This capability is required for read pipelining. We will only read in as
many records as will fit in the read buffer (and the network can provide
in one go). The bigger the buffer the more records we can process in
parallel.
Reviewed-by: NTim Hudson <tjh@openssl.org>

Read pipelining is controlled in a slightly different way than with write
pipelining. While reading we are constrained by the number of records that
the peer (and the network) can provide to us in one go. The more records
we can get in one go the more opportunity we have to parallelise the
processing.

There are two parameters that affect this:
* The number of pipelines that we are willing to process in one go. This is
controlled by max_pipelines (as for write pipelining)
* The size of our read buffer. A subsequent commit will provide an API for
adjusting the size of the buffer.

Another requirement for this to work is that "read_ahead" must be set. The
read_ahead parameter will attempt to read as much data into our read buffer
as the network can provide. Without this set, data is read into the read
buffer on demand. Setting the max_pipelines parameter to a value greater
than 1 will automatically also turn read_ahead on.

Finally, the read pipelining as currently implemented will only parallelise
the processing of application data records. This would only make a
difference for renegotiation so is unlikely to have a significant impact.
Reviewed-by: NTim Hudson <tjh@openssl.org>

Use the new pipeline cipher capability to encrypt multiple records being
written out all in one go. Two new SSL/SSL_CTX parameters can be used to
control how this works: max_pipelines and split_send_fragment.

max_pipelines defines the maximum number of pipelines that can ever be used
in one go for a single connection. It must always be less than or equal to
SSL_MAX_PIPELINES (currently defined to be 32). By default only one
pipeline will be used (i.e. normal non-parallel operation).

split_send_fragment defines how data is split up into pipelines. The number
of pipelines used will be determined by the amount of data provided to the
SSL_write call divided by split_send_fragment. For example if
split_send_fragment is set to 2000 and max_pipelines is 4 then:
SSL_write called with 0-2000 bytes == 1 pipeline used
SSL_write called with 2001-4000 bytes == 2 pipelines used
SSL_write called with 4001-6000 bytes == 3 pipelines used
SSL_write_called with 6001+ bytes == 4 pipelines used

split_send_fragment must always be less than or equal to max_send_fragment.
By default it is set to be equal to max_send_fragment. This will mean that
the same number of records will always be created as would have been
created in the non-parallel case, although the data will be apportioned
differently. In the parallel case data will be spread equally between the
pipelines.
Reviewed-by: NTim Hudson <tjh@openssl.org>

Avoids modifying certificate reference count, and thereby avoids locking.
Reviewed-by: NEmilia Käsper <emilia@openssl.org>
Reviewed-by: NRich Salz <rsalz@openssl.org>

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

Without this, the peer certificate would never be deleted, resulting in
a memory leak.
Reviewed-by: NEmilia Käsper <emilia@openssl.org>
Reviewed-by: NRich Salz <rsalz@openssl.org>

Disabled by default, but can be enabled by setting the
ct_validation_callback on a SSL or SSL_CTX.
Reviewed-by: NBen Laurie <ben@openssl.org>
Reviewed-by: NRich Salz <rsalz@openssl.org>

This patch provides getters for default_passwd_cb and userdata for SSL
and SSL_CTX. The getter functions are required to port Python's ssl module
to OpenSSL 1.1.0.
Reviewed-by: NRichard Levitte <levitte@openssl.org>
Reviewed-by: NMatt Caswell <matt@openssl.org>

Implementation experience has shown that the original plan for async wait
fds was too simplistic. Originally the async logic created a pipe internally
and user/engine code could then get access to it via API calls. It is more
flexible if the engine is able to create its own fd and provide it to the
async code.

Another issue is that there can be a lot of churn in the fd value within
the context of (say) a single SSL connection leading to continually adding
and removing fds from (say) epoll. It is better if we can provide some
stability of the fd value across a whole SSL connection. This is
problematic because an engine has no concept of an SSL connection.

