Revert the check for NaN in %f format

Unfortunately -Ofast seems to break that check.

Fixes #11994

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/12003)

undeprecate SSL_CTX_load_verify_locations and X509_STORE_load_locations

The underlying functions remain and these are widely used.
This undoes the deprecation part of PR8442

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/12001)

Make BIO_do_connect() and friends handle multiple IP addresses

Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11971)

add FFDH to speed command

the openssl speed command could not benchmark FFDH speed, but it could
benchmark ECDH, making comparisons between the two hard

this commit adds this feature

fixes #9475

Signed-off-by: Hubert Kario <hubert@kario.pl>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/10887)

bio printf: Avoid using rounding errors in range check

There is a problem casting ULONG_MAX to double which clang-10 is warning about.
ULONG_MAX typically cannot be exactly represented as a double.  ULONG_MAX + 1
can be and this fix uses the latter, however since ULONG_MAX cannot be
represented exactly as a double number we subtract 65535 from this number,
and the result has at most 48 leading one bits, and can therefore be
represented as a double integer without rounding error.  By adding
65536.0 to this number we achive the correct result, which should avoid the
warning.

The addresses a symptom of the underlying problem: we print doubles via an
unsigned long integer.  Doubles have a far greater range and should be printed
better.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/11955)

DOCS: add openssl-core_names.h(7)

A CAVEATS section is present in this manual.  That section name is
borrowed from OpenBSD, where mdoc(7) explains it like this:

    CAVEATS
    Common misuses and misunderstandings should be explained in this
    section.

Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/11963)

DOCS: add openssl-core_numbers.h(7)

Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/11963)

EVP_EncryptInit.pod: fix example

Signed-off-by: Patrick Steuer <patrick.steuer@de.ibm.com>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11976)

RSA: Do not set NULL OAEP labels

As of the previous commit, when a zero-length (string) parameter
is present in the parameters passed to a provider for a given operation,
we will produce an object corresponding to that zero-length parameter,
indicating to the underlying cryptographic operation that the parameter
was passed.  However, rsa_cms_decrypt() was relying on the previous
behavior, and unconditionally tried to call
EVP_PKEY_CTX_set0_rsa_oaep_label() even when the implicit default label
was used (and thus the relevant local variable was still NULL).
In the new setup that distinguishes present-but-empty and absent
more clearly, it is an error to attempt to set a NULL parameter,
even if it is zero-length.

Exercise more caution when setting parameters, and do not call
EVP_PKEY_CTX_set0_rsa_oaep_label() when there is not actually a
label provided.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11920)

params: do not ignore zero-length strings

Prior to this commit, if a string (or octet string) parameter
was present but indicated it was zero-length, we would return success
but with a NULL output value.  This can be problematic in cases where
there is a protocol-level distinction between parameter-absent and
parameter-present-but-zero-length, which is uncommon but can happen.

Since OPENSSL_malloc() returns NULL for zero-length allocation requests,
make a dummy allocation for this case, to give a signal that the string
parameter does exist but has zero length.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11920)

test HKDF with empty IKM

Add an extra EVP test that provides empty input key material.  It
currently fails, since we lose the information about "key present but
zero length" as we deserialize parameters in the provider.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11920)

Add a test for fetching EVP_PKEY style algs without a provider

Following on from the previous commit, add a test to check that we fail
to create an EVP_PKEY_CTX if an algorithm is not available in any provider,
*unless* it is an algorithm that has no provider support.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11826)

Fail if we fail to fetch the EVP_KEYMGMT

If we failed to fetch an EVP_KEYMGMT then we were falling back to legacy.
This is because some algorithms (such as MACs and KDFs used via an old
style EVP_PKEY) have not been transferred to providers.

Unfortunately this means that you cannot stop some algorithms from being
used by not loading the provider.

For example if you wanted to prevent RSA from being used, you might expect
to just not load any providers that make it available. Unfortunately that
doesn't work because we simply fall back to legacy if we fail to fetch
the EVP_KEYMGMT.

