jsteube [Thu, 31 Dec 2015 14:31:21 +0000 (15:31 +0100)]
Fix the GPU Platform fix
jsteube [Thu, 31 Dec 2015 14:25:34 +0000 (15:25 +0100)]
Fix OpenCL platform selection
jsteube [Thu, 31 Dec 2015 08:32:10 +0000 (09:32 +0100)]
Update 9300 default tmto for NV
jsteube [Wed, 30 Dec 2015 22:06:45 +0000 (23:06 +0100)]
Fix some scrypt default values
jsteube [Wed, 30 Dec 2015 20:53:01 +0000 (21:53 +0100)]
Updated old RC4 code in Kerberos 5, increased performance
jsteube [Wed, 30 Dec 2015 20:30:29 +0000 (21:30 +0100)]
Fix speed -m 9700, 9710, 9800 and 9810 for NV, also increases some on AMD
jsteube [Wed, 30 Dec 2015 15:54:08 +0000 (16:54 +0100)]
Declare word_buf_r as __constant for _a3 kernel
AMD's runtime compiler does that automatically, NV's does not or with different (but worse) thresholds
jsteube [Tue, 29 Dec 2015 23:29:39 +0000 (00:29 +0100)]
Fix -m 7800 for NV
jsteube [Tue, 29 Dec 2015 23:13:59 +0000 (00:13 +0100)]
replace swap32 and swap64 with faster versions for NV
jsteube [Mon, 28 Dec 2015 21:07:09 +0000 (22:07 +0100)]
Fix TC whirlpool speed for NV
jsteube [Mon, 28 Dec 2015 10:29:07 +0000 (11:29 +0100)]
Add allocatable memory location per GPU
jsteube [Mon, 28 Dec 2015 08:20:18 +0000 (09:20 +0100)]
prepare scrypt tmto value for manual tuning
jsteube [Sun, 27 Dec 2015 20:22:54 +0000 (21:22 +0100)]
revert switch to rotate ()
jsteube [Sun, 27 Dec 2015 19:15:24 +0000 (20:15 +0100)]
Speedup -m 5300
jsteube [Sun, 27 Dec 2015 19:01:10 +0000 (20:01 +0100)]
Speedup -m 5400#
jsteube [Sun, 27 Dec 2015 18:23:55 +0000 (19:23 +0100)]
Testchange for speed
jsteube [Sun, 27 Dec 2015 16:13:17 +0000 (17:13 +0100)]
Fix -m 8000 for NV
Jens Steube [Sun, 27 Dec 2015 09:37:32 +0000 (10:37 +0100)]
Merge branch 'GetRidOfCUDA' of https://github.com/hashcat/oclHashcat into GetRidOfCUDA
Jens Steube [Sun, 27 Dec 2015 09:37:13 +0000 (10:37 +0100)]
Fix bug in test.pl, did not handle zero length passwords in -m 7700
Jens Steube [Sun, 27 Dec 2015 09:01:47 +0000 (10:01 +0100)]
Merge pull request #33 from philsmd/GetRidOfCUDA
limit the salt length of -m 22 = Juniper Netscreen/SSG (ScreenOS) to 10
philsmd [Sun, 27 Dec 2015 08:48:52 +0000 (09:48 +0100)]
limit the salt length of -m 22 = Juniper Netscreen/SSG (ScreenOS) to 10
Jens Steube [Sat, 26 Dec 2015 16:10:03 +0000 (17:10 +0100)]
Fix a bug in the implementation of GOST R 34.11-94, zero length passwords were not cracked
Jens Steube [Sat, 26 Dec 2015 09:14:42 +0000 (10:14 +0100)]
Fix LM bug for both NV and AMD
jsteube [Thu, 24 Dec 2015 11:10:03 +0000 (12:10 +0100)]
Workaround buggy NV OpenCL runtime
jsteube [Thu, 24 Dec 2015 11:05:43 +0000 (12:05 +0100)]
Revert "Workaround buggy NV OpenCL runtime"
This reverts commit
90eba9cd2e1994b6eb01a858e4603936e2c50e83.
