Reviewed-by: NRichard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5108)

On Windows, we sometimes see a behavior with SO_REUSEADDR where there
remains lingering listening sockets on the same address and port as a
newly created one.

To avoid this scenario, we don't create a new proxy port for each new
client run.  Instead, we create one proxy socket when the proxy object
is created, and close it when destroying that object.
Reviewed-by: NBernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/5095)

Reviewed-by: NAndy Polyakov <appro@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5060)

Reviewed-by: NAndy Polyakov <appro@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5060)

Because OPENSSL_SYS_CYGWIN will keep OPENSSL_SYS_UNIX defined, there's
no point having checks of this form:

    #if (defined(OPENSSL_SYS_UNIX) || defined(OPENSSL_SYS_CYGWIN))
Reviewed-by: NAndy Polyakov <appro@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5060)

More to the point, Cygwin is a POSIX API.  In our library, the use of
a POSIX API is marked by defining the macro OPENSSL_SYS_UNIX.
Therefore, that macro shouldn't be undefined when building for Cygwin.
Reviewed-by: NAndy Polyakov <appro@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5060)

On Windows, we sometimes see a behavior with SO_REUSEADDR where there
remains lingering listening sockets on the same address and port as a
newly created one.

An easy solution is not to use ReuseAddr on Windows.

Thanks Bernd Edlinger for the suggestion.
Reviewed-by: NBernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/5103)

CLA: trivial
Reviewed-by: NMatt Caswell <matt@openssl.org>
Reviewed-by: NTim Hudson <tjh@openssl.org>
Reviewed-by: NRich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/4861)

On Windows, it seems that doing so in a forked (pseudo-)process
sometimes affects the parent, and thereby hides all the results that
are supposed to be seen by the running test framework (the "ok" and
"not ok" lines).

It turns out that our redirection isn't necessary, as the test
framework seems to swallow it all in non-verbose mode anyway.
It's possible that we did need this at some point, but the framework
has undergone some refinement since then...
Reviewed-by: NRich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5100)

We use the first we can of the following IO::Socket modules to create
sockets:

- IO::Socket::INET6
- IO::Socket::IP
- IO::Socket::INET

The last of them doesn't support IPv6, so if that's the one available,
we must force the s_client and s_server processes to use IPv4.
Reviewed-by: NBernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/5092)

Because the related PR/commits are merged in 2018...
Reviewed-by: NAndy Polyakov <appro@openssl.org>
Reviewed-by: NMatt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/4976)

To avoid check failure, make dummy RETURN VALUES sections in the docs
which have no real functions decribed inside...
Reviewed-by: NAndy Polyakov <appro@openssl.org>
Reviewed-by: NMatt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/4976)

All missing sections are added.
Reviewed-by: NAndy Polyakov <appro@openssl.org>
Reviewed-by: NMatt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/4976)

I received this error from a TLSProxy test:

Failed to start up server (localhost,4443): Transport endpoint is not
connected

So, extend the timeout before we give up trying to connect to the server.
Reviewed-by: NRichard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5088)

Commit 9f944291 changed the semantics of BN_copy() to additionally
copy the BN_FLG_CONSTTIME flag if it is set. This turns out to be
ill advised as it has unintended consequences. For example calling
BN_mod_inverse_no_branch() can sometimes return a result with the flag
set and sometimes not as a result. This can lead to later failures if we
go down code branches that do not support constant time, but check for
the presence of the flag.

The original commit was made due to an issue in BN_MOD_CTX_set(). The
original PR fixed the problem in that function, but it was changed in
review to fix it in BN_copy() instead. The solution seems to be to revert
the BN_copy() change and go back to the originally proposed way.
Reviewed-by: NPaul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/5080)

Fixes #5076

Since do_rand_drbg_init() allocates three locks, it needs to ensure
that OPENSSL_init_crypto() is called, otherwise these resources are
not cleaned up properly.
Reviewed-by: NMatt Caswell <matt@openssl.org>
Reviewed-by: NBen Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/5083)

The same kind of failure that has already been observed on the
s_client can sometimes also be observed on s_server, so we need to add
the same kind of 50ms delay as was previously added on s_client.

Ref: git commit cb2e10f2:
Reviewed-by: NMatt Caswell <matt@openssl.org>
Reviewed-by: NRich Salz <rsalz@openssl.org>
Reviewed-by: NBernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/5079)

Only when building the main shared libraries

Fixes #5075
Reviewed-by: NRich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5077)

This was identified for Windows almost two years ago for VC and
msys/mingw.  It seems that Cygwin suffers from the same issue, and
since Cygwin doesn't define OPENSSL_SYS_WINDOWS, we need to make a
special case to have a 50ms pause before closing the TLS connection.

