jsteube
5ae5a4bc25
Reorder device mapping
2016-01-15 17:23:07 +01:00
Jens Steube
63e06f582b
Prepare fix for datatypes, not final
2016-01-15 17:16:43 +01:00
Jens Steube
2e61685f48
Fix outputting wrong password for weak-hash-check in case user defined -r or -g
2016-01-14 23:03:23 +01:00
Jens Steube
ed91e69ae3
Fix out of bounds access in -m 7500
2016-01-14 22:13:06 +01:00
Jens Steube
f3b85a6363
Fix some clang compiler warnings
2016-01-14 21:45:11 +01:00
jsteube
76612ac031
Fix more missing casts to uint in rule-engine
2016-01-14 20:54:41 +01:00
jsteube
497b41fee4
Fix more missing casts to uint
2016-01-14 20:44:11 +01:00
jsteube
d3981ee66a
Fix missing cast to uint
2016-01-14 20:42:07 +01:00
Jens Steube
7bfee268d4
Fixed some heap buffer overflow
2016-01-14 20:30:38 +01:00
Jens Steube
245301c9b4
Started optimizing some of the OpenCL kernel for latest AMD Catalyst 15.12:
...
- Replaced SBOX for DES:
replaced JtR's * Bitslice DES S-boxes making use of a vector conditional select operation (e.g., vsel on PowerPC with AltiVec).
with JtR's * Bitslice DES S-boxes for x86 with MMX/SSE2/AVX and for typical RISC architectures.
Performance increased for DEScrypt from 355MH/s to 405MH/s and for LM from 11100MH/s to 12000MH/s
BTW, the same effect can be seen with non-maxwell GPU's
- Remove some volatile keywords no longer needed thanks to fixed catalyst bugs
- Fix weak-hash-check parameter for use with tools/test.sh
2016-01-14 19:44:47 +01:00
Jens Steube
24b5aa6226
Merge pull request #128 from philsmd/pr-set-iter-for-dcc2
...
set default iteration count for -m 2100 = DCC2 to 10240
2016-01-14 09:02:13 +01:00
philsmd
cbe6be9246
set default iteration count for -m 2100 = DCC2 to 10240
2016-01-14 00:14:43 +01:00
philsmd
88cbb45ca5
add skipped devices to the --benchmark output
2016-01-13 23:38:07 +01:00
philsmd
7b53ccab70
minor: remove quiet_sav, was just intended for debugging
2016-01-13 23:06:03 +01:00
Jens Steube
49d0767aa8
Run weak-hash checks only in straight-attack mode, this greatly reduces code complexity
2016-01-13 22:22:46 +01:00
Jens Steube
cfff6220fc
Fix -m 1500 in combination with new weak-hash-check
2016-01-13 21:39:34 +01:00
philsmd
4ed6b2c10a
HMS: nvapi does not need a dll (windows + NV)
2016-01-13 21:12:19 +01:00
jsteube
e7e6c4da17
Do not use OpenCL CPU devices by default to avoid slow GPU synchronization, user needs to explicitly enable them using --opencl-device-type
...
If a platform like pocl is filtered by any of the filter mechanism do not request the user to use --force
2016-01-13 20:27:26 +01:00
Jens Steube
d3dc57896e
Merge pull request #123 from philsmd/pr-tab-typo-fix
...
cosmetic: replace tab by spaces
2016-01-13 19:56:38 +01:00
philsmd
182819af11
cosmetic: replace tab by spaces
2016-01-13 19:05:28 +01:00
philsmd
7081e3cdf6
Makefile: remove kernels/ directory when running make clean
2016-01-13 18:16:56 +01:00
Jens Steube
52d71de0d4
Added support to utilize multiple different OpenCL platforms in parallel, ex: AMD + NV
2016-01-13 17:10:40 +01:00
Jens Steube
0ff49c5b44
Fix segfault in related to weak-hash-check on CPU devices
...
Move typedef for bool datatype to common.h
2016-01-12 17:18:01 +01:00
Jens Steube
54ed58e28b
Merge pull request #119 from philsmd/pr-fix-adl-mapping-CPU
...
