code.

Remove (now inaccurate) in line documentation.
Reviewed-by: Emilia Käsper <emilia@openssl.org>
(cherry picked from commit 33f653adf3bff5b0795e22de1f54b7c5472252d0)

Callback revision.

Use "parse" and "add" for function and callback names instead of
"first" and "second".

Change arguments to callback so the extension type is unsigned int
and the buffer length is size_t. Note: this *will* break existing code.
Reviewed-by: Emilia Käsper <emilia@openssl.org>
(cherry picked from commit de2a9e38f39eacc2e052d694f5b5fa5b7e734abc)

Remove serverinfo checks.

Since sanity checks are performed for all custom extensions the
serverinfo checks are no longer needed.
Reviewed-by: Emilia Käsper <emilia@openssl.org>
(cherry picked from commit 707b026d7871eb12c23671c975e6a15a8c331785)

Conflicts:

ssl/ssl3.h
ssl/t1_lib.c

Add custom extension sanity checks.

Reject attempts to use extensions handled internally.

Add flags to each extension structure to indicate if an extension
has been sent or received. Enforce RFC5246 compliance by rejecting
duplicate extensions and unsolicited extensions and only send a
server extension if we have sent the corresponding client extension.
Reviewed-by: Emilia Käsper <emilia@openssl.org>
(cherry picked from commit 28ea0a0c6a5e4e217c405340fa22a8503c7a17db)

Custom extension revision.

Use the same structure for client and server custom extensions.

Add utility functions in new file t1_ext.c.
Use new utility functions to handle custom server and client extensions
and remove a lot of code duplication.
Reviewed-by: Emilia Käsper <emilia@openssl.org>
(cherry picked from commit ecf4d660902dcef6e0afc51d52926f00d409ee6b)

Conflicts:

ssl/ssl_lib.c
ssl/ssl_locl.h
ssl/t1_lib.c

Revision of custom extension code.

Move custom extension structures from SSL_CTX to CERT structure.

This change means the form can be revised in future without binary
compatibility issues. Also since CERT is part of SSL structures
so per-SSL custom extensions could be supported in future as well as
per SSL_CTX.
Reviewed-by: Rich Salz <rsalz@openssl.org>
Reviewed-by: Emilia Käsper <emilia@openssl.org>
(cherry picked from commit b83294fe3022b9d5d525ccdcfeb53d39c25b05bd)

Conflicts:

ssl/ssl.h
ssl/ssl_cert.c
ssl/ssl_locl.h

Constant-time utilities

Pull constant-time methods out to a separate header, add tests.

Reviewed-by: Bodo Moeller <bodo@openssl.org>
(cherry picked from commit 5a3d21c0585064292bde5cd34089e120487ab687)

Conflicts:
ssl/s3_cbc.c
test/Makefile

RT2400: ASN1_STRING_to_UTF8 missing initializer

Reviewed-by: Tim Hudson <tjh@openssl.org>
(cherry picked from commit f9fb43e176ad2a914108cd2b403425dc1ebc7262)

RT2308: Add extern "C" { ... } wrapper

Add the wrapper to all public header files (Configure
generates one).  Don't bother for those that are just
lists of #define's that do renaming.

Reviewed-by: Tim Hudson <tjh@openssl.org>
Cherry-pick of commit 17e80c6bd05de7406a65116f34ed59665607d8d5

Explicitly check for empty ASN.1 strings in d2i_ECPrivateKey

The old code implicitly relies on the ASN.1 code returning a \0-prefixed buffer
when the buffer length is 0. Change this to verify explicitly that the ASN.1 string
has positive length.

Reviewed-by: Dr Stephen Henson <steve@openssl.org>
(cherry picked from commit 82dc08de54ce443c2a9ac478faffe79e76157795)

RT3065: automatically generate a missing EC public key

When d2i_ECPrivateKey reads a private key with a missing (optional) public key,
generate one automatically from the group and private key.

Reviewed-by: Dr Stephen Henson <steve@openssl.org>
(cherry picked from commit ed383f847156940e93f256fed78599873a4a9b28)

RT3065: ec_private_key_dont_crash

This change saves several EC routines from crashing when an EC_KEY is
missing a public key. The public key is optional in the EC private key
format and, without this patch, running the following through `openssl
ec` causes a crash:

-----BEGIN EC PRIVATE KEY-----
MBkCAQEECAECAwQFBgcIoAoGCCqGSM49AwEH
-----END EC PRIVATE KEY-----

Reviewed-by: Dr Stephen Henson <steve@openssl.org>
(cherry picked from commit b391570bdeb386d4fd325917c248d593d3c43930)

RT2210: Add missing EVP_cleanup to example

I also removed some trailing whitespace and cleaned
up the "see also" list.