jsteube [Thu, 24 Dec 2015 11:01:39 +0000 (12:01 +0100)]
Workaround buggy NV OpenCL runtime
jsteube [Thu, 24 Dec 2015 09:32:22 +0000 (10:32 +0100)]
Update timeout-patch url to wiki
jsteube [Wed, 23 Dec 2015 17:35:38 +0000 (18:35 +0100)]
speedup -m 5800 for NV
jsteube [Wed, 23 Dec 2015 17:02:01 +0000 (18:02 +0100)]
Added option --gpu-platform to select a single OpenCL platform in case multiple OpenCL platforms are present
jsteube [Wed, 23 Dec 2015 14:51:55 +0000 (15:51 +0100)]
- Fixed PHY memory handling for scrypt based algorithms
- Bring back kernel exec timeout checking for NV
jsteube [Wed, 23 Dec 2015 10:24:41 +0000 (11:24 +0100)]
Updated algorithm used to automatically select an ideal --scrypt-tmto value for NV
jsteube [Tue, 22 Dec 2015 22:56:12 +0000 (23:56 +0100)]
Another small increased bcrypt speed for NV
jsteube [Tue, 22 Dec 2015 22:46:40 +0000 (23:46 +0100)]
Remove unused stuff from bcrypt
jsteube [Tue, 22 Dec 2015 22:43:39 +0000 (23:43 +0100)]
Increase bcrypt speed for NV
jsteube [Tue, 22 Dec 2015 21:02:17 +0000 (22:02 +0100)]
Fixed other md5crypt based algos for NV
jsteube [Tue, 22 Dec 2015 20:26:11 +0000 (21:26 +0100)]
Fix md5crypt for NV
Jens Steube [Tue, 22 Dec 2015 18:16:43 +0000 (19:16 +0100)]
Merge pull request #30 from philsmd/GetRidOfCUDA
fix OpenCL memory problem (exhaustion): -4 error should be gone
philsmd [Tue, 22 Dec 2015 18:03:35 +0000 (19:03 +0100)]
fix OpenCL memory problem (exhaustion): -4 error should be gone
jsteube [Tue, 22 Dec 2015 10:45:32 +0000 (11:45 +0100)]
Fix unrolling with SHA512
jsteube [Tue, 22 Dec 2015 01:21:27 +0000 (02:21 +0100)]
Revert "increase bcrypt again a bit"
This reverts commit
c315ea94a2bd1d019584a1b3bf338246c68149f3.
Broke functionality
jsteube [Tue, 22 Dec 2015 01:17:48 +0000 (02:17 +0100)]
increase bcrypt again a bit
jsteube [Mon, 21 Dec 2015 22:49:22 +0000 (23:49 +0100)]
Fix 64 bit based rotates for use with ForceWare >= 358.x
jsteube [Mon, 21 Dec 2015 21:02:52 +0000 (22:02 +0100)]
Update changes to inform people about the good results with ForceWare 358.09
jsteube [Mon, 21 Dec 2015 20:17:42 +0000 (21:17 +0100)]
Fix indent
jsteube [Mon, 21 Dec 2015 20:14:16 +0000 (21:14 +0100)]
add __bfe stub for AMD
jsteube [Mon, 21 Dec 2015 20:00:52 +0000 (21:00 +0100)]
Slightly increase blowfish based algorithms for NV
Jens Steube [Mon, 21 Dec 2015 19:21:53 +0000 (20:21 +0100)]
