Don't use '-z defs' with Clang's sanitizers

The clang documentation in all sanitizers we currently use says this:

    When linking shared libraries, the {flavor}Sanitizer run-time is
    not linked, so -Wl,-z,defs may cause link errors (don’t use it
    with {flavor}Sanitizer)

(in our case, {flavor} is one of Address, Memory, or UndefinedBehavior)

Therefore, we turn off that particular flag specifically when using
the sanitizers.

Fixes #8735

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8749)

providers/common/digests/sha2.c: forward declare all dispatched functions

Forward declare the dispatched functions using typedefs from
core_numbers.h.  This will ensure that they have correct signatures.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8747)

Providers: for the digest_final operation, pass a output buffer size

This allows the provider digest_final operation to check that it
doesn't over-run the output buffer.

The EVP_DigestFinal_ex function doesn't take that same parameter, so
it will have to assume that the user provided a properly sized buffer,
but this leaves better room for future enhancements of the public API.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8747)

Fix calling convention bug in ecp_nistz256_ord_sqr_mont

The rep parameter takes an int in C, but the assembly implementation
looks at the upper bits. While it's unlikely to happen here, where all
calls pass a constant, in other scenarios x86_64 compilers will leave
arbitrary values in the upper half.

Fix this by making the C prototype match the assembly. (This aspect of
the calling convention implies smaller-than-word arguments in assembly
functions should be avoided. There are far fewer things to test if
everything consistently takes word-sized arguments.)

This was found as part of ABI testing work in BoringSSL.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/8108)

Don't use coordinate blinding when scalar is group order

This happens in ec_key_simple_check_key and EC_GROUP_check.
Since the the group order is not a secret scalar, it is
unnecessary to use coordinate blinding.

Fixes: #8731
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8734)

Further harmonisation of manual page HISTORY sections

A couple of minor tweaks to match the style introduced in #7854:
- BIO_connect: remove line break to make more grep friendly
- SSL_CTX_new: harmoise the format of the HISTORY section

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/8729)

Update various man pages to place HISTORY section after SEE ALSO

SEE ALSO before HISTORY is the more common pattern in OpenSSL manual
pages and seems to be the prevalent order based on sampling my system
manual pages.

Fixes #8631

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/8729)

Add a check for history section location to find-doc-nits

Check that the HISTORY section is located after the SEE ALSO section,
this is a much more frequent order in OpenSSL manual pages (and UNIX
manual pages in general).

Also check that SEE ALSO comes after EXAMPLES, so that the tool can
ensure the correct manual section sequence.

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/8729)

Make check_example_location() in find-doc-nits generic

Change to check_section_location(), a generic function to ensure that
section SECTION appears before section BEFORE in the man pages.

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/8729)

Deprecate AES_ige_encrypt() and AES_bi_ige_encrypt()

These undocumented functions were never integrated into the EVP layer
and implement the AES Infinite Garble Extension (IGE) mode and AES
Bi-directional IGE mode. These modes were never formally standardised
and usage of these functions is believed to be very small. In particular
AES_bi_ige_encrypt() has a known bug. It accepts 2 AES keys, but only
one is ever used. The security implications are believed to be minimal,
but this issue was never fixed for backwards compatibility reasons.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8710)

Add CMAC speed measurements

usage: openssl speed -cmac aes128

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8721)

Add prediction resistance capability to the DRBG reseeding process.

Refer to NIST SP 800-90C section 5.4 "Prediction Resistance.l"

This requires the seed sources to be approved as entropy sources, after
which they should be considered live sources as per section 5.3.2 "Live
Entropy Source Availability."

Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/8647)

ec key validation checks updated

Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8564)

doc fixups

Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8555)

Separate the lookup test

This fixes the "verifying the alias" case.
Actually, while working on it, I realized that conceptually we were
testing the 2 different behaviours of `EC_GROUP_check_named_curve()` at
the same time, and actually not in the proper way.

I think it's fair to assume that overwriting the curve name for an
existing group with `NID_undef` could lead to the unexpected behaviour
we were observing and working around.
Thus I decided to separate the lookup test in a dedicated simpler test
that does what the documentation of `EC_GROUP_check_named_curve()`
suggests: the lookup functionality is meant to find a name for a group
generated with explicit parameters.

