Cyclic0007/wolfssl
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
25,655 Commits 12 Branches 184 Tags
f119086d3e1f9e8f8bfe46519e492f6ca2dfd646
Commit Graph

8641 Commits

Author SHA1 Message Date
JacobBarthelmeh
fe7d458d29 random.c is also locked in FIPS v6 2025-06-24 16:08:25 -06:00
David Garske
978a29da0b Merge pull request #8898 from cconlon/getpidOptionsH 
Add HAVE_GETPID to options.h if getpid detected
 2025-06-23 17:11:55 -07:00
David Garske
caf8494d65 Merge pull request #8911 from gojimmypi/pr-allow-only-rsa 
Allow configuration with only RSA cipher suites
 2025-06-23 11:18:27 -07:00
Daniel Pouzzner
b361c62372 Merge pull request #8903 from dgarske/cadate_calist 
Expose API to access "store" error code and depth for cert failure callback
 2025-06-23 10:08:41 -05:00
gojimmypi
afa22dfc2b Allow configuration with only RSA cipher suites 2025-06-21 14:54:10 -07:00
David Garske
1be303866e Merge pull request #8908 from douzzer/20250620-clang-tidy-and-cppcheck-fixes-and-workarounds 
20250620-clang-tidy-and-cppcheck-fixes-and-workarounds
 2025-06-20 15:07:09 -07:00
David Garske
f30c54abdd Merge pull request #8894 from SparkiDev/ppc32_sha256_asm 
PPC 32 ASM: SHA-256
 2025-06-20 14:29:47 -07:00
Daniel Pouzzner
7977a605c5 src/internal.c: in FreeSskeArgs(), move nullness check on args to the start, and make it unconditional, to resolve nullPointerRedundantChecks. 2025-06-20 15:04:07 -05:00
David Garske
b98cf8882b Remove HAVE_LIGHTY from the client_ca_names feature. 2025-06-20 11:29:02 -07:00
David Garske
9b50708741 Fix to expose API to access "store" error code and error depth for cert failure callback (from set_verify). Useful for C# wrapper or clients that cannot directly dereference X509_STORE. Fixes for building with WOLFSSL_EXTRA and WOLFSSL_NO_CA_NAMES (and added new tests). Added example in CSharp TLS client for overriding a begin date error (useful if date is not set). 2025-06-19 14:49:00 -07:00
Chris Conlon
cdd02f9665 Add check for reseed in ssl.c for HAVE_SELFTEST, similar to old FIPS bundles that do not have older random.c files 2025-06-18 17:21:55 -06:00
David Garske
27176a5eeb Merge pull request #8870 from kareem-wolfssl/zd20030 
Various minor fixes.
 2025-06-18 08:55:07 -07:00
Sean Parkinson
c39f1fe721 PPC 32 ASM: SHA-256 
Pure and inline  ASM for the PowerPC 32-bit.
 2025-06-18 21:23:15 +10:00
David Garske
7d77446964 Merge pull request #8882 from rizlik/dtls13_always_transmit_explicit_ack 
dtls13: always send ACKs on detected retransmission
 2025-06-17 11:35:07 -07:00
David Garske
6b68797b4f Merge pull request #8883 from JacobBarthelmeh/rng 
account for Intel RDRAND build without HAVE_HASHDRBG
 2025-06-17 11:33:16 -07:00
Kareem
2366718d5a Add args->input free in FreeSskeArgs. 
This free is redundant in most cases but it covers the specific
case of using async, exiting SendServerKeyExchange early due to
WANT_WRITE or WC_PENDING_E, then later freeing the async context
without calling SendServerKeyExchange again.
 2025-06-17 10:12:06 -07:00
David Garske
5e6c1ba05f Merge pull request #8879 from julek-wolfssl/openssh-10.0p2 
Updates for OpenSSH 10.0p2
 2025-06-17 09:36:45 -07:00
Marco Oliverio
e82c099bec fix indentation 2025-06-16 18:42:17 +02:00
JacobBarthelmeh
ce61f0d517 account for Intel RDRAND build without HAVE_HASHDRBG 2025-06-16 09:04:50 -06:00
Marco Oliverio
b1b49c9ffb dtls13: always send ACKs on detected retransmission 
Otherwise the connection can stall due the indefinite delay of an explicit ACK,
for exapmle:

 -> client sends the last Finished message
<- server sends the ACK, but the ACK is lost
 -> client rentrasmit the Finished message
 - server delay sending of the ACK until a fast timeout
 -> client rentrasmit the Finished message quicker than the server timeout
 - server resets the timeout, delaying sending the ACK
 -> client rentrasmit the Finished...
 2025-06-16 14:19:32 +02:00
Marco Oliverio
509491f554 dtls13: wolfSSL_is_init_finished true after last server ACK 
Do not consider the handshake finished until the last server ACK.
This way the application knows where to switch from
wolfSSL_negotiate/wolfSSL_connect to wolfSSL_read/wolfSSL_write.
 2025-06-16 14:19:31 +02:00
Juliusz Sosinowicz
37554a13db Updates for OpenSSH 10.0p2 
- random.c: use getrandom when available and fall back to direct file access
- openssh.yml: run more tests
- openssh.yml: add 10.0p2 and 9.9p2
- configure.ac: detect if `getrandom` is available on the system
- configure.ac: openssh requires WC_RNG_SEED_CB to always use `getrandom` so that the RNG doesn't get killed by SECCOMP
 2025-06-13 18:06:19 +02:00
Josh Holtrop
8bde5e6982 Fix printing empty names in certificates 
The empty-issuer-cert.pem certificate was created with:

    wolfssl genkey rsa -size 2048 -out mykey -outform pem -output KEY
    wolfssl req -new -days 3650 -key mykey.priv -out empty-issuer-cert.pem -x509

Prior to this fix this command would error printing the certificate:

    wolfssl x509 -inform pem -in empty-issuer-cert.pem -text
 2025-06-13 11:22:52 -04:00
David Garske
2fc1110a13 Merge pull request #8587 from lealem47/gh8574 
Fix bug in ParseCRL_Extensions
 2025-06-12 12:09:52 -07:00
David Garske
6571f42cb9 Merge pull request #8867 from JacobBarthelmeh/rng 
Improvements to RNG and compatibility layer
 2025-06-11 14:31:53 -07:00
JacobBarthelmeh
ae87afa677 Merge pull request #8857 from miyazakh/tsip_fix 
fix TSIP TLS example program
 2025-06-10 16:26:34 -06:00
JacobBarthelmeh
47cf634965 add a way to restore previous pid behavior 2025-06-10 16:12:09 -06:00
JacobBarthelmeh
fbbb6b7707 add mutex locking and compat layer FIPS case 2025-06-10 14:37:11 -06:00
JacobBarthelmeh
31490ab813 add sanity checks on pid with RNG 2025-06-10 14:37:11 -06:00
JacobBarthelmeh
2d892f07eb Merge pull request #8861 from gasbytes/psk-handshake-failure-fix 
tls13: clear tls1_3 on downgrade
 2025-06-10 10:24:17 -06:00
JacobBarthelmeh
047f0bb5fc Merge pull request #8847 from gojimmypi/pr-platformio-cert-bundles 
Improve PlatformIO Certificate Bundle Support
 2025-06-10 10:23:07 -06:00
JacobBarthelmeh
94f5948f20 Merge pull request #8858 from rizlik/dtls13_set_epoch_fix 
dtls13: move Dtls13NewEpoch into DeriveTls13Keys
 2025-06-10 09:48:58 -06:00
Marco Oliverio
3e6703e1fb fixup! dtls13: move Dtls13NewEpoch into DeriveTls13Keys 2025-06-09 19:20:06 +02:00
Marco Oliverio
1024bf0109 fixup! dtls13: move Dtls13NewEpoch into DeriveTls13Keys 2025-06-09 18:00:23 +02:00
Reda Chouk
92b6e2f2e9 tls13: clear tls1_3 on downgrade 
Unset ssl->options.tls1_3 whenever we drop to TLS 1.2 so PSK
handshakes don’t hit -326 VERSION_ERROR.
 2025-06-09 17:12:56 +02:00
Marco Oliverio
59ff71f936 fixup! dtls13: move Dtls13NewEpoch into DeriveTls13Keys 2025-06-09 16:11:17 +02:00
Andrew Hutchings
5e6cb2b0b6 Allow trusted_ca_keys with TLSv1.3 
It is possible that the client will provied `trusted_ca_keys` during a
TLSv1.3 connection with 1.2 downgrade. wolfSSL would error with
`EXT_NOT_ALLOWED`. The TLSv1.3 spec states that it can be provided and
should be ignored.