Fix blowfish based algos for AMD, NV needs testing
jsteube [Mon, 21 Dec 2015 16:06:17 +0000 (17:06 +0100)]
Fix -m 9000 speed (AMD ONLY)
Jens Steube [Mon, 21 Dec 2015 13:50:13 +0000 (14:50 +0100)]
Fix AMD speeds @ sha512crypt
jsteube [Mon, 21 Dec 2015 13:29:04 +0000 (14:29 +0100)]
Fixed some speeds:
1800
3200
7100
7200
7400
8200
9300
9000
9600
11300
11600
jsteube [Mon, 21 Dec 2015 11:40:34 +0000 (12:40 +0100)]
Fix package script
jsteube [Mon, 21 Dec 2015 11:31:23 +0000 (12:31 +0100)]
We do not depend on CUDA any longer, so get rid of it :)
Gabriele 'Matrix' Gristina [Wed, 16 Dec 2015 12:47:46 +0000 (13:47 +0100)]
Update for BUILD.md
Make the build documentation like hashcat
jsteube [Mon, 21 Dec 2015 11:20:08 +0000 (12:20 +0100)]
We do not depend on CUDA any longer, so get rid of it :)
jsteube [Mon, 21 Dec 2015 11:01:38 +0000 (12:01 +0100)]
Fix caching system for use with AMD and NV, drop BINARY_KERNEL define
jsteube [Mon, 21 Dec 2015 09:17:54 +0000 (10:17 +0100)]
Workaround missing clEnqueueFillBuffer() support in NVidia's OpenCL runtime
jsteube [Sun, 20 Dec 2015 21:53:14 +0000 (22:53 +0100)]
Fix Lotus Notes/Domino 8 for NV
jsteube [Sun, 20 Dec 2015 21:48:36 +0000 (22:48 +0100)]
Fix Lotus Notes/Domino 6 for NV
jsteube [Sun, 20 Dec 2015 21:46:59 +0000 (22:46 +0100)]
Fix Lotus Notes/Domino 5 for NV
jsteube [Sun, 20 Dec 2015 21:32:40 +0000 (22:32 +0100)]
Fix Android/PIN for NV
jsteube [Sun, 20 Dec 2015 21:09:55 +0000 (22:09 +0100)]
Fix bsdicrypt for NV
jsteube [Sun, 20 Dec 2015 21:08:59 +0000 (22:08 +0100)]
Revert "Fix bsdicrypt for NV"
This reverts commit
284990e4d3196f5a50793185daf7fe31b8358b88.
Accidentially changed oclHashcat.c
jsteube [Sun, 20 Dec 2015 21:07:44 +0000 (22:07 +0100)]
Fix bsdicrypt for NV
Jens Steube [Sun, 20 Dec 2015 16:02:02 +0000 (17:02 +0100)]
Remove some unused variables in scrypt kernel
Jens Steube [Sun, 20 Dec 2015 15:59:02 +0000 (16:59 +0100)]
Remove old macro, not needed anymore
jsteube [Sun, 20 Dec 2015 15:09:00 +0000 (16:09 +0100)]
Fix keccak speed on NV OpenCL runtime
jsteube [Sat, 19 Dec 2015 17:30:34 +0000 (18:30 +0100)]
Fix -m 11700 and -m 11800
jsteube [Sat, 19 Dec 2015 17:21:41 +0000 (18:21 +0100)]
Fix -m 6500
jsteube [Sat, 19 Dec 2015 17:18:54 +0000 (18:18 +0100)]
Fix -m 10700
jsteube [Sat, 19 Dec 2015 17:18:28 +0000 (18:18 +0100)]
Revert "Fix -m 10700"
This reverts commit
15da53da386a4253eeaefc66911e7525e2f3699e.