In case an alternative alias is returned by the lookup instead of the
expected nid, to avoid doing comparisons between `EC_GROUP`s with
different `EC_METHOD`s, the workaround is to retrieve the `ECPARAMETERS`
of the "alias group" and create a new explicit parameters group to use
in `EC_GROUP_cmp()`.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8555)

EC_GROUP_set_curve() might fail for arbitrary params

Setting arbitrary `p`, `a` or `b` with `EC_GROUP_set_curve()` might fail
for some `EC_GROUP`s, depending on the internal `EC_METHOD`
implementation, hence the block of tests verifying that
`EC_GROUP_check_named_curve()` fails when any of the curve parameters is
changed is modified to run only if the previous `EC_GROUP_set_curve()`
call succeeds.

`ERR_set_mark()` and `ERR_pop_to_mark()` are used to avoid littering the
thread error stack with unrelated errors happened during
`EC_GROUP_set_curve()`.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8555)

added code to validate EC named curve parameters

Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8555)

Added Test::ok_nofips, Test::is_nofips & Test::isnt_nofips methods.
Used to check that a test fails in fips mode i.e. ok_nofips(run(...))

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8661)

Reseeding without derivation function is not supported in FIPS mode.

Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/8648)

Fixed linux_x86_icc compiler errors in EC code related to __uint128_t/__int128_t

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8697)

Avoid creating invalid rsa pss params

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8621)

Fix typos

CLA: trivial

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8714)

EVP_*Update: ensure that input NULL with length 0 isn't passed

Even with custome ciphers, the combination in == NULL && inl == 0
should not be passed down to the backend cipher function.  The reason
is that these are the values passed by EVP_*Final, and some of the
backend cipher functions do check for these to see if a "final" call
is made.

Fixes #8675

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8676)

Make X509_set_sm2_id consistent with other setters

This commit makes the X509_set_sm2_id to 'set0' behaviour, which means
the memory management is passed to X509 and user doesn't need to free
the sm2_id parameter later. API name also changes to X509_set0_sm2_id.

Document and test case are also updated.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8626)

test/params_test.c : Adjust tests to check utf8_ptr sizes

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8703)

Params API: {utf8,octet}_ptr need to know the data size

When the purpose is to pass parameters to a setter function, that
setter function needs to know the size of the data passed.  This
remains true for the pointer data types as well.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8703)

Fix crash in X509_STORE_CTX_get_by_subject

If using a custom X509_LOOKUP_METHOD then calls to
X509_STORE_CTX_get_by_subject may crash due to an incorrectly initialised
X509_OBJECT being passed to the callback get_by_subject function.

Fixes #8673

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8698)

Add a legacy provider and put MD2 in it

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8541)

Use the right NID when putting a method in the store

When we attempt to fetch a method with a given NID we will ask the
providers for it if we don't already know about it. During that process
we may be told about other methods with a different NID. We need to
make sure we don't confuse the two.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8541)

Configurations/10-main.conf: Don't inherit assembler in Cygwin-common

The targets Cygwin-x86 and Cygwin-x86_64 are the ones that should do
this.

Fixes #8684

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8685)

Params: add OSSL_PARAM_construct_end()

OSSL_PARAM_END is a macro that can only be used to initialize an
OSSL_PARAM array, not to assign an array element later on.  For
completion, we add an end constructor to facilitate that kind of
assignment.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8704)

EC keygen updates + changed ecdsa_sign to use BN_secure_new

Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8557)

Coverity: hkdf ENV_MD_size() is an int that can be negative

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8692)

Avoid alignment problems in params API.