Instead we should fail *unless* the key type is one of those legacy key
types that we have not transferred.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11826)

Update gost-engine commit to match the API changes

[extended tests]

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11970)

util/mkpod2html.pl: Fix unbalanced quotes

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11969)

Fix errtest for older compilers

Some older compilers use "unknown function" if they dont support __func, so the
test using ERR_PUT_error needed to compensate for this when comparing against the
expected value.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11967)

ossl_shim: use the correct ticket key call back.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11966)

fips: add AES OFB mode ciphers to FIPS provider.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11908)

fips: add AES CFB mode ciphers to FIPS provider.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11908)

Avoid undefined behavior with unaligned accesses

Fixes: #4983

[extended tests]

Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/6074)

Rename EVP_PKEY_cmp() to EVP_PKEY_eq() and EVP_PKEY_cmp_parameters() to EVP_PKEY_parameters_eq()

Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11953)

crypto/cms: add CAdES-BES signed attributes validation

for signing certificate V2 and signing certificate extensions.

CAdES: lowercase name for now internal methods.

crypto/cms: generated file changes.

Add some CHANGES entries.

[extended tests]

Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/8098)

PROV: Use rsa_padding_add_PKCS1_OAEP_mgf1_with_libctx() in RSA-OAEP

Fixes #11904

Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11959)

doc: fix trace category names

The `ENGINE_CONF` and `PROVIDER_CONF` trace categories were merged
into a single `CONF` category (see bc362b9b7202 and 71849dff56d6).

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/11806)

Clean up some doc nits

Mostly "No items in =over/=back list"

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/11902)

Fix X509_PUBKEY_cmp(), move to crypto/x509/x_pubkey.c, rename, export, and document it

Fixes #11870

Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11894)

Constify X509_PUBKEY_get(), X509_PUBKEY_get0(), and X509_PUBKEY_get0_param()

Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11894)

PROV: Fix RSA-OAEP memory leak

The OAEP label wasn't freed when the operation context was freed.

Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/11927)

Update core_names.h fields and document most fields.

Renamed some values in core_names i.e Some DH specific names were changed to use DH instead of FFC.
Added some strings values related to RSA keys.
Moved set_params related docs out of EVP_PKEY_CTX_ctrl.pod into its own file.
Updated Keyexchange and signature code and docs.
Moved some common DSA/DH docs into a shared EVP_PKEY-FFC.pod.
Moved Ed25519.pod into EVP_SIGNATURE-ED25519.pod and reworked it.

Added some usage examples. As a result of the usage examples the following change was also made:
ec allows OSSL_PKEY_PARAM_USE_COFACTOR_ECDH as a settable gen parameter.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11610)

Fix ERR_print_errors so that it matches the documented format in doc/man3/ERR_error_string.pod

Fixes #11743

The ouput format had 2 issues that caused it not to match the expected documented format:
(1) At some point the thread id printing was changed to use the OPENSSL_hex2str method which puts ':' between hex bytes.
    An internal function that skips the seperator has been added.
(2) The error code no longer exists. So this was completely removed from the string. It is now replaced by ::

As an example:
  00:77:6E:52:14:7F:00:00:error:asn1 encoding routines:asn1_check_tlen:wrong tag:crypto/asn1/tasn_dec.c:1135:
Is now:
  00776E52147F0000:error::asn1 encoding routines:asn1_check_tlen:wrong tag:crypto/asn1/tasn_dec.c:1135:

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11789)

ossl_shim: add deprecation guards around the -use-ticket-callback option.

The ticket callback is deprecated in 3.0 and can't be used in a no-deprecated
build.

[extended tests]

Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/11944)

Coverity 1463830: Resource leaks (RESOURCE_LEAK)

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/11941)

Update the gost-engine submodule

Fixes #11949
[extended tests]

Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/11951)

Fix omissions in providers/common/der/build.info

Dependencies on generated files must be declared explicitly.  When
refactoring the DER code in providers/common/der, a few of those
dependency declaration were omitted, which may lead to build errors in
a parallel build.

