Delay setting the sig algs until after the cert_cb has been called

Otherwise the sig algs are reset if SSL_set_SSL_CTX() gets called.

Fixes #7244

Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/7257)

crypto/bn/asm/x86_64-gcc.c: remove unnecessary redefinition of BN_ULONG

This module includes bn.h via other headers, so it picks up the
definition from there and doesn't need to define them locally (any
more?).  Worst case scenario, the redefinition may be different and
cause all sorts of compile errors.

Fixes #7227

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/7287)

/dev/crypto engine: add missing RC4 parameter

Fixes #7280

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/7281)

Add some missing ciphers in 'enc' document

The original issue is #7273 and this commit fixes part of that issue.

[skip ci]

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/7275)

util/mkdef.pl, util/add-depends.pl: don't lowercase file names

It turns out to be detrimental on some file systems that may or may not
be case sensitive (such as NTFS, which has a case sensitive mode).

Fixes #7172

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7172)

crypto/ui/ui_openssl.c: make sure to recognise ENXIO and EIO too

These both indicate that the file descriptor we're trying to use as a
terminal isn't, in fact, a terminal.

Fixes #7271

Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/7272)

Reset TLS 1.3 ciphers in SSL_CTX_set_ssl_version()

Historically SSL_CTX_set_ssl_version() has reset the cipher list
to the default.  Splitting TLS 1.3 ciphers to be tracked separately
caused a behavior change, in that TLS 1.3 cipher configuration was
preserved across calls to SSL_CTX_set_ssl_version().  To restore commensurate
behavior with the historical behavior, set the ciphersuites to the default as
well as setting the cipher list to the default.

Closes: #7226

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7270)

Add a GMAC demonstration program.

Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/7249)

ssl/ssl_ciph.c: make set_ciphersuites static

Fixes #7252

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/7253)

Trivial test improvements

This commit reuses a variable instead of reevaluating the expression
and updates an outdated comment in the EVP test.

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/7242)

Fixed typos in hkdf documentation.

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/7236)

Add missing include file.
Specifically, include e_os.h to pick up alloca definition for WIN32.

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7234)

Add a compile time test to verify that openssl/rsa.h and complex.h can
coexist.

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7233)

Use 'i' as parameter name not 'I'.

The latter causes problems when complex.h is #included.

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7233)

Update RAND_DRBG.pod

Fixed a minor typo while reading the documentation.
I agree that this contribution is trivial can be freely used.

CLA: trivial

Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
(Merged from https://github.com/openssl/openssl/pull/7221)

Improve SSL_shutdown() documentation

Reviewed-by: Ben Kaduk <kaduk@mit.edu>
GH: #7188

VMS: only use the high precision on VMS v8.4 and up

It simply isn't available on older versions.

Issue submitted by Mark Daniels

Fixes #7229

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/7230)

(cherry picked from commit d6d6aa3521e207c2727bbd1e5c97772502d15739)

VMS libtestutil: look for lower case "main"

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7208)

VMS: turn on name mangling for all our programs

With the change to have separate object files by intent, VMS name
mangling gets done differently.  While we previously had that for
libraries only, we must now turn that on generally for our programs,
because some of them depend in internal libraries where mangled names
are all that there is.

Dynamic modules are still built with non-mangled names, which is good
enough to show that it's possible to build with our public libraries
using our public headers.

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7208)

VMS build: fix a misspelled 'bin_cflags' and a wrongly coded 'NO_INST_'

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7208)

Make some return checks consistent with others

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/7209)

Don't allow -early_data with other options where it doesn't work

-early_data is not compatible with -www, -WWW, -HTTP or -rev.

Fixes #7200

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7206)

Add an explicit cast to time_t

Caused a compilation failure in some environments

Fixes #7204

Reviewed-by: Viktor Dukhovni <viktor@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7205)

Update the documentation on libobj2shlib / obj2shlib

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7198)

VMS: stop trying to build shared libraries from static ones

The possibility to do this was killed when we started producing object
file names with encoded intention (and possibly different builds), and
leads to build errors.

With that, 'libobj2shlib' is renamed to 'obj2shlib' to reflect this
design change.  The old name is still used if the new one isn't
available, for the sake of backward compatibility.

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7198)

Replace the public RAND_DRBG_USED_FLAGS #define by an internal constant

The new DRBG API added the aforementioned #define. However, it is
used internally only and having it defined publicly does not serve
any purpose except causing potential version compatibility problems.

