Jens Steube [Wed, 24 Feb 2016 18:53:17 +0000 (19:53 +0100)]
Merge pull request #239 from philsmd/master
this fixes the beta issue #163 (-d and --opencl-platforms not within valid range)
philsmd [Wed, 24 Feb 2016 18:50:05 +0000 (19:50 +0100)]
this fixes the beta issue #163 (-d and --opencl-platforms not within valid range)
Jens Steube [Wed, 24 Feb 2016 14:56:02