This commit refactors things to introduce an ASYNC_WAIT_CTX which acts as a
proxy for an SSL connection down at the engine layer.
Reviewed-by: NRichard Levitte <levitte@openssl.org>

Simplifies calling code.  Also fixed up any !ptr tests that were
nearby, turning them into NULL tests.
Reviewed-by: NRichard Levitte <levitte@openssl.org>

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

The ssl_set_masks() function no longer depends on the cipher. This
also means there is no need to set the masks for each cipher in
ssl3_choose_cipher.
Reviewed-by: NRich Salz <rsalz@openssl.org>

Add -DBIO_DEBUG to --strict-warnings.
Remove comments about outdated debugging ifdef guards.
Remove md_rand ifdef guarding an assert; it doesn't seem used.
Remove the conf guards in conf_api since we use OPENSSL_assert, not assert.
For pkcs12 stuff put OPENSSL_ in front of the macro name.
Merge TLS_DEBUG into SSL_DEBUG.
Various things just turned on/off asserts, mainly for checking non-NULL
arguments, which is now removed: camellia, bn_ctx, crypto/modes.
Remove some old debug code, that basically just printed things to stderr:
  DEBUG_PRINT_UNKNOWN_CIPHERSUITES, DEBUG_ZLIB, OPENSSL_RI_DEBUG,
  RL_DEBUG, RSA_DEBUG, SCRYPT_DEBUG.
Remove OPENSSL_SSL_DEBUG_BROKEN_PROTOCOL.
Reviewed-by: NRichard Levitte <levitte@openssl.org>

As handshake_func is a function pointer, it should compare to NULL
Signed-off-by: NRich Salz <rsalz@openssl.org>
Reviewed-by: NMatt Caswell <matt@openssl.org>

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

Remove support for static ECDH ciphersuites. They require ECDH keys
in certificates and don't support forward secrecy.
Reviewed-by: NViktor Dukhovni <viktor@openssl.org>

Add utility macros REF_ASSERT_NOT and REF_PRINT_COUNT
This is also RT 4181
Reviewed-by: NRichard Levitte <levitte@openssl.org>

The new init functions can fail if the library has already been stopped. We
should be able to indicate failure with a 0 return value.
Reviewed-by: NRich Salz <rsalz@openssl.org>

Man, there were a lot of renamings :)
Reviewed-by: NRichard Levitte <levitte@openssl.org>

Building on the pervious commit to auto initialise and de-initialise libssl
Reviewed-by: NRichard Levitte <levitte@openssl.org>

As documented both SSL_get0_dane_authority() and SSL_get0_dane_tlsa()
are expected to return a negative match depth and nothing else when
verification fails.  However, this only happened when verification
failed during chain construction.  Errors in verification of the
constructed chain did not have the intended effect on these functions.

This commit updates the functions to check for verify_result ==
X509_V_OK, and no longer erases any accumulated match information
when chain construction fails.  Sophisticated developers can, with
care, use SSL_set_verify_result(ssl, X509_V_OK) to "peek" at TLSA
info even when verification fail.  They must of course first check
and save the real error, and restore the original error as quickly
as possible.  Hiding by default seems to be the safer interface.

Introduced X509_V_ERR_DANE_NO_MATCH code to signal failure to find
matching TLSA records.  Previously reported via X509_V_ERR_CERT_UNTRUSTED.

This also changes the "-brief" output from s_client to include
verification results and TLSA match information.

Mentioned session resumption in code example in SSL_CTX_dane_enable(3).
Also mentioned that depths returned are relative to the verified chain
which is now available via SSL_get0_verified_chain(3).

Added a few more test-cases to danetest, that exercise the new
code.

Resolved thread safety issue in use of static buffer in
X509_verify_cert_error_string().

Fixed long-stating issue in apps/s_cb.c which always sets verify_error
to either X509_V_OK or "chain to long", code elsewhere (e.g.
s_time.c), seems to expect the actual error.  [ The new chain
construction code is expected to correctly generate "chain
too long" errors, so at some point we need to drop the
work-arounds, once SSL_set_verify_depth() is also fixed to
propagate the depth to X509_STORE_CTX reliably. ]
Reviewed-by: NRich Salz <rsalz@openssl.org>

Deprecate undocumented SSL_cache_hit(). Make SSL_session_reused() into a
real function.
Reviewed-by: NRichard Levitte <levitte@openssl.org>

Previous commit 7bb196a7 attempted to "fix" a problem with the way
SSL_shutdown() behaved whilst in mid-handshake. The original behaviour had
SSL_shutdown() return immediately having taken no action if called mid-
handshake with a return value of 1 (meaning everything was shutdown
successfully). In fact the shutdown has not been successful.