ZD 19936
 2025-06-09 08:31:54 +01:00
Marco Oliverio
c1c1929e55 dtls13: move Dtls13NewEpoch into DeriveTls13Keys 
Dlts13NewEpoch saves the keys currently derived in the ssl object.
Moving Dtls13NewEpoch inside DeriveTls13Keys avoid the risk of using the wrong
keys when creating a new Epoch.

This fixes at least he following scenario:

- Client has encryption epoch != 2 in the handshake (eg. due to rtx)

- Client derives traffic0 keys after receiving server Finished message

- Client set encryption epoch to 2 again to send the Finished message, this
   override the traffic key computed

- Client creates the new epoch with the wrong key
 2025-06-09 02:35:29 +02:00
Hideki Miyazaki
1f8efc3c14 fix TSIP example 
fix Client Certificate Verify using RSA sign/verify
 2025-06-07 12:38:18 +09:00
gojimmypi
3254f56d32 Improve PlatformIO Certificate Bundle Support 2025-06-06 15:48:07 -07:00
JacobBarthelmeh
0bac2c2b34 Merge pull request #8846 from lealem47/zd20027 
Don't include AEAD nonce in decrypted data size
 2025-06-06 15:43:20 -06:00
JacobBarthelmeh
369f9f0339 Merge pull request #8849 from holtrop/reseed-drbg-in-rand-poll 
Reseed DRBG in RAND_poll()
 2025-06-06 11:55:46 -06:00
JacobBarthelmeh
f4821eb0f4 Merge pull request #8827 from SparkiDev/ml_kem_codepoints 
ML_KEM IDs backward compat
 2025-06-06 11:06:15 -06:00
JacobBarthelmeh
570c1fc390 Merge pull request #8824 from JeremiahM37/tlsCurveFix 
tls fix for set_groups
 2025-06-06 10:47:06 -06:00
Josh Holtrop
0c12337194 Reseed DRBG in RAND_poll() 2025-06-06 12:20:58 -04:00
JacobBarthelmeh
bfc55d9016 Merge pull request #8848 from julek-wolfssl/gh/8841 
dtlsProcessPendingPeer: correctly set the current peer
 2025-06-06 09:52:35 -06:00
Daniel Pouzzner
efc36655e6 src/internal.c: add pedantic-compatible NO_TLS codepath for cipher_names[] and GetCipherNamesSize(). 2025-06-06 18:02:19 +04:00
Sean Parkinson
7eca4fb331 ML_KEM IDs backward compat 
Allow backward compatibilitly of Hybrid ML_KEM codepoints in TLS with
version before wolfSSL 5.8.0.
When WOLFSSL_ML_KEM_USE_OLD_IDS is defined, it will accept the old
codepoints for P256 with ML-KEM-512, P384 with ML-KEM-768, P521 with
ML-KEM-10124. (Others combinations were not know pre 5.8.0.)
Both old client with new server and new client with new server work with
old codepoints.
 2025-06-06 09:17:40 +10:00
JacobBarthelmeh
3ecc58cc0e Merge pull request #8842 from julek-wolfssl/zd/19966 
ALT_NAMES_OID: Mark IP address as WOLFSSL_V_ASN1_OCTET_STRING
 2025-06-05 17:07:47 -06:00
Juliusz Sosinowicz
736a5e1f89 dtlsProcessPendingPeer: correctly set the current peer 2025-06-06 00:12:38 +02:00