Fixes #7182

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/7190)

Fix a possible recursion in SSLfatal handling

Fixes: #7161 (hopefully)

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7175)

Update README.md

Fixes a minor typo that would cause the linker to complain about not finding -lFuzzer

CLA: trivial

Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7197)

minor fixes for Windows

- fix to use secure URL in generated Windows resources
- fix a potentially uninitialized variable
- fix an unused variable warning

CLA: trivial

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/7189)

crypto/sm2/sm2_sign.c: ensure UINT16_MAX is properly defined

Fixes #7186

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7193)

Build files: Separate 'lib' intent from 'shlib' intent

This is in preparation for having separate CFLAGS variables for static
and for shared library builds.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7159)

Configure: Name object files according to the product they are part of

This will allow to have different object files for different products,
even if they share the same source code, and possibly different builds
for those different object files.

For example, one can have something like this:

    SOURCES[libfoo]=cookie.c
    INCLUDES[libfoo]=include/foo
    SOURCES[libbar]=cookie.c
    INCLUDES[libbar]=include/bar

This would mean that the object files and libraries would be build
somewhat like this:

    $(CC) -Iinclude/foo -o libfoo-lib-cookie.o cookie.c
    $(AR) $(ARFLAGS) libfoo.a libfoo-lib-cookie.o
    $(CC) -Iinclude/bar -o libbar-lib-cookie.o cookie.c
    $(AR) $(ARFLAGS) libbar.a libbar-lib-cookie.o

Reviewed-by: Paul Dale <paul.dale@oracle.com>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7159)

Configure: DON'T trickle down includes from products to sources

Instead, use the include settings from the products later in the process,
making it possible to have different includes for two different libraries
that share the same source code.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7159)

Limit the number of AES-GCM keys allowed in TLS.  A new error is raised if this
limit is ever reached.

This is a FIPS 140-2 requirement from IG A.5 "Key/IV Pair Uniqueness
Requirements from SP 800-38D".

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7129)

Speed for HMACs.

Add support for HMAC over any evp supported digest.

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/6945)

Add a note to CHANGES indicating that AES-XTS now enforces two different
keys.

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7120)

FIPS 140-2 IG A.9 XTS key check.

Add a check that the two keys used for AES-XTS are different.

One test case uses the same key for both of the AES-XTS keys.  This causes
a failure under FIP 140-2 IG A.9.  Mark the test as returning a failure.

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7120)

The next version in master is at least 1.1.2, not 1.1.1x

The OMC hasn't yet decided what the next release version will be, but
it's at least going to 1.1.2, so we set that value for the moment.

Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/7180)

Prepare for 1.1.1a-dev

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

Prepare for 1.1.1 release

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

Update copyright year

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7176)

CAPI engine: add support for RSA_NO_PADDING

Since the SSL code started using RSA_NO_PADDING, the CAPI engine became
unusable.  This change fixes that.

Fixes #7131

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7174)

Check the return value from ASN1_INTEGER_set

Found by Coverity

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7169)

Validate the SM2 digest len before use

Fixes a Coverity complaint.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7170)

Don't cast an int * to a size_t *

If sizeof(int) != sizeof(size_t) this may not work correctly.

Fixes a Coverity issue.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7168)

More updates to CHANGES and NEWS for the 1.1.1 release

Reviewed-by: Ben Kaduk <kaduk@mit.edu>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7167)

Updates NEWS for the 1.1.1 release

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7164)

Update the pyca-cryptography submodule

Hopefully this will resolve spurious travis failures.

[extended tests]

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7163)

Add a sentence in CHANGES to note SM2 support

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7160)

test/evp_extra_test.c: fix null pointer dereference

It's actually not a real issue but caused by the absence of the default case
which does not occur in reality but which makes coverity see a code path where
pkey remains unassigned.

Reported by Coverity Scan (CID 1423323)
[extended tests]

Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7158)

crypto/rsa/rsa_pss.c: silence coverity warning

Reported by Coverity Scan (CID 1439138)
[extended tests]

Reviewed-by: Ben Kaduk <kaduk@mit.edu>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7156)

test/dhtest.c: fix resource leak

Reported by Coverity Scan (CID 1439136)
[extended tests]

Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7155)

ASN.1 DER: Make INT32 / INT64 types read badly encoded LONG zeroes

The deprecated ASN.1 type LONG / ZLONG (incorrectly) produced zero
length INTEGER encoding for zeroes.  For the sake of backward
compatibility, we allow those to be read without fault when using the
replacement types INT32 / UINT32 / INT64 / UINT64.

Fixes #7134

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7144)

TESTS: add test of decoding of invalid zero length ASN.1 INTEGER zero

Confirms #7134

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/7153)

SipHash: add separate setter for the hash size

This was originally part of SipHash_Init.  However, there are cases
where there isn't any key material to initialize from when setting the
hash size, and we do allow doing so with a EVP_PKEY control.  The
solution is to provide a separate hash_size setter and to use it in
the corresponding EVP_PKEY_METHOD.

