Michael Baentsch [Tue, 29 Jul 2025 10:32:00 +0000 (12:32 +0200)]
Remove unnecessary OPENSSL_NO_RSA remnants
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com> Reviewed-by: Paul Dale <ppzgs1@gmail.com> Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/28113)
Michael Baentsch [Mon, 28 Jul 2025 04:52:49 +0000 (06:52 +0200)]
fuzz/dtlsserver.c: Remove incorrect ifdef guard
Reviewed-by: Kurt Roeckx <kurt@roeckx.be> Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/28102)
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com> Reviewed-by: Viktor Dukhovni <viktor@openssl.org> Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/28105)
Only report generic error if provider did not put an error on the error queue
Commit 72351b0d18078170af270418b2d5e9fc579cb1af added code to unconditionally
put a generic error onto the error stack, if key generation, encryption,
decryption, sign, or verify fails to ensure that there is an error entry
on the error queue, even if the provider did not itself put a specific error
onto the queue.
However, this can hide error details if an application just looks at the very
last error entry and checks for specific errors. Now, the generic error is
always the last entry, and the application won't find the expected error
entry, although it would be there as second last entry. This can lead to
different application behavior in error situations than before this change.
To fix this, only add the generic error entry if the provider did not itself
add an error entry onto the queue. That way, there always is an error on the
error queue in case of a failure, but no behavior change in case the provider
emitted the error entry itself.
Closes: https://github.com/openssl/openssl/issues/27992 Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com> Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/28073)
Make ERR_count_to_mark() available to providers via 'in' dispatch array
Functions like ERR_set_mark(), ERR_clear_last_mark(), and ERR_pop_to_mark()
are already passed to the a provider via the 'in' dispatch array of the
provider initialization function (although the documentation did not
mention them).
Also pass ERR_count_to_mark() to the provider the same way, and update
the documentation to mention all four functions.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com> Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/28073)
The commit "Remove HARNESS_OSSL_PREFIX manipulation in the test harness"
forced all the output to be processed by the test harness, which means
that any process that keeps the stdout FD open prevents the run() call
from finishing, as was the case in the test_external_cf_quiche test that
ran quiche server in the background, but retaining the std{in,out,err}
descriptors. Avoid that by explicitly redirecting them to a log file.
Reported-by: Tomas Mraz <tomas@openssl.org> Fixes: 70c05fcde53cf "Remove HARNESS_OSSL_PREFIX manipulation in the test harness" Signed-off-by: Eugene Syromiatnikov <esyr@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Tim Hudson <tjh@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/28125)
Also add support for ML-KEM in CMS (draft-ietf-lamps-cms-kyber).
Add the -recip_kdf and -recip_ukm parameters to `openssl cms -encrypt`
to allow the user to specify the KDF algorithm and optional user
keying material for each recipient.
A provider may indicate which RecipientInfo type is supported
for a key, otherwise CMS will try to figure it out itself. A
provider may also indicate which KDF to use in KEMRecipientInfo
if the user hasn't specified one.
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/27681)
Anton Moryakov [Fri, 25 Jul 2025 12:43:15 +0000 (15:43 +0300)]
crypto: evp: fix potential null pointer dereference in EVP_DigestSignUpdate in m_sigver.c
Static analysis detected that EVP_DigestSign() could lead to null pointer
dereference in EVP_DigestSignUpdate() when pctx->pmeth is NULL. The issue
occurs in the legacy code path where pmeth is accessed without prior null
check.
This fix adds a proper null check for pctx->pmeth in the legacy section
of EVP_DigestSignUpdate() to prevent the crash when the function is called
through EVP_DigestSign() with improperly initialized context.
The check is placed in EVP_DigestSignUpdate() rather than EVP_DigestSign()
to maintain proper separation of concerns and follow OpenSSL's architectural
patterns where lower-level functions handle their own parameter validation.
Fixes potential crash in signature operations with legacy providers.
