|
9d0bdc7e
|
2019-12-04T22:23:24
|
|
run the test always
|
|
1e65c0bf
|
2019-10-11T00:06:45
|
|
Introduction of a fast but slightly over-estimating radix_size
|
|
5f3063fb
|
2019-11-22T03:14:59
|
|
correction of typo in test.c
|
|
642032dd
|
2019-11-22T02:56:48
|
|
wrong sign in mp_div_recursive
|
|
bb1e56e7
|
2019-11-12T16:20:07
|
|
Fix compilation of demo/test.c using C++
|
|
9f2cd043
|
2019-11-12T15:30:34
|
|
more unnecessary type-casts, correct type-case for malloc
|
|
e9d54e98
|
2019-11-12T14:44:39
|
|
remove unnecessary type-casts
|
|
a91eb5a0
|
2019-11-12T14:29:02
|
|
fix build (hopefully)
|
|
9cfb6c32
|
2019-11-06T11:52:42
|
|
use type-cast in stead of UINT64_C
|
|
02aa95c2
|
2019-11-06T11:29:17
|
|
Fix wrong use of uLL suffix
|
|
f6a7bedb
|
2019-10-29T20:52:29
|
|
suffix _u32 -> _n of mp_(expt|log|root) functions, use int for now
|
|
4f00e75b
|
2019-11-06T16:51:51
|
|
make mp_div_3 private
|
|
e8e65119
|
2019-11-05T19:18:32
|
|
add COMPILE_LTO to test link time optimization
the library performs best with lto since many small functions
can be inlined across objects.
|
|
bbb17808
|
2019-10-30T14:19:50
|
|
mp_radix off-by-one error and other related code-cleanup
|
|
40342807
|
2019-11-01T20:08:41
|
|
explicit operator precedence
|
|
146becbd
|
2019-10-31T19:26:55
|
|
literal suffix
|
|
5c335f84
|
2019-10-30T17:26:27
|
|
rename mul/sqr functions for consistency, comba instead of fast suffix
|
|
b9977adf
|
2019-10-29T08:44:51
|
|
use uint8_t instead of unsigned char
|
|
af376edc
|
2019-10-29T16:37:56
|
|
no more MP_WUR in timing.c
|
|
4500d066
|
2019-10-29T14:41:33
|
|
also no MP_WUR in mtest_opponent()
|
|
b250ec44
|
2019-10-29T13:24:34
|
|
clean-up test.c
* no more `MP_WUR` in test.c
* clean-up console output
|
|
04ee1e75
|
2019-10-28T15:07:58
|
|
improve demo/timing a bit
* less verbose output on console
* allow changing name for logs
* pre-heat caches before starting the timing
|
|
bf9507a9
|
2019-10-24T22:02:29
|
|
replace mp_bool by stdbool
* This gives the advantage that static analysis **understands** bool,
but complains about using an enum type instead of bool.
* If stdbool.h is not desired, true/false/bool can be replaced using sed
as in the no-stdint-h branch.
* We already include stdint.h and stdbool.h is not more harmful than
this header
|
|
27e142bc
|
2019-10-24T17:52:03
|
|
remove unnecessary == MP_YES/MP_NO comparisons
|
|
14642642
|
2019-10-22T18:01:08
|
|
add prefix to cutoff variables
|
|
3180c66c
|
2019-10-23T18:17:20
|
|
also use MP_YES/NO in tests
|
|
9edd185f
|
2019-10-04T17:41:09
|
|
Addition of fast division (recursive divrem only)
|
|
a29aa59b
|
2019-10-21T19:12:22
|
|
use size_t for mp_radix_size
|
|
485be9de
|
2019-10-20T17:07:43
|
|
explicit condition
|
|
3995ece5
|
2019-10-20T17:05:55
|
|
const parameter
|
|
315975db
|
2019-10-20T16:58:40
|
|
remove useless initialization
|
|
7f05df0a
|
2019-10-20T17:01:14
|
|
const parameter
|
|
7a68f128
|
2019-10-19T16:24:39
|
|
Execute move.sh - Rename files from bn_* to match the function names.
