Commit Graph

9965 Commits

Author SHA1 Message Date
David Garske 703bd6d0ba Merge pull request #8975 from JeremiahM37/mldsa_static_mem
ML-DSA Static Memory Fix
2025-07-09 08:22:51 -07:00
Sean Parkinson d6a72e2480 PPC32 ARM ASM SHA-256: SPE impl, tidy up original
Implement using SPE instructions that allow for 64-bit registers as a
vector of 2 32-bit values.
Tidy up original implementation to not use stack.
2025-07-09 18:34:33 +10:00
JeremiahM37 88da86e900 ML DSA Static Memory Fix 2025-07-08 17:51:42 -06:00
Sean Parkinson f0041cd761 SP int: fix 8 bit words and sp_clamp_ct
Need to cast to sp_size_t as it may be bigger than the word type
sp_int_digit.
2025-07-07 18:43:29 +10:00
Hideki Miyazaki ee8be22a3f Fix Qt nightly jenkins failure
PBKDF1 encrpted key
2025-07-07 15:10:41 +09:00
Daniel Pouzzner a40b56ccb5 Merge pull request #8964 from douzzer/20250703-linuxkm-fixes
20250703-linuxkm-fixes

Note, final commit reviewed by @SparkiDev, and earlier commit reviewed by @philljj.
2025-07-04 08:45:55 -05:00
Daniel Pouzzner ef3a1a28d9 linuxkm/linuxkm_wc_port.h, linuxkm/module_hooks.c, and wolfcrypt/src/wc_port.c: fixes for spinlocks on CONFIG_ARM64;
wolfcrypt/src/wc_port.c: include random.h, for Entropy_Init().
2025-07-03 22:09:34 -05:00
Sean Parkinson d1893dbdec ARM 32-bit ASM AES: fixup 32-bit code to not use B.EQ
Changes made for Green Hills Aarch64 got into the 32-bit code.
2025-07-04 11:37:06 +10:00
Daniel Pouzzner 688bc168de wolfcrypt/src/random.c: small stack refactor of noise[] in wc_Entropy_Get(). 2025-07-03 18:30:46 -05:00
David Garske fb691fac94 Merge pull request #8947 from SparkiDev/mldsa_openssl_der
ML-DSA/Dilithium: support OpenSSL format
2025-07-03 16:10:55 -07:00
Sean Parkinson 41eef2ef71 CodeQL: o is larger type and could cause issues
Make 'o' sp_size_t as the callers are passing 0 or explicit cast to
sp_size_t
2025-07-04 09:04:39 +10:00
Sean Parkinson 519d1430d0 ML-DSA/Dilithium: support OpenSSL format
Support DER private key format.
2025-07-04 07:54:26 +10:00
David Garske 3fe84bf3c0 Merge pull request #8961 from douzzer/20250703-fixes-for-multi-test-reports
20250703-fixes-for-multi-test-reports
2025-07-03 11:07:46 -07:00
JacobBarthelmeh 01de7cc04b Merge pull request #8955 from anhu/signed
Explicitly declare dilithium_coeff_eta2[] as signed
2025-07-03 10:25:46 -06:00
JacobBarthelmeh 7abaa131d3 Merge pull request #8954 from dgarske/asm_introspection
Added introspection for Intel and ARM assembly speedups
2025-07-03 10:22:19 -06:00
philljj c0837cb073 Merge pull request #8943 from douzzer/20250617-linuxkm-get_random_bytes
20250617-linuxkm-get_random_bytes
2025-07-03 11:22:08 -05:00
Daniel Pouzzner bdd2056645 wolfcrypt/test/test.c: fix gate in dh_test() (fixes disable-sha256). 2025-07-03 10:19:07 -05:00
Daniel Pouzzner a1fa897572 wolfcrypt/src/dilithium.c: fix cast flubs in dilithium_encode_gamma1_19_bits() (fixes quantum-safe-wolfssl-all-gcc-latest-m32). 2025-07-03 10:17:51 -05:00
David Garske 6be8a3710d Merge pull request #8937 from miyazakh/tsip_cryptcb_ut
Fix TSIP port using crypto callback
2025-07-02 17:42:51 -07:00
Hideki Miyazaki b60a05f45e Fix TSIP port using crypto callback
- Add unit test using cb
2025-07-03 08:23:24 +09:00
David Garske 59061aebec Fix issue with benchmark help options and descriptions not lining up due to new -aead_set_key added in #8160 on April 14, 2025. 2025-07-02 14:58:11 -07:00
Anthony Hu f7ea8fca67 Explicitly declare dilithium_coeff_eta2[] as signed 2025-07-02 15:50:51 -04:00
Daniel Pouzzner dc05c4c01b wolfcrypt/src/random.c and wolfssl/wolfcrypt/random.h: refactor DRBG_internal.reseedCtr as a word64 if WORD64_AVAILABLE, to accommodate max reseed count per NIST SP 800-90A Rev. 1;
wolfssl/wolfcrypt/settings.h: if WOLFSSL_LINUXKM && LINUXKM_LKCAPI_REGISTER && WORD64_AVAILABLE, set default WC_RESEED_INTERVAL to max allowed (2^48);