jsteube [Sat, 19 Dec 2015 17:17:19 +0000 (18:17 +0100)]
Fix -m 10700
jsteube [Thu, 17 Dec 2015 18:19:31 +0000 (19:19 +0100)]
Fix -m 1800 for NV
jsteube [Wed, 16 Dec 2015 16:01:14 +0000 (17:01 +0100)]
Fixed empty line problem
jsteube [Wed, 16 Dec 2015 15:21:10 +0000 (16:21 +0100)]
Fix -m 8500 kernel
jsteube [Wed, 16 Dec 2015 15:15:21 +0000 (16:15 +0100)]
Fix -m 7800 by adding amd_bytealign emulation
jsteube [Wed, 16 Dec 2015 14:45:05 +0000 (15:45 +0100)]
fixed -m 6900
jsteube [Wed, 16 Dec 2015 14:38:25 +0000 (15:38 +0100)]
fix -m 5500
Jens Steube [Wed, 16 Dec 2015 14:27:34 +0000 (15:27 +0100)]
fix -m 3100
jsteube [Wed, 16 Dec 2015 12:27:34 +0000 (13:27 +0100)]
Fix -m 3000 kernel
jsteube [Wed, 16 Dec 2015 11:53:52 +0000 (12:53 +0100)]
Fix -m 1500 for -a 0 and -a 1 attack mode
jsteube [Wed, 16 Dec 2015 11:06:18 +0000 (12:06 +0100)]
final cleanup for -m 1500
jsteube [Wed, 16 Dec 2015 10:52:32 +0000 (11:52 +0100)]
Fix -m 1500 for NV
Fix C++ function header
jsteube [Wed, 16 Dec 2015 10:18:57 +0000 (11:18 +0100)]
Removed useless comments
jsteube [Wed, 16 Dec 2015 10:11:31 +0000 (11:11 +0100)]
Fix use of LOP3
jsteube [Wed, 16 Dec 2015 09:19:17 +0000 (10:19 +0100)]
Prepare for LOP3 integration
Jens Steube [Wed, 16 Dec 2015 09:00:33 +0000 (10:00 +0100)]
Fix invalid VENDOR_ID for AMD
jsteube [Tue, 15 Dec 2015 18:28:27 +0000 (19:28 +0100)]
update packages for source kernel distribution
jsteube [Tue, 15 Dec 2015 18:18:45 +0000 (19:18 +0100)]
Cleanup makefile
Remove unused CUDA library defs
jsteube [Tue, 15 Dec 2015 17:58:02 +0000 (18:58 +0100)]
It's safe now to use AMD's OpenCL includes, tested on NV
Jens Steube [Tue, 15 Dec 2015 17:36:29 +0000 (18:36 +0100)]
Merge pull request #25 from philsmd/GetRidOfCUDA
change in library loading: libnvidia-ml now does load dynamically
philsmd [Tue, 15 Dec 2015 19:34:07 +0000 (20:34 +0100)]
change in library loading: libnvidia-ml now does load dynamically only on NVidia systems
jsteube [Tue, 15 Dec 2015 16:46:00 +0000 (17:46 +0100)]
More kernel fixes for function calls and vector datatypes
jsteube [Tue, 15 Dec 2015 15:50:21 +0000 (16:50 +0100)]
Fix more append_* functions in kernels
Jens Steube [Tue, 15 Dec 2015 15:44:40 +0000 (16:44 +0100)]
Merge pull request #24 from philsmd/GetRidOfCUDA
fixed hardware monitor for amd/nv
philsmd [Tue, 15 Dec 2015 17:41:11 +0000 (18:41 +0100)]
fixed hardware monitor for amd/nv (one problem that is still there since last commit: dynamic loading of libnvidia-ml)
jsteube [Tue, 15 Dec 2015 12:42:37 +0000 (13:42 +0100)]
Fix append_* function calls
jsteube [Tue, 15 Dec 2015 12:23:32 +0000 (13:23 +0100)]
More cleanup
jsteube [Tue, 15 Dec 2015 11:04:22 +0000 (12:04 +0100)]
- Dropped all vector code since new GPU's are all scalar, makes the code much easier
- Some performance on low-end GPU may drop because of that, but only for a few hash-modes
- Dropped scalar code (aka warp) since we do not have any vector datatypes anymore
- Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4
- Still need to fix kernels to new function names, needs to be done manually
- Temperature Management needs to be rewritten partially because of conflicting datatypes names
- Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id)
- Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD)
- First tests working for -m 0, for example
- Great performance increases in general for NV so far
- Tested amp_* and markov_* kernel
- Migrated special NV optimizations for rule processor
Jens Steube [Sun, 13 Dec 2015 17:46:21 +0000 (18:46 +0100)]
Revert "Fixed a bug where oclHashcat rejected to load a rule which calls 15 functions although it is supported"
This reverts commit
f230ed73dc142a4a6e5ed5c61afef976d51aa4b6.
Some kernel require too much constant memory and do not compile anymore
Jens Steube [Sun, 13 Dec 2015 13:54:39 +0000 (14:54 +0100)]
Fixed a bug where oclHashcat rejected to load a rule which calls 15 functions although it is supported