fumiama/blake2b-simd
1
0
Fork 0
mirror of https://github.com/fumiama/blake2b-simd.git synced 2026-06-05 02:00:26 +08:00
Code Activity
54 Commits 1 Branch 0 Tags
master
Commit Graph

54 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
源文雨
a5dcaba541 feat: impl. encoding.BinaryMarshaler/Unmarshaler 2025-02-28 13:59:19 +09:00
源文雨
4481822068 add commandline tool 2022-04-12 19:01:31 +08:00
源文雨
c99573b3b2 adapt go1.17 2022-04-12 17:23:18 +08:00
Harshavardhana
3f5f724cb5 build: Enable windows builds. (#23) 2016-07-22 23:10:19 -07:00
Harshavardhana
c50cace0dc internal: use arrays instead of make for simple values. (#22) 
Fixes #16
 2016-07-22 02:38:12 -07:00
Frank
326c321661 Changed asm2plan9s link to minio (#21) 2016-07-21 11:14:46 -07:00
Frank
959f72aaa5 Restyled AVX and SSE version using asm2plan9s and asmfmt (#20) 2016-07-08 07:47:17 -07:00
Frank
25efc542f2 Styling of comparison benchmark (#14) 2016-07-06 01:29:24 -07:00
Frank
b899a74a3a Added performance comparison to other hashing techniques (#12) 2016-07-05 10:48:44 -07:00
Frank
d2af7cb658 Add support for SSE (#10) 2016-07-04 10:40:46 -07:00
Harshavardhana
4cf7fe6339 Fix README.md with new numbers. (#8) 2016-07-03 23:29:06 +02:00
Harshavardhana
cf4f8e0c34 Fix benchmarks to report proper values. (#7) 
```
$ go test -run=NONE -bench .
PASS
BenchmarkHash64-4  	 1000000	      1036 ns/op	  61.77 MB/s
BenchmarkHash128-4 	 2000000	       801 ns/op	 159.67 MB/s
BenchmarkHash1K-4  	  500000	      2464 ns/op	 415.53 MB/s
BenchmarkHash8K-4  	  200000	     11212 ns/op	 730.60 MB/s
BenchmarkHash32K-4 	   30000	     40766 ns/op	 803.80 MB/s
BenchmarkHash128K-4	   10000	    163170 ns/op	 803.28 MB/s
ok  	github.com/minio/blake2b-simd	10.298s
```
 2016-07-03 21:58:02 +02:00
Frank
00562011ad Added AVX2 support (#6) 
* Added AVX2 support
 2016-07-03 11:54:51 -07:00
Frank
f0521e8972 Improve formatting for long instructions (#4) 2016-06-30 15:40:31 -07:00
Frank
cf89911846 Move message loop into assembly (#3) 2016-06-29 11:18:44 -07:00
Harshavardhana
94258bb4c0 doc: Add platform specific details 2016-06-28 17:21:01 -07:00
Harshavardhana
9d80c8bfe5 doc: Add comparative benchmark results. 2016-06-28 17:06:07 -07:00
Harshavardhana
0b3e695ecc Add travis build CI. 2016-06-28 09:28:56 -07:00
frankw
f41b7a312a asm: Add new defines 
- Add UNDIAGONALIZE define
- Add DIAGONALIZE define
- Add G1 define
- Add G2 define
- Add LOAD_SHUFFLE define

Additionally check for AVX support.
 2016-06-28 09:28:38 -07:00
Harshavardhana
3a46db1cb4 Detect SSE automatically and enable corresponding optimized code path. 
With this change, following happens.

```
if sse {
   compressSSE()
   return
}
compressGeneric()
```

compressGeneric is used as a fallback when SSE is not detected.
 2016-06-28 02:51:14 -07:00
frankw
83ff4cf5f4 Full test cases 2016-06-28 10:55:07 +02:00
frankw
c82983850e Update license header 2016-06-28 10:43:31 +02:00
frankw
f6a3512276 Update README for repo name change 2016-06-28 10:23:10 +02:00
frankw
1953a0fe1e Removed functionality/code for comparing (during dev) to reference code 2016-06-28 10:01:56 +02:00
frankw
0cb28af64e Added initial test vectors 2016-06-28 09:39:20 +02:00
frankw
ef4414c329 Round 10 (last one!) 2016-06-28 09:38:27 +02:00
frankw
6a2ebd9a03 Round 9 2016-06-28 08:52:13 +02:00
frankw
d2c63e4fcd Round 8 2016-06-27 22:42:48 +02:00
frankw
c26c95931c Round 7 2016-06-27 19:27:03 +02:00
frankw
c6883ff2b3 Round 6 2016-06-27 14:28:05 +02:00
frankw
2b06d39333 Round 5 2016-06-27 13:17:42 +02:00
frankw
2f87bcb1b1 Round 4 2016-06-27 11:55:37 +02:00
frankw
95d67c9edb Small optimization of using VPALINGR instead of 2 VPUNPCK?QDQ's 2016-06-26 23:55:49 +02:00
frankw
b7b013f822 Round 3 2016-06-26 23:54:23 +02:00
frankw
33bf120400 Round 12 (identical to Round 2) 2016-06-26 20:50:15 +02:00
frankw
ff8ba2ccd0 Round 2 completed 2016-06-26 20:47:17 +02:00
frankw
7759a2b074 G2 macro for Round 2 2016-06-26 20:10:27 +02:00
frankw
55b55e023b G1 macro for Round 2 2016-06-26 19:45:07 +02:00
frankw
e1903c8762 Round 11 (identical to Round 1) 2016-06-26 19:39:00 +02:00
frankw
4d9461fef6 Round 1 completed 2016-06-25 09:57:20 +02:00
frankw
d71b79de75 G1 macro after DIAGONALIZE added 2016-06-25 09:51:53 +02:00
frankw
53bb6668a3 UNDIAGONALIZE macro 2016-06-25 09:04:36 +02:00
frankw
b0e1a3d003 DIAGONALIZE macro 2016-06-25 08:57:01 +02:00
frankw
c5dac2387d G2 macro completed 2016-06-25 08:42:30 +02:00
frankw
100130f468 Comment about approach 2016-06-24 23:42:10 +02:00
frankw
07dd36a646 First half of G2 macro completed 2016-06-24 23:30:25 +02:00
frankw
2fe67e0d78 Add comment for register use 2016-06-24 22:51:51 +02:00
frankw
5ba9ce6a04 G1 macro completed 2016-06-24 22:38:20 +02:00
frankw
4d0b10ee48 Reordering of iv loading 2016-06-24 22:14:30 +02:00
frankw
207d895ad9 First half of G1 macro working for Round 1 2016-06-24 22:14:06 +02:00