linuxkm/lkcapi_sha_glue.c: handle NO_LINUXKM_DRBG_GET_RANDOM_BYTES, for build-time override control of LINUXKM_DRBG_GET_RANDOM_BYTES, and handle WOLFSSL_LINUXKM_USE_GET_RANDOM_USER_KRETPROBE, for separate opt-in control of the buggy wc_get_random_bytes_user_kretprobe_enter().
2025-07-02 14:25:05 -05:00
Daniel Pouzzner b3944a73c2 linuxkm/lkcapi_sha_glue.c:
* implement interception of _get_random_bytes() and get_random_bytes_user() (implicitly intercepts /dev/random and /dev/urandom):

    * get_crypto_default_rng()
    * get_default_drbg_ctx()
    * wc__get_random_bytes()
    * wc_get_random_bytes_user()
    * wc_extract_crng_user()
    * wc_mix_pool_bytes()
    * wc_crng_reseed()
    * wc_get_random_bytes_by_kprobe()
    * wc_get_random_bytes_user_kretprobe_enter()
    * wc_get_random_bytes_user_kretprobe_exit()

    * add LINUXKM_DRBG_GET_RANDOM_BYTES sections to wc_linuxkm_drbg_startup() and wc_linuxkm_drbg_cleanup()

    * add linuxkm/patches/*/WOLFSSL_LINUXKM_HAVE_GET_RANDOM_CALLBACKS-*.patch, initially for versions:
      * 5.10.17
      * 5.10.236
      * 5.15
      * 5.17
      * 6.1.73
      * 6.12
      * 6.15

    * remove "*.patch" from .gitignore.

    * add linuxkm/patches/regen-patches.sh.

  * in wc_linuxkm_drbg_ctx_clear(), check lock count before freeing.

  * in get_drbg() and put_drbg(), use migrate_disable(), not DISABLE_VECTOR_REGISTERS().

  * in wc_linuxkm_drbg_generate(), explicitly DISABLE_VECTOR_REGISTERS() for the crypto_default_rng.

  * in wc_linuxkm_drbg_generate(), add DRBG reinitialization code to handle RNG_FAILURE_E.  This handles the situation where a DRBG was instantiated in a vector-ops-allowed context, caching a vectorized SHA256 ethod, but later used in a no-vector-ops-allowed context.

  * in wc_linuxkm_drbg_seed(), add DISABLE_VECTOR_REGISTERS() wrapper around wc_RNG_DRBG_Reseed() for crypto_default_rng.

linuxkm/x86_vector_register_glue.c:

  * add crash recovery logic to wc_linuxkm_fpu_state_assoc_unlikely()

  * in wc_linuxkm_fpu_state_assoc(), when wc_linuxkm_fpu_states is null, don't call wc_linuxkm_fpu_state_assoc_unlikely() if !assume_fpu_began.

  * in can_save_vector_registers_x86(), save_vector_registers_x86(), and restore_vector_registers_x86(), check for hard interrupt context first, to return early failure if current->pid is unusable.

