Make basic AES ciphers available from within the FIPS providers

These ciphers were already provider aware, and were available from the
default provider. We move them into the FIPS provider too.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/9038)

Write a test for receiving a KeyUpdate (update requested) while writing

Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/8773)

Defer sending a KeyUpdate until after pending writes are complete

If we receive a KeyUpdate message (update requested) from the peer while
we are in the middle of a write, we should defer sending the responding
KeyUpdate message until after the current write is complete. We do this
by waiting to send the KeyUpdate until the next time we write and there is
no pending write data.

This does imply a subtle change in behaviour. Firstly the responding
KeyUpdate message won't be sent straight away as it is now. Secondly if
the peer sends multiple KeyUpdates without us doing any writing then we
will only send one response, as opposed to previously where we sent a
response for each KeyUpdate received.

Fixes #8677

Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/8773)

Remove file reference from property documentation.

Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/9066)

Add the content type attribute to additional CMS signerinfo.

Fixes #8923

Found using the openssl cms -resign option.
This uses an alternate path to do the signing which was not adding the required signed attribute
content type. The content type attribute should always exist since it is required is there are
any signed attributes.
As the signing time attribute is always added in code, the content type attribute is also required.
The CMS_si_check_attributes() method adds validity checks for signed and unsigned attributes
e.g. The message digest attribute is a signed attribute that must exist if any signed attributes
exist, it cannot be an unsigned attribute and there must only be one instance containing a single
value.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8944)

Fix a typo in the property grammar that creates an ambiguous parse.

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/9065)

Windows: Add missing quotes in build file

All invokations of $(PERL) need to be quoted, in case it contains
spaces.  That was forgotten in one spot.

Fixes #9060

Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/9062)

Add sparse array of const pointer type

Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/8840)

Fix various mistakes in ec_GFp_nistp_recode_scalar_bits comment.

Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/9050)

Fix CHECK_BSD_STYLE_MACROS for OpenBSD and cryptodev-linux

CLA: trivial

Fixes #8911

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/8866)

Update X509_STORE_new.pod

Minor typo.

CLA: trivial

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/8866)

Constify OSSL_PROVIDER getter input parameters

Some OSSL_PROVIDER getters took a non-const OSSL_PROVIDER parameter.
There's no reason to do so.

Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/9054)

Update EVP_VerifyInit.pod

Method name correction.

CLA: trivial

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/9052)

issue-8998: Ensure that the alert is generated and reaches the remote

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/9041)

Certificate Management Protocol (CMP, RFC 4210) extension to OpenSSL
Also includes CRMF (RFC 4211) and HTTP transfer (RFC 6712)

CMP and CRMF API is added to libcrypto, and the "cmp" app to the openssl CLI.
    Adds extensive man pages and tests.  Integration into build scripts.

Incremental pull request based on OpenSSL commit 1362190b1b78 of 2018-09-26

3rd chunk: CMP ASN.1 structures (in crypto/cmp/cmp_asn.c) and related files

Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8669)

Add man 7 page about properties.

Add a page about properties in the man7 section of the public documentation.

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/9011)

Join the x509 and x509v3 directories

This has been long overdue.

Note that this does not join the X509 and X509V3 error modules, that
will be too many macro changes at this stage.

Fixes #8919

Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8925)

Fuzz: add a few more types into Fuzzing for ESS

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8117)

CAdES: Fix SignerInfo attribute construction order.

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8117)

crypto/conf: openssl_config_int() returns unitialized value

openssl_config_int() returns the uninitialized variable `ret`
when compiled with OPENSSL_SYS_UEFI.

Fixes #9026

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/9029)

Typo BIO_SOCK_REUSADDR => BIO_SOCK_REUSEADDR

CLA: trivial

Reviewed-by: Paul Yang <yang.yang@baishancloud.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/9021)

Test of uniformity of BN_rand_range output.

Rework the test so that it fails far less often.

A number of independent tests are executed and 5% are expected to fail.
The number of such failures follows a binomial distribution which permits
a statistical test a 0.01% expected failure rate.

There is a command line option to enable the stochastic range checking.
It is off by default.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8830)

fix timeout in 30-test_evp_pkey_dparam

enabling the 'enable-crypto-mdebug' option and running parameter generation
causes timeouts.
Loading pregenerated params is more suited for these tests.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/9022)

Configurations/unix-Makefile.tmpl: not -v for rm

Not all Unixen know the -v option

Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/9012)

Configure: link AIX modules correctly

Use -bnoentry, not -bexpall

Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/9012)

The SHA256 is not a mandatory digest for DSA.