HMS fix: only copy the ADL adapter info for GPU devices
2016-01-12 08:49:25 +01:00
philsmd
b47452e343
HMS fix: only copy the ADL adapter info for GPU devices, for CPU devices we leave it AS-IS (zeroed out)
2016-01-12 01:21:11 +01:00
philsmd
2e1e3318ec
fixed % and C (degrees celcius) in status display
2016-01-12 01:02:23 +01:00
Jens Steube
730044e26e
Added option --opencl-device-types to filter for specific OpenCL device types
2016-01-12 00:44:28 +01:00
philsmd
02cecce247
fixed hardware monitor: mapping between supported ADL devices and available GPU (and CPUs) fixed
2016-01-12 00:30:28 +01:00
jsteube
93427c073c
Prepare for some HSM updates related to device_type
2016-01-11 23:49:10 +01:00
Jens Steube
87dca98a51
Fix segfault if a weak hash is detected for some fast hashes
2016-01-10 21:37:45 +01:00
Jens Steube
1d628ca6e0
The weak-hash-check feature didn't work for algorithms that made use of some automatic optimizer flags
...
We'll enforce a weak-hash-check on an _a0 kernel for them
2016-01-10 21:02:03 +01:00
jsteube
3cdee2e06e
We need an esalt to hold the unused IV otherwise the unique salt count becomes incorrect in multihash sitations
2016-01-09 21:35:29 +01:00
jsteube
f0a84a2410
Added new hash mode -m 13000 = RAR5
2016-01-09 20:34:12 +01:00
jsteube
332b3c35e5
Added new hash mode -m 12900 = Android FDE (Samsung DEK)
2016-01-09 00:49:54 +01:00
jsteube
e8229af09b
Fix for -m 1500, -m 3000: Missing "defined" keyword
...
Fix for -m 1000: MD4_H1 copy/paste error
Fix for -m 8900, -m 9300: Invalid value for device_processor_cores for CPU devices
Fix for -m 9100: Variable salt2 initializer needed some clauses for clearness
Temporary limit gpu_accel for CPU devices to 1 for development phase
Mark pocl as too bleeding edge for production use, recommend native drivers
Remove workarounds for pocl
Rename VENDOR_ID_UNKNOWN to VENDOR_ID_GENERIC in host code
Rename IS_UNKNOWN to IS_GENERIC in kernel code
2016-01-07 20:14:34 +01:00
Jens Steube
6dab39e69c
Merge pull request #64 from philsmd/master
...
added --opencl-platform check, reject numbers > number of OpenCL platforms
2016-01-07 20:09:50 +01:00
philsmd
82ac8cbcfe
additional --opencl-platform value check
2016-01-07 17:15:28 +01:00
philsmd
6fcc3f2728
typo
2016-01-07 16:48:50 +01:00
philsmd
7b7caf3842
added additional --opencl-platform check (reject numbers > number of OpenCL platforms)
2016-01-07 16:35:45 +01:00
jsteube
36df3ef329
A CPU is not a GPU
2016-01-05 23:35:06 +01:00
Jens Steube
51a1868903
Shorten some helptext
2016-01-05 12:28:56 +01:00
philsmd
8bf3bdbf1b
the parameter --opencl-platform takes a number, not a string
2016-01-05 10:29:05 +01:00
jsteube
16294fe94e
Next release version will be v2.10
2016-01-05 08:30:18 +01:00
jsteube
331188167c
Replace the substring GPU to a more appropriate "device" or "kernel" substring depending on the context
2016-01-05 08:26:44 +01:00
jsteube
894140b816
Generate a more easy to read filename for cached kernels
2016-01-04 20:56:15 +01:00
jsteube
8baf705f6a
Preparation for distinguish between OpenCL device types at runtime (mostly for HMS)
2016-01-04 20:12:34 +01:00
jsteube
8b50b60484
More fixes for general OpenCL platforms
2016-01-04 19:23:21 +01:00
jsteube
c4b1c8e7ca
small fix for unknown opencl platforms
2016-01-04 16:13:10 +01:00
jsteube
3b589e3aac
Prepare for allow other OpenCL platforms thans AMD and NVidia
2016-01-04 13:17:20 +01:00
philsmd
f2b8f008c0
pr #53 did not copy the example .hash and .dict files to the doc folder
2016-01-03 23:29:39 +01:00