Some cleanup and extensive used of build.info variables is done while
at it, to avoid unnecessary repetition.

Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11906)

Drop special case of time interval calculation for VMS

The existing special case code is broken and it is not needed
anymore as times() and _SC_CLK_TCK should be supported
on the supported VMS versions.

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/11905)

Revert "Guard use of struct tms with #ifdef __TMS"

The __TMS might be necessary on VMS however there is no such
define on glibc even though the times() function is fully
supported.

Fixes #11903

This reverts commit db71d315479762eefbf2bcda8be3b44b1867133f.

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/11905)

Fix auto-gen names in .gitignore

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11916)

s_client: Fix -proxy flag regression

s_client: connection via an HTTP proxy broke somewhere prior to openssl-3.0.0-alpha2.

openssl s_client -connect <target> -proxy <proxy_host:proxy_port>
Results in s_client making a TCP connection to proxy_host:proxy_port and then issuing an HTTP CONNECT to the proxy, instead of the target.

Fixes https://github.com/openssl/openssl/issues/11879

Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11880)

Fix coverity issues in EC after #11807

This should fix 2 issues detected by Coverity and introduced with
https://github.com/openssl/openssl/pull/11807

- CID 1463577:  Memory - corruptions  (ARRAY_VS_SINGLETON)
- CID 1463573:  Memory - corruptions  (ARRAY_VS_SINGLETON)

In practice the tests seem to show that they both aren't real issues,
yet I believe this small change should appease the scanner and at the
same time improve clarity for the reader.

Here is the original report:

```
** CID 1463577:  Memory - corruptions  (ARRAY_VS_SINGLETON)

________________________________________________________________________________________________________
*** CID 1463577:  Memory - corruptions  (ARRAY_VS_SINGLETON)
/crypto/ec/ec_lib.c: 1123 in EC_POINT_mul()
1117
1118         if (group->meth->mul != NULL)
1119             ret = group->meth->mul(group, r, g_scalar, point != NULL
1120                                    && p_scalar != NULL, &point, &p_scalar, ctx);
1121         else
1122             /* use default */
   CID 1463577:  Memory - corruptions  (ARRAY_VS_SINGLETON)
   Passing "&point" to function "ec_wNAF_mul" which uses it as an array. This might corrupt or misinterpret adjacent memory locations.
1123             ret = ec_wNAF_mul(group, r, g_scalar, point != NULL
1124                               && p_scalar != NULL, &point, &p_scalar, ctx);
1125
1126     #ifndef FIPS_MODULE
1127         BN_CTX_free(new_ctx);
1128     #endif

** CID 1463573:  Memory - corruptions  (ARRAY_VS_SINGLETON)

________________________________________________________________________________________________________
*** CID 1463573:  Memory - corruptions  (ARRAY_VS_SINGLETON)
/crypto/ec/ec_lib.c: 1123 in EC_POINT_mul()
1117
1118         if (group->meth->mul != NULL)
1119             ret = group->meth->mul(group, r, g_scalar, point != NULL
1120                                    && p_scalar != NULL, &point, &p_scalar, ctx);
1121         else
1122             /* use default */
   CID 1463573:  Memory - corruptions  (ARRAY_VS_SINGLETON)
   Passing "&p_scalar" to function "ec_wNAF_mul" which uses it as an array. This might corrupt or misinterpret adjacent memory locations.
1123             ret = ec_wNAF_mul(group, r, g_scalar, point != NULL
1124                               && p_scalar != NULL, &point, &p_scalar, ctx);
1125
1126     #ifndef FIPS_MODULE
1127         BN_CTX_free(new_ctx);
1128     #endif
```

Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/11919)

Move decl of OSSL_CRMF_CERTID_dup from {crmf,cmp}_local.h to include/openssl/crmf.h

fixes #11818

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11911)

Re-introduce legacy EVP_PKEY types for provided keys

EVP_PKEYs with provider side internal keys got the key type
EVP_PKEY_NONE.  This turned out to be too disruptive, so we try
instead to find a matching EVP_PKEY_ASN1_METHOD and use whatever
EVP_PKEY type it uses.

