Document the new split_send_frag, max_pipelines and read_buf options.
Reviewed-by: NTim Hudson <tjh@openssl.org>

Previously s_client and s_server relied on using SSL_pending() which does
not take into account read_ahead. For read pipelining to work, read_ahead
gets set automatically. Therefore s_client and s_server have been
converted to use SSL_has_pending() instead.
Reviewed-by: NTim Hudson <tjh@openssl.org>

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>

Add the options min_send_frag and max_pipelines to s_server and s_client
in order to control pipelining capabilities. This will only have an effect
if a pipeline capable cipher is used (such as the one provided by the
dasync engine).
Reviewed-by: NTim Hudson <tjh@openssl.org>

This replaces SHUTDOWN/SHUTDOWN2 with BIO_closesocket.
Reviewed-by: NRichard Levitte <levitte@openssl.org>

in s_server cmd:
specifying -trace option, falls through and turn-on security_debug
Signed-off-by: NRich Salz <rsalz@openssl.org>
Reviewed-by: NMatt Caswell <matt@openssl.org>

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

- srtp_profiles variable is defined when building with SRTP, keeping
the variable usage also under ifndef OPENSSL_NO_SRTP
- alpn help option was kept under ifndef OPENSSL_NO_SRTP
Reviewed-by: NKurt Roeckx <kurt@openssl.org>
Reviewed-by: NRich Salz <rsalz@openssl.org>

Now that BIO_sock_nbio is available, use it in the apps.
Reviewed-by: NRichard Levitte <levitte@openssl.org>

The SRP user database lookup method SRP_VBASE_get_by_user had confusing
memory management semantics; the returned pointer was sometimes newly
allocated, and sometimes owned by the callee. The calling code has no
way of distinguishing these two cases.

Specifically, SRP servers that configure a secret seed to hide valid
login information are vulnerable to a memory leak: an attacker
connecting with an invalid username can cause a memory leak of around
300 bytes per connection.

Servers that do not configure SRP, or configure SRP but do not configure
a seed are not vulnerable.

In Apache, the seed directive is known as SSLSRPUnknownUserSeed.

To mitigate the memory leak, the seed handling in SRP_VBASE_get_by_user
is now disabled even if the user has configured a seed.

Applications are advised to migrate to SRP_VBASE_get1_by_user. However,
note that OpenSSL makes no strong guarantees about the
indistinguishability of valid and invalid logins. In particular,
computations are currently not carried out in constant time.
Reviewed-by: NRich Salz <rsalz@openssl.org>

This is also RT 4137
Signed-off-by: NRich Salz <rsalz@akamai.com>
Reviewed-by: NAndy Polyakov <appro@openssl.org>

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

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

On some platforms, the implementation is such that a signed char
triggers a warning when used with is*() functions.  On others, the
behavior is outright buggy when presented with a char that happens
to get promoted to a negative integer.

The safest thing is to cast the char that's used to an unsigned char.
Reviewed-by: NAndy Polyakov <appro@openssl.org>

    * added missing help option messages
    * ecdh_single option is removed as it is a no-op and not an option
    supported in earlier versions
    * ssl_ctx_security_debug() was invoked before ctx check for NULL
    * trusted_first option can be removed, as it is always enabled in 1.1.
    But not removed the option, require confirmation.
Signed-off-by: NRich Salz <rsalz@openssl.org>
Reviewed-by: NMatt Caswell <matt@openssl.org>

This builds on the previous commit to auto initialise/deinitialise
libcrypto.
Reviewed-by: NRichard Levitte <levitte@openssl.org>

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

The DTLSv1_listen function exposed details of the underlying BIO
abstraction and did not properly allow for IPv6. This commit changes the
"peer" argument to be a BIO_ADDR and makes it a first class function
(rather than a ctrl) to ensure proper type checking.
Reviewed-by: NRichard Levitte <levitte@openssl.org>

s_socket.c gets brutally cleaned out and now consists of only two
functions, one for client and the other for server.  They both handle
AF_INET, AF_INET6 and additionally AF_UNIX where supported.  The rest
is just easy adaptation.

Both s_client and s_server get the new flags -4 and -6 to force the
use of IPv4 or IPv6 only.

