Fix potential double-free

The `sk` variable is assigned to `s->session->peer_chain`.
If `ssl3_digest_cached_records()` were to fail, then `sk` would still be
non-NULL, and subsequently freed on the error return. When the session
is freed, it will then attempt to free `s->session->peer_chain`,
resulting in a double-free (of `sk`).

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16309)

Multiple fixes for getting pub key from legacy DH PKEY

There were multiple issues with getting OSSL_PKEY_PARAM_PUB_KEY
from a legacy EVP_PKEY DH and DHX keys.

Fixes #16247

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16253)

Correct documentation errors in regards to UTF8 params

This fixes numerous bugs in documentation in regards to UTF8
params and their sizes. The returned size should always be without the
terminating NUL byte. On the other hand on the requestor side
the size of the buffer should include the NUL byte if it expects it
being included in the returned string.

Also make this clear in the EVP_PKEY_get_group_name() documentation
which uses utf8 string params under the hood.

Fixes #16287

Reviewed-by: Viktor Dukhovni <viktor@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16296)

Add documentation about the multilib postfix and libdir

Fixes #16244

Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/16281)

Allow small RSA exponents in the default provider

Fixes #16255

Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16285)

Disclaimer about the default provider activation added to config

Fixes #16249

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16280)

genpkey: -quiet doesn't take an argument

Fixes #16238

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16240)

Sort SSL_OP names in documentation

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16236)

Add missing SSL_OP flags

Add missing SSL_OP flags. Correct the list of flags set by SSL_OP_ALL.

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16236)

EVP_CIPHER_CTX_set_key_length: Raise error when key length is not settable

If key length is different from the existing key length and it is not
a settable parameter, raise an error.

Fixes #16277

Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/16279)

dsatest: Properly detect failure in generate/sign/verify

Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/16268)

Set FFC_PARAM_FLAG_VALIDATE_LEGACY on params generated with FIPS 186-2 gen

Fixes #16261

Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/16268)

aes_v8_xts_encrypt is present only on 64bit arm builds

Fixes #16273

Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16275)

Omitted signature_algorithms extension alerts updated

Fixes #15484

Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16217)

MacOS: Add an include of <CommonCrypto/CommonCryptoError.h>

The include is added before <CommonCrypto/CommonRandom.h>,
as required by older releases of the macOS developer tools.

Fixes #16248

CLA: trivial

Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16258)

doc: remove errant claim that these are not FIPS okay

Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16239)

doc: Fix ECX FIPS documentation

Both Ed448 and Ed25519 were omitted from the signature list.
X448 and X25519 were flagged as not FIPS valid which wasn't correct.

Fixes #16234

Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16239)

Avoid freeing the conf lhashes in X509_V3_EXT*_add_conf

Fixes #16226

Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16227)

Fix CMP app TLS connection not respecting vpm options like -crl_check

Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16225)

Fix test case for a2i_IPADDRESS

Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16201)

Test case for a2i_IPADDRESS

Unit test to show that a2i_IPADDRESS("1.2.3.4.test.example") ignores
trailing data.

See: https://github.com/openssl/openssl/issues/12649
See: https://bugs.python.org/issue41556
Signed-off-by: Christian Heimes <christian@python.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16201)

Fix ipv4_from_asc behavior on invalid Ip addresses

sscanf() call in ipv4_from_asc does not check that
the string is terminated immediately after the last digit.

Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16201)

X509_STORE_CTX_get_error: Fix some minor documentation issues

Original PR by Eric Valcik
(https://github.com/openssl/openssl/pull/12302)

Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16219)

Fix VS2019 compile error C4703: potentially uninitialized local pointer variable used.

encode_key2text.c(689): error C4703: potentially uninitialized local pointer variable 'modulus_label' used
encode_key2text.c(691): error C4703: potentially uninitialized local pointer variable 'exponent_label' used

CLA: trivial

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/12845)

cms: Fix handling of -rctform option

Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/16216)

cms: Do not try to check binary format on stdin

Fixes #16195

Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/16216)

req: Avoid segfault when -modulus is used

Fixes #16196

Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/16215)

[doc/man3] documentation: BN_cmp manpage updates

Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/16214)

test: add -macopt hexkey: to dgst command tests

Comparison checksum generated using 1.1.1f.