To make internal coding easier, we introduce a few internal macros to
distinguish what can be expected from a EVP_PKEY:

- evp_pkey_is_blank(), to detect an unassigned EVP_PKEY.
- evp_pkey_is_typed(), to detect that an EVP_PKEY has been assigned a
  type, which may be an old style type number or a EVP_KEYMGMT method.
- evp_pkey_is_assigned(), to detect that an EVP_PKEY has been assigned
  an key value.
- evp_pkey_is_legacy(), to detect that the internal EVP_PKEY key is a
  legacy one, i.e. will be handled via an EVP_PKEY_ASN1_METHOD and an
  EVP_PKEY_METHOD.
- evp_pkey_is_provided(), to detect that the internal EVP_PKEY key is
  a provider side one, i.e. will be handdled via an EVP_KEYMGMT and
  other provider methods.

This also introduces EVP_PKEY_KEYMGMT, to indicate that this EVP_PKEY
contains a provider side key for which there are no known
EVP_PKEY_ASN1_METHODs or EVP_PKEY_METHODs, i.e. these can only be
handled via EVP_KEYMGMT and other provider methods.

Fixes #11823

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11913)

GOST external tests

[extended tests]

Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11792)

Fix the parameter types of the CRYPTO_EX_dup function type.

This fixes a strict aliasing issue in ui_dup_method_data.

The parameter type of CRYPTO_EX_dup's from_d parameter
is in fact void **, since it points to a pointer.

This function is rarely used, therefore fix the param type
although that may be considered an API breaking change.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/2986)

Adjust length of some strncpy() calls

This fixes warnings detected by -Wstringop-truncation.

Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/11878)

Prevent use after free of global_engine_lock

If buggy application calls engine functions after cleanup of engines
already happened the global_engine_lock will be used although
already freed.

See for example:
https://bugzilla.redhat.com/show_bug.cgi?id=1831086

Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/11896)

Coverity 1463571: Null pointer dereferences (FORWARD_NULL)

Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
(Merged from https://github.com/openssl/openssl/pull/11892)

Coverity 1463574: Null pointer dereferences (REVERSE_INULL)

Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
(Merged from https://github.com/openssl/openssl/pull/11892)

Coverity 1463576: Error handling issues (CHECKED_RETURN)

Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
(Merged from https://github.com/openssl/openssl/pull/11892)

Coverity 1463258: Incorrect expression (EVALUATION_ORDER)

Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
(Merged from https://github.com/openssl/openssl/pull/11892)

There is no -signreq option in CA.pl

CLA: trivial

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11876)

STORE: Make try_decode_PrivateKey() ENGINE aware

This function only considered the built-in and application
EVP_PKEY_ASN1_METHODs, and is now amended with a loop that goes
through all loaded engines, using whatever table of methods they each
have.

Fixes #11861

Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11872)

rsa_padding_add_PKCS1_OAEP_mgf1_with_libctx(): fix check of |md|

In the FIPS module, the code as written generate an unconditional
error.

Fixes #11865

Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11869)

Refactor the provider side DER constants and writers

This splits up all the providers/common/der/*.c.in so the generated
portion is on its own and all related DER writing routines are in
their own files.  This also ensures that the DIGEST consstants aren't
reproduced in several files (resulting in symbol clashes).

Finally, the production of OID macros is moved to the generated header
files, allowing other similar macros, or DER constant arrays, to be
built on top of them.

Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11868)

deprecate EC_POINT_make_affine and EC_POINTs_make_affine

Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11874)

t1_trce: Fix remaining places where the 24 bit shift overflow happens

[extended tests]

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11857)

Avoid potential overflow to the sign bit when shifting left 24 places

Although there are platforms where int is 64 bit, 2GiB large BIGNUMs
instead of 4GiB should be "big enough for everybody".

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11857)

Cast the unsigned char to unsigned int before shifting left

This is needed to avoid automatic promotion to signed int.

