jsteube
373d2b9882
Remove barrier from DCC1, high-end maxwells didn't like it
2016-05-09 21:51:36 +02:00
jsteube
c79bed3b7d
Prepare for a more dynamic #pragma unroll use
2016-05-09 21:32:12 +02:00
jsteube
34c3557d50
Fix compiler warnings
2016-05-09 19:23:37 +02:00
Jens Steube
941e016378
Fix -m 9000 performance for AMD
2016-05-09 09:40:57 +02:00
jsteube
a0221cd368
Fix broken -m 1500 and -m 3000
2016-05-09 09:17:59 +02:00
jsteube
6dac6b409e
Add compiler kernel hints for algorithms with fixed workgroup size
2016-05-09 00:58:04 +02:00
jsteube
71c31fca59
Rewrote sha256crypt from scratch
2016-05-08 22:56:36 +02:00
jsteube
fd77a098c5
Another some unroll to reduce register pressure
2016-05-08 18:57:44 +02:00
jsteube
bd1f555c9c
Remove some unroll to reduce register pressure
2016-05-08 18:29:54 +02:00
jsteube
303cfcae15
Enable unrolling of SHA512 with HMAC on NV
2016-05-08 16:07:40 +02:00
Jens Steube
7f04a89f77
Optimize out the swap32() of sha256 in 7zip
2016-05-08 15:33:32 +02:00
jsteube
16af77af18
Extended password length up to 32 for 7zip
2016-05-08 14:39:44 +02:00
jsteube
6d027e77f2
Optimize some macros
...
Use rotate() instead of inline asm; breaks Titan speed
2016-05-08 11:09:44 +02:00
jsteube
89dcaeb76d
Replace SHL+ADD with MAD
2016-05-07 23:51:37 +02:00
jsteube
8141c7fb60
Little bit inline assembly for the older nvidia cards
2016-05-07 20:45:16 +02:00
jsteube
006f5252b7
Optimize a few modes for hashcat_tuning.hctab for budget NV cards
...
Little experiment with MD4 based optimizations on -m 900 -m 1000 and -m 1100
Fix benchmark in case user fixes -u and -n values
2016-05-07 13:15:21 +02:00
Jens Steube
c40bf412e5
Fix -m 8500: volatile no longer required
2016-05-06 11:44:28 +02:00
Jens Steube
72e3821a4c
Simplify auto-tuning and benchmark routines
...
Decrease the time it takes to run a benchmark
Removed --benchmark-repeat, it creates no advantage
Fix some wording related to drivers
Dropped special 64-bit rotate() handling for NV
Cleanup SHA384
Cleanup try_run()
2016-05-05 23:21:15 +02:00
Jens Steube
8316210233
Unlock kernel_threads > 64
2016-05-04 00:52:53 +02:00
jsteube
dbe2d96618
Fix -m 1500 multihash and benchmark
...
Fix -m 3000 multihash and benchmark
Fix -m 3100 multihash and benchmark
Fix -m 7700 multihash and benchmark
Fix -m 8500 multihash and benchmark
Fix -m 11500 multihash and benchmark
2016-05-03 10:57:23 +02:00
jsteube
c7b67376a8
Fix -m 200 multihash and benchmark
2016-05-03 10:39:09 +02:00
jsteube
518fcb387a
Fix -m 10100 multihash and benchmark
2016-05-03 10:23:32 +02:00
jsteube
c0a38846e3
Fix -m 5100 multihash and benchmark
2016-05-03 10:14:53 +02:00
Jens Steube
0b3743ce94
- Added inline declaration to functions from simd.c, common.c, rp.c and types_ocl.c to increase performance
...
- Dropped static declaration from functions in all kernel to achieve OpenCL 1.1 compatibility
- Added -cl-std=CL1.1 to all kernel build options
- Created environment variable to inform NVidia OpenCL runtime to not create its own kernel cache
- Created environment variable to inform pocl OpenCL runtime to not create its own kernel cache
2016-05-01 23:15:26 +02:00
jsteube
9b3d18f87d
SIMD for slow hashes prototype
2016-05-01 18:34:59 +02:00
Jens Steube
012275e873
Merge branch 'master' of https://github.com/hashcat/oclHashcat
2016-04-29 03:29:25 +02:00
Jens Steube
d7a90f9579
Fix missing multihash support for -m 3100
2016-04-29 03:28:55 +02:00
Jens Steube
abce366d00
Fixed -m 13500 kernels and tests
2016-04-28 22:23:02 +02:00
Jens Steube
1ccb643fba
Fix variable name typo in -m 13500
2016-04-27 14:03:28 +02:00
Jens Steube
aef8dc2b01
First working version of -m 13500
2016-04-26 22:45:23 +02:00
Jens Steube
9294aaccae
Merge pull request #311 from fgaudreault/master
...
Adding parser and basic kernels for -m 13500
2016-04-26 20:12:33 +02:00
Jens Steube
7cbce12ea7
Add platform vendor_id detection
2016-04-26 13:59:14 +02:00
Jens Steube
15d7e41d68
Fix broken #ifdef in types_ocl.c
2016-04-25 07:10:02 +02:00
Jens Steube
81b19ddc7d
Remove no longer needed allx() macro
2016-04-25 07:05:47 +02:00
Jens Steube
c0d0ef60a5
Replace minus with select give a small boost in speed for -m 1500 and -m 3000
2016-04-24 20:52:21 +02:00
Jens Steube
50ea3a88dd
Merge pull request #305 from neheb/master
...
Disable AMD optimizations for LLVM
2016-04-24 15:18:17 +02:00
Jens Steube
01baa89bc0
Fix -m 3000 cracking performance
...
Fix -m 1500 cracking performance
2016-04-23 22:29:29 +02:00
Jens Steube
7ab180c868
Fix -m 200 cracking performance
2016-04-23 20:52:16 +02:00
Mangix
530651e96a
Disable AMD optimizations for LLVM
2016-04-22 20:22:28 -07:00
Francois Gaudreault
2df81367df
Adding parser and basic kernels for -m 134
2016-04-21 13:22:05 -04:00
jsteube
ad01afeeb8
Fix -m 5000 performance on NV after change to new standard
2016-04-21 09:01:31 +02:00
Fist0urs
9e8b7a1d9a
-m 13400 Increased max kdb size to 300KB
2016-04-18 18:27:51 +02:00
Jens Steube
23efa03209
Cleanup -m 131xx kernels to latest standard
2016-04-18 16:52:11 +02:00
Jens Steube
3066fbf00d
Cleanup -m 126xx kernels to latest standard
2016-04-18 16:44:54 +02:00
Jens Steube
6f0ea7e8fe
Cleanup -m 118xx kernels to latest standard
2016-04-18 16:37:16 +02:00
Jens Steube
610ae882a2
Cleanup -m 117xx kernels to latest standard
2016-04-18 16:25:55 +02:00
Jens Steube
76d74bdc04
Cleanup -m 115xx kernels to latest standard
2016-04-18 15:51:48 +02:00
Jens Steube
b6ba4d489a
Cleanup -m 114xx kernels to latest standard
2016-04-18 15:16:23 +02:00
Jens Steube
ee0b1d2c21
Cleanup -m 112xx kernels to latest standard
2016-04-18 14:39:52 +02:00
Jens Steube
79b3a1b7ca
Cleanup -m 111xx kernels to latest standard
2016-04-18 13:41:17 +02:00
