Remove handling of NULL sig param in ossl_ecdsa_deterministic_sign

The handling of sig=NULL was broken in this function, but since it
is only used internally and was never called with sig=NULL, it is
better to return an error in that case.

Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23529)

(cherry picked from commit 294782f3b5c4b81d682e6e8608bb6e851177494d)

Fix handling of NULL sig parameter in ECDSA_sign and similar

The problem is, that it almost works to pass sig=NULL to the
ECDSA_sign, ECDSA_sign_ex and DSA_sign, to compute the necessary
space for the resulting signature.
But since the ECDSA signature is non-deterministic
(except when ECDSA_sign_setup/ECDSA_sign_ex are used)
the resulting length may be different when the API is called again.
This can easily cause random memory corruption.
Several internal APIs had the same issue, but since they are
never called with sig=NULL, it is better to make them return an
error in that case, instead of making the code more complex.

Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23529)

(cherry picked from commit 1fa2bf9b1885d2e87524421fea5041d40149cffa)

Fix openssl req with -addext subjectAltName=dirName

The syntax check of the -addext fails because the
X509V3_CTX is used to lookup the referenced section,
but the wrong configuration file is used, where only
a default section with all passed in -addext lines is available.
Thus it was not possible to use the subjectAltName=dirName:section
as an -addext parameter.  Probably other extensions as well.

This change affects only the syntax check, the real extension
was already created with correct parameters.

Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23669)

(cherry picked from commit 387418893e45e588d1cbd4222549b5113437c9ab)

Bump actions/setup-python from 5.0.0 to 5.1.0

Bumps [actions/setup-python](https://github.com/actions/setup-python) from 5.0.0 to 5.1.0.
- [Release notes](https://github.com/actions/setup-python/releases)
- [Commits](https://github.com/actions/setup-python/compare/v5.0.0...v5.1.0)

---
updated-dependencies:
- dependency-name: actions/setup-python
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>
CLA: trivial

Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23977)

(cherry picked from commit de85587911dcd41dc3546b348acf9c9f15dd7c3d)

Explicitly state what -keys does

Signed-off-by: Simo Sorce <simo@redhat.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/23919)

(cherry picked from commit 693c479a2ca671e0dfca8d1ad14e789169b982ff)

Prepare for 3.3 beta 2

Reviewed-by: Neil Horman <nhorman@openssl.org>
Release: yes

Prepare for release of 3.3 beta 1

Reviewed-by: Neil Horman <nhorman@openssl.org>
Release: yes

make update

Reviewed-by: Neil Horman <nhorman@openssl.org>
Release: yes

Copyright year updates

Reviewed-by: Neil Horman <nhorman@openssl.org>
Release: yes

Update CHANGES.md and NEWS.md with changes that have occurred since 3.2

Release: yes

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23998)

Fix threadstest wrapping again

Stochastic failures in the RCU test on MACOSX are occuring.  Due to beta
release, disabling this test on MACOSX until post 3.3 release

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23967)

(cherry picked from commit 1967539e212c17139dc810096da987c8100b1ba2)

Update gost-engine submodule to fix the CI

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/23773)

(cherry picked from commit e7d5d61b66ee5a1d0827b3c92dc4d484fb9c91fe)

apps/req,crl: exit with 1 on verification failure

Fixes #23771

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/23773)

(cherry picked from commit 6af739b79ba50bd42ac8934747ab5c8b996f16b6)

Add NULL check before accessing PKCS7 encrypted algorithm

Printing content of an invalid test certificate causes application crash, because of NULL dereference:

user@user:~/openssl$ openssl pkcs12 -in test/recipes/80-test_pkcs12_data/bad2.p12 -passin pass: -info
MAC: sha256, Iteration 2048
MAC length: 32, salt length: 8
PKCS7 Encrypted data: Segmentation fault (core dumped)

Added test cases for pkcs12 bad certificates

Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23632)

(cherry picked from commit a4cbffcd8998180b98bb9f7ce6065ed37d079d8b)

Replace size_t with int and add the check for the EVP_MD_get_size()

Replace the type of "digest_size" with int to avoid implicit conversion when it is assigned by EVP_MD_get_size().
Moreover, add the check for the "digest_size".

