45139582f3
sha3: optimizations and cleanup
...
* complate reset of the SHA-3 code. Affects mostly the code in sha3.go
* fixes a bug which causes SHAKE implementation to crash
* implementation of Read()/Write() avoid unnecessary buffering as much
as possible
* NOTE: at some point I've done separated implementation for SumXXX,
functions, but after optimizing implementation of Read/Write/Sum, the
gain wasn't that big
Current speed on Initial speed on i7-8665U@1.90
BenchmarkPermutationFunction 1592787 736 ns/op 271.90 MB/s 0 B/op 0 allocs/op
BenchmarkSha3Chunk_x01/SHA-3/224 98752 11630 ns/op 176.02 MB/s 0 B/op 0 allocs/op
BenchmarkSha3Chunk_x01/SHA-3/256 92508 12447 ns/op 164.46 MB/s 0 B/op 0 allocs/op
BenchmarkSha3Chunk_x01/SHA-3/384 76765 15206 ns/op 134.62 MB/s 0 B/op 0 allocs/op
BenchmarkSha3Chunk_x01/SHA-3/512 54333 21932 ns/op 93.33 MB/s 0 B/op 0 allocs/op
BenchmarkSha3Chunk_x16/SHA-3/224 10000 102161 ns/op 160.37 MB/s 0 B/op 0 allocs/op
BenchmarkSha3Chunk_x16/SHA-3/256 10000 106531 ns/op 153.80 MB/s 0 B/op 0 allocs/op
BenchmarkSha3Chunk_x16/SHA-3/384 8641 137272 ns/op 119.35 MB/s 0 B/op 0 allocs/op
BenchmarkSha3Chunk_x16/SHA-3/512 6340 189124 ns/op 86.63 MB/s 0 B/op 0 allocs/op
BenchmarkShake_x01/SHAKE-128 167062 7149 ns/op 188.83 MB/s 0 B/op 0 allocs/op
BenchmarkShake_x01/SHAKE-256 151982 7748 ns/op 174.24 MB/s 0 B/op 0 allocs/op
BenchmarkShake_x16/SHAKE-128 12963 87770 ns/op 186.67 MB/s 0 B/op 0 allocs/op
BenchmarkShake_x16/SHAKE-256 10000 105554 ns/op 155.22 MB/s 0 B/op 0 allocs/op
BenchmarkCShake/cSHAKE-128 109148 10940 ns/op 187.11 MB/s 0 B/op 0 allocs/op
BenchmarkCShake/cSHAKE-256 90324 13211 ns/op 154.94 MB/s 0 B/op 0 allocs/op
PASS
2020-08-29 02:04:26 +01:00
7dcb72bf74
remove shake
2020-08-25 22:39:56 +01:00
ffd7590213
improve comment
...
Initial speed on i7-8665U
> go test -bench=. -test.cpu=1
goos: linux
goarch: amd64
pkg: github.com/henrydcase/nobs/hash/sha3
BenchmarkPermutationFunction 1634836 732 ns/op 273.18 MB/s
BenchmarkSha3_512_MTU 78438 15340 ns/op 88.00 MB/s
BenchmarkSha3_384_MTU 108807 11025 ns/op 122.45 MB/s
BenchmarkSha3_256_MTU 136902 8767 ns/op 153.98 MB/s
BenchmarkSha3_224_MTU 143377 8355 ns/op 161.57 MB/s
BenchmarkShake128_MTU 163569 7108 ns/op 189.94 MB/s
BenchmarkShake256_MTU 156534 7643 ns/op 176.64 MB/s
BenchmarkShake256_16x 10000 112109 ns/op 146.14 MB/s
BenchmarkShake256_1MiB 204 5877014 ns/op 178.42 MB/s
BenchmarkSha3_512_1MiB 100 10967026 ns/op 95.61 MB/s
PASS
ok github.com/henrydcase/nobs/hash/sha3 13.855s
2020-08-25 17:09:40 +01:00
516ea4f5e8
cleanup
2020-08-25 12:32:22 +01:00
68ba33e34f
sha3: remove s390
2020-08-25 12:11:08 +01:00
b56c355c8d
adds cycle count. fixes csidh which provides 128 not 512 bits of security ( #38 )
2020-08-25 11:22:53 +01:00
c30f61923a
adds cycle count. fixes csidh which provides 128 not 512 bits of security
2020-08-25 11:21:11 +01:00
a02b9a77a0
mkem: add csidh
2020-08-25 11:19:07 +01:00
2500d74484
export more symbols from common
2020-05-16 22:37:41 +00:00
a152c09fd5
sike: move common ( #33 )
...