CLA: trivial Signed-off-by: Anton Moryakov <ant.v.moryakov@gmail.com> Reviewed-by: Tomas Mraz <tomas@openssl.org> Reviewed-by: Paul Dale <ppzgs1@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/28095)
Reviewed-by: Viktor Dukhovni <viktor@openssl.org> Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com> Reviewed-by: Paul Dale <ppzgs1@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/28101)
Neil Horman [Sat, 26 Jul 2025 13:28:01 +0000 (09:28 -0400)]
Set *sk_resp to NULL when freeing.
Its possible for get_ocsp_resp_from_responder to return OK after having
freed *sk_resp without setting the freed pointer to NULL, leading us to
set a garbage pointer in other code.
Reviewed-by: Viktor Dukhovni <viktor@openssl.org> Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com> Reviewed-by: Paul Dale <ppzgs1@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/28101)
Reviewed-by: Viktor Dukhovni <viktor@openssl.org> Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com> Reviewed-by: Paul Dale <ppzgs1@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/28101)
Neil Horman [Mon, 14 Jul 2025 13:12:17 +0000 (09:12 -0400)]
reduce lock contention when adding objects to ADDED_OBJ hash table
Reviewed-by: Tomas Mraz <tomas@openssl.org> Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Paul Dale <ppzgs1@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/28081)
Neil Horman [Sun, 13 Jul 2025 12:51:41 +0000 (08:51 -0400)]
move added creation to happen outside of write lock
We have a global ADDED_OBJ hashtable, and we check to create it
on every object add with the write lock potentially held. move that
creation to a RUN_ONCE routine so that its always created before the
write lock is held
Reviewed-by: Tomas Mraz <tomas@openssl.org> Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Paul Dale <ppzgs1@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/28081)
Check the code (source files and .in files) for undocumented ENV
variable names. The variable name should be documented in openssl-env or
in a designated man page in the "ENVIRONMENT" section.
Fixes: ee0bf38e8709 "improve and move text on OPENSSL_TRACE from doc/man1/openssl.pod to doc/man7/openssl-env.pod" Signed-off-by: Eugene Syromiatnikov <esyr@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org> Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/28025)
It seems that its usage has been changed since the initial mention
in 6d594fdf52c4 "OSSL_trace_set_channel.pod and openssl.pod:
fix missing/inconsistent category items".
Signed-off-by: Eugene Syromiatnikov <esyr@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org> Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/28025)
The manual page describes not only environment variables that pertain
the library, but some of the applications as well (OPENSSL,
OPENSSL_CONFIG, and TSGET are possible examples). Also, the effect
of the variables is not limited to the changes in default paths.
Signed-off-by: Eugene Syromiatnikov <esyr@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org> Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/28025)
Sort HTTP_PROXY, HTTPS_PROXY, and NO_PROXY in lexicographical order, add
mentions of the lowercase variants of the variables (as they are already
mentioned in OSSL_HTTP_parse_url(3)), in the same order, despite
it is not a lexicographical one, do minimise confusion.
Signed-off-by: Eugene Syromiatnikov <esyr@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org> Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/28025)
doc/man7/openssl-env: sort the capability envvars/links lexicographically
Put the mentions of OPENSSL_*cap environment variables and man page
references in lexicographical order. Also, add an Oxford comma before
"and" in the man page references.
Signed-off-by: Eugene Syromiatnikov <esyr@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org> Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/28025)
As the environment variable is no longer used by the test harness,
its support can be removed from the openssl command, along
with the possible unintended consequences of prefixing the standard
output of the openssl command with an arbitrary string set
in the environment.
Signed-off-by: Eugene Syromiatnikov <esyr@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org> Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/28025)
Remove HARNESS_OSSL_PREFIX manipulation in the test harness
The aim of HARNESS_OSSL_PREFIX environment variable is to avoid contaminating
TAP producer's output with stanzas that can be interpreted by a TAP producer
by prefixing them with comment; this can be achieved by processing
the output within the runner instead, as it already does for non-standard
prefixes; it also has the added benefit of alleviating the need
to reset it for the external tests in order to avoid messing
with their output checks.
