toddouska
|
a4fa4d5253
|
Merge pull request #431 from JacobBarthelmeh/master
sanity checks on wolfSSL_dtls_get_peer arguments
|
2016-05-26 14:46:58 -07:00 |
|
dgarske
|
806a0bbaa7
|
Merge pull request #430 from toddouska/sesstick_type_typo
fix dynamic type session ticket typos
|
2016-05-26 11:35:46 -07:00 |
|
Jacob Barthelmeh
|
5e1f06ce52
|
sanity checks on wolfSSL_dtls_get_peer arguments
|
2016-05-25 10:57:00 -06:00 |
|
toddouska
|
669cdfc357
|
fix dynamic type session ticket typos
|
2016-05-25 09:47:54 -07:00 |
|
John Safranek
|
b3d068dc40
|
Merge pull request #418 from toddouska/dyntickets
Dyntickets
|
2016-05-24 11:00:14 -07:00 |
|
toddouska
|
9ab5ec4a36
|
Merge pull request #428 from dgarske/eccmathfixes
Fixed bug with "ecc_projective_add_point"
|
2016-05-23 11:46:54 -07:00 |
|
toddouska
|
61f0af5cf3
|
Merge pull request #429 from dgarske/wolfcryptinit
Added missing wolfCrypt_Init() to wolfCrypt test application
|
2016-05-23 09:23:42 -07:00 |
|
David Garske
|
5eb7768d00
|
Added missing wolfCrypt_Init() to wolfCrypt tests.
|
2016-05-22 18:18:06 -07:00 |
|
David Garske
|
15cd06b7ca
|
Fixed bug with "ecc_projective_add_point" that was not protecting the case where Q == R and using the R variable directly for the result. Now if Q == R then swap P and Q and using a local x,y,z, is only required when used with ALT_ECC_SIZE.
|
2016-05-22 17:31:36 -07:00 |
|
JacobBarthelmeh
|
eb6153f028
|
Merge pull request #425 from shihrer/master
Fixed scripts/benchmark.test, replaced == with -eq
|
2016-05-20 14:49:43 -06:00 |
|
JacobBarthelmeh
|
f369f8a434
|
Merge pull request #426 from jrblixt/master
Changes to date
|
2016-05-20 14:48:04 -06:00 |
|
John Blixt
|
f1860d6263
|
Changes to date
|
2016-05-20 09:48:03 -06:00 |
|
Michael Shihrer
|
fa94f5ffe9
|
Replaced == with -eq to fix benchmark.test scripting error
|
2016-05-20 09:43:30 -06:00 |
|
toddouska
|
504081e602
|
Merge pull request #421 from ejohnstown/limit-country-code
Limit Country Name size
|
2016-05-19 16:05:15 -07:00 |
|
toddouska
|
1bb5c48080
|
Merge pull request #423 from dgarske/bench-ecc-encrypt-decrypt
Added a benchmark for ECC encrypt/decrypt
|
2016-05-19 14:47:03 -07:00 |
|
David Garske
|
c1507957c4
|
Added a benchmark for ECC encrypt/decrypt when enabled via "HAVE_ECC_ENCRYPT" or "--enable-eccencrypt --enable-hkdf".
|
2016-05-19 11:42:00 -07:00 |
|
John Safranek
|
bae0fe9b63
|
MakeCertReq() was not checking return codes correctly for the SetFoo()
functions.
|
2016-05-18 15:14:23 -07:00 |
|
John Safranek
|
5c8daa0ac6
|
1. SetName() should return error if country code isn't 2 bytes.
2. MakeCert() was not checking return codes correctly for the SetFoo()
functions.
3. Added error code for invalid country code length.
|
2016-05-18 15:04:40 -07:00 |
|
John Safranek
|
03e6f7cca3
|
RFC 5280 Appendix A.1 states that the Country Name in a certificate
shall have a size of 2 octets. Restrict country name length to 2 or 0.
|
2016-05-18 10:39:18 -07:00 |
|
toddouska
|
b8c0802e3c
|
Merge pull request #414 from JacobBarthelmeh/DTLS-MultiCore
Dtls multi core
|
2016-05-17 17:39:18 -07:00 |
|
Jacob Barthelmeh
|
1b278edfd0
|
fix unused functions, make WOLFSSL first parameter, add comments
|
2016-05-16 23:48:26 -06:00 |
|
Jacob Barthelmeh
|
c8576566cc
|
add public wolfSSL_dtls_export and api tests
|
2016-05-14 12:49:09 -06:00 |
|
toddouska
|
185e60ad19
|
based on pr#316; turns isDynamic ticket off on some failure cases, move an XFREE out of a Lock that isn't required
|
2016-05-13 14:11:46 -07:00 |
|
toddouska
|
1ea01a7ad8
|
Merge branch 'master' of https://github.com/NickolasLapp/wolfssl into dyntickets
|
2016-05-13 09:07:55 -07:00 |
|
Jacob Barthelmeh
|
3897f78073
|
truncated hmac export and sanity checks
|
2016-05-13 09:11:38 -06:00 |
|
toddouska
|
67ab30088b
|
Merge pull request #416 from dgarske/mathmemleakfix
Fix mem leak with fast math disabled due to recent ecc_projective_add_point changes
|
2016-05-12 15:23:19 -07:00 |
|
David Garske
|
2ad9d41641
|
Performance improvement for fast math mp_clear to use fp_zero (memset) instead of fp_clear(ForceZero). Added new mp_forcezero function for clearing/free'ing sensitive private key data. Changed ECC and RSA to use mp_forcezero to explicitly handle private key clearing.