Reviewed-by: Emilia Kasper <emilia@openssl.org>
(cherry picked from commit 7b3e11c54466f1da8b707c932e308d345fd61101)

RT2724: Remove extra declaration

Extra SSL_get_selected_srtp_profile() declaration in ssl/srtp.h
causes -Werror builds to fail.

Cherry-picked from 3609b02305c3678525930ff9bacb566c0122ea2a

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

RT1744: SSL_CTX_set_dump_dh() doc feedback

The description of when the server creates a DH key is
confusing.  This cleans it up.
(rsalz: also removed trailing whitespace.)

Reviewed-by: Viktor Dukhovni <viktor@openssl.org>

RT1804: fix EXAMPLE in EVP_EncryptInit.pod

The EXAMPLE that used FILE and RC2 doesn't compile due to a
few minor errors.  Tweak to use IDEA and AES-128. Remove
examples about RC2 and RC5.

Reviewed-by: Emilia Kasper <emilia@openssl.org>

Typo fixes to evp documentation.

This patch was submitted by user "Kox" via the wiki

Reviewed-by: Tim Hudson <tjh@openssl.org>
(cherry picked from commit 2dd8cb3b9593f528d9537aa6a003d5c93df1e3c5)

RT3060: Limit the number of empty records.

Limit the number of empty records that will be processed consecutively
in order to prevent ssl3_get_record from never returning.

Reported by "oftc_must_be_destroyed" and George Kadianakis.

Reviewed-by: Bodo Moeller <bodo@openssl.org>

RT3061: Don't SEGFAULT when trying to export a public DSA key as a private key.

Reviewed-by: Viktor Dukhovni <viktor@openssl.org>
Reviewed-by: Rich Salz <rsalz@openssl.org>

Improve EVP_PKEY_sign documentation

Clarify the intended use of EVP_PKEY_sign. Make the code example compile.

Reviewed-by: Dr Stephen Henson <steve@openssl.org>
(cherry picked from commit d64c533a207f7b6d86c3bc8ffb053e5f4d0c1ca0)

define inline for Visual Studio

In Visual Studio, inline is available in C++ only, however __inline is available for C, see
http://msdn.microsoft.com/en-us/library/z8y1yy88.aspx

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Dr Stephen Henson <steve@openssl.org>
(cherry picked from commit f511b25a7370c775dc9fd6198dbacd1706cf242b)

Fix build when BSAES_ASM is defined but VPAES_ASM is not

Reviewed-by: Andy Polyakov <appro@openssl.org>
(cherry picked from commit da92be4d68bec81030838e3228ef0238c565af85)

bn/asm/rsaz-*.pl: allow spaces in Perl path name.

RT: 2835

Reviewed-by: Dr. Stephen Henson <steve@openssl.org>
(cherry picked from commit 15735e4f0e81d535cda0ad7ab52a0ed64b644cd0)

sha1-mb-x86_64.pl: add commentary.

Reviewed-by: Emilia Kasper <emilia@openssl.org>
(cherry picked from commit e608273a8094a95a5703c26a428a007497e74392)

crypto/evp/e_aes_cbc_hmac_sha[1|256].c: fix compiler warnings.

Reviewed-by: Dr. Stephen Henson <steve@openssl.org>
(cherry picked from commit 2893a302a9b6a70161d1859d985a52af11b2195d)

sha1-mb-x86_64.pl: fix typo.

Reviewed-by: Emilia Kasper <emilia@openssl.org>
(cherry picked from commit 55eb14da201cc35fe744a08718f5c2efb97f6155)

Fixed out-of-bounds read errors in ssl3_get_key_exchange.

PR#3450

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

Fix use after free bug.

Reviewed-by: Stephen Henson <steve@openssl.org>
Reviewed-by: Emilia Käsper <emilia@openssl.org>
(cherry picked from commit 5afa57fb7b17aa51cfba1ffa94e900fc7a5f0e04)

Further improve/fix ec_GFp_simple_points_make_affine (ecp_smpl.c) and
group_order_tests (ectest.c).  Also fix the EC_POINTs_mul documentation (ec.h).

