570a8e1b36
Whitespace check-in
2019-06-09 12:12:46 -04:00
955ac6fe24
Rework SSE2 and AVX2 loads and stores
2019-06-09 04:29:40 -04:00
8c78985de2
Add ModularArithmetic::operator=
2019-06-09 02:56:30 -04:00
c1f4d17e10
Cleanup BERGeneralDecoder constructors
...
For real this time...
2019-06-09 02:03:06 -04:00
55fe6a2191
Cleanup BERGeneralDecoder constructors
2019-06-09 02:00:53 -04:00
8fab1c3677
Revert changes for lgtm findings
...
This broke SunCC to the point of no repair. SunCC is using AVX2 instructions for C++ and SSE2. Man this compiler sucks...
2019-06-09 01:49:44 -04:00
3ce1823fd1
Fix SunCC compile
...
Sun's compiler is mostly braindead.
2019-06-09 00:00:22 -04:00
6a11f00768
Clear lgtm findings
2019-06-08 12:59:14 -04:00
43b01973b1
Clear lgtm findings
...
We did some refactoring and added sse_simd.h. Over time more SSE functions will likely move into sse_simd.h
2019-06-08 11:00:11 -04:00
afffba7b7b
Add -mtune=native option to makefile
...
Added for Solaris 11 .3 on SPARC64
2019-06-07 13:02:25 -04:00
ed4996f652
Cleanup governor.sh script
2019-06-05 23:21:00 -04:00
29a30b74b0
Update head comments
2019-06-05 23:09:58 -04:00
17fa3031b3
Clear lgtm finding in ECP class
2019-06-05 10:52:45 -04:00
d8122cec16
Avoid ARM rev on Aarch64
...
This broke Aarch64
2019-06-04 21:17:13 -04:00
6c009ddf43
Remove dummy operator= in ModularArithmetic
2019-06-04 19:05:33 -04:00
994c98b6c0
Fix GCC compile on AIX
...
In file included from test.cpp:31:0:
validate.h:213:93: error: operator '||' has no right operand
#elif (_POSIX_C_SOURCE >= 1 || _XOPEN_SOURCE || _BSD_SOURCE || _SVID_SOURCE || _POSIX_SOURCE)
2019-06-04 09:45:33 -04:00
3afb1f1099
Fix GCC compile on AIX
...
In file included from test.cpp:31:0:
validate.h:213:93: error: operator '||' has no right operand
#elif (_POSIX_C_SOURCE >= 1 || _XOPEN_SOURCE || _BSD_SOURCE || _SVID_SOURCE || _POSIX_SOURCE)
2019-06-04 09:37:28 -04:00
198b081df5
Add assert to RemainingLength member function
2019-06-04 09:29:01 -04:00
238578a808
Clear lgtm findings
...
The lgtm service asks for scoped calls to IsolatedInitialize. The code is a tad bit more readable, so we will clear the findings.
2019-06-04 05:31:46 -04:00
462bcc859d
Update Poly1305 nonce length check
2019-06-04 04:39:19 -04:00
a644008679
Cleanup ASN.1 encoders and decoders (GH #851 )
2019-06-04 02:49:16 -04:00
ab538471b1
Update documentation
2019-06-04 02:46:43 -04:00
f330c0eca8
Clear warning on missing assignment operator
2019-06-03 23:17:15 -04:00
c76711237a
Call base class IsolatedInitialize
2019-06-03 06:45:58 -04:00
3faf716d73
Call base class IsolatedInitialize
2019-06-03 06:44:58 -04:00
0a20141f36
Clear warning for assignment operator in HuffmanNode
2019-06-03 06:32:24 -04:00
54d48ac1f4
Add header guard for serpentp.h
2019-06-03 06:17:58 -04:00
342cdb9589
Clear lgtm warning on unsafe functions
2019-06-03 05:41:58 -04:00
959494871f
Guard use of volatile cast in TEA and XTEA
2019-06-02 05:29:08 -04:00
9538f2d715
Fix compile on iOS
2019-06-01 08:05:39 -04:00
f00b0427aa
Exit setenv script if not sourced
2019-06-01 07:43:32 -04:00
1a0d7c516d
Disable BMI2 code paths in Integer class (GH #850 )
2019-06-01 00:14:19 -04:00
8e27c6b3fa
Fix OS X compile with Clang later than 6.0
...
It looks like Apple Clang 6.0 does not provide _blsr_u32 and friends. Later versions of Clang defines _blsr_u32 and it causes a stream of warnings.
2019-05-31 23:33:17 -04:00
392ec3465e
Fix OS X compile with Clang 6.0
2019-05-31 23:03:24 -04:00
4952fa489d
Use BMI2 when available for MultiplyWordsLoHi, MulAcc and friends
...
Using BMI2 saves about 0.03 ms on a Core i5 6400 @ 2.7 GHz. It is small but measurable. It also gives GCC more freedom in selecting memory or register operands
2019-05-31 09:42:39 -04:00
fb0bef1eb6
Use BMI2 when available for MultiplyWordsLoHi, MulAcc and friends
...
Using BMI2 saves about 0.03 ms on a Core i5 6400 @ 2.7 GHz. It is small but measurable. It also gives GCC more freedom in selecting memory or register operands
2019-05-31 05:37:35 -04:00
aed6e935d6
Update documentation
2019-05-30 23:22:47 -04:00
1f2be58434
Update Doxygen comments
2019-05-30 22:30:22 -04:00
0dd07252d2
Update comments
2019-05-28 20:19:34 -04:00
e40de18538
Update comments
2019-05-28 20:18:58 -04:00
b61da7acfe
Update cryptest-autotools script
2019-05-28 06:36:01 -04:00
a2e7c26f6c
Add 16-bit rev16 and rbit
2019-05-28 02:50:58 -04:00
623059f28c
Use ARM rev and rbit when appropriate
...
We tried to add a ByteReverse(word64) and BitReverse(word64) overloads but GCC was producing bad code for it.
2019-05-27 14:27:12 -04:00
81da61fe7b
Breakout sha_block_data_order and sha_block_data_order_neon (GH #847 )
2019-05-26 22:10:26 -04:00
d4b533a60f
Add Thumb and Arm versions of CPU_ProbeARMv7()
2019-05-26 17:06:55 -04:00
7cd8d0278a
Remove VFP_ABI_PUSH and VFP_ABI_POP macros
2019-05-26 02:14:03 -04:00
9a3c1e351d
Clear Valgrind finding in IncrementCounterByOne
...
The single buffer IncrementCounterByOne generated a Valgrind finding on ARM. This commit uses the same pattern for both overloads in case Valgrind wants to fire on the two-buffer version.
2019-05-25 19:49:49 -04:00
ca11105a40
Clear sign comparison warning under GCC
2019-05-25 17:00:32 -04:00
f90c162b68
Fix IncrementCounterByOne
...
This was introduced earlier in the day when clearing a Valgrind finding. It tested good with the self tests. However, we double process byte[0] if there's a carry.
2019-05-25 07:02:32 -04:00
fc10a7f1ea
Fix SHA512 on ARM benchmarks
...
This was a mistake when porting from Cryptogams to Crypto++. The macros VFP_ABI_PUSH and VFP_ABI_POP needed to be defined because they save and restore SIMD register state. They were originally missing during the port. The benchmarks would hang because the doubles we used for benchmarking were blown away in sha512_block_data_order_neon.
2019-05-25 06:23:19 -04:00
Jeffrey Walton