Fixes #11853

[extended tests]

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11857)

Fix egd and devrandom source configs

./config --with-rand-seed=egd

need to defines OPENSSL_RAND_SEED_EGD and OPENSSL_NO_EGD
so get rid of OPENSSL_NO_EGD (compiles but I did not really test EGD)

./config --with-rand-seed=devrandom

does not work since wait_random_seeded works under the assumption
that OPENSSL_RAND_SEED_GETRANDOM is supposed to be enabled as well,
that is usually the case, but not when only devrandom is enabled.
Skip the wait code in this special case.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/11848)

Add OSSL_PROVIDER_do_all()

This allows applications to iterate over all loaded providers.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11858)

Update early data exchange scenarios in doc

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/11816)

Update limitation of psk_client_cb and psk_server_cb in usage with TLSv1.3

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/11816)

Test for the SSL_OP_IGNORE_UNEXPECTED_EOF option

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11735)

Introducing option SSL_OP_IGNORE_UNEXPECTED_EOF

Partially fixes #11209.

Before OpenSSL 3.0 in case when peer does not send close_notify,
the behaviour was to set SSL_ERROR_SYSCALL error with errno 0.
This behaviour has changed. The SSL_OP_IGNORE_UNEXPECTED_EOF restores
the old behaviour for compatibility's sake.

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11735)

Use {module,install}-mac, not -checksum

As the documentation points out, these fipsmodule.cnf fields are a MAC,
not a digest or checksum.  Rename them to be correct.

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11369)

Revise fips_install.pod

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11369)

Revise x509v3_config.pod

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11369)

Revise fips_config.pod

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11369)

Add missing pragma weak declaration to lhash.h

The missing symbol caused a linker failure on solaris x86_64.

Fixes #11796

Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11798)

deprecate EC precomputation functionality

Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/11851)

New Russian TLS 1.2 implementation

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11442)

GOST-related objects changes

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11442)

Constants for new GOST TLS 1.2 ciphersuites

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11442)

CORE: query for operations only once per provider (unless no_store is true)

When a desired algorithm wasn't available, we didn't register anywhere
that an attempt had been made, with the result that next time the same
attempt was made, the whole process would be done again.

To avoid this churn, we register a bit for each operation that has
been queried in the libcrypto provider object, and test it before
trying the same query and method construction loop again.

If course, if the provider has told us not to cache, we don't register
this bit.

Fixes #11814

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11842)

Use _get0_ functions instead of _get_.

Fix build error on some platforms

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/10757)

s_client: Show cert algorithms & validity period

Add certificate validity period (v) and public key & signature algorithms (a) to the "Certificate Chain" output.

Eg:
Certificate chain
 0 s:C = US, ST = California, L = Mountain View, O = Google LLC, CN = www.google.com
   i:C = US, O = Google Trust Services, CN = GTS CA 1O1
   a:PKEY: id-ecPublicKey, 256 (bit); sigalg: RSA-SHA256
   v:NotBefore: Dec  3 14:49:26 2019 GMT; NotAfter: Feb 25 14:49:26 2020 GMT
 1 s:C = US, O = Google Trust Services, CN = GTS CA 1O1
   i:OU = GlobalSign Root CA - R2, O = GlobalSign, CN = GlobalSign
   a:PKEY: rsaEncryption, 2048 (bit); sigalg: RSA-SHA256
   v:NotBefore: Jun 15 00:00:42 2017 GMT; NotAfter: Dec 15 00:00:42 2021 GMT

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/10757)

Fix some places where X509_up_ref is used
without error handling.

This takes up the ball from #11278
without trying to solve everything at once.

[extended tests]

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
(Merged from https://github.com/openssl/openssl/pull/11850)

TTY_get() in crypto/ui/ui_openssl.c open_console() can also return errno 1 (EPERM, Linux)

Signed-off-by: Maxim Zakharov <5158255+Maxime2@users.noreply.github.com>
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11767)

Delete the sslprovider test

This was added before the changes to the sslap/ssl_new/ssl_old tests which
run those tests with a non-default library context. It no longer adds
anything that those tests don't already do, so it can be deleted.