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/16212)

ctrls: add missing control string translation for key -> priv for HMAC

Fixes #16200

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/16212)

evp_test: add TLS 1.3 KDF test suite

Reviewed-by: Tomas Mraz <tomas@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/16203)

test: add test cases for TLS 1.3 KDF

Reviewed-by: Tomas Mraz <tomas@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/16203)

update doc/build.info

Reviewed-by: Tomas Mraz <tomas@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/16203)

fips: add power up test for TLS 1.3 KDF

The power up known answer test for the TLS 1.3 KDF does just the first step
to derive the "client_early_traffic_secret" using the two modes of the KDF.

Reviewed-by: Tomas Mraz <tomas@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/16203)

doc: add TLS 1.3 KDF to the FIPS provider list of algorithms.

Fix link to TLS1 PRF.

Reviewed-by: Tomas Mraz <tomas@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/16203)

doc: reorder the string and int extract/expand param values

Reviewed-by: Tomas Mraz <tomas@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/16203)

doc: add links to new KDF

Reviewed-by: Tomas Mraz <tomas@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/16203)

doc: add documentation for TLS13_KDF

Reviewed-by: Tomas Mraz <tomas@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/16203)

provider: add TLS13_KDF to the default and FIPS providers

Reviewed-by: Tomas Mraz <tomas@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/16203)

tls/prov: move the TLS 1.3 KDF code to providers

This function needs to be power up tested as part of the FIPS validation and
thus it needs to be inside the provider boundary.  This is realised by
introducing a new KDF "TLS13-KDF" which does the required massaging of
parameters but is otherwise functionally equivalent to HKDF.

Reviewed-by: Tomas Mraz <tomas@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/16203)

doc: add missing link directive in X942 KDF

Reviewed-by: Tomas Mraz <tomas@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/16203)

CI: remove spurious blank lines

Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16174)

ci: specific gcc explicitly on the basic-gcc CI build

GitHub Actions default to clang not gcc so this is necessary now.

Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16174)

ci: separate the config dump from the configuration command

This avoids using the shell's `&&` and shortens the lines a bit.

Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16174)

Document necessary error code processing

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16187)

Add oid_section to sysdefault.cnf to test adding new oids

Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16210)

Prevent recursive call of OPENSSL_INIT_LOAD_CONFIG

If objects are added in a config file the OPENSSL_INIT_LOAD_CONFIG
will be called recursively which results in hang in RUN_ONCE.

Fixes #16186

Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16210)

apps/pkeyutl.c: call ERR_print_errors() on all errors, including Signature Verification Failure

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16209)

Use copy.pl to install the fips module on Windows

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16208)

Windows, VMS: Do install_fips on install if fips is enabled

Also fix some inconsistencies and minor bugs related to the
install_fips target on Windows and VMS.

Fixes #16194

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16208)

doc: replace markdown backticks with perlpod syntax

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16190)

doc: use the documented =item markers

The generated lists[1] look weird when using a dash as the list item
character. Perlpod documents[2] '*' for unordered lists and '1.' (note
the period) for ordered lists. Use these characters instead.

[1] e.g. https://www.openssl.org/docs/manmaster/man7/migration_guide.html#New-Algorithms
[2] https://perldoc.perl.org/perlpod

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16190)

Add config_diagnostics to our configuration files.

The change to a more configuration based approach to enable FIPS mode
operation highlights a shortcoming in the default should do something
approach we've taken for bad configuration files.

Currently, a bad configuration file will be automatically loaded and
once the badness is detected, it will silently stop processing the
configuration and continue normal operations. This is good for remote
servers, allowing changes to be made without bricking things. It's bad
when a user thinks they've configured what they want but got something
wrong and it still appears to work.

Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/16171)

If we have passed the private key, don't copy it implicitly

Fixes #16197

Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16199)

Redefine getpid() -> _getpid() only for MSVC

This was introduced in 814b5133e for MSVC. C++Builder doesn't need it.

Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/16192)

changes: remove duplicate entry

Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/16180)

doc: document that config_diagnostics is sensible but involves risk

Also:

- add this option to the sample configurations in the documentation.
- note that it is a sensible choice when using FIPS via config

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16172)

demo: add GMAC demonstration program

Test vector from NIST's GCM mode suite.

Fixes #14123

Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/16113)

Fix dtls timeout dead code

Delete dtls timeout dead code in dtls1_handle_timeout

Fix: #15559

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/16151)

Prepare for 3.0 beta 3

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

Prepare for release of 3.0 beta 2

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

make update

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/16176)

Update fingerprints.txt

Add Paul Dale as an approved release signer.

Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16175)

Add missing session timeout calc

Fixes #16142

Add missing session timeout calculation in `ssl_get_new_session()`

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16145)

test: add a comment indication that a bad MAC is intentional

This permits negative testing of FIPS module load failure.

Also changed the MAC to all zeros to make it even clearer.

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16168)

Add some testing for the case where the FIPS provider fails to load

Ensure we get correct behaviour in the event that an attempt is made
to load the fips provider but it fails to load.

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16168)

Don't try and load the config file while already loading the config file

Calls to the API function EVP_default_properties_enable_fips() will
automatically attempt to load the default config file if it is not
already loaded. Therefore this function should not be called from inside
code to process the config file.

Fixes #16165

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16168)

Ensure any default_properties still apply even in the event of a provider load failure

We don't treat a failure to load a provider as a fatal error. If it is fatal then
we give up attempting to load the config file - including reading any default
properties. Additionally if an attempt has been made to load a provider then we
disable fallback loading.

Fixes #16166

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16168)

Don't leak the OSSL_LIB_CTX in the event of a failure to load the FIPS module

Ensure we free the OSSL_LIB_CTX on the error path.

Fixes #16163

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16168)

ci: disable async for the SH4 build and reenable the associated test

The platform doesn't seem to have support for this.

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16144)

ci: get rid of no-asm flag to m68k cross compiles

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16144)

ci: add the param conversion tests to the cross compiles.

There was a failure because an "inf" values was being read as a "NaN" not an
infinity.

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16144)

QEMU: include test runs for most cross compilation targets

For the cross compiles where the tests couldn't be run, most are capable
of being run when statically linked.  For these, a shared with FIPS build
but not test run is also included to maximise compilation coverage.
The builds take a couple of minutes so the impact of these extra jobs
isn't great.

The test failures for test_includes, test_store and test_x509_store
across several platforms are related the the OPENSSL_DIR_read() call.
This gets a "Value too large for defined data type" error calling the
standard library's readdir() wrapper.  That is, the failure is during
the translation from the x86-64 structure to the 32 bit structure.
I've tried tweaking the include defines to use larger fields but couldn't
figure out how to make it work.  The most prudent fix is to ignore these
tests for these platforms.

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16144)

test: handle not a number (NaN) values in the param conversion test.

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16144)

Add a test case for EVP_MD_meth_dup() and EVP_CIPHER_meth_dup()

Check that EVP_MD_meth_free() and EVP_CIPHER_meth_free() does actually
free the data.

Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16159)

Fix EVP_MD_meth_dup and EVP_CIPHER_meth_dup

Make sure the origin is set correctly when duping an EVP_MD or EVP_CIPHER.

Fixes #16157

Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16159)

Mark the EVP_PKEY_METHOD arg as const on some EVP_PKEY_meth_get_*() funcs

Most EVP_PKEY_meth_get_*() functions mark the EVP_PKEY_METHOD argument as
const. But 3 did not. We fix those to be consistent.