Ref: git commit cb2e10f2Reviewed-by: NBernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/5074)

Reviewed-by: NRichard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5072)

Reviewed-by: NRichard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5072)

Reviewed-by: NRichard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5071)

This to avoid a clash with a #defined symbol on Solaris from time.h.
Reviewed-by: NTim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5064)

The data argument of SSL_dane_tlsa_add is used read-only, so it
should be const.
Reviewed-by: NViktor Dukhovni <viktor@openssl.org>
Reviewed-by: NRich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5056)

- Avoid creating repeated year ranges <current_year>-<current_year> for
  documents which were added in the current year.

- Use extended regular expressions for better readability (less quoting)

- Use a shebang line which is more portable
Reviewed-by: NRich Salz <rsalz@openssl.org>
Reviewed-by: NRichard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5044)

In the case of a protocol version alert being sent by a peer the record
version number may not be what we are expecting. In DTLS records with an
unexpected version number are silently discarded. This probably isn't
appropriate for alerts, so we tolerate a mismatch in the minor version
number.

This resolves an issue reported on openssl-users where an OpenSSL server
chose DTLS1.0 but the client was DTLS1.2 only and sent a protocol_version
alert with a 1.2 record number. This was silently ignored by the server.
Reviewed-by: NViktor Dukhovni <viktor@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5018)

Reviewed-by: NRich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5047)

Found with gcc 4.8.4
Reviewed-by: NBen Kaduk <kaduk@mit.edu>
Reviewed-by: NRichard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5045)

Reviewed-by: NTim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5038)

Reviewed-by: NAndy Polyakov <appro@openssl.org>
Reviewed-by: NRichard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/4994)

Closes: https://github.com/openssl/openssl/issues/4897Signed-off-by: NChristian Heimes <christian@python.org>
Reviewed-by: NTim Hudson <tjh@openssl.org>
Reviewed-by: NMatt Caswell <matt@openssl.org>
Reviewed-by: NRich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/4898)

This avoids situations where third party client is unable to recognize
that the client certificate was issued by the same CA with name differring
only by case or insignificant characters.
Reviewed-by: NMatt Caswell <matt@openssl.org>
Reviewed-by: NTim Hudson <tjh@openssl.org>
Reviewed-by: NRich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/4731)

In man1/engine.pod and man3/ENGINE_add.pod
Reviewed-by: NRichard Levitte <levitte@openssl.org>
Reviewed-by: NAndy Polyakov <appro@openssl.org>
Reviewed-by: NRich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/4956)

Signed-off-by: NPatrick Steuer <patrick.steuer@de.ibm.com>
Reviewed-by: NTim Hudson <tjh@openssl.org>
Reviewed-by: NBen Kaduk <kaduk@mit.edu>
Reviewed-by: NRich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/4958)

Many of the sub-commands under apps/ accept cipher or digest arguments like
"-sha256". These are implemented using a catchall flag that runs the result
through opt_md() or opt_cipher(). That means any unrecognized flag, including
typos, gets sent to those two functions, producing confusing error messages like
below:

    $ ./apps/openssl req -x590
    req: Unrecognized digest x590
    req: Use -help for summary.

This change switches these two functions to say "Unrecognized flag X" instead.
The new message deliberately leaves off the "-" from the flag name, because
there are some cases where opt_md() and opt_cipher() are passed a flag value
instead (for instance, openssl ca -md). I think the new message is generic
enough that it can serve both cases with improved clarity.

CLA: trivial
Reviewed-by: NAndy Polyakov <appro@openssl.org>
Reviewed-by: NRich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/4972)

Reviewed-by: NRichard Levitte <levitte@openssl.org>
Reviewed-by: NAndy Polyakov <appro@openssl.org>
Reviewed-by: NRich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5000)

CLA: trivial
Reviewed-by: NRich Salz <rsalz@openssl.org>
Reviewed-by: NBen Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/5029)

Extract the RSA key using EVP_PKEY_get0.  Type is checked externally to be either EVP_PKEY_RSA_PSS or EVP_PKEY_RSA.
Reviewed-by: NRichard Levitte <levitte@openssl.org>
Reviewed-by: NRich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/4389)

Reviewed-by: NRichard Levitte <levitte@openssl.org>
Reviewed-by: NRich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/4389)

Reviewed-by: NRichard Levitte <levitte@openssl.org>
Reviewed-by: NRich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5036)