* makes common reusable
* exports some more symbols from common
* remove kem for a moment
2020-05-16 20:14:48 +00:00
55957bbf5e
sike: move common ( #32 )
...
* makes common reusable
* exports some more symbols from common
2020-05-16 18:51:34 +00:00
ab962715d5
Fixes cSIDH key generation when run in the loop
2020-05-14 11:53:23 +00:00
bc32024729
sidh: updates ( #31 )
2020-05-14 08:51:20 +00:00
f5a7daf2bb
sidh: update to p434
2020-05-14 00:02:32 +00:00
91945fde1f
csidh: cosmettic updates
2020-05-13 23:48:43 +00:00
7d891c7eb8
support go 1.14 ( #29 )
...
NOBS now supports Go 1.14 for
* x86-64
* ARM
2020-03-05 11:19:51 +00:00
d0692c81f0
Remove BS from README.md
2020-02-13 10:27:42 +00:00
48ea6a583a
Remove BS from README.md
2020-02-13 10:27:18 +00:00
c5bff4fa11
Remove BS from README.md
2020-02-13 10:25:47 +00:00
2a73461591
remove crapy x448
2020-02-13 10:17:54 +00:00
7efbbf4745
cSIDH-511: ( #26 )
...
Implementation of Commutative Supersingular Isogeny Diffie Hellman,
based on "A faster way to CSIDH" paper (2018/782).
* For fast isogeny calculation, implementation converts a curve from
Montgomery to Edwards. All calculations are done on Edwards curve
and then converted back to Montgomery.
* As multiplication in a field Fp511 is most expensive operation
the implementation contains multiple multiplications. It has
most performant, assembly implementation which uses BMI2 and
ADOX/ADCX instructions for modern CPUs. It also contains
slower implementation which will run on older CPUs
* Benchmarks (Intel SkyLake):
BenchmarkGeneratePrivate 6459 172213 ns/op 0 B/op 0 allocs/op
BenchmarkGenerateKeyPair 25 45800356 ns/op 0 B/op 0 allocs/op
BenchmarkValidate 297 3915983 ns/op 0 B/op 0 allocs/op
BenchmarkValidateRandom 184683 6231 ns/op 0 B/op 0 allocs/op
BenchmarkValidateGenerated 25 48481306 ns/op 0 B/op 0 allocs/op
BenchmarkDerive 19 60928763 ns/op 0 B/op 0 allocs/op
BenchmarkDeriveGenerated 8 137342421 ns/op 0 B/op 0 allocs/op
BenchmarkXMul 2311 494267 ns/op 1 B/op 0 allocs/op
BenchmarkXAdd 2396754 501 ns/op 0 B/op 0 allocs/op
BenchmarkXDbl 2072690 571 ns/op 0 B/op 0 allocs/op
BenchmarkIsom 78004 15171 ns/op 0 B/op 0 allocs/op
BenchmarkFp512Sub 224635152 5.33 ns/op 0 B/op 0 allocs/op
BenchmarkFp512Mul 246633255 4.90 ns/op 0 B/op 0 allocs/op
BenchmarkCSwap 233228547 5.10 ns/op 0 B/op 0 allocs/op
BenchmarkAddRdc 87348240 12.6 ns/op 0 B/op 0 allocs/op
BenchmarkSubRdc 95112787 11.7 ns/op 0 B/op 0 allocs/op
BenchmarkModExpRdc 25436 46878 ns/op 0 B/op 0 allocs/op
BenchmarkMulBmiAsm 19527573 60.1 ns/op 0 B/op 0 allocs/op
BenchmarkMulGeneric 7117650 164 ns/op 0 B/op 0 allocs/op
* Go code has very similar performance when compared to C
implementation.