* git blame <renamed-file> is not affected
* git log --follow <renamed-file> can be used to show log across renames
|
|
36fca2f9
|
2019-09-09T02:58:18
|
|
remove support for 8-bit (MP_8BIT)
|
|
d4f6b43f
|
2019-10-11T00:29:20
|
|
use of mp_ilogb in mp_radix_size
|
|
45a3bf76
|
2019-10-16T09:21:19
|
|
remove deprecated functions
|
|
cb1b2dc8
|
2019-10-15T14:04:32
|
|
mp_log_u32: return uint32_t
|
|
c9d41fba
|
2019-10-15T10:40:40
|
|
rename mp_ilogb() to mp_log_u32()
|
|
1cc02f2d
|
2019-10-14T15:09:42
|
|
fix build of timing.c
|
|
b4f7d8d4
|
2019-10-09T20:41:17
|
|
test: use malloc/free
tests are library consumers, MP_MALLOC is an internal macro
|
|
6e3dae56
|
2019-10-08T21:47:57
|
|
add MP_BUF error
|
|
b6813a41
|
2019-09-27T03:40:08
|
|
replace mp_export/import by mp_pack/unpack
|
|
71d1b7b9
|
2019-09-08T23:29:44
|
|
make mp_to_radix return the count of characters of the converted number
|
|
d68e2877
|
2019-10-02T05:47:04
|
|
bugfix for an edgecase
|
|
abdb0334
|
2019-09-25T00:29:19
|
|
Refactored functions to read and write binaries and added "maxlen"
|
|
4ebdaca2
|
2019-09-05T14:24:42
|
|
"fix" warnings
|
|
baff51e9
|
2019-09-05T14:23:22
|
|
allow to run specific timing tests
|
|
47ab9c60
|
2019-09-05T14:19:17
|
|
fix printf
|
|
f8d64c64
|
2019-09-10T16:05:27
|
|
create `mtest_opponent` instead of alt version of `test`
|
|
234d474f
|
2019-09-07T15:01:46
|
|
only run tests if the module is available
|
|
75c52a89
|
2019-09-05T17:12:05
|
|
be nice and tell when it's hex output
did you know that 0x35 is equal to 53? ... ;-)
|
|
54661db6
|
2019-09-05T17:10:48
|
|
allow RSA superclass to be built with tests
|
|
9eecab7b
|
2019-09-05T16:41:36
|
|
print jenkins seed on startup
I just had a case around where the `mp_prime_rand()` testcase failed
but I couldn't reproduce.
I'm pretty sure that was caused by all my changes and partial rebuilds
but anyways it'll give us in the future the possibility to debug such
a case.
|
|
72f92519
|
2019-09-03T10:59:32
|
|
fix last occurrences of `mp_toradix()`
|
|
8bb11ded
|
2019-09-02T21:05:04
|
|
included tests for mp_prime_next_prime
|
|
6cd570d0
|
2019-09-02T17:52:36
|
|
use `mp_to_radix` etc.
|
|
8c1b296e
|
2019-04-08T23:48:39
|
|
add feature detection macro MP_HAS
|
|
ca89e9c0
|
2019-07-03T13:10:05
|
|
rename mp_root/mp_expt to mp_*_u32
|
|
71696549
|
2019-05-29T20:36:18
|
|
deprecate mp_expt_d and mp_n_root in favor of mp_expt and mp_root
|
|
0db64fc8
|
2019-07-03T14:45:14
|
|
Rename bn_get_magxx -> bn_get_mag_uxx. Documentation updated too
|
|
ea3353b6
|
2019-06-13T15:44:00
|
|
Simplify test-cases, by using mp_set_l. Also fix use of types in 2 other files
|
|
5302b907
|
2019-06-06T21:06:06
|
|
use MP_SIZEOF_BITS
|
|
dfbb5ecd
|
2019-06-06T16:22:54
|
|
MSVC compile: higher warning-level -Wall, but warning-free
|
|
c7314fa4
|
2019-05-25T06:42:01
|
|
deprecate mp_n_root_ex and mp_expt_d_ex
These functions were introduced to give some timing guarantees.
However the guarantees are too weak to be useful.
The functions seem to be unused essentially by downstream users.
|
|
0e1a0e65
|
2019-05-24T11:27:03
|
|
also test mp_shrink()
|
|
bcec605a
|
2019-05-22T15:34:49
|
|
deprecate mp_prime_is_divisible and ltm_prime_tab
* it is an implementation detail used for prime testing
* there is upcoming work by @czurnieden regarding a generalised prime sieve
* furthermore remove jacobi test (replaced by kronecker)
|
|
0a9f5e9b
|
2019-05-24T10:21:54
|
|
use u32 i64 suffix
|
|
d7e1cb4a
|
2019-05-24T00:15:11
|
|
another proposal: use _ul and _l for the long macros
|
|
b58c9966
|
2019-05-23T23:39:47
|
|
use int32 and uint32 instead of sint and uint
|
|
3289c959
|
2019-05-20T10:41:06
|
|
introduce various mp_set_sint/mp_set_uint functions with precise types
|
|
1af0de1f
|
2019-05-20T19:11:25
|
|
deprecate mp_tc_(and|or|xor) in favor of mp_(and|or|xor)
* same behavior for positive numbers
* generalisation for negative numbers, treating them as two complement
* improve algorithm, iterate once over the digits, manually perform two complement
* simplify mp_add_d, mp_sub_d
* functions are safe in case of a==c or b==c
* renamed mp_tc_div_2d to mp_signed_rsh (signed right shift)
|
|
0b840b78
|
2019-05-17T20:53:24
|
|
make mp_word private
mp_word is an internal type and it is problematic if it is exposed
in the public api. See for example #216 - MSVC does not support 128 bit
mp_words. But it is perfectly ok to use those internally in the library,
as long as the library is compiled with GCC.