Also, the default host "localhost" in s_client is removed.  It's not
certain that this host is set up for both IPv4 and IPv6.  For example,
Debian has "ip6-localhost" as the default hostname for [::1].  The
better way is to default |host| to NULL and rely on BIO_lookup() to
return a BIO_ADDRINFO with the appropriate loopback address for IPv4
or IPv6 as indicated by the |family| parameter.
Reviewed-by: NKurt Roeckx <kurt@openssl.org>

The control commands that previously took a struct sockaddr * have
been changed to take a BIO_ADDR * instead.
Reviewed-by: NKurt Roeckx <kurt@openssl.org>

PACKET contents should be read-only. To achieve this, also
- constify two user callbacks
- constify BUF_reverse.
Reviewed-by: NRich Salz <rsalz@openssl.org>

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

Some users want to disable SSL 3.0/TLS 1.0/TLS 1.1, and enable just
TLS 1.2.  In the future they might want to disable TLS 1.2 and
enable just TLS 1.3, ...

This commit makes it possible to disable any or all of the TLS or
DTLS protocols.  It also considerably simplifies the SSL/TLS tests,
by auto-generating the min/max version tests based on the set of
supported protocols (425 explicitly written out tests got replaced
by two loops that generate all 425 tests if all protocols are
enabled, fewer otherwise).
Reviewed-by: NRichard Levitte <levitte@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: NRichard Levitte <levitte@openssl.org>

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

Rename BUF_{strdup,strlcat,strlcpy,memdup,strndup,strnlen}
to OPENSSL_{strdup,strlcat,strlcpy,memdup,strndup,strnlen}
Add #define's for the old names.
Add CRYPTO_{memdup,strndup}, called by OPENSSL_{memdup,strndup} macros.
Reviewed-by: NTim Hudson <tjh@openssl.org>

s_server was trying to set the ECDH curve when no-ec was defined. This also
highlighted the fact that the -no_ecdhe option to s_server is broken, and
doesn't make any sense any more (ECDHE is on by default and the only way it
can be disabled is through the cipherstring). Therefore this commit removes
the option.
Reviewed-by: NKurt Roeckx <kurt@openssl.org>

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

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

MR: #364

In theory the pthreads approach for Thread Local Storage should be more
portable.

This also changes some APIs in order to accommodate this change. In
particular ASYNC_init_pool is renamed ASYNC_init_thread and
ASYNC_free_pool is renamed ASYNC_cleanup_thread. Also introduced ASYNC_init
and ASYNC_cleanup.
Reviewed-by: NRich Salz <rsalz@openssl.org>

s_server already had the ability to wait on an async file descriptor. This
adds it to s_client too.
Reviewed-by: NRich Salz <rsalz@openssl.org>

0 is a valid file descriptor so SSL_get_async_wait_fd should instead return
-1 on error.
Reviewed-by: NRich Salz <rsalz@openssl.org>

If an async event occurs during a renegotiation in SSL_read then s_server
was looping around, detecting we were in init and calling
init_ssl_connection instead of re-calling SSL_read.
Reviewed-by: NRich Salz <rsalz@openssl.org>

Implement the ASYNC_JOB as a local thread pool. Remove the API support
for global pools.
Reviewed-by: NRich Salz <rsalz@openssl.org>

Initial API implemented for notifying applications that an ASYNC_JOB
has completed. Currently only s_server is using this. The Dummy Async
engine "cheats" in that it notifies that it has completed *before* it
pauses the job. A normal async engine would not do that.

Only the posix version of this has been implemented so far, so it will
probably fail to compile on Windows at the moment.
Reviewed-by: NRich Salz <rsalz@openssl.org>

It is expensive to create the ASYNC_JOB objects due to the "makecontext"
call. This change adds support for pools of ASYNC_JOB objects so that we
don't have to create a new ASYNC_JOB every time we want to use one.
Reviewed-by: NRich Salz <rsalz@openssl.org>

The s_server option -WWW was not async aware, and therefore was not
handling SSL_ERROR_WANT_ASYNC conditions. This commit fixes that.
Reviewed-by: NRich Salz <rsalz@openssl.org>

A new -async option is added which activates SSL_MODE_ASYNC. Also
SSL_WANT_ASYNC errors are handled appropriately.
Reviewed-by: NRich Salz <rsalz@openssl.org>

continue on from previous commits but in the apps directory
Reviewed-by: NKurt Roeckx <kurt@openssl.org>