Signed-off-by: Eugene Syromiatnikov <esyr@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org> Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/28025)
Nikola Pajkovsky [Thu, 17 Jul 2025 10:45:49 +0000 (12:45 +0200)]
Pairwise check for DH keys import as part of FIPS
For DH key import, it appears the PCT and assurances are implemented in
the source but are not be performed by default.
Signed-off-by: Nikola Pajkovsky <nikolap@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Viktor Dukhovni <viktor@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/28058)
Nikola Pajkovsky [Wed, 16 Jul 2025 12:37:58 +0000 (14:37 +0200)]
Update dh_pub to be pairwise consistent with dh_priv
Signed-off-by: Nikola Pajkovsky <nikolap@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Viktor Dukhovni <viktor@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/28058)
Co-authored-by: Andrew Dinh <andrewd@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/28023)
Co-authored-by: Andrew Dinh <andrewd@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/28023)
ACK manager must avoid infinite probe time when waiting handshake confirmation
According to RFC 9002, section 6.2.2.1 the client the client must keep PTO (probe
time out) armed if it has not seen HANDSHAKE_DONE quic message from server.
Not following RFC spec here may cause the QUIC session to stale during TLS handshake.
Fixes openssl/project#1266
Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/28023)
Richard Levitte [Wed, 16 Jul 2025 13:08:48 +0000 (15:08 +0200)]
Add a test of 'openssl storeutl' with a BER-encoded PKCS#12 file
The test file (test-BER.p12) was given to us by David von Oheimb
Co-Authored-By: David von Oheimb <david.von.oheimb@siemens.com> Reviewed-by: Tomas Mraz <tomas@openssl.org> Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/28016)
Richard Levitte [Thu, 10 Jul 2025 15:55:50 +0000 (17:55 +0200)]
Fix OSSL_STORE to consider cached info in the EOF check.
OSSL_STORE_load() called OSSL_STORE_eof() before checking if there is
cached OSSL_STORE_INFO to consider. To fix this issue, the cached info
check is moved to OSSL_STORE_eof(), as that seems to make most common
sense.
This solves an issue with PKCS#12 files, where the cached info was never
considered because the underlying file IO layer signaled that EOF is
reached.
Fixes #28010
Reviewed-by: Tomas Mraz <tomas@openssl.org> Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/28016)
martin [Fri, 7 Feb 2025 13:22:41 +0000 (14:22 +0100)]
Add support for TLS 1.3 OCSP multi-stapling for server certs
Co-authored-by: Michael Krueger 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/20945)
AntonMoryakov [Mon, 2 Jun 2025 10:14:28 +0000 (13:14 +0300)]
sm2: sm2_sign.c: check EC_KEY_get0_private_key() for NULL in sm2_sig_gen()
Static analysis revealed that sm2_sig_gen() dereferences the return value
of EC_KEY_get0_private_key() without checking for NULL. This could lead to
a crash if the private key is unset.
This patch adds a NULL check and raises ERR_R_PASSED_NULL_PARAMETER if the
key is missing.
Issue found by static analyzer:
> Return value of EC_KEY_get0_private_key() is dereferenced without checking for NULL (11/12 checked)
CLA: trivial Signed-off-by: Anton Moryakov <ant.v.moryakov@gmail.com> Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com> Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/27741)
Daniel Van Geest [Wed, 23 Jul 2025 08:30:57 +0000 (10:30 +0200)]
openssl rand command should use the loaded library context
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com> Reviewed-by: Paul Dale <ppzgs1@gmail.com> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/28075)
Julian Zhu [Thu, 26 Jun 2025 06:14:45 +0000 (14:14 +0800)]
LoongArch: Add SHA-512 assembly implementation for better performance on small-size data
Signed-off-by: Julian Zhu <jz531210@gmail.com> Reviewed-by: Paul Yang <paulyang.inf@gmail.com> Reviewed-by: Paul Dale <ppzgs1@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/27904)
Julian Zhu [Fri, 20 Jun 2025 07:59:08 +0000 (15:59 +0800)]
LoongArch: Add SHA-256 assembly implementation for better performance
It can get about 25%~ better performance than compiler-generated code (gcc version 15.1.1).