Fixes: 29ce1066bc ("Update the demos/README file because it is really old. New demos should provide best practice for API use. Add demonstration for computing a SHA3-512 digest - digest/EVP_MD_demo")
Signed-off-by: Jiasheng Jiang <jiasheng@purdue.edu>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23924)

(cherry picked from commit 87e747000fef07c9ec43877bc5e9f2ca34f76a3b)

Replace unsigned with int

Replace the type of "digest_length" with int to avoid implicit conversion when it is assigned by EVP_MD_get_size().
Otherwise, it may pass the following check and cause the integer overflow error when EVP_MD_get_size() returns negative numbers.
Signed-off-by: Jiasheng Jiang <jiasheng@purdue.edu>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23922)

(cherry picked from commit f13ddaab69def0b453b75a8f2deb80e1f1634f42)

fix demos/sslecho/main.c so it builds on OpenBSD too

trying to build `demos/sslecho/main.c` shipped by current openssl
fails with error as follows:
```
cc -I../../include -g -Wall   -c -o main.o main.c
main.c:35:24: error: variable has incomplete type 'struct sockaddr_in'
    struct sockaddr_in addr;
                       ^
main.c:35:12: note: forward declaration of 'struct sockaddr_in'
    struct sockaddr_in addr;
           ^
main.c:46:32: error: use of undeclared identifier 'INADDR_ANY'
        addr.sin_addr.s_addr = INADDR_ANY;
                               ^
main.c:152:24: error: variable has incomplete type 'struct sockaddr_in'
    struct sockaddr_in addr;
                       ^
main.c:152:12: note: forward declaration of 'struct sockaddr_in'
    struct sockaddr_in addr;
           ^
3 errors generated.
gmake: *** [<builtin>: main.o] Error 1
```

including `netinet/in.h` fixes the build

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23936)

(cherry picked from commit 01eaf203856bfbb63051f8ecf56eae2d21132496)

Correct OSSL_sleep for NonStop PUT model by introducing sleep().

This fix also removes SPT model support as it was previously deprecated.
Upcoming threading models on the platform should be supportable without change
to this method.

Fixes: #23923
Fixes: #23927
Fixes: #23928
Signed-off-by: Randall S. Becker <randall.becker@nexbridge.ca>
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23926)

(cherry picked from commit 4a9e48f727ce7ad924c53a55b301e426d7e43863)

Don’t use the recvmmsg dgram method on Android <5

recvmmsg and sendmmsg were only added to Android’s C library in version 5, starting with API Level 21.

Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23754)

(cherry picked from commit 24109dca5a793d58c68a346db5b21746079ec317)

Fixed a typo and grammar in openssl-ts.pod

Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23913)

(cherry picked from commit f1c14f1853d2df94e339208eed1df823c2238389)

fix uploading artifacts for paramertrized jobs

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23933)

(cherry picked from commit 395ab201a7f99ebe2b1598890c9a43081867d226)

Make counters in rcu/rw threads torture test 64 bit

Its possible in some conditions for the rw/rcu torture tests to wrap the
counter, leading to false positive failures, make them 64 bits to avoid
this

Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23724)

(cherry picked from commit b50c174ee3b11f916285046d52574ba653745083)

Add M1 macOS runner to some workflows

Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23724)

(cherry picked from commit ada9d8c785cce8e75a88675622dd5ec79e9aa6d7)

riscv: Fix remaining asm checks

There are additional asm checks which don't check for OPENSSL_CPUID_OBJ
causing the build to still fail on riscv [1], so fix them in the same
manner as ff279597692f9f19dca5b147944d3d96f2e109f8

[1] https://bugs.gentoo.org/923956
Fixes: https://github.com/openssl/openssl/issues/22871
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23752)

Update FIPS hmac key documentation

The documentation is slightly incorrect about the FIPS hmac key.

CLA: trivial

Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23846)

(cherry picked from commit 53ef123f48d402aff7c27f8ec15191cb1cde4105)

Move artifact upload code into the shell script

Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23551)

(cherry picked from commit 9abcf116962e9a117717c751de93846f11da16cd)

Experimental support for uploading qlog artifacts

Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23551)

(cherry picked from commit f2db70962cacc2602bc614d51e0610085c99e999)

Prepare for 3.3 alpha 2

Reviewed-by: Neil Horman <nhorman@openssl.org>
Release: yes

Prepare for release of 3.3 alpha 1

Reviewed-by: Neil Horman <nhorman@openssl.org>
Release: yes

make update

Reviewed-by: Neil Horman <nhorman@openssl.org>
Release: yes

Copyright year updates

Reviewed-by: Neil Horman <nhorman@openssl.org>
Release: yes

Fix unbounded memory growth when using no-cached-fetch

When OpenSSL has been compiled with no-cached-fetch we do not cache
algorithms fetched from a provider. When we export an EVP_PKEY to a
provider we cache the details of that export in the operation cache for
that EVP_PKEY. Amoung the details we cache is the EVP_KEYMGMT that we used
for the export. When we come to reuse the key in the same provider that
we have previously exported the key to, we check the operation cache for
the cached key data. However because the EVP_KEYMGMT instance was not
cached then instance will be different every time and we were not
recognising that we had already exported the key to the provider.

