CLA: trivial

Reviewed-by: Tomas Mraz <[email protected]>
Reviewed-by: Matt Caswell <[email protected]>
(Merged from openssl#13632)

(cherry picked from commit 74c8dd1)

Reviewed-by: Shane Lontis <[email protected]>
(Merged from openssl#13635)

(cherry picked from commit 7eea331)

A servername cb may change the available certificates, so if we have one
set then we cannot rely on the configured certificates to determine if we
are capable of negotiating TLSv1.3 or not.

Fixes openssl#13291

Reviewed-by: Tomas Mraz <[email protected]>
(Merged from openssl#13305)

If an SNI callback has been set then we may have no certificuates suitable
for TLSv1.3 use configured for the current SSL_CTX. This should not prevent
us from negotiating TLSv1.3, since we may change the SSL_CTX by the time we
need a suitable certificate.

Reviewed-by: Tomas Mraz <[email protected]>
(Merged from openssl#13305)

Define B_ENDIAN on PowerPC because it is a big endian architecture. With
this change the BN* related tests pass.

Fixes: openssl#12199

Signed-off-by: Sebastian Andrzej Siewior <[email protected]>

Reviewed-by: Richard Levitte <[email protected]>
Reviewed-by: Paul Dale <[email protected]>
Reviewed-by: Tomas Mraz <[email protected]>
(Merged from openssl#12371)

(cherry picked from commit 52c6c12)

Reviewed-by: Richard Levitte <[email protected]>
(Merged from openssl#13686)

(cherry picked from commit 4159ebc)

This is a backport of the documentation from openssl#13620.

Reviewed-by: David von Oheimb <[email protected]>
Reviewed-by: Tomas Mraz <[email protected]>
(Merged from openssl#13691)

'check-update' runs a 'make update' to check that it wasn't forgotten.

'check-docs' runs 'make doc-nits'.  We have that as a separate job to
make it more prominent.

Reviewed-by: Tomas Mraz <[email protected]>
Reviewed-by: Kurt Roeckx <[email protected]>
(Merged from openssl#13701)

(cherry picked from commit 8175476)

When X509_ATTRIBUTE_create() receives an invalid NID (e.g., -1), return
failure rather than silently constructing a broken X509_ATTRIBUTE object
that might cause NULL pointer accesses later on.  This matters because
X509_ATTRIBUTE_create() is used by API functions like PKCS7_add_attribute(3)
and the NID comes straight from the user.

This bug was found while working on LibreSSL documentation.

Reviewed-by: Theo Buehler <[email protected]>

CLA: trivial

Reviewed-by: Matt Caswell <[email protected]>
Reviewed-by: Tomas Mraz <[email protected]>
(Merged from openssl#12052)

(cherry picked from commit c4b2c53)

Backport of openssl#13394

Reviewed-by: Tomas Mraz <[email protected]>
Reviewed-by: Ben Kaduk <[email protected]>
(Merged from openssl#13637)

This backports openssl#13764.

Reviewed-by: Tim Hudson <[email protected]>
(Merged from openssl#13769)

I.e.:

    error: out of range immediate fixup value

This fix is identical to one of the changes made in 3405db9, which I
discovered right after taking a quick stab at fixing this.

CLA: trivial
Fixes openssl#7878

Reviewed-by: Kurt Roeckx <[email protected]>
Reviewed-by: Tomas Mraz <[email protected]>
(Merged from openssl#13757)

Our free functions should be able to deal with the case where the object
being freed is NULL. This turns out to not be quite the case for DTLS
related objects.

Fixes openssl#13649

Reviewed-by: Tomas Mraz <[email protected]>
(Merged from openssl#13655)

(cherry picked from commit d0afb30)

Fixes openssl#13183

From the original issue report, before this commit, on master and on
1.1.1, the issue can be detected with the following steps:

- Start with a default SSL_CTX, initiate a TLS 1.3 connection with SNI,
  "Accept" count of default context gets incremented
- After servername lookup, "Accept" count of default context gets
  decremented and that of SNI context is incremented
- Server sends a "Hello Retry Request"
- Client sends the second "Client Hello", now again "Accept" count of
  default context is decremented. Hence giving a negative value.

This commit fixes it by adding a check on `s->hello_retry_request` in
addition to `SSL_IS_FIRST_HANDSHAKE(s)`, to ensure the counter is moved
only on the first ClientHello.

CLA: trivial

Reviewed-by: Matt Caswell <[email protected]>
Reviewed-by: Paul Dale <[email protected]>
(Merged from openssl#13297)

Reviewed-by: Tomas Mraz <[email protected]>
Reviewed-by: Nicola Tuveri <[email protected]>
(Merged from openssl#13772)

Reviewed-by: Ben Kaduk <[email protected]>
Reviewed-by: Matt Caswell <[email protected]>
(Merged from openssl#13821)

(cherry picked from commit b57ec73)

when possible using the getauxval equivalent which has similar ids as Linux, instead of bad instructions catch approach.

Reviewed-by: Ben Kaduk <[email protected]>
Reviewed-by: Matt Caswell <[email protected]>
(Merged from openssl#13650)

(cherry picked from commit 5eb24fb)

Pull in check from openssl#10878
Move disabling of pic, threads and statics up higher before they
are checked.

Fixes openssl#12772

Reviewed-by: Tomas Mraz <[email protected]>
Reviewed-by: Richard Levitte <[email protected]>
(Merged from openssl#12773)

Fixes openssl#13840

Reviewed-by: Richard Levitte <[email protected]>
(Merged from openssl#13857)

(cherry picked from commit 4369a88)

…o due to invalid cert

This is the backport of openssl#13755 to v1.1.1.
Fixes openssl#13698

Reviewed-by: Tomas Mraz <[email protected]>
(Merged from openssl#13756)

...in case the candidate issuer cert is identical to the target cert.

Fixes openssl#13739

Reviewed-by: Tomas Mraz <[email protected]>
(Merged from openssl#13749)

There were a number of older style references to the pass phrase
options section, now streamlined with the current openssl(1).

Fixes openssl#13883

Reviewed-by: Kurt Roeckx <[email protected]>
(Merged from openssl#13886)

Also add a new no-deprecated CI build to test it.

Fixes openssl#13896

Reviewed-by: David von Oheimb <[email protected]>
(Merged from openssl#13902)

SRP_Calc_client_key calls BN_mod_exp with private data. However it was
not setting BN_FLG_CONSTTIME and therefore not using the constant time
implementation. This could be exploited in a side channel attack to
recover the password.

Since the attack is local host only this is outside of the current OpenSSL
threat model and therefore no CVE is assigned.

Thanks to Mohammed Sabt and Daniel De Almeida Braga for reporting this
issue.

Reviewed-by: Tomas Mraz <[email protected]>
(Merged from openssl#13889)

Fixes openssl#13910

CLA: trivial

Reviewed-by: Tim Hudson <[email protected]>
Reviewed-by: Matt Caswell <[email protected]>
Reviewed-by: Tomas Mraz <[email protected]>
(Merged from openssl#13911)

(cherry picked from commit 6857058)

There are some options that seem to belong to the legacy build.

Reviewed-by: Dmitry Belyavskiy <[email protected]>
(Merged from openssl#13903)

(cherry picked from commit adcaebc)

At this point, we have transitioned completely from Travis to GitHub Actions

Reviewed-by: Tim Hudson <[email protected]>
Reviewed-by: Paul Dale <[email protected]>
(Merged from openssl#13941)

This (re-)allows RSA-PSS signers

Fixes openssl#13931

Reviewed-by: Tomas Mraz <[email protected]>
(Merged from openssl#13982)

Fixes openssl#13815

Reviewed-by: Richard Levitte <[email protected]>
Reviewed-by: Paul Dale <[email protected]>
(Merged from openssl#14029)

(cherry picked from commit af403db)

This folder "../apps/include" is accidentally created.
This prevents this glitch.

Fixes 19b4fe5 ("Add a CMAC test")

Reviewed-by: Tomas Mraz <[email protected]>
(Merged from openssl#14051)

CLA: trivial

Reviewed-by: Paul Dale <[email protected]>
Reviewed-by: Richard Levitte <[email protected]>
Reviewed-by: Matthias St. Pierre <[email protected]>
(Merged from openssl#14078)

The options listed in the array @disablables are regular expressions.
For most of them, it's not visible, but there are a few.