This also fixes a number of run-checker build failures which were failing
in this test if TLSv1.2 was disabled.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11832)

Fix small documentation issues

Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11835)

deprecate EC_POINTs_mul function

Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11807)

Update documentation following changes of various types

The previous commit changed the types of various objects passed between
the core and providers. Therefore the documentation needs to be updated
to reflect that.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11758)

Maintain strict type discipline between the core and providers

A provider could be linked against a different version of libcrypto than
the version of libcrypto that loaded the provider. Different versions of
libcrypto could define opaque types differently. It must never occur that
a type created in one libcrypto is used directly by the other libcrypto.
This will cause crashes.

We can "cheat" for "built-in" providers that are part of libcrypto itself,
because we know that the two libcrypto versions are the same - but not for
other providers.

To ensure this does not occur we use different types names for the handful
of opaque types that are passed between the core and providers.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11758)

CORE: Fix a couple of bugs in algorithm_do_this()

The call of ossl_provider_query_operation() used |data->operation_id|,
when |cur_operation| should be used.

If any ossl_provider_query_operation() call returned NULL, the loop
was stopped, when it should just continue on to the next operation.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11837)

Test TLSv1.3 out-of-band PSK with all 5 ciphersuites

Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/11785)

Fix crash in early data send with out-of-band PSK using AES CCM

Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/11785)

Fix all MD036 (emphasis used instead of heading)

The main fixes were errors in itemized lists "*)" instead of "*"

Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/11770)

Add "md-nits" make target

Also fix a nit in recent CHANGES.md update.

Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/11770)

Add 'methods' parameter to setup_engine() in apps.c for individual method defaults

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/4277)

Clean up macro definitions of openssl_fdset() in apps.h and sockets.h

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/4277)

Use OSSL_STORE for load_{,pub}key() and load_cert() in apps/lib/apps.c

This also adds the more flexible and general load_key_cert_crl()
as well as helper functions get_passwd(), cleanse(), and clear_free()
to be used also in apps/cmp.c etc.

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/11755)

Nit-fix: remove whitespace in doc/man3/EVP_PKEY_fromdata.pod causing warning

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/11755)

Guard use of struct tms with #ifdef __TMS like done earlier in apps/lib/apps.c

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/11755)

Build: make apps/progs.c depend on configdata.pm

Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/11824)

SSL: refactor ssl_cert_lookup_by_pkey() to work with provider side keys

Fixes #11720

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11828)

Ignore some auto-generated DER files

Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/11830)

Prepare for 3.0 alpha 3

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

Prepare for release of 3.0 alpha 2

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/11839)

If SOURCE_DATE_EPOCH is defined, use it for copyright year

Using the date from SOURCE_DATE_EPOCH instead of the current date makes
it possible to reproduce a build that was built on a different year:
https://reproducible-builds.org/specs/source-date-epoch/

This is fixing an issue we had while building Tor Browser:
https://trac.torproject.org/projects/tor/ticket/33535

CLA: trivial

Reviewed-by: Paul Dale <paul.dale@oracle.com>
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/11296)

Correct alignment calculation in ssl3_setup_write

The alignment calculation in ssl3_setup_write incorrectly results in an
alignment allowance of
(-SSL3_RT_HEADER_LENGTH) & (SSL3_ALIGN_PAYLOAD - 1) bytes. This equals 3
in almost all cases. The maximum alignment actually used in do_ssl3_write
is (SSL3_ALIGN_PAYLOAD - 1). This equals 7 bytes in almost all cases. So
there is a potential to overrun the buffer by up to 4 bytes.

Fortunately, the encryption overhead allowed for is 80 bytes which
consists of 16 bytes for the cipher block size and 64 bytes for the MAC
output. However the biggest MAC that we ever produce is HMAC-384 which is
48 bytes - so we have a headroom of 16 bytes (i.e. more than the 4 bytes
of potential overrun).

Thanks to Nagesh Hegde for reporting this.

Fixes #11766

Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/11768)