Results from sidh_torturer (4e2996e12d68364761064341cbe1d1b47efafe23)
github.com:henrydcase/sidh-torture/csidh
| TestName |Go | C |
|------------------|----------|----------|
|TestSharedSecret | 57.95774 | 57.91092 |
|TestKeyGeneration | 62.23614 | 58.12980 |
|TestSharedSecret | 55.28988 | 57.23132 |
|TestKeyGeneration | 61.68745 | 58.66396 |
|TestSharedSecret | 63.19408 | 58.64774 |
|TestKeyGeneration | 62.34022 | 61.62539 |
|TestSharedSecret | 62.85453 | 68.74503 |
|TestKeyGeneration | 52.58518 | 58.40115 |
|TestSharedSecret | 50.77081 | 61.91699 |
|TestKeyGeneration | 59.91843 | 61.09266 |
|TestSharedSecret | 59.97962 | 62.98151 |
|TestKeyGeneration | 64.57525 | 56.22863 |
|TestSharedSecret | 56.40521 | 55.77447 |
|TestKeyGeneration | 67.85850 | 58.52604 |
|TestSharedSecret | 60.54290 | 65.14052 |
|TestKeyGeneration | 65.45766 | 58.42823 |
On average Go implementation is 2% faster.
2019-11-25 15:03:29 +00:00
15f6ee16b9
SHA-3: Fixes crash when cloning Shake state
2019-05-26 17:29:15 +01:00
9b3c0190b0
Updates P34 strategy calculation
2019-05-23 18:32:28 +01:00
7298b650cc
Adds go.mod ( #21 )
...
* Reset Makefile after adding go.mod
* Remove ``build`` directory
* Simiplifies makefile
* shake: Make xorIn copyOut platform specific
2019-05-15 18:03:35 +01:00
49bf0db8fd
SHAKE: Don't use function pointers ( #20 )
...
* xorIn and copyOut function pointers cause input and output data
to be moved to heap. This degrades performance of calling code.
* This change removes usage of those function pointers. We will always
use unaligned implementation as it's faster (but may crash on some
systems)
* Benchmark compares generic vs unaligned xorIn and copyOut
benchmark old ns/op new ns/op delta
BenchmarkPermutationFunction-4 463 815 +76.03%
BenchmarkShake128_MTU-4 4443 8180 +84.11%
BenchmarkShake256_MTU-4 4739 9060 +91.18%
BenchmarkShake256_16x-4 71886 132629 +84.50%
BenchmarkShake256_1MiB-4 3695138 6649012 +79.94%
BenchmarkCShake128_448_16x-4 21210 24611 +16.03%
BenchmarkCShake128_1MiB-4 3009342 3396496 +12.87%
BenchmarkCShake256_448_16x-4 26034 27785 +6.73%
BenchmarkCShake256_1MiB-4 3654713 3829404 +4.78%
2019-05-14 17:08:33 +01:00
e6439f96ab
Adds cSHAKE with 0 alloc interface ( #19 )
2019-05-14 01:19:29 +01:00
6f9706df01
CTR-DRBG: Use hardware acceleration on X86 ( #18 )
...
benchmark old ns/op new ns/op delta
BenchmarkInit-4 3403 397 -88.33%
BenchmarkRead-4 14535 1560 -89.27%
2019-04-09 23:50:21 +01:00
71624cdc4c
Improvements to makefile
2019-04-09 17:30:30 +01:00
b184944242
Nits for SIDH
2019-04-09 17:09:34 +01:00
08f7315b64
DRBG: Speed improvements
...