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16128)

KTLS: AES-CCM in TLS-1.3 is broken on 5.x kernels, disable it

Fixes #16089

Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16120)

Test ktls in non-default options CI build

Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16120)

Drop no-ktls from runchecker daily build as it has no effect

Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16120)

ECDSA_SIG_set0(): Clarify documentation and fix formatting errors

Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16129)

ECDSA_SIG_set0: r and s parameters cannot be NULL

Fixes #7731

Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16129)

DOCS: Move the description of EVP_PKEY_get0_description()

It appears to have been misplaced

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16063)

EVP: Add EVP_PKEY_get0_provider() and EVP_PKEY_CTX_get0_provider()

Fixes #16058

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16063)

Fix potential problems with EVP_PKEY_CTX_new() with engine set

If an engine is non-NULL in EVP_PKEY_CTX_new() call an assert might
have been incorrectly triggered or the engine might be finished
without being inited.

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16137)

do_sigver_init: Add missing ERR_clear_last_mark()

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16138)

ci: QEMU based cross compiled testing

With a little set up, Debian provides an ability to use QEMU to execute
programs compiled for other architectures. Using this, most of our cross
compilation CI builds can be executed.

This PR does this.

Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16133)

ci: reinstate the passwd tests for the no-cached-fetch run.

By selectively skipping the high round test cases, the out of memory problem
can be avoided.

partially fixes #16127

Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16132)

Makefile: Avoid changing LIBDIR based on whether it already exists

unix-Makefile.tmpl checks if the target LIBDIR exists on the build
machine or not and based on the result modify the final LIBDIR.
This should be avoided, build results should not depend on the build
machine root filesystem layout. It makes the build results unstable.

The fix simply removes the dir existence test from the unix-Makefile.tmpl.

Fixes: openssl#16121
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16122)

DSA/RSA_print(): Fix potential memory leak

Fixes #10777

Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16130)

Add a test for custom EVP_PKEY_METHODs

Adds a test for using custom EVP_PKEY_METHODs without an ENGINE. As part
of this we also test having a custom EVP_PKEY_METHOD that wraps a built-in
EVP_PKEY_METHOD. We do this for both legacy and provided keys.

Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16118)

Fix custom EVP_PKEY_METHOD implementations where no engine is present

It is possible to have a custom EVP_PKEY_METHOD implementation without
having an engine. In those cases we were failing to use that custom
implementation.

Fixes #16088

Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16118)

Update our EVP_PKEY_METHODs to get low level keys via public APIs

It is possible to call built-in EVP_PKEY_METHOD functions with a provided
key. For example this might occur if a custom EVP_PKEY_METHOD is in use
that wraps a built-in EVP_PKEY_METHOD. Therefore our EVP_PKEY_METHOD
functions should not assume that we are using a legacy key. Instead we
get the low level key using EVP_PKEY_get0_RSA() or other similar functions.
This "does the right thing" if the key is actually provided.

Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16118)

OSSL_HTTP_open(): Fix memory leak on TLS connect failure via proxy

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16119)

ci: omit tests that consume too much memory

The SSL API tests and the passwd command test trigger memory leakage in the
address sanitizer.

Fixes #16116

Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16125)

tls_process_{client,server}_certificate(): allow verify_callback return > 1

Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/13937)

SSL_CTX_set_cert_verify_callback.pod: various corrections and clarifications

- Make clear the callback is called whenever a peer certificate has been received,
  which is independent of the verification mode.
- Make clear that a return value > 1 always leads to handshake failure.
- Make clear that in server mode also return values <= 0 lead to handshake failure.
- For client mode replace the incorrect formulation "if B<SSL_VERIFY_PEER> is set"
  by what is actually implemented: "if the verification mode is not B<SSL_VERIFY_NONE>".
- Refer to X509_STORE_CTX_set_error() rather than to internal error variable.

Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/13937)

Add testcases for SSL_key_update() corner case calls

Test that SSL_key_update() is not allowed if there are writes pending.
Test that there is no reset of the packet pointer in ssl3_setup_read_buffer().

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16085)

test: include all DRBG tests in FIPS mode

Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/16096)