This causes us to re-export the key to the same provider everytime the key
is used. Since this consumes memory we end up with unbounded memory growth.

The fix is to be more intelligent about recognising that we have already
exported key data to a given provider even if the EVP_KEYMGMT instance is
different.

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Paul Dale <ppzgs1@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/23841)

Try to fix intermittent CI failures in quic_multistream test

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23807)

plug potential memory leak in error code path

Function `module_add()` may leak stack of modules when
it fails to initialize newly added module.

Fixes #23835

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23836)

Raise the AFL_MAP_SIZE to accommodate future growth

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23851)

82-test_ocsp_cert_chain.t: Just ignore unrecognized lines in server output

There might be warnings from AFL fuzz checker
or other warnings that we do not care about.

For success it is just required that cert_status: ocsp response sent:
is present.

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23851)

Minor docfix for OSSL_HTTP_REQ_CTX_set_max_response_hdr_lines(3)

Also removal of duplicate assignment and addition of comment
in test/http_test.c

Follow up change to PR #23781

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23837)

Set AFL_MAP_SIZE to avoid crash in the AFL CI job

Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23843)

Add a test using the bandwidth limit filter

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23588)

Add support for bandwidth limitation in noisydgram BIO filter

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23588)

bio_f_noisy_dgram_filter(): Fix typo

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23588)

SSL_add_dir_cert_subjects_to_stack(): Documented return values

In the man page for SSL_add_dir_cert_subjects_to_stack(), the functions
returning int have undocumented return values.

Fixes #23171

Signed-off-by: Shakti Shah <shaktishah33@gmail.com>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23433)

Add Content Type OID for id-ct-rpkiSignedPrefixList

References: draft-ietf-sidrops-rpki-prefixlist
Title: "A profile for Signed Prefix Lists for Use in the Resource Public Key Infrastructure (RPKI)"

OID assigned under 'SMI Security for S/MIME CMS Content Type (1.2.840.113549.1.9.16.1)'
https://www.iana.org/assignments/smi-numbers/smi-numbers.xhtml#security-smime-1

Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23426)

apps/x509.c: No warning reading from stdin if redirected

Fixes #22893

Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23526)

Fix ASLR to be smaller during asan/tsan/ubsan runs

Recently asan/tsan/ubsan runs have been failing randomly.  It appears
that a recent runner update may have led to the Address Space Layout
Randomization setting in the linux kernel of ubuntu-latest runner
getting set to too high a value (it defaults to 30).  Such a setting
leads to the possibility that a given application will have memory
mapped to an address space that the sanitizer code typically uses to do
its job.  Lowering this value allows a/t/ubsan to work consistently
again

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23842)

Avoid a memcpy in dtls_get_reassembled_message()

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23828)

Fix dasync_rsa_decrypt to call EVP_PKEY_meth_get_decrypt

Signed-off-by: Vladimirs Ambrosovs <rodriguez.twister@gmail.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23825)

Implement KAT for KBKDF with KMAC128

Reviewed-by: Paul Dale <ppzgs1@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23745)

Update tls13ccstest.c, removal of deadcode

tst has been already checked for invalid value in the start of the function with switch statement.

Checked again here, so removed deadcode

CLA: trivial

Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23813)

Add fuzzing for DTLS

Update the fuzz corpora submodule with the DTLS fuzz corpus.

Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23585)

s_server: test ocsp with "-cert_chain"

Add a test to exercise the use of s_server with "-cert_chain" to
construct an ocsp request.

This new functionality was added in PR #22192.

Testing:

  make V=1 TESTS='test_ocsp_cert_chain' test

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23101)

Limit the number of http headers when receiving the http response

Change introduces a default limit on HTTP headers we expect to receive
from server to 256. If limit is exceeded http client library indicates
HTTP_R_RESPONSE_TOO_MANY_HDRLINES error. Application can use
OSSL_HTTP_REQ_CTX_set_max_response_hdr_lines() to change default.
Setting limit to 0 implies no limit (current behavior).

Fixes #22264

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23781)

Add check for xor_get_aid()

Add check for the return value of xor_get_aid() in order to avoid NULL pointer deference.