* CTR-DRBG doesn't call "NewCipher" for block encryption
* Changes API of CTR-DRBG, so that read operation implementes io.Reader
Benchmark results:
----------------------
benchmark old ns/op new ns/op delta
BenchmarkInit-4 1118 3579 +220.13%
BenchmarkRead-4 5343 14589 +173.05%
benchmark old allocs new allocs delta
BenchmarkInit-4 15 0 -100.00%
BenchmarkRead-4 67 0 -100.00%
benchmark old bytes new bytes delta
BenchmarkInit-4 1824 0 -100.00%
BenchmarkRead-4 9488 0 -100.00%
2019-04-09 14:37:59 +01:00
e66cc99401
Improves comment
2019-02-19 14:44:11 +00:00
b47a731959
Run tests on ARM64 ( #11 )
2019-02-16 21:29:20 +00:00
90f8cba329
SIDH: Update ( #9 )
...
* Change license to BSD-3
* SIDH: Multiple developlemnts
2018-12-03 23:07:01 +00:00
ea2ffa2d61
PERF: sidh-p503: Split sub and add into 2 uops instead of 3 ( #8 )
...
The performance improvement comes from the fact that on Skylake
"add mem, reg" splits into 2 uops - one arithmetic uop and another one
for loading a value from mem.
However, changing operand order to "add reg, mem" splits into 3 uops:
one for arithmetic op, one for load and one additional one for storing
the result back.
Using separated instruction for loading/storing helps to parallelize
execution (load/store and arithmetic instruction is done in parallel
if possible)
For details, see: https://www.agner.org/optimize/instruction_tables.pdf
New: BenchmarkFp503StrongReduce-4 300000000 5.57 ns/op
Old: BenchmarkFp503StrongReduce-4 200000000 8.60 ns/op
This just improves one function, but more functions can be improved
2018-11-18 20:57:29 +00:00
e9ddb6fb45
sidh/csidh: use SEE for performing CSWAP ( #6 )
...
* Makefile
* makefile: tools for profiling
* sidh: use SIMD for performing CSWAP
Loads data into 128-bit XMM registers and performs conditional swap.
This is probably less useful for SIDH, but will be useful for cSIDH
2018-10-29 15:41:09 +00:00
a456dc4dd9
readme: License
2018-10-25 15:22:28 +01:00
ae57368c7b
License BS for sha3
2018-10-25 15:22:28 +01:00
00c16fe97e
License bulshit
2018-10-25 15:22:28 +01:00
65bbafeef5
script used for calculating sliding window startegy in SIDH P34
2018-10-25 15:22:28 +01:00
0531c3479b
Update README.md
2018-10-25 15:22:28 +01:00
1e34845d00
complate rewrite for SIDH and SIKE. adds p503 ( #5 )
2018-10-25 15:22:28 +01:00
d6fc82531f
Doc
2018-10-25 15:22:28 +01:00
b769c88767
Improves some comments and hardcodes precomputed value ( #4 )
...
* Improves some comments and hardcodes precomputed value
* Tests curve coefficients recovery
2018-10-25 15:22:28 +01:00
51688dc4bb
makefile: adds bench target
2018-10-25 15:18:54 +01:00
35e326cf2c
Merge branch 'master' of github.com:henrydcase/nobscrypto
2018-08-03 14:39:10 +01:00
10fb1a7164
x448: Export shared secret size
...
Changes x448Bytes variable to SharedSecretSize
2018-08-03 14:37:38 +01:00
c88bbf0f75
x448: Export shared secret size ( #3 )
...
Changes x448Bytes variable to SharedSecretSize
2018-08-03 14:36:45 +01:00
2ff456da90
Temporarily adds simple x448 implementation
2018-08-02 23:45:28 +01:00
fc932264c3
Merge pull request #2 from henrydcase/x448
...
Temporarily adds simple x448 implementation
2018-08-02 23:44:22 +01:00
22e3d2373f
adds code coverage
2018-07-31 20:26:50 +01:00