|
2016-05-12 11:50:34 -07:00 |
|
Jacob Barthelmeh
|
8d66ba2923
|
refactor export/import of cipher specs struct
|
2016-05-12 10:06:15 -06:00 |
|
Jacob Barthelmeh
|
fbf39a761a
|
refactor export/import of key struct
|
2016-05-12 09:23:38 -06:00 |
|
Jacob Barthelmeh
|
f9ce2f2677
|
verify cipher suite is valid on import
|
2016-05-12 09:09:07 -06:00 |
|
John Safranek
|
39833768cf
|
Merge pull request #417 from toddouska/stdio
make sure stdio.h can be swtiched off if not available
|
2016-05-11 14:55:05 -07:00 |
|
toddouska
|
cc16a543dd
|
make sure stdio.h can be swtiched off if not available
|
2016-05-11 10:11:25 -07:00 |
|
David Garske
|
440956f8d4
|
Fixed new issue with ecc_projective_add_point not free'ing the local x, y, z if fast math was disabled. Formatting cleanup in integer.c.
|
2016-05-11 08:47:30 -07:00 |
|
toddouska
|
3bff6f10e3
|
Merge pull request #411 from dgarske/mathfixes
Math fixes
|
2016-05-10 14:45:20 -07:00 |
|
Jacob Barthelmeh
|
8c45cb1938
|
add DTLS session export option
|
2016-05-10 13:27:45 -06:00 |
|
David Garske
|
2fb4506922
|
iFixes to FP_MAX_BITS_ECC calculation. Alignment check against digit_bits is based on max ecc bits times two. If alignment check fails we add a digit_bit to make sure we have enough room.
|
2016-05-10 12:20:39 -07:00 |
|
David Garske
|
d71d0f2cb4
|
Fix with fast math disabled so ecc_projective_add_point uses temp local variable for x,y,z result.
|
2016-05-09 13:29:25 -07:00 |
|
David Garske
|
8f6352725a
|
Fixed math for FP_MAX_BITS_ECC calculations. Error in alignment check. Altered non-aligned formula to be (max bits * 2) + digit, then 8-bit aligned. Cleanup of the example user_settings.h.
|
2016-05-09 10:34:37 -07:00 |
|
David Garske
|
8c9b8a596a
|
Fixed calculation of max ECC bits with ALT_ECC_SIZE defined so it only allocates what is required. For 8-bit aligned curve sizes its double the max bits. For un-aligned curves sized, like ECC521, its 521 8-bit aligned, doubled, plus digit bit.
|
2016-05-09 09:51:11 -07:00 |
|
David Garske
|
44b1f98b39
|
Fixed issue with ALT_ECC_SIZE and default value for FP_MAX_BITS_ECC so its based on max enabled ECC curve bits.
|
2016-05-09 09:51:11 -07:00 |
|
David Garske
|
880b2e454b
|
Refactor of the ecc_projective_add_point and ecc_projective_dbl_point functions to eliminate duplicate versions. Modified new single functions to work with normal, fast and alt_ecc_size math options. Careful use of mp_clear to retain original performance.
|
2016-05-09 09:51:11 -07:00 |
|
David Garske
|
fe58db2a07
|
Fixed typo with new "eccshamir" configure option.
|
2016-05-09 09:51:11 -07:00 |
|
David Garske
|
1b602d783c
|
Fast math correction of "sizeof" to use (). Updates to tfm and ecc comments.
|
2016-05-09 09:51:11 -07:00 |
|
David Garske
|
5cbc4bdf29
|
Added new "WOLFSSL_DEBUG_MATH", which enables use of "mp_dump" to display information about an mp_int.
|
2016-05-09 09:51:11 -07:00 |
|
David Garske
|
a4782fcf01
|
Fix in fast math version of ecc_projective_dbl_point to use a local for x,y,z since ecc_point fp_int's are reduced size and cause math issues with ALT_ECC_SIZE enabled. Added local stack variable cleanups for ecc_projective_add_point.
|
2016-05-09 09:51:11 -07:00 |
|
David Garske
|
0ddbe0e60e
|
Enhancement to RSA math function "_fp_exptmod" (non timing resistant version) to support WOLFSSL_SMALL_STACK, which moves the allocation of the 64 fp_int's from the stack to the heap.
|
2016-05-09 09:51:11 -07:00 |
|
David Garske
|
fa5dd01001
|
Fixes/improvements to the wolfCrypt ECC tests. Fixed bug with sharedA/sharedB being too small when BENCH_EMBEDDED is used and curve size over 256 bit. Added error message for ECC test failures, to show the curve size used. Fix to wc_ecc_verify_hash test to use digest that is not all zeros as that doesn't work correctly for non-Shamir ECC math. Changed return code for wc_ecc_check_check so its unique.
|
2016-05-09 09:51:11 -07:00 |
|
David Garske
|
a5d27853fa
|
Fixes to fp_mul and fp_div to clear any excess digits on the destination. Added compile-time check to confirm FP_SIZE is compatible with TFM_ acceleration defines enabled. Updated comments in other places where excess digits are cleared.
|
2016-05-09 09:51:11 -07:00 |
|
David Garske
|
7c3fbd7644
|
Fix for fp_copy() when used with ALT_ECC_SIZE so any excess digits on the destination that we didn't write to are set to zero.
|
2016-05-09 09:51:11 -07:00 |
|
David Garske
|
9001036e09
|
Fixes memory leak in the wc_RsaFunction if failure happens when using normal math (not fast math) and RSA_LOW_MEM is not defined.
|
2016-05-09 09:51:11 -07:00 |
|