Fixes #7143

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7145)

TESTS: add SipHash tests with digestsize controls

Confirms #7143

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7154)

SipHash: make it possible to control the hash size through string controls

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7154)

test/evp_test.c: make it possible to use controls with MAC tests

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7154)

Do not reset SNI data in SSL_do_handshake()

PR #3783 introduce coded to reset the server side SNI state in
SSL_do_handshake() to ensure any erroneous config time SNI changes are
cleared. Unfortunately SSL_do_handshake() can be called mid-handshake
multiple times so this is the wrong place to do this and can mean that
any SNI data is cleared later on in the handshake too.

Therefore move the code to a more appropriate place.

Fixes #7014

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Viktor Dukhovni <viktor@openssl.org>
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/7149)

Simplify SSL_get_servername() to avoid session references

Ideally, SSL_get_servername() would do exactly as it is documented
and return exactly what the client sent (i.e., what we currently
are stashing in the SSL's ext.hostname), without needing to refer
to an SSL_SESSION object.  For historical reasons, including the
parsed SNI value from the ClientHello originally being stored in the
SSL_SESSION's ext.hostname field, we have had references to the
SSL_SESSION in this function.  We cannot fully excise them due to
the interaction between user-supplied callbacks and TLS 1.2 resumption
flows, where we call all callbacks but the client did not supply an
SNI value.  Existing callbacks expect to receive a valid SNI value
in this case, so we must fake one up from the resumed session in
order to avoid breakage.

Otherwise, greatly simplify the implementation and just return the
value in the SSL, as sent by the client.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7115)

Restore historical SSL_get_servername() behavior

Commit 1c4aa31d79821dee9be98e915159d52cc30d8403 modified the state machine
to clean up stale ext.hostname values from SSL objects in the case when
SNI was not negotiated for the current handshake.  This is natural from
the TLS perspective, since this information is an extension that the client
offered but we ignored, and since we ignored it we do not need to keep it
around for anything else.

However, as documented in https://github.com/openssl/openssl/issues/7014 ,
there appear to be some deployed code that relies on retrieving such an
ignored SNI value from the client, after the handshake has completed.
Because the 1.1.1 release is on a stable branch and should preserve the
published ABI, restore the historical behavior by retaining the ext.hostname
value sent by the client, in the SSL structure, for subsequent retrieval.

[extended tests]

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7115)

Ensure certificate callbacks work correctly in TLSv1.3

The is_tls13_capable() function should not return 0 if no certificates
are configured directly because a certificate callback is present.

Fixes #7140

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/7141)

Remove a reference to SSL_force_post_handshake_auth()

That function was removed in favour of SSL_set_post_handshake_auth().
Update the docs accordingly.

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/7139)

Test that we can handle a PHA CertificateRequest after we sent close_notify

Even though we already sent close_notify the server may not have recieved
it yet and could issue a CertificateRequest to us. Since we've already
sent close_notify we can't send any reasonable response so we just ignore
it.

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7114)

Test that we can process a KeyUpdate received after we sent close_notify

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7114)

Process KeyUpdate and NewSessionTicket messages after a close_notify

If we've sent a close_notify then we are restricted about what we can do
in response to handshake messages that we receive. However we can sensibly
process NewSessionTicket messages. We can also process a KeyUpdate message
as long as we also ignore any request for us to update our sending keys.

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7114)

Add missing SM2err and fix doc nits

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7113)

Allow EVP_MD_CTX_set_pkey_ctx to accept NULL pctx

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7113)

Add a SM2(7) man page

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7113)

Update document for SM2 stuffs

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7113)

Add test case for SM2 evp verification

This test case is originally submitted in #6757, by Jack Lloyd. The test
case has been modified to use the a different method to set the ID when
computing the Z hash of SM2 signature.

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Yang <yang.yang@baishancloud.com>
(Merged from https://github.com/openssl/openssl/pull/7113)

Support setting SM2 ID

zero-length ID is allowed, but it's not allowed to skip the ID.

Fixes: #6534

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7113)

Make SM2 ID stick to specification

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7113)

Support pmeth->digest_custom

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7113)

Introduce EVP_MD_CTX_set_pkey_ctx

Thus users can use this function to set customized EVP_PKEY_CTX to
EVP_MD_CTX structure.

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7113)

Remove unnecessary sm2_za.c

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7113)

fuzz/driver.c: appease -Wmissing-prototypes

Signed-off-by: Patrick Steuer <patrick.steuer@de.ibm.com>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/7138)

Support EdDSA in apps/speed

This addresses issue #6922.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7073)

Avoid SEGV when giving X509_sign a NULL private key.

Put a NULL check back in to avoid dereferencing the NULL pointer.