The #7408 implemented mandatory digest checking in TLS.
However this broke compatibility of DSS support with GnuTLS
which supports only SHA1 with DSS.

There is no reason why SHA256 would be a mandatory digest
for DSA as other digests in SHA family can be used as well.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/9015)

Configure: Remove extra warning and sanitizer options from CXXFLAGS

We add the extra warning and sanitizer options to check our code,
which is entirely in C.  We support C++ compilers uniquely for the
sake of certain external test suites, and those projects can probably
sanitize their own code themselves.

[extended tests]

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/9013)

Add d2i_KeyParams/i2d_KeyParams API's.

Convert EVP_PKEY Parameters to/from binary.
This wraps the low level i2d/d2i calls for DH,DSA and EC key parameters
in a similar way to Public and Private Keys.
The API's can be used by applications (including openssl apps) that only
want to use EVP_PKEY without needing to access low level key API's.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8903)

Changed ssl layer to use EVP_KDF API for TLS1_PRF and HKDF.

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/9010)

doc/man3/X509_LOOKUP_meth_new.pod: clarify the requirements

The documentation of what a X509_LOOKUP implementation must do was
unclear and confusing.  Most of all, clarification was needed that it
must store away the found objects in the X509_STORE.

Fixes #8707

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8755)

Added Conforming To section to EVP_KDF_TLS1_PRF documentation.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/9008)

OCSP: fix memory leak in OCSP_url_svcloc_new method.
Add a few coverage test case.

Fixes #8949

[extended tests]

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8959)

Fix input checks wrt legacy code

In all legacy code ctx->cipher is dereferenced without checks, so it
makes no sense to jump there is ctx->cipher is NULL as it will just lead
to a crash. Catch it separately and return an error.

This is simlar to the fix in d2c2e49eab69c7446c1c2c7227f63f8618ca99a5

Signed-off-by: Simo Sorce <simo@redhat.com>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/9002)

crypto/evp/evp_key.c: #define BUFSIZ if <stdio.h> doesn't #define it

CLA: trivial

Fixes #8904

Commit 48feaceb53fa ("Remove the possibility to disable the UI module
entirely", 2017-07-03) made the BUFSIZ references in "evp_key.c"
unconditional, by deleting the preprocessing directive "#ifndef
OPENSSL_NO_UI". This breaks the build when compiling OpenSSL for edk2
(OPENSSL_SYS_UEFI), because edk2's <stdio.h> doesn't #define BUFSIZ.

Provide a fallback definition, like we do in "crypto/ui/ui_util.c" (from
commit 984d6c605216, "Fix no-stdio build", 2015-09-29).

Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8922)

Use fixed length for formatting standard cipher names

Example with patch:
```
$ openssl ciphers -stdname 'TLS_AES_256_GCM_SHA384:ECDHE-ECDSA-CHACHA20-POLY1305'
TLS_AES_256_GCM_SHA384                        - TLS_AES_256_GCM_SHA384         TLSv1.3 Kx=any      Au=any   Enc=AESGCM(256) Mac=AEAD
TLS_CHACHA20_POLY1305_SHA256                  - TLS_CHACHA20_POLY1305_SHA256   TLSv1.3 Kx=any      Au=any   Enc=CHACHA20/POLY1305(256) Mac=AEAD
TLS_AES_128_GCM_SHA256                        - TLS_AES_128_GCM_SHA256         TLSv1.3 Kx=any      Au=any   Enc=AESGCM(128) Mac=AEAD
TLS_ECDHE_ECDSA_WITH_CHACHA20_POLY1305_SHA256 - ECDHE-ECDSA-CHACHA20-POLY1305  TLSv1.2 Kx=ECDH     Au=ECDSA Enc=CHACHA20/POLY1305(256) Mac=AEAD
```

Example without patch:
```
$ openssl ciphers -stdname 'TLS_AES_256_GCM_SHA384:ECDHE-ECDSA-CHACHA20-POLY1305'
TLS_AES_256_GCM_SHA384 - TLS_AES_256_GCM_SHA384  TLSv1.3 Kx=any      Au=any  Enc=AESGCM(256) Mac=AEAD
TLS_CHACHA20_POLY1305_SHA256 - TLS_CHACHA20_POLY1305_SHA256 TLSv1.3 Kx=any      Au=any  Enc=CHACHA20/POLY1305(256) Mac=AEAD
TLS_AES_128_GCM_SHA256 - TLS_AES_128_GCM_SHA256  TLSv1.3 Kx=any      Au=any  Enc=AESGCM(128) Mac=AEAD
TLS_ECDHE_ECDSA_WITH_CHACHA20_POLY1305_SHA256 - ECDHE-ECDSA-CHACHA20-POLY1305 TLSv1.2 Kx=ECDH     Au=ECDSA Enc=CHACHA20/POLY1305(256) Mac=AEAD
```

CLA: Trivial

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8999)

Update format string for ciphers to account for newer ciphers

* Cipher name: from 23 to 30 (example: ECDHE-ECDSA-AES128-GCM-SHA256)
* Fixed length for TLS version (examples: TLSv1, TLSv1.3)
* Au length from 4 to 5 (example: ECDSA)

Example (without patch):
```
$ openssl ciphers -v 'ECDHE-ECDSA-AES128-GCM-SHA256:ECDHE-ECDSA-AES128-SHA'
TLS_AES_256_GCM_SHA384  TLSv1.3 Kx=any      Au=any  Enc=AESGCM(256) Mac=AEAD
TLS_CHACHA20_POLY1305_SHA256 TLSv1.3 Kx=any      Au=any  Enc=CHACHA20/POLY1305(256) Mac=AEAD
TLS_AES_128_GCM_SHA256  TLSv1.3 Kx=any      Au=any  Enc=AESGCM(128) Mac=AEAD
ECDHE-ECDSA-AES128-GCM-SHA256 TLSv1.2 Kx=ECDH     Au=ECDSA Enc=AESGCM(128) Mac=AEAD
ECDHE-ECDSA-AES128-SHA  TLSv1 Kx=ECDH     Au=ECDSA Enc=AES(128)  Mac=SHA1
```

Example (with patch):
```
$ openssl ciphers -v 'ECDHE-ECDSA-AES128-GCM-SHA256:ECDHE-ECDSA-AES128-SHA'
TLS_AES_256_GCM_SHA384         TLSv1.3 Kx=any      Au=any   Enc=AESGCM(256) Mac=AEAD
TLS_CHACHA20_POLY1305_SHA256   TLSv1.3 Kx=any      Au=any   Enc=CHACHA20/POLY1305(256) Mac=AEAD
TLS_AES_128_GCM_SHA256         TLSv1.3 Kx=any      Au=any   Enc=AESGCM(128) Mac=AEAD
ECDHE-ECDSA-AES128-GCM-SHA256  TLSv1.2 Kx=ECDH     Au=ECDSA Enc=AESGCM(128) Mac=AEAD
ECDHE-ECDSA-AES128-SHA         TLSv1   Kx=ECDH     Au=ECDSA Enc=AES(128)  Mac=SHA1
```

CLA: trivial

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8999)

issue-8973: Added const to parameters for values that were not altered

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8995)

Doc update: minor typo in CMS_verify.pod
CLA: trivial

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8952)

Cleaned up tls1_prf and added comments.

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8987)

Added algorithm description comments to HKDF.

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8994)

Fix a crash in the speed command with wrap ciphers

e.g. openssl speed -evp id-aes256-wrap-pad
was crashing because the return code from EVP_CipherInit_ex
was ignored.
Not going to allow that cipher mode because wrap ciphers
produces more bytes output than the input length
and EVP_Update_loop is not really prepared for that.

Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8739)

Add more commentary about recursive Provider intialisation in the FIPS module

In addition this commit ensures that the "provctx" value is defaulted to the current
library context when we are recurively initialising the FIPS provider when already inside
the FIPS module.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8728)

Move where include path for providers/common/include gets specified

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8728)

Make some EVP code available from within the FIPS module

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8728)

Make core code available within the FIPS module

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8728)

Add the NIST CAVS test vectors for CCM

This imports all of the NIST CAVS test vectors for CCM (SP800-38C) and
coverts them for use within evp_test. This commit also adds a script to
convert the .rsp CAVS files into the evp_test format.

Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8981)

Avoid trailing space in "openssl version -o".

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8963)

Always show application and library versions in "openssl version".

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8963)

Fix GOST OID

CLA: trivial

Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/8979)

EVP_Digest*: enable SHA3 pre-hashing for DSA

Fixes #7323

Signed-off-by: Patrick Steuer <patrick.steuer@de.ibm.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8970)

EVP_Digest*: enable SHA3 pre-hashing for ECDSA

Fixes #8957

Signed-off-by: Patrick Steuer <patrick.steuer@de.ibm.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8970)

s390x assembly pack: allow specifying the tag after aad in aes-ccm

67c81ec311 forgot about s390x

Signed-off-by: Patrick Steuer <patrick.steuer@de.ibm.com>
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8971)

Optional property query support.

Add the possibility of a property query clause to be optional by preceding
it with a question mark.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8943)

Change default RSA, DSA and DH size to 2048 bit

Fixes: #8737

Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
Reviewed-by: Richard Levitte <levitte@openssl.org>
GH: #8741

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

This reverts commit dcb982d792d6064ed3493e79749208d8c257ff04.

This change is causing extended tests to fail.

[extended tests]

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8964)

Configure: for '-z defs', also check $config{cflags}

When sanitize options are added as 'enable-msan' or similar, the
-fsanitize C flags is set in $config{cflags} rather than
$config{CFLAGS}, so we need to check both.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8965)

Change SSL parameter SSL_session_reused const

This function only returns a status and does not modify the parameter.
Since similar function are already taking const parameters, also
change this function to have a const parameter.

Fixes #8934

CLA: trivial
Signed-off-by: Arne Schwabe <arne@rfc2549.org>
Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8945)

Params conversion tests.

Add ranged checked OSSL_PARAM conversions between the native types.  A
conversion is legal only if the given value can be exactly represented
by the target type.

Includes a test case that reads a stanza test case file and verified that param
conversions are processed properly.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8733)

Clear CRMF vs CMP confusion

In the development of the CRMF sub-system, there seems to have been
some confusion as to what configuration option should be used.
'no-crmf' was added, but the C macro guards were using OPENSSL_NO_CMP
rather than OPENSSL_NO_CRMF...

In fact, we want 'no-cmp', but since the CRMF code is part of CMP, we
need 'no-crmf' to depend on 'no-cmp'.  We do this by making 'crmf' a
silent "option" that get affected by 'cmp' by way of %disable_cascades.
This allows options to be "aliases" for a set of other ones, silent or
not.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8897)

Revert "ppc assembly pack: always increment CTR IV as quadword"

The 32 bit counter behaviour is necessary and was intentional.

This reverts commit e9f148c9356b18995298f37bafbf1836a3fce078.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8958)

Configure: let platform->dsoext() default with platform->shlibextsimple()

We still use '.so' as a last resort...

Fixes #8950

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8951)

ppc assembly pack: always increment CTR IV as quadword

The kernel self-tests picked up an issue with CTR mode. The issue was
detected with a test vector with an IV of
FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFD: after 3 increments it should wrap
around to 0.

There are two paths that increment IVs: the bulk (8 at a time) path,
and the individual path which is used when there are fewer than 8 AES
blocks to process.

In the bulk path, the IV is incremented with vadduqm: "Vector Add
Unsigned Quadword Modulo", which does 128-bit addition.

In the individual path, however, the IV is incremented with vadduwm:
"Vector Add Unsigned Word Modulo", which instead does 4 32-bit
additions. Thus the IV would instead become
FFFFFFFFFFFFFFFFFFFFFFFF00000000, throwing off the result.

Use vadduqm.

This was probably a typo originally, what with q and w being
adjacent.

CLA: trivial

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8942)

Fix typo in NOTES.PERL

CLA: trivial

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8905)

Added X963KDF API

X963 KDF is used for CMS ec keyagree Recipient Info.
The X963 KDF that is used by CMS EC Key Agreement has been moved
into a EVP_KDF object. This KDF is almost identical to the the SSKDF
hash variant, so it has been implemented inside the SSKDF code with
its own method table.

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8902)

EVP_FETCH: remove the need to transport the legacy NID through construction

Now that the legacy NID isn't used as a main index for fetched
algorithms, the legacy NID was just transported around unnecessarily.
This is removed, and the legacy NID is simply set by EVP_{API}_fetch()
after the construction process is done.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8878)

util/find-doc-nits: Fine tune detection of POD markup in NAME section

POD markup is only forbidden in the actual names, while permitted in
the description.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8878)

EVP_FETCH: deal with names without pre-defined NIDs

We didn't deal very well with names that didn't have pre-defined NIDs,
as the NID zero travelled through the full process and resulted in an
inaccessible method.  By consequence, we need to refactor the method
construction callbacks to rely more on algorithm names.

We must, however, still store the legacy NID with the method, for the
sake of other code that depend on it (for example, CMS).

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8878)

Make the generic EVP fetching mechanism use the namenum map

This avoids using the ASN1_OBJECT database, which is bloated for the
purpose of a simple number <-> name database.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8878)

Create internal number<->name mapping API

This can be used as a general name to identity map.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8878)

Ignore entropy from RAND_add()/RAND_seed() in FIPS mode [fixup]

Small correction to RAND_DRBG(7) (amends 3a50a8a91ad1)

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8909)

Ignore entropy from RAND_add()/RAND_seed() in FIPS mode

The functions RAND_add() and RAND_seed() provide a legacy API which
enables the application to seed the CSPRNG.

But NIST SP-800-90A clearly mandates that entropy *shall not* be provided
by the consuming application, neither for instantiation, nor for reseeding.

The provided random data will be mixed into the DRBG state as additional
data only, and no entropy will accounted for it.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8722)

Test skip option.

Provide C test cases with the option to skip tests and subtests.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8695)

Skip the no derivation functions when in FIPS mode because they are not
applicable.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8695)

Fixed 32bit issue for kdf_opts

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8901)

Don't use '-z defs' if sanitizers are used

There are quite a number of sanitizers for clang that aren't
documented in the clang user documentation.  This makes it impossible
to be selective about what sanitizers to look at to determine if
'-z defs' should be used of not.

Under these circumstances, the sane thing to do is to just look for
any sanitizer specification and not use '-z defs' if there's one
present.

Fixes #8735

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8892)

crypto/ppccap.c: Fix which hwcap value used to check for HWCAP_ARCH_3_00

CLA: trivial

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8889)

Squashed commit of the following:

Add new option '-http_server_binmode' which allows the server to open and send
binary files as well as text.

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8811)

ossl_bsearch(): New generic internal binary search utility function

OBJ_bsearch_ and OBJ_bsearch_ex_ are generic functions that don't
really belong with the OBJ API, but should rather be generic utility
functions.  The ending underscore indicates that they are considered
internal, even though they are declared publicly.

Since crypto/stack/stack.c uses OBJ_bsearch_ex_, the stack API ends up
depending on the OBJ API, which is unnecessary, and carries along
other dependencies.

Therefor, a generic internal function is created, ossl_bsearch().
This removes the unecessary dependencies.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8899)

Allow specifying the tag after AAD in CCM mode

This change allows to pass the authentication tag after specifying
the AAD in CCM mode. This is already true for the other two supported
AEAD modes (GCM and OCB) and it seems appropriate to match the
behavior.

GCM and OCB also support to set the tag at any point before the call
to `EVP_*Final`, but this won't work for CCM due to a restriction
imposed by section 2.6 of RFC3610: The tag must be set before
actually decrypting data.

This commit also adds a test case for setting the tag after supplying
plaintext length and AAD.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/7243)

EVP_EncryptUpdate, EVP_EncryptFinal_ex: don't branch on uninitialized memory

If ctx->cipher->cupdate/ctx->cipher->cfinal failed, 'soutl' is left
uninitialized.

This patch incorporates the same logic as present in EVP_DecryptUpdate and
EVP_DecryptFinal_ex: only branch on 'soutl' if the preceding call succeeded.

Bug found by OSS-Fuzz.

Signed-off-by: Guido Vranken <guidovranken@gmail.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8874)

Coverity CID 1444950: Control flow issues

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8888)

Coverity CID 1444951: Null pointer dereferences

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8888)

Coverity CID 1444952: Null pointer dereferences

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8888)

Coverity CID 1444953: Null pointer dereferences

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8888)

Coverity CID 1444954: Integer handling issues

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8888)

Coverity CID 1444955: Null pointer dereferences

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8888)

Coverity CID 1444956: Integer handling issues

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8888)

Coverity CID 1444957: Error handling issues

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8888)

Coverity CID 1444958: Null pointer dereferences

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8888)

Coverity CID 1444959: Integer handling issues

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8888)

Coverity CID 1444960: Error handling issues

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8888)

Coverity CID 1444961: Integer handling issues

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8888)

Coverity CID 1444963: Null pointer dereferences

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8888)

Reject obviously invalid DSA parameters during signing

Fixes #8875

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/8876)

ssl: Add SSL_sendfile

This commit adds the SSL_sendfile call, which allows KTLS sockets to
transmit file using zero-copy semantics.

Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8727)

Linux ktls sendfile

This commit introduces support for Linux KTLS sendfile.
Sendfile semantics require the use of a kernel TLS socket to construct the TLS
record headers, encrypt and authenticate the data.
KTLS sendfile improves performance by avoiding the copy of file data into user
space, which is required today.

Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8727)

man: fix a typo and a grammar nit in EVP_PKEY_meth_new(3)

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8893)

Change the digest string from "md" to "digest"

Conform to other modules which were changed at the last minute and this
discrepancy was not noticed.
Retain "md" as an alias so not to break 3rd party backports/tests scripts.

Signed-off-by: Simo Sorce <simo@redhat.com>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/8783)

Added generated files for EVP_KDF changes

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8808)

Added EVP_KDF (similiar to the EVP_MAC)

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8808)