Using a byte buffer causes problems for device that don't handle unaligned
reads.  Instead use the properly aligned variable that was already pointed at.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8696)

s_client starttls: fix handling of multiline reply

Fixes #8645

Reviewed-by: Paul Dale <paul.dale@oracle.com>
Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8654)

coverity resource leak fixes in apps/pkeyutl

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8691)

fix --strict-warnings build

ISO C90 forbids specifying subobject to initialize

Signed-off-by: Patrick Steuer <patrick.steuer@de.ibm.com>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8693)

fix --strict-warnings build

C++ style comments are not allowed in ISO C90

Signed-off-by: Patrick Steuer <patrick.steuer@de.ibm.com>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8693)

Coverity: fix two minor NPD issues.
Found by Coverity.

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8274)

Fix big endian param API tests.

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/8629)

Fix the allocation size in EVP_OpenInit and PEM_SignFinal

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8620)

EVP configuration section: add 'default_properties' command

The value of the 'default_properties' command is simply passed to
EVP_set_default_properties().

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8681)

EVP_set_default_properties(): New function to set global properties

EVP_MD_fetch() can be given a property query string.  However, there
are cases when it won't, for example in implicit fetches.  Therefore,
we also need a way to set a global property query string to be used in
all subsequent fetches.  This also applies to all future algorithm
fetching functions.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8681)

Convert the ENGINE_CONF trace calls to use CONF instead

Additionally, merge ENGINE_CONF into CONF.

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/8680)

Add a bit of tracing in the core conf module runner

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/8680)

Rename the PROVIDER_CONF trace to CONF

Other configuration modules may have use for tracing, and having one
tracing category for each of them is a bit much.  Instead, we make one
category for them all.

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/8680)

OPENSSL_init_crypto(): check config return code correctly

It was assumed that the config functionality returned a boolean.
However, it may return a negative number on error, so we need to take
that into account.

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8679)

Fix number clash: EVP_F_AESNI_XTS_INIT_KEY vs EVP_F_EVP_MD_BLOCK_SIZE

The three macros EVP_F_AESNI_XTS_INIT_KEY, EVP_F_AES_T4_XTS_INIT_KEY
and EVP_F_AES_XTS_INIT_KEY are affected.

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8682)

Move the AES-XTS mode duplicated key check into the init_key function rather
than the update call.  The means an earlier error return at the cost of some
duplicated code.

Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/8625)

Complain if there are missing symbols when creating a provider .so file

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8537)

Test that we can use the FIPS provider

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8537)

Add a no-fips Configure option

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8537)

Create a FIPS provider and put SHA256 in it

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8537)

trace: add PROVIDER_CONF trace category

Fixes #8667

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8672)

Correct the documentation about SSL_CIPHER_description()

There are some ciphersuites that were introduced in TLSv1.0/TLSv1.1 but
are backwards compatible with SSLv3.

Fixes #8655

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8658)

Skip the correct number of tests if SM2 is disabled

Fixes no-sm2 (and also no-sm3 and no-ec)

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8650)

For provider tests, don't define a OPENSSL_NO_ macro

Since the macro to indicate if the test provider module is available
is local to the test programs, it's better to use a name that isn't
as easily confused with a library feature disabling macro that one
would expect to find in opensslconf.h.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8664)

Document the 'no-module' configuration option

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8665)

Correct EVP_F_EVP_MD_BLOCK_SIZE number

The number that was used was already taken

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8656)

Use AI_ADDRCONFIG hint with getaddrinfo if available.

This prevents failure of openssl s_server socket binding to wildcard
address on hosts with disabled IPv6.

Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8550)

Modify OCSP to use alt MD for cert IDs in responses

Modify openssl OCSP utility to produce certIDs in responses using other
hash algorithms (e.g. SHA256).

Added option -rcid for this purpose.

Reviewed-by: Ben Kaduk <kaduk@mit.edu>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/5274)

Add a test for EVP_MD_block_size()

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8604)

Support EVP_MD_block_size() with providers

Fixes #8565

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8604)

Add an EVP_MD_CTX_md() test

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8614)

Ensure EVP_MD_CTX_md returns the EVP_MD originally used

Fixes #8613

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8614)

replaced snprintf with BIO version (for windows builds)

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8651)

Replumbing: add documentation for the provider configuration module

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8549)

Add test for the provider configuration module

We reuse test/provider_internal_test.c and test/p_test.c,
and get it loaded one more time via the configuration file
test/provider_internal_test.conf