Signed-off-by: Julian Zhu <jz531210@gmail.com> Reviewed-by: Paul Yang <paulyang.inf@gmail.com> Reviewed-by: Paul Dale <ppzgs1@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/27904)
Most of the GitHub Actions can be replaced with a couple of commands.
This will reduce dependencies bloating and some attacks surface.
This commit replaces all GitHub Actions which are not from GitHub
and Cygwin for Windows jobs with plain commands.
Reviewed-by: Saša Nedvědický <sashan@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/27467)
Pauli [Wed, 9 Jul 2025 02:50:02 +0000 (12:50 +1000)]
evp_test: add a new global "Test-Entropy" line to allow deterministic `random` input.
When specified, this directive replaces the public and private DRBGs
with a generator that returns the specified bytes. If more bytes are
requested than are specified, the generator loops around the supplied
bytes to satisfy the request.
ECX and ED were sharing the same code for the getter, whilst also
maintaining seperate gettable tables.
The code has been reworked so that common code is shared, and algorithm
specific code is seperated out to make this clearer.
This fixes:
(1) The security category not being in the gettable table for ED
(2) The fips indicator for ED. i.e There is no fips indicator for ED
(previously there was no gettable, but the get() still tried to return
unapproved).
Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Paul Dale <ppzgs1@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/28039)
Raise PROV_R_NULL_OUTPUT_BUFFER if shsec is NULL in ml_kem_encapsulate()
Signed-off-by: lanming <lanming@huawei.com> Reviewed-by: Tomas Mraz <tomas@openssl.org> Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Paul Dale <ppzgs1@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/28009)
quarckster is the owner of the /.github/workflows/ directory.
Reviewed-by: Richard Levitte <levitte@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Tim Hudson <tjh@openssl.org> Reviewed-by: Paul Dale <ppzgs1@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/28060)
Dmitry Misharov [Thu, 27 Mar 2025 15:49:20 +0000 (16:49 +0100)]
pin GitHub Actions revisions from untrusted vendors
Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Viktor Dukhovni <viktor@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/27181)
evp_pkey_type: Make base_id_conversion table static
This commit fixes CI failures (when configured with --strict-warnings)
Signed-off-by: Norbert Pocs <norbertp@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Paul Dale <ppzgs1@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/28061)
Norbert Pocs [Tue, 17 Jun 2025 10:43:12 +0000 (12:43 +0200)]
Deprecate ASN1_METH related tests
ASN1 tests had to be turned off, but the biggest change is the
ssl_test_ctx, where the NID resolution does not equal to the old one and
a little hack had to be used to make the test work.
Signed-off-by: Norbert Pocs <norbertp@openssl.org> Reviewed-by: Saša Nedvědický <sashan@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/27727)
Signed-off-by: Norbert Pocs <norbertp@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/27961)
Signed-off-by: Norbert Pocs <norbertp@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/27961)
Signed-off-by: Norbert Pocs <norbertp@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/27961)
Signed-off-by: Norbert Pocs <norbertp@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/27961)
Signed-off-by: Norbert Pocs <norbertp@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/27961)
Signed-off-by: Norbert Pocs <norbertp@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/27961)
Signed-off-by: Norbert Pocs <norbertp@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/27961)
crypto/bn/bn_lib.c: optimize - seems to bring not much benefit
Signed-off-by: Norbert Pocs <norbertp@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/27961)
Signed-off-by: Norbert Pocs <norbertp@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/27961)
Signed-off-by: Norbert Pocs <norbertp@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Paul Dale <ppzgs1@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/28036)
Signed-off-by: Norbert Pocs <norbertp@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Paul Dale <ppzgs1@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/28036)
This was a false positive in a test.
The code has been reordered to make the flow clearer.
Reviewed-by: Tim Hudson <tjh@openssl.org> Reviewed-by: Paul Dale <ppzgs1@gmail.com> Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/28030)
If ctx->key is already set and the passed in key is NULL then ctx->key
should not be set to NULL.