  * in save_vector_registers_x86(), tweak logic around WC_FPU_INHIBITED_FLAG, adding local_bh_disable()...local_bh_enable() to provide for safe recursion.

wolfcrypt/src/random.c: optimization: in Hash_df(), for WOLFSSL_LINUXKM, don't put digest[WC_SHA256_DIGEST_SIZE] in the heap, keep it on the stack.

wolfssl/wolfcrypt/types.h: add WOLFSSL_NO_ASM no-op definitions for DISABLE_VECTOR_REGISTERS() and REENABLE_VECTOR_REGISTERS().

configure.ac:

* move --enable-linuxkm and --enable-linuxkm-defaults initial detection early, so that HMAC_COPY_DEFAULT picks it up.

* add ENABLED_ENTROPY_MEMUSE_DEFAULT, and enable it by default when ENABLED_LINUXKM_DEFAULTS.

* update linuxkm-lkcapi-register help message.

linuxkm/linuxkm_wc_port.h:

* add my_kallsyms_lookup_name().

* add preempt_count, _raw_spin_lock_irqsave, _raw_spin_trylock, _raw_spin_unlock_irqrestore, and _cond_resched, to wolfssl_linuxkm_pie_redirect_table, and add spin_unlock_irqrestore() macro to mask native inline.

* move linuxkm mutex wrappers from wolfcrypt/src/wc_port.c to linuxkm_wc_port.h, make them inlines, and add new default spinlock-based implementation, with old method now gated on WOLFSSL_LINUXKM_USE_MUTEXES.

* change malloc() and realloc() wrappers from GFP_KERNEL to GFP_ATOMIC.

linuxkm/lkcapi_glue.c: make misc.h/misc.c inclusion unconditional, and trim now-redundant inclusions out of lkcapi_dh_glue.c and lkcapi_ecdh_glue.c.
2025-07-02 14:25:05 -05:00
Daniel Pouzzner b25d484a4e linuxkm/lkcapi_sha_glue.c: implement mutex-free sync mechanism for wc_linuxkm_drbg_ctx in new get_drbg(), get_drbg_n(), and put_drbg();
linuxkm/x86_vector_register_glue.c: implement support for WC_FPU_INHIBITED_FLAG, and an `int inhibit_p` argument to save_vector_registers_x86();

wolfcrypt/src/random.c: implement linuxkm support for RDSEED and HAVE_ENTROPY_MEMUSE;

wolfssl/wolfcrypt/error-crypt.h and wolfcrypt/src/error.c: add WC_ACCEL_INHIBIT_E "Crypto acceleration is currently inhibited";