To support different platform standards regarding module
extensions, we generate test/provider_internal_test.conf

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8549)

Replumbing: add a configuration module for providers

This configuration module supports a configuration structure pretty
much like the engine configuration module, i.e. something like this:

    openssl_conf = openssl_init

    [openssl_init]
    providers = provider_section

    [provider_section]
    # Configure the provider named "foo"
    foo = foo_section
    # Configure the provider named "bar"
    bar = bar_section

    [foo_section]
    # Override name given in the provider section
    identity = myfoo
    # The exact path of the module.  This is platform specific
    module_path = /opt/openssl/modules/foo.so
    # Whether it should be automatically activated.  Value is unimportant
    activate = whatever

    # Anything else goes as well, and becomes parameters that the
    # provider can get
    what = 1
    # sub-sections will be followed as well
    ever = ever_section

    [ever_section]
    cookie = monster

All the configurations in a provider section and its sub-sections
become parameters for the provider to get, i.e. the "foo" provider
will be able to get values for the following keys (with associated
values shown):

    identity            => myfoo
    module_path         => /opt/openssl/modules/foo.so
    activate            => whatever
    what                => 1
    ever.cookie         => monster

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8549)

Replumbing: add functionality to set provider parameters

Provider parameters are parameters set by the core that the provider
can retrieve.  The primary use it to support making OpenSSL
configuration data available to the provider.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8549)

AES-XTS block limit.

Limit the number of AES blocks in a data unit to 2^20 or less.
This corresponds to the mandates in IEEE Std 1619-2018 and NIST SP 800-38E.

Note: that this is a change from IEEE Std 1619-2007 which only recommended
this limit.

Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/8627)

Revert "crypto/rand/rand_win.c: include "e_os.h" to get the default _WIN32_WINNT"

I turns out that this made crypto/rand/rand_win.c to never build with
BCrypt support unless the user sets _WIN32_WINNT.  That wasn't the
intent.

This reverts commit cc8926ec8fcecae89ceab91ef753de93e49568f9.

Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/8641)

rand_win.c: loosen version requirements for BCryptGenRandom

BCryptGenRandom() is available for Windows Vista and newer versions, see
https://docs.microsoft.com/en-us/windows/desktop/api/bcrypt/nf-bcrypt-bcryptgenrandom

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8639)

crypto/rand/rand_win.c: include "e_os.h" to get the default _WIN32_WINNT

This helps decide if the BCrypt API should be used or not.

Fixes #8635

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8638)

Correct the checks of module availability in provider test programs

Previously, the macro OPENSSL_NO_SHARED was defined of the test/p_test
module wasn't built, but the provider test programs didn't check that
macro.  We rename it to OPENSSL_NO_MODULE, since that name describes
the situation more than OPENSSL_NO_SHARED does, and use it.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8623)

Configuration / build: make it possible to disable building of modules

While we're at it, sort out inconsistencies with the build of modules:
- not building shared libraries means not building dynamic engines.
  However, other modules may still be built.
- not having DSO functionality doesn't mean not to build modules (even
  though we can't use them from apps linked with libraries that are
  built this way).

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8623)

Build cleanup: don't use SHARED_SOURCE with modules

SHARED_SOURCE is reserved for products that are expected to come in
dual shared / non-shared form, i.e. the routine libraries like
libcrypto and libssl, to distinguish source that should only appear in
their shared form.

Modules are always shared, so there's no need for them to have this
type of distinction.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8623)

Restore the "heartbeats" configuration option among the deprecated

Removing the option entirely would break builds unnecessarily, so
let's make it deprecated.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8632)

add documentation

Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7848)

apps: print Kernel receive side TLS in s_client and s_server

Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7848)

sslapitest: add test ktls Rx

Add a unit-test for ktls receive side.

Change-Id: I890588681d05fba419f644f6d903be6dc83c9ed5
Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7848)

ssl: Linux TLS Rx Offload

This patch adds support for the Linux TLS Rx socket option.
It completes the previous patch for TLS Tx offload.
If the socket option is successful, then the receive data-path of the TCP
socket is implemented by the kernel.
We choose to set this option at the earliest - just after CCS is complete.