Reviewed-by: emilia@openssl.org

Fix SRP authentication ciphersuites.

The addition of SRP authentication needs to be checked in various places
to work properly. Specifically:

A certificate is not sent.
A certificate request must not be sent.
Server key exchange message must not contain a signature.
If appropriate SRP authentication ciphersuites should be chosen.
Reviewed-by: Matt Caswell <matt@openssl.org>
(cherry picked from commit 8f5a8805b82d1ae81168b11b7f1506db9e047dec)

Test SRP authentication ciphersuites.

Reviewed-by: Matt Caswell <matt@openssl.org>
(cherry picked from commit 193c1c07165b0042abd217274a084b49459d4443)

Only use FIPS EC methods in FIPS mode.

Reviewed-by: Tim Hudson <tjh@openssl.org>
(cherry picked from commit 1433cac53c93f9f109290389f60b17078a572d3d)

Check SRP parameters early.

Check SRP parameters when they are received so we can send back an
appropriate alert.
Reviewed-by: Kurt Roeckx <kurt@openssl.org>

Fix SRP buffer overrun vulnerability.

Invalid parameters passed to the SRP code can be overrun an internal
buffer. Add sanity check that g, A, B < N to SRP code.

Thanks to Sean Devlin and Watson Ladd of Cryptography Services, NCC
Group for reporting this issue.

Fix SRP ciphersuite DoS vulnerability.

If a client attempted to use an SRP ciphersuite and it had not been
set up correctly it would crash with a null pointer read. A malicious
server could exploit this in a DoS attack.

Thanks to Joonas Kuorilehto and Riku Hietamäki from Codenomicon
for reporting this issue.

CVE-2014-2970
Reviewed-by: Tim Hudson <tjh@openssl.org>

Fix race condition in ssl_parse_serverhello_tlsext

CVE-2014-3509
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Dr. Stephen Henson <steve@openssl.org>

Fix OID handling:

- Upon parsing, reject OIDs with invalid base-128 encoding.
- Always NUL-terminate the destination buffer in OBJ_obj2txt printing function.

CVE-2014-3508

Reviewed-by: Dr. Stephen Henson <steve@openssl.org>
Reviewed-by: Kurt Roeckx <kurt@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>

Fix DTLS anonymous EC(DH) denial of service

CVE-2014-3510

Reviewed-by: Dr. Stephen Henson <steve@openssl.org>

Fix protocol downgrade bug in case of fragmented packets

CVE-2014-3511

Reviewed-by: Emilia Käsper <emilia@openssl.org>
Reviewed-by: Bodo Möller <bodo@openssl.org>

Remove some duplicate DTLS code.

In a couple of functions, a sequence number would be calculated twice.

Additionally, in |dtls1_process_out_of_seq_message|, we know that
|frag_len| <= |msg_hdr->msg_len| so the later tests for |frag_len <
msg_hdr->msg_len| can be more clearly written as |frag_len !=
msg_hdr->msg_len|, since that's the only remaining case.

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Emilia Käsper <emilia@openssl.org>

Applying same fix as in dtls1_process_out_of_seq_message. A truncated DTLS fragment would cause *ok to be clear, but the return value would still be the number of bytes read.

Problem identified by Emilia Käsper, based on previous issue/patch by Adam
Langley.

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

Fix return code for truncated DTLS fragment.

Previously, a truncated DTLS fragment in
|dtls1_process_out_of_seq_message| would cause *ok to be cleared, but
the return value would still be the number of bytes read. This would
cause |dtls1_get_message| not to consider it an error and it would
continue processing as normal until the calling function noticed that
*ok was zero.

I can't see an exploit here because |dtls1_get_message| uses
|s->init_num| as the length, which will always be zero from what I can
see.

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Emilia Käsper <emilia@openssl.org>

Fix memory leak from zero-length DTLS fragments.

The |pqueue_insert| function can fail if one attempts to insert a
duplicate sequence number. When handling a fragment of an out of
sequence message, |dtls1_process_out_of_seq_message| would not call
|dtls1_reassemble_fragment| if the fragment's length was zero. It would
then allocate a fresh fragment and attempt to insert it, but ignore the
return value, leaking the fragment.