linuxkm/module_hooks.c and linuxkm/x86_vector_register_glue.c: remove broken and bit-rotten WOLFSSL_COMMERCIAL_LICENSE and LINUXKM_FPU_STATES_FOLLOW_THREADS code paths.
2025-07-02 14:25:05 -05:00
David Garske 221330df0b Added introspection for Intel and ARM assembly speedups (useful for benchmarking output). Added STM32F439ZI benchmarks.
`--enable-armasm` : `Assembly Speedups: ARMASM ALIGN`
`--enable-intelasm` : `Assembly Speedups: INTELASM ALIGN X86_64_BUILD`
2025-07-02 10:57:24 -07:00
jordan 9e811b5bd5 wolfcrypt misc: avoid frivolous initialization. 2025-07-02 10:46:38 -05:00
jordan 9ac480a60d linuxkm fedora: fix uninitialized build errors. 2025-07-02 10:00:28 -05:00
Daniel Pouzzner a8fc68d81b wolfcrypt/src/random.c: in Hash_DRBG_Generate(), gate the verbose reseed message on DEBUG_WOLFSSL or DEBUG_DRBG_RESEEDS, use WOLFSSL_MSG_EX(), and refactor the condition from drbg->reseedCtr == RESEED_INTERVAL to drbg->reseedCtr >= WC_RESEED_INTERVAL.
also some unrelated cleanup in .wolfssl_known_macro_extras.
2025-07-01 13:05:00 -05:00
JacobBarthelmeh 8fa0f6b3df Merge pull request #8944 from SparkiDev/evp_hmac_copy_hash_fix
EVP HMAC: get working with WOLFSSL_HMAC_COPY_HASH
2025-07-01 09:50:53 -06:00
Sean Parkinson 7c4de54e73 EVP HMAC: get working with WOLFSSL_HMAC_COPY_HASH
Get the EVP layer working with the wolfSSL HMAC implementation when
WOLFSSL_HMAC_COPY_HASH is defined.
This define hashes the ipad and opad into temporary hashes and copies
the required hash into the working hash when needed. Uses more memory
but is faster when starting a new hash with the same key.
2025-07-01 13:14:26 +10:00
Sean Parkinson 574de4b234 Memory allocation failure testing fixes
Fixes for test code to cleanup on failure properly.
pkcs7.c: when streaming, free the decrypting content when adding data to
the stream fails.
2025-07-01 11:50:42 +10:00
Daniel Pouzzner 018ee9754f Merge pull request #8608 from anhu/2akid
Check for duplicate extensions in a CRL
2025-06-27 22:25:27 -05:00
Daniel Pouzzner d1c1bca9e4 Merge pull request #8914 from dgarske/stm32n6
Added support for STM32N6
2025-06-27 22:19:01 -05:00
Daniel Pouzzner 4421f8bd84 Merge pull request #8934 from dgarske/macos_typo
Fix minor code typos for macos signal and types.h max block size
2025-06-27 14:11:31 -05:00
David Garske 9a2c8840e2 Merge pull request #8933 from SparkiDev/armv7a_asm_branch_fix
ARMv7a ASM: fix branch instruction
2025-06-27 08:45:52 -07:00
David Garske 295d90655b Merge pull request #8929 from SparkiDev/regression_fixes_18
Regression testing
2025-06-27 08:44:43 -07:00
David Garske 20a2ec0fc1 Fix typo from MacOS signal fix improvement in PR #8928. 2025-06-27 07:43:55 -07:00
Sean Parkinson edacf0434c Merge pull request #8869 from rlm2002/coverityTests
Coverity: parameter checking and NULL assignment
2025-06-27 09:49:21 +10:00
Sean Parkinson a7430b3f70 Merge pull request #8930 from kojiws/check_shift_counts
Clarify the len range on SetShortInt()
2025-06-27 09:46:42 +10:00
Sean Parkinson f713882c54 Merge pull request #8928 from dgarske/macos
Implement proper MacOS dispatch for conditional signal/wait
2025-06-27 09:35:30 +10:00
Sean Parkinson 81e1eb4600 ARMv7a ASM: fix branch instruction
Branch instructions got changed for 64-bit to be B.<cond>.
32-bit must be B<cond>.
Return them to this form.
2025-06-27 09:26:28 +10:00
Anthony Hu a0cd18daea Add back a removed comment and give RFC reference. 2025-06-26 16:08:21 -04:00
Ruby Martin 8ab08f7b17 check length in wc_oid_sum()
add MAX_OID_SZ to known macro extras
2025-06-26 09:20:26 -06:00
Sean Parkinson f1cb4d579c Regression testing
Fixes to get WOLFSSL_PUBLIC_MP testing passing.
Fix DH constant time agreement:
  - implement constant time encoding to big-endian byte array in TFM
- only force x to be zero for SP math as others implementations ensure
unused words are zero
- exponentiate in constant time to the smallest number of words
possible
- no need to encode into separate buffer anymore as encoding is
constant time and front padded
- make requested_sz be the maximum size for the parameters and check
against agreeSz
- update agreeSz to be the maximum valid size instead of filling all
the buffer which may be many times too big
- fix SP result to front pad when doing constant time
2025-06-26 21:21:05 +10:00
Koji Takeda b734c47cc9 Check the len range stricter 2025-06-26 17:48:52 +09:00
Daniel Pouzzner 6fb1c54c29 Merge pull request #8854 from dgarske/renesas_rx_tsip_aesctr
Added Renesas RX TSIP AES CTR support
2025-06-25 22:20:03 -05:00
Daniel Pouzzner d6d124bb85 Merge pull request #8774 from SparkiDev/armv8_ghs
Armv8 (Aarch64) ASM fixes for Green Hills compiler
2025-06-25 21:46:48 -05:00
Daniel Pouzzner 29f534f3b0 Merge pull request #8836 from SparkiDev/lms_serialize_state
LMS: Allow state to be saved with private key
2025-06-25 21:34:42 -05:00
David Garske 6b7fe091bf Implement proper MacOS dispatch for conditional signal/wait. Note: this logic was pulled from wolfMQTT and is well established. 2025-06-25 17:14:12 -07:00