Reviewed-by: Tim Hudson <tjh@openssl.org> Reviewed-by: Paul Dale <ppzgs1@gmail.com> Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/28030)
Reviewed-by: Richard Levitte <levitte@openssl.org> Reviewed-by: Paul Dale <ppzgs1@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/28029)
the test shows that sec mem is ok-ish up to the number of available cores,
and when the sec mem lock gets contended, performance goes down rapidly.
Tested on Apple M4 Pro.
[1] https://github.com/openssl/openssl/pull/27625
Signed-off-by: Nikola Pajkovsky <nikolap@openssl.org> Reviewed-by: Viktor Dukhovni <viktor@openssl.org> Reviewed-by: Kurt Roeckx <kurt@roeckx.be> Reviewed-by: Tomas Mraz <tomas@openssl.org> Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Paul Dale <ppzgs1@gmail.com> Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/28011)
Nikola Pajkovsky [Fri, 11 Jul 2025 07:18:12 +0000 (09:18 +0200)]
ci: enable lms only on master
a new config option cannot be enabled globally because the option
was not backported to the older versions.
Signed-off-by: Nikola Pajkovsky <nikolap@openssl.org> Reviewed-by: Saša Nedvědický <sashan@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/28019)
Neil Horman [Tue, 24 Jun 2025 17:50:09 +0000 (13:50 -0400)]
Add lock contention checking to our pthreads implementation
Something we're missing in our ability to do performance monitoring
and diagnosis in openssl is the ability to check for lock contention.
While some tools exist for this (valgrinds drd tool for example), they
really only measure the time spent in critical sections, not the
instances in which they are contended. For that we need something more
specific.
This patch introduces the REPORT_RWLOCK_CONTENTION macro. When openssl
is built with:
./Configure -rdynamic -fno-omit-frame-pointer -DREPORT_RWLOCK_CONTENTION
We can now get output sent to a log file that looks like the following:
Which tells us when a thread blocked because someone else was already
holding the lock, how long it was blocked for, and where the blocking
call originated from via its backtrace.
I think this should enable us to better determine where our contended
locking paths are for a given application, and give us some insight on
how to fix them.
Currently its linux only (as the backtrace functionality only exists
there, and there are few warts (like the need to use a file pointer
rather than a bio to record the log, see comments), but I think its
enough to give us a useful diagnostic tool to help drive some
performance improvements.
Fixes openssl/project#1237
Reviewed-by: Saša Nedvědický <sashan@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org> Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/27884)
Nikola Pajkovsky [Thu, 10 Jul 2025 07:48:15 +0000 (09:48 +0200)]
Make the lock in CRYPTO_secure_actual_size a read lock
there is no operations within critical section that would
require write lock.
Signed-off-by: Nikola Pajkovsky <nikolap@openssl.org> Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Neil Horman <nhorman@openssl.org> Reviewed-by: Paul Dale <ppzgs1@gmail.com> Reviewed-by: Viktor Dukhovni <viktor@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/28014)
Michael Baentsch [Mon, 30 Jun 2025 07:33:46 +0000 (09:33 +0200)]
Add note about use of EVP_PKEY in different libctxs
Co-authored-by: Shane Lontis <slontis@oracle.com> Reviewed-by: Tim Hudson <tjh@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/26309)
Theo Buehler [Sun, 6 Jul 2025 11:55:52 +0000 (13:55 +0200)]
Provide X509_CRL_get0_tbs_sigalg()
X509_CRL_get0_tbs_sigalg() corresponds to X509_get0_tbs_sigalg() and
retrieves the AlgorithmIdentifier inside the TBSCertList which is not
currently accessible in any sane way from public API.
This PR adds X509_get0_tbs_sigalg() to the public API, documents it,
adds a simple regress check so there is coverage and mentions the
addition in CHANGES.md.
On top of that, fix a typo in .gitignore and clean up some order
inconsistencies in X509_get0_signature.pod.
Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Tim Hudson <tjh@openssl.org> Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/27971)
projects / thirdparty / openssl.git / log