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7146)

RSA padding Zeroization fixes

Reviewed-by: Paul Yang <yang.yang@baishancloud.com>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/7090)

Key zeroization fix for EVP_SealInit + added simple test

Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/7105)

Harmonize the error handling codepath

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/7121)

Fix segfault in RSA_free() (and DSA/DH/EC_KEY)

`RSA_free()` and friends are called in case of error from
`RSA_new_method(ENGINE *e)` (or the respective equivalent functions).

For the rest of the description I'll talk about `RSA_*`, but the same
applies for the equivalent `DSA_free()`, `DH_free()`, `EC_KEY_free()`.

If `RSA_new_method()` fails because the engine does not implement the
required method, when `RSA_free(RSA *r)` is called,
`r->meth == NULL` and a segfault happens while checking if
`r->meth->finish` is defined.

This commit fixes this issue by ensuring that `r->meth` is not NULL
before dereferencing it to check for `r->meth->finish`.

Fixes #7102 .

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/7121)

New openssl subject parser hard to debug

-subj 'subject=C = US, ST = A, L = root, O = Hewlett Packard Enterprise Company, OU = Remote Device Access, CN = Hewlett Packard Enterprise Remote Device Access Test Local CA, emailAddress = rda@hpe.com'
was a valid subject in openssl 1.0. Error received in 1.1 is:

problems making Certificate Request

Not very informative, I only figured this out because I compiled the
code and added logging.

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/7098)

hkdf zeroization fix

Reviewed-by: Paul Yang <yang.yang@baishancloud.com>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/7106)

key zeroisation fix for p12

Reviewed-by: Paul Yang <yang.yang@baishancloud.com>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/7109)

key zeroisation for pvkfmt now done on all branch paths

Reviewed-by: Paul Yang <yang.yang@baishancloud.com>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/7107)

key zeroization fix for a branch path of tls13_final_finish_mac

Reviewed-by: Paul Yang <yang.yang@baishancloud.com>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/7110)

openssl req: don't try to report bits

With the introduction of -pkeyopt, the number of bits may change
without |newkey| being updated.  Unfortunately, there is no API to
retrieve the information from a EVP_PKEY_CTX either, so chances are
that we report incorrect information.  For the moment, it's better not
to try to report the number of bits at all.

Fixes #7086

Reviewed-by: Paul Yang <yang.yang@baishancloud.com>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/7096)

Clarify the return value of SSL_client_version()

The SSL_client_version() function returns the value held in the
legacy_version field of the ClientHello. This is never greater than
TLSv1.2, even if TLSv1.3 later gets negotiated.

Fixes #7079

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/7095)

Add a test for RSA key exchange with both RSA and RSA-PSS certs

Check that we use an RSA certificate if an RSA key exchange ciphersuite
is being used and we have both RSA and RSA-PSS certificates configured.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7099)

Don't use an RSA-PSS cert for RSA key exchange

If we have selected a ciphersuite using RSA key exchange then we must
not attempt to use an RSA-PSS cert for that.

Fixes #7059

Reviewed-by: Paul Dale <paul.dale@oracle.com>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7099)

Test creation of tickets when using a TLSv1.3 PSK

Add a test to check that we create the correct number of tickets after a
TLSv1.3 PSK.

Reviewed-by: Viktor Dukhovni <viktor@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7097)

Send a NewSessionTicket after using an external PSK

Treat a connection using an external PSK like we would a resumption and
send a single NewSessionTicket afterwards.

Fixes #6941

Reviewed-by: Viktor Dukhovni <viktor@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7097)

Clarify the EVP_DigestSignInit docs

They did not make it clear how the memory management works for the |pctx|
parameter.

Fixes #7037

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7042)

Ignore EPIPE when sending NewSessionTickets in TLSv1.3

If a client sends data to a server and then immediately closes without
waiting to read the NewSessionTickets then the server can receive EPIPE
when trying to write the tickets and never gets the opportunity to read
the data that was sent. Therefore we ignore EPIPE when writing out the
tickets in TLSv1.3

Fixes #6904

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/6944)

Fix example in crl(1) man page

The default input format is PEM, so explicit "-inform DER" is needed to
read DER-encoded CRL.

CLA: trivial

Reviewed-by: Paul Yang <yang.yang@baishancloud.com>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7094)

Fixed a comment that referenced the wrong method

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/7100)

Fix HMAC SHA3-224 and HMAC SHA3-256.

Added NIST test cases for these two as well.

Additionally deprecate the public definiton of HMAC_MAX_MD_CBLOCK in 1.2.0.

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/6972)

demos/evp: add make clean

Add make clean for evp demos and remove whitespace from a line.

CLA: trivial

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/7072)