|
|
03ca7bcd
|
2019-05-20T23:33:35
|
|
added tests for Karatsuba and Toom-Cook 3-way
|
|
51cda5b8
|
2019-05-19T09:55:14
|
|
use correct amount of MR trials in tests
|
|
cb3ff8c2
|
2019-05-19T09:54:25
|
|
improve coverage
|
|
150c47cc
|
2019-05-19T16:36:53
|
|
refactor literal suffix with u lowercase
|
|
f55f0a9e
|
2019-05-18T09:28:33
|
|
fix some usages of enum types
|
|
56e7a401
|
2019-05-13T02:46:11
|
|
deprecate mp_jacobi
|
|
198418ba
|
2019-05-14T20:39:33
|
|
bugfix: don't revert to platform rand source, but to jenkins source!
|
|
4534056c
|
2019-05-13T00:22:18
|
|
use enums mp_err, mp_ord, mp_bool, mp_sign
* MP_USE_ENUMS enables enums
* Wc++-compat catches some implicit conversions if MP_USE_ENUMS is defined
* 100% backwards compatible API/ABI if MP_USE_ENUMS is not defined
|
|
f85bd31e
|
2019-05-12T17:06:40
|
|
fix undefined behavior in labs
|
|
2a2e2716
|
2019-05-12T11:33:17
|
|
move jenkins prng to bn_s_mp_rand_jenkins.c
|
|
8d71a9e3
|
2019-05-11T00:19:28
|
|
travis: add -fsanitize=undefined
|
|
7365442a
|
2019-05-10T23:59:46
|
|
No grow necessary in mp_set_int* functions
* mp_set_int* always return MP_OKAY
* remove return checks for mp_set_int*
* introduce MP_MIN_PREC
|
|
adf9605d
|
2019-05-09T03:28:49
|
|
add warn_unused_result, found one missing check!
|
|
0669e92e
|
2019-05-08T20:18:53
|
|
deprecate mp_prime_random_ex in favor of mp_prime_rand
We can use the internal mp_rand generator now, since it generates
numbers which are safe for crypto purposes.
|
|
1a12486d
|
2019-05-08T21:08:16
|
|
add mp_rand test
|
|
eedf1624
|
2019-03-07T15:01:45
|
|
allow filtering of tests executed
|
|
44d03a6f
|
2019-05-07T09:45:38
|
|
deprecate LTM_PRIME_* macros in favor of MP_PRIME_*
|
|
250751be
|
2019-05-07T09:32:59
|
|
tommath.h: do not expose stdlib.h
|
|
e45f75fd
|
2019-04-13T08:46:57
|
|
deprecate DIGIT_BIT, use MP_DIGIT_BIT
|
|
e8f56cc4
|
2019-04-05T11:37:49
|
|
enable -Wconversion and -Wsign-conversion on travis
* no changes to the library code
* conversion issues in the demo testsuite fixed
* add CONV_WARNINGS and enable the warnings only for clang-7 (for now)
* disable Wsystem-headers if Wconversion is enabled, to avoid warnings from the system headers
|
|
fbfcb661
|
2019-04-12T14:56:29
|
|
apply rename
|
|
35311aea
|
2019-04-03T23:50:57
|
|
added function mp_ilogb
|
|
27ca25cb
|
2019-04-06T18:55:43
|
|
added multiplcation balancing for the Toom-Cook algorithms
|
|
70a30ae7
|
2019-04-07T02:29:28
|
|
remove footers
|
|
984d3ff6
|
2019-04-01T03:41:26
|
|
changed seed to make nth-root usable
|
|
0ed98cf1
|
2019-04-04T12:55:39
|
|
removed tracing
|
|
769218cc
|
2019-04-03T18:59:09
|
|
additional functions mp_incr and mp_decr
|
|
cb1eb161
|
2019-04-04T07:58:07
|
|
run `make astyle`
[skip ci]
|