For example, "algor" could be NULL if the allocation of X509_ALGOR_new() fails. As a result, i2d_X509_ALGOR() will return 0 and "ctx->aid" will be an invalid value NULL.

Fixes: f4ed6eed2c ("SSL_set1_groups_list(): Fix memory corruption with 40 groups and more")
Signed-off-by: Jiasheng Jiang <jiasheng@purdue.edu>
Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23764)

Improve the documentation on -cert_chain and -status_verbose options

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/22192)

Use the untrusted certificate chain to create a valid certificate ID for OCSP_request

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/22192)

Doc: fix style

CLA: trivial

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23805)

PKCS7: Remove one of the duplicate checks

There are two consecutive identical checks "if (i <= 0)".
We can remove one of them to make the code clear.

CLA: trivial

Signed-off-by: Jiasheng Jiang <jiasheng@purdue.edu>
Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23741)

Add reformatting commit to .git-blame-ignore-revs

CLA: trivial

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23748)

Make the generated params_idx.c file deterministic if run multiple
times.

Fixes #23672

There are many name/value pairs currently that have duplicate names e.g.

    'CAPABILITY_TLS_GROUP_MAX_TLS' =>           "tls-max-tls",
    'CAPABILITY_TLS_SIGALG_MAX_TLS' =>          "tls-max-tls",

Stripping the .pm file down to just the above entries and running
multiple times gives different results for the produce_decoder.

On multiple runs any iterations over the unordered hash table keys using
foreach my $name (keys %params) results in a different order on multiple
runs. Because of this the mapping from the hash 'value' back to the
'key' will be different.

Note that the code also uses another mechanism in places that uses
"name1" => "value"
"name2" => "*name1"
Rather than fix all the strings the change done was to sort the keys. If
we were to chose to fix the strings then the perl code should be changed
to detect duplicates.

Reviewed-by: Paul Dale <ppzgs1@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
(Merged from https://github.com/openssl/openssl/pull/23688)

FAQ.md should be removed

the page the link refers to does not exist.
Anyone objects to delete file?

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23719)

Fix BIO_get_new_index() to return an error when it is exhausted.

Fixes #23655

BIO_get_new_index() returns a range of 129..255.

It is set to BIO_TYPE_START (128) initially and is incremented on each
call.
>= 256 is reserved for the class type flags (BIO_TYPE_DESCRIPTOR) so it
should error if it reaches the upper bound.

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23732)

Bring SSL_group_to_name docs in line with API definition

docs say the SSL object in this function is const, but the api doesn't
qualify it as such.  Adjust the docs to match the definition

Reviewed-by: Ben Kaduk <kaduk@mit.edu>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23785)

Add CHANGES

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23584)

QUIC MULTISTREAM TEST: Test write buffer statistics queries

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23584)

QUIC: Add stream write buffer queries

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23584)

Try to fix intermittent CI failures in sslapitest

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
(Merged from https://github.com/openssl/openssl/pull/23705)

QLOG: Fix indentation

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23598)

QUIC: Define error code for stateless reset

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23598)

QUIC: Add documentation for QUIC error codes

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23598)

QUIC: Uniform changes for QUIC error code definitions rename

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23598)

QUIC: Make QUIC transport error codes public

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23598)

Dont run the self-hosted workflows when not available

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23678)

Gate setting of ipi_spec_dst on not building for freebsd

some variants of FreeBSD (notably Dells OneFS) implement IP_PKTINFO
partially, and as such the build breaks for those variants.
specifically, it supports IP_PKTINFO, but the in_pktinfo struct has no
defined ipi_spec_dst field.  Work around this by gating the setting of
that variable on not building for FreeBSD

Fixes #23739

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23753)

interop tests: Fedora 39 config, simplify updates

Imitating Fedora 39 configuration in openssl.cnf with
SECLEVEL lowered to 0 in order to be able to run
TLS 1.3 tests with TLS_AES_128_CCM_8_SHA256.

In order to make updating smoother, check out specific tag rather
than the branch. This way, "old" tests can be fetched until PR
pointing to "new" tests is merged, so backwards-incompatible
changes can be done when needed.

Files specific for openssl upstream moved to separate
directory.

CLA: trivial

Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23747)

Fix a memory leak on successful load of CRL

Fixes #23693

Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23695)

QUIC QLOG: Fix ANSI

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23744)

QUIC QLOG: Fix use of sprintf

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23744)

Enable qlog support by default

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23744)

Document that unknown groups and sigalgs marked with ? are ignored

Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/23050)

Add test for ignoring unknown sigalgs and groups marked with ?

Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/23050)

Allow ignoring unknown sigalgs and groups in the configuration

Related to #20789

Signature algorithms and groups in the configuration that are
preceded with ? character and are unknown to libssl are just ignored.
The handling for them is similar to handling of ciphers.
I.e., there should be a failure only in case the configuration produces
no valid sigalgs or groups.

Also ignore duplicate sigalgs and groups as such confiuration errors
should not be fatal.

Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/23050)

apps/cmp: improve -reqin option to read fallback public key from first request message file given

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/21660)

crypto/cmp: add OSSL_CMP_MSG_get0_certreq_publickey(); fix coding style nit

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/21660)

apps/cmp: extend documentation and diagnostics for using -reqin in special situations

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/21660)

apps/cmp: add -reqout_only option for dumping/saving just the initial CMP request message

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/21660)

apps/cmp.c: refactor to fix some coding style nits and more convenient source-level debugging

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/21660)

cmperr.h: use free reason value 106 rather than 197 for CMP_R_UNEXPECTED_SENDER

Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/21660)

Dump out qlog json if it is malformed

We're still seeing periodic failures in qlog from malformed json output,
so lets try to catch it.

Modify the verify-qlog.py script to, in the event of an exception in
json.loads, to replay the entire json file to the console, followed by
an exception indicating what line it died trying to parse.

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23715)

Implement PCT for EDDSA

According to FIPS 140-3 IG 10.3.A Additonal Comment 1, a PCT shall be
performed consistent with the intended use of the keys.

This commit implements PCT for EDDSA via performing sign and verify
operations after key generated.

Also use the same pairwise test logic in EVP_PKEY_keygen and
EVP_PKEY_pairwise_check for EDDSA in FIPS_MODULE.

Add OSSL_SELF_TEST_DESC_PCT_EDDSA to OSSL_PROVIDER-FIPS page.

Reviewed-by: Paul Dale <ppzgs1@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23408)

demos/http3: Use `SSL_write_ex2()` together with `SSL_WRITE_FLAG_CONCLUDE`

These calls were introduced by PR #23343.

Change also does a minor tweak to Makefile so CFLAGS and LDFLAGS variables
from the environment are respected.

Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23602)

Fix off by one issue in buf2hexstr_sep()

Fixes #23363

Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23404)

Improve documentation of standard IANA cipher suite names.

Reviewed-by: Paul Yang <kaishen.yy@antfin.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23539)

Add appropriate lower bound checks for GeneralizedTime and UTCTime

ITU-T X.690 / ISO/IEC 8825-1 section 11.7 and section 11.8
impose specific constraints on how GeneralizedTime and UTCTime
can be encoded in BER/CER/DER. Following from these constraints
a minimum length can be derived.

Checking the length in this context can potentially help prevent
applications from interpreting an invalid GeneralizedTime as a
valid UTCTime.

Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23483)

apps/engine: add EC to list of capabilities

openssl engine -c wasn't showing if an engine implemented EC

cla: trivial

Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23538)

Ensure `$(MAKE)` commands and `CFLAGS` are appropriately quoted in the Makefile.

If a user's `make` command came from a path that contained a space then both the
`$(MAKE)` variable (and parts of the generated `CFLAGS`, when building for iOS)
would not be properly quoted and the build would fail.

Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23663)

Apply the AES-GCM unroll8 optimisation to Microsoft Azure Cobalt 100

Performance improvements range from 18% to 32%.

Change-Id: Ifb89eeac3c0625a582a25ff07cf7f9c9ec8f5ba6

Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23651)

chachap10-ppc.pl: Fix truncated relocation

Fix error: relocation truncated to fit: R_PPC64_REL14 (stub)
against symbol `ChaCha20_ctr32_vsx_8x'

CLA: trivial

Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23618)

SSL_set1_groups_list(): Fix memory corruption with 40 groups and more

Fixes #23624

The calculation of the size for gid_arr reallocation was wrong.
A multiplication by gid_arr array item size was missing.

Testcase is added.

Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23625)

s_cb.c: Add missing return value checks

Return value of function 'SSL_CTX_ctrl', that is called from
SSL_CTX_set1_verify_cert_store() and SSL_CTX_set1_chain_cert_store(),
is not checked, but it is usually checked for this function.

CLA: trivial

Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23647)

(cherry picked from commit 6f794b461c6e16c8afb996ee190e084cbbddb6b8)

Minor wording fixes related to no-atexit

Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/23642)

(cherry picked from commit 66e6f72c3e4221580a7f456ddeaa5027f0bbb8b7)