Commit 7bb196a7 changed that to send a close_notify anyway and then
return. This seems to be causing some problems for some applications so
perhaps a better (much simpler) approach is revert to the previous
behaviour (no attempt at a shutdown), but return -1 (meaning the shutdown
was not successful).

This also fixes a bug where SSL_shutdown always returns 0 when shutdown
*very* early in the handshake (i.e. we are still using SSLv23_method).
Reviewed-by: NViktor Dukhovni <viktor@openssl.org>

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

Signed-off-by: NRich Salz <rsalz@openssl.org>
Reviewed-by: NKurt Roeckx <kurt@openssl.org>

CRIME protection: disable compression by default, even if OpenSSL is
compiled with zlib enabled. Applications can still enable compression by
calling SSL_CTX_clear_options(ctx, SSL_OP_NO_COMPRESSION), or by using
the SSL_CONF library to configure compression. SSL_CONF continues to
work as before:

SSL_CONF_cmd(ctx, "Options", "Compression") enables compression.

SSL_CONF_cmd(ctx, "Options", "-Compression") disables compression (now
no-op by default).

The command-line switch has changed from -no_comp to -comp.
Reviewed-by: NRich Salz <rsalz@openssl.org>

This extends the existing async functionality to SSL_shutdown(), i.e.
SSL_shutdown() can now casuse an SSL_ERROR_WANT_ASYNC error to be returned
from SSL_get_error() if async mode has been enabled.
Reviewed-by: NViktor Dukhovni <viktor@openssl.org>

The SSL and SSL_CTX structures are reference counted. However since libssl
was made opaque there is no way for users of the library to manipulate the
reference counts. This adds functions to enable that.
Reviewed-by: NStephen Henson <steve@openssl.org>

Signed-off-by: NRich Salz <rsalz@openssl.org>
Reviewed-by: NKurt Roeckx <kurt@openssl.org>

Calling SSL_shutdown while in init previously gave a "1" response, meaning
everything was successfully closed down (even though it wasn't). Better is
to send our close_notify, but fail when trying to receive one.

The problem with doing a shutdown while in the middle of a handshake is
that once our close_notify is sent we shouldn't really do anything else
(including process handshake/CCS messages) until we've received a
close_notify back from the peer. However the peer might send a CCS before
acting on our close_notify - so we won't be able to read it because we're
not acting on CCS messages!
Reviewed-by: NViktor Dukhovni <viktor@openssl.org>

Also report an SSL_dane_enable error when the basedomain is an
invalid SNI name.  Avoid side-effects when such a name is valid
with X509_VERIFY_PARAM_set1_host(), as e.g. with an empty name, by
setting the SNI name first.
Reviewed-by: NRich Salz <rsalz@openssl.org>

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

It turns out that -pause calls the undocumented function SSL_set_debug.
That just sets flag inside the SSL structure.  That flag, despite
the command is never used.  So remove the flag, the field, and the
function.
Reviewed-by: NRichard Levitte <levitte@openssl.org>

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

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

These now take and return unsigned long, and get is constified.
Updated related documentation and util/ssleay.num
Reviewed-by: NMatt Caswell <matt@openssl.org>

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

Also tweak some of the code in demos/bio, to enable interactive
testing of BIO_s_accept's use of SSL_dup.  Changed the sconnect
client to authenticate the server, which now exercises the new
SSL_set1_host() function.
Reviewed-by: NRichard Levitte <levitte@openssl.org>

The protocol selection code is now consolidated in a few consecutive
short functions in a single file and is table driven.  Protocol-specific
constraints that influence negotiation are moved into the flags
field of the method structure.  The same protocol version constraints
are now applied in all code paths.  It is now much easier to add
new protocol versions without reworking the protocol selection
logic.

In the presence of "holes" in the list of enabled client protocols
we no longer select client protocols below the hole based on a
subset of the constraints and then fail shortly after when it is
found that these don't meet the remaining constraints (suiteb, FIPS,
security level, ...).  Ideally, with the new min/max controls users
will be less likely to create "holes" in the first place.
Reviewed-by: NKurt Roeckx <kurt@openssl.org>

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