This allows an attacker to exhaust the memory of a DTLS peer.

Fixes CVE-2014-3507

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Emilia Käsper <emilia@openssl.org>

Fix DTLS handshake message size checks.

In |dtls1_reassemble_fragment|, the value of
|msg_hdr->frag_off+frag_len| was being checked against the maximum
handshake message size, but then |msg_len| bytes were allocated for the
fragment buffer. This means that so long as the fragment was within the
allowed size, the pending handshake message could consume 16MB + 2MB
(for the reassembly bitmap). Approx 10 outstanding handshake messages
are allowed, meaning that an attacker could consume ~180MB per DTLS
connection.

In the non-fragmented path (in |dtls1_process_out_of_seq_message|), no
check was applied.

Fixes CVE-2014-3506

Wholly based on patch by Adam Langley with one minor amendment.

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

Added comment for the frag->reassembly == NULL case as per feedback from Emilia

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

Avoid double free when processing DTLS packets.

The |item| variable, in both of these cases, may contain a pointer to a
|pitem| structure within |s->d1->buffered_messages|. It was being freed
in the error case while still being in |buffered_messages|. When the
error later caused the |SSL*| to be destroyed, the item would be double
freed.

Thanks to Wah-Teh Chang for spotting that the fix in 1632ef74 was
inconsistent with the other error paths (but correct).

Fixes CVE-2014-3505

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Emilia Käsper <emilia@openssl.org>

make update

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

Fix error discrepancy.

We can't rename ssleay_rand_bytes to md_rand_bytes_lock as this will cause
an error code discrepancy. Instead keep ssleay_rand_bytes and add an
extra parameter: since ssleay_rand_bytes is not part of the public API
this wont cause any binary compatibility issues.
Reviewed-by: Kurt Roeckx <kurt@openssl.org >

Update $default_depflags to match current defaults.

Clean up CHANGES files: If a change is already present in 1.0.1f or 1.0.1h,
don't list it again under changes between 1.0.1h and 1.0.2.

Simplify and fix ec_GFp_simple_points_make_affine
(which didn't always handle value 0 correctly).

Reviewed-by: emilia@openssl.org

Avoid multiple lock using FIPS DRBG.

Don't use multiple locks when SP800-90 DRBG is used outside FIPS mode.

PR#3176
Reviewed-by: Rich Salz <rsalz@openssl.org>

Add conditional unit testing interface.

Don't call internal functions directly call them through
SSL_test_functions(). This also makes unit testing work on
Windows and platforms that don't export internal functions
from shared libraries.

By default unit testing is not enabled: it requires the compile
time option "enable-unit-test".
Reviewed-by: Geoff Thorpe <geoff@openssl.org>
(cherry picked from commit e0fc7961c4fbd27577fb519d9aea2dc788742715)

Conflicts:

ssl/heartbeat_test.c
ssl/ssl.h
util/mkdef.pl

Prepare for 1.0.2-beta3-dev

Reviewed-by: Stephen Henson <steve@openssl.org>

Prepare for 1.0.2-beta2 release

Reviewed-by: Stephen Henson <steve@openssl.org>

make update

Reviewed-by: Stephen Henson <steve@openssl.org>

update $default_depflags

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

CHANGES: mention new platforms.

Reviewed-by: Dr. Stephen Henson <steve@openssl.org>

"EC_POINT_invert" was checking "dbl" function pointer instead of "invert".

PR#2569

Reviewed-by: Rich Salz <rsalz@openssl.org>
(cherry picked from commit cba11f57ce161fd301a72194827327128191de7e)

Remove old unused and unmaintained demonstration code.

Reviewed-by: Dr. Stephen Henson <steve@openssl.org>
(cherry picked from commit 62352b8138018775a4c085a105fccd9cdcb6323f)

sha1-ppc.pl: shave off one cycle from BODY_20_39
and improve performance by 10% on POWER[78].

Reviewed-by: Kurt Roeckx <kurt@openssl.org>
(cherry picked from commit 5c3598307ebbf5a88d1c39fbb2629536e443a5dd)

Minor documentation update removing "really" and a
statement of opinion rather than a fact.

Reviewed-by: Dr. Stephen Henson <steve@openssl.org>
Reviewed-by: Rich Salz <rsalz@openssl.org>
(cherry picked from commit c8d133e4b6f1ed1b7ad3c1a6d2c62f460e26c050)