Change-Id: I59741e04d89dddca7fb138e88fffcc1259b30132
Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7848)

bio: Linux TLS Rx Offload

Add support for Linux TLS Rx offload in the BIO layer.

Change-Id: I79924b25dd290a873d69f6c8d429e1f5bb2c3365
Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7848)

Linux ktls Rx infrastructure

Introduce the infrastructure for supproting receive side Linux Kernel TLS
data-path.

Change-Id: I71864d8f9d74a701cc8b0ad5536005f3c1716c1c
Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7848)

Rework DSO API conditions and configuration option

'no-dso' is meaningless, as it doesn't get any macro defined.
Therefore, we remove all checks of OPENSSL_NO_DSO.  However, there may
be some odd platforms with no DSO scheme.  For those, we generate the
internal macro DSO_NONE aand use it.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/#8622)

Fixed typo in enc.c warning

"warning: iv not use by this cipher" -> "warning: iv not used by this cipher"
CLA: trivial

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8608)

fixed public range check in ec_GF2m_simple_oct2point

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/8607)

Implement the param types that aren't explicitly lengthened (e.g. int) in terms
of those that are (e.g. int32_t).

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8616)

issue-8493: Fix for filenames with newlines using openssl dgst

The output format now matches coreutils *dgst tools.

[ edited to remove trailing white space ]

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8578)

Fix a memleak in apps/verify

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8465)

Add documents for SM2 cert verification

This follows #8321 which added the SM2 certificate verification feature.
This commit adds the related docs - the newly added 2 APIs and options
in apps/verify.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8465)

Add test cases for SM2 cert verification

This follows #8321 which added the SM2 certificate verification feature.
This commit adds some test cases for #8321.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8465)

trace: rename the default trace category from 'ANY' to 'ALL'

It seems more intuitive to set `OPENSSL_TRACE=all` instead of
`OPENSSL_TRACE=any` to obtain trace output for all categories.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8552)

trace: apps/openssl: print the correct category name

Previously, if the openssl application was run with OPENSSL_TRACE=any,
all trace output would just show 'ANY' as the category name, which was
not very useful. To get the correct category name printed in the trace
output, the openssl application now registers separate channels for
each category.

The trace API is unchanged, it is still possible for an application to
register a single channel for the 'ANY' category to see all outputt,
if it does not need this level of detail.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8552)

trace: fix out-of-bound memory access

When OSSL_trace_get_category_num() is called with an unknown category
name, it returns -1. This case needs to be considered in order to
avoid out-of-bound memory access to the `trace_channels` array.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8552)

trace: don't pretend success if it's not enabled

Partially reverts d33d76168fb7 Don't fail when tracing is disabled

Commit d33d76168fb7 fixed the problem that the initialization of
libcrypto failed when tracing was disabled, because the unoperational
ossl_trace_init() function returned a failure code. The problem was
fixed by changing its return value from failure to success.

As part of the fix the return values of other unimplemented trace API
functions (like OSSL_trace_set_channel(),OSSL_trace_set_callback())
was changed from failure to success, too. This change was not necessary
and is a bit problematic IMHO, because nobody expects an unimplemented
function to pretend it succeeded.

It's the application's duty to handle the case correctly when the trace
API is not enabled (i.e., OPENSSL_NO_TRACE is defined), not the API's job
to pretend success just to prevent the application from failing.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8552)

s390x assembly pack: import poly from cryptogams repo

>=20% faster than present code.

Signed-off-by: Patrick Steuer <patrick.steuer@de.ibm.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8560)

Remove heartbeats completely

Fixes #4856

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/1928)

Windows, VMS: build fixes

The naming of generated assembler wasn't done quite right.  There are
assembler files that are generated from a perl script, and there are
those who are not.  Only the former must be renamed to the platform
specific asm extension.

Furthermore, we need to make sure that 'OSSL_provider_init' isn't case
sensitive on VMS, to allow for the least surprise for provider
builders.

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8529)

Correctly initialise PACKET to zero in the tests to avoid possible problems
with padding bytes.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8611)