Windows XP doesn't support setting socket handles to be non-inheritable,
but the rio_notifier attempts to do so. WSASocketA will there return
an error when the NO_INHERIT flag is set. In that case, just retry the
call without the flag.
Fixes#26943
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26970)
The current retirement code for rcu qp's has a race condition,
which can cause use-after-free errors, but only if more than
3 QPs are allocated, which is not the default configuration.
This fixes an oversight in commit 5949918f9a ("Rework and
simplify RCU code")
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26952)
Make CRYPTO_atomic_add consistent with
CRYPTO_atomic_load_int and set the
reader_idx under write_lock since there
is no CRYPTO_atomic_store_int.
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26963)
- Apply doc nits suggested by Viktor from https://github.com/openssl/openssl/pull/26762
- Update CHANGES.md & NEWS.md saying there is now support for QUIC server
- Added copyright header in: test/radix/quic_ops.c
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26944)
The change checks for all HTTP methods in ssl_record, not only GET, POST,
PUT and HEAD. (additionally PATCH, DELETE, OPTIONS and TRACE)
CLA: trivial
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26968)
The way we're currently handling SAN URIs does not allow for userinfo,
meaning the name constraint check on such URIs will fail. Fix this by
skipping over the userinfo component:
authority = [ userinfo "@" ] host [ ":" port ]
(per RFC 3986).
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
Reviewed-by: Viktor Dukhovni <viktor@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/25861)
And drop the rest.
The ubuntu-20.04 CI runners are discontinued.
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26971)
AIX (at least for 7.1) defines some macros for "events" and "revents" which
interferes with our own use of these names.
Fixes#24236
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26933)
The FIPS providers that support ECX (3.0.x & maybe 3.1.x) do not support ECX
KEM so there is little point to testing these algorithms under FIPS. Consequently,
they are being tested only with the default provider.
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Viktor Dukhovni <viktor@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26946)
The evppkey_rsa.txt data were only tested against the default provider.
Change this so that they are tested against the FIPS provider too.
Also add the RSA KEM tests.
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Viktor Dukhovni <viktor@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26946)
This updates the openssl documentation link to the one currently in use,
and removes the standards.txt section as that URL leads to the normal
documentation page and there is no "standards" page in the openssl
documentation site.
CLA: trivial
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26759)
Also specify whether server or client preference
Reviewed-by: Viktor Dukhovni <viktor@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26897)
Reviewed-by: Viktor Dukhovni <viktor@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26897)
Signed-off-by: Ondrej Moris <omoris@redhat.com>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26899)
We use the coreutils format since 3.4.0.
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26907)
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Viktor Dukhovni <viktor@openssl.org>
Reviewed-by: Paul Dale <ppzgs1@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/26925)
When displaying distinguished names the control characters
are escaped by default.
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26932)
The quic_multistream_test occasionally fails script_84, specifically
failing on:
OP_CHECK2(check_write_buf_stat, 0, 0)
which fails due to the send stream buffer not reading zero after data is
sent on the stream
However, the send stream is culled of pending data, not after the stream
is sent, but rather only after the peer sends an ack confirming that the
data has been received. There is no guarantee that ACK will be sent
immediately, so occasionally timing discrepancies result in the test not
getting that ack by the time we check the send stream buffer.
We couldmodify the script to wait longer, or repeatedly tick the quic
stack to wait for that ack to be collected, but since its perfectly
valid for that data to live in the ring buffer for a period, and that
any true erroneous keeping of that data beyond its ack point would
manifest as any number of other duplicate transmissions, it seems more
sane to just remove the check.
Fixesopenssl/project#1117
Reviewed-by: Saša Nedvědický <sashan@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26939)
We move ssl_err.c out of libssl and into libcrypto. This file is entirely
self contained and is used to load error strings into the libcrypto error
tables. By moving this file into libcrypto, libssl can be unloaded safely
without having dangling references to this error information.
Fixes#26672
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Saša Nedvědický <sashan@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26931)
- The decoder should consider fewer options based on
more precise tracking of the desired input type
(DER, PVK, MSBLOB), algorithm (RSA, EC, ...),
input structure (SPKI, P8, ...).
How much this affects actual use-cases is harder to estimate, we'll just
have to run before/after perf tests.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26927)
EVP_DecodeUpdate() should not produce zeros for input padding `=` signs to avoid writing to non-allocated memory regions.
To achieve this:
- Add `eof` parameter to `evp_decodeblock_int` function in `openssl/crypto/evp`. The parameter should either contain the number of the input padding characters to ignore or `-1` if the function has to count them.
- Use precalculated `eof` in `EVP_DecodeUpdate` to fix its behaviour.
- Use `eof = -1` in `EVP_DecodeFinal` to count it in `evp_decodeblock_int`.
- Do not ignore padding in `EVP_DecodeBlock` (`eof = 0`) because it should write padding zeros according to the documentation.
- Add the HISTORY section to EVP_EncodeInit documentation to describe the fix.
Other changes:
- Update AUTHORS.md
- Update the copyright date in the documentation.
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Viktor Dukhovni <viktor@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26678)
When both seed and key are provided compare the full ML-KEM private key
with the seed keygen output, not just the public key.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26905)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Paul Dale <ppzgs1@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/26908)
Reviewed-by: Saša Nedvědický <sashan@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26916)
twice.
Fixes#26862
This only happens when using the FIPS provider, since it needs to export
the key.
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Paul Dale <ppzgs1@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Viktor Dukhovni <viktor@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26891)
Don't expect success with dated FIPS modules.
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26904)
The decoders in some cases failed to capture or propagate
information about what is being decoded, causing more work
happen to try unrelated decoders as a fallback.
We now try harder to keep track of the expected object (private key or
public key, if known), and the algorithm determined from the OID of a
PKCS8 object or SPKI. This leads in many cases to fewer decoder
invocations. With so many more algorithms now, trying every decoder
is increasingly best avoided.
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26892)
Added a python script to convert the json files into evp_test data.
Added a EVP_TEST_METHOD "KeyFromData" that can test failures when
loading raw keys. (The existing "PrivateKeyRaw" and "PublicKeyRaw"
were not fit for this purpose).
Reviewed-by: Paul Dale <ppzgs1@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26885)
Also move the deprecated curves to the end of the list, and order the
soon most preferred groups first.
Reviewed-by: Paul Dale <ppzgs1@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26873)
Fixes#26876
The issue here is that the pbkdf2 'lower_bounds_checks' currently errors by default
in FIPS mode if iterations < 1000.
i.e. the "pkcs5" flag = 0 triggers an error..
Turning the flag on means the FIPS indicator is triggered (which is probably correct behaviour)
Not sure testing the fips state here is a good idea (i.e. taking a TSAN hit).
Reviewed-by: Paul Dale <ppzgs1@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26887)