However, configdata.pm didn't quite treat them that way, which meant
that the few that are visibly regular expressions, there's a
difference between that and the corresponding the key in %disabled,
which is never a regular expression.

To correctly display the enabled and disabled options with --dump,
we must therefore go through a bit of Perl gymnastics to get the
output correct enough, primarly so that disabled features don't look
enabled.

Fixes openssl#13790

Reviewed-by: Paul Dale <[email protected]>
(Merged from openssl#14081)

'no-tests' wasn't entirely respected by test/build.info.

Reviewed-by: Paul Dale <[email protected]>
(Merged from openssl#14081)

This field has not been used since openssl#3858 was merged in 2017 when we
moved to a table-based lookup for certificate type properties instead of
an index-based one.

Reviewed-by: Kurt Roeckx <[email protected]>
(Merged from openssl#13991)

(cherry picked from commit 3bc0b62)

DCL has a total command line limitation that's too easily broken by
them.

We solve them by creating separate message scripts and using them.

Fixes openssl#13789

Reviewed-by: Tomas Mraz <[email protected]>
(Merged from openssl#13834)

This mostly clarifies details.

Fixes openssl#13789

Reviewed-by: Tomas Mraz <[email protected]>
(Merged from openssl#13834)

The OpenSSL public API function X509_issuer_and_serial_hash() attempts
to create a unique hash value based on the issuer and serial number data
contained within an X509 certificate. However it fails to correctly
handle any errors that may occur while parsing the issuer field (which
might occur if the issuer field is maliciously constructed). This may
subsequently result in a NULL pointer deref and a crash leading to a
potential denial of service attack.

The function X509_issuer_and_serial_hash() is never directly called by
OpenSSL itself so applications are only vulnerable if they use this
function directly and they use it on certificates that may have been
obtained from untrusted sources.

CVE-2021-23841

Reviewed-by: Richard Levitte <[email protected]>
Reviewed-by: Paul Dale <[email protected]>
(cherry picked from commit 8130d65)

Provide a certificate with a bad issuer and check that
X509_issuer_and_serial_hash doesn't crash.

Reviewed-by: Richard Levitte <[email protected]>
Reviewed-by: Paul Dale <[email protected]>
(cherry picked from commit 55869f5)

Reduce code copying by factoring out common code into a separate function.

Reviewed-by: Paul Dale <[email protected]>

This also fixes the public function RSA_padding_check_SSLv23.

Commit 6555a89 changed the padding check logic in RSA_padding_check_SSLv23
so that padding is rejected if the nul delimiter byte is not immediately
preceded by at least 8 bytes containing 0x03. Prior to that commit the
padding is rejected if it *is* preceded by at least 8 bytes containing 0x03.

Presumably this change was made to be consistent with what it says in
appendix E.3 of RFC 5246. Unfortunately that RFC is in error, and the
original behaviour was correct. This is fixed in later errata issued for
that RFC.

This has no impact on libssl for modern versions of OpenSSL because
there is no protocol support for SSLv2 in these versions. However
applications that call RSA_paddin_check_SSLv23 directly, or use the
RSA_SSLV23_PADDING mode may still be impacted. The effect of the original
error is that an RSA message encrypted by an SSLv2 only client will fail to
be decrypted properly by a TLS capable server, or a message encrypted by a
TLS capable client will fail to decrypt on an SSLv2 only server. Most
significantly an RSA message encrypted by a TLS capable client will be
successfully decrypted by a TLS capable server. This last case should fail
due to a rollback being detected.

Thanks to D. Katz and Joel Luellwitz (both from Trustwave) for reporting
this issue.

CVE-2021-23839

Reviewed-by: Paul Dale <[email protected]>

We test all three cases:
- An SSLv2 only client talking to a TLS capable server
- A TLS capable client talking to an SSLv2 only server
- A TLS capable client talking to a TLS capable server (should fail due
to detecting a rollback attack)

Reviewed-by: Paul Dale <[email protected]>

CVE-2021-23840

Reviewed-by: Paul Dale <[email protected]>

Reviewed-by: Richard Levitte <[email protected]>

Reviewed-by: Richard Levitte <[email protected]>

Reviewed-by: Richard Levitte <[email protected]>

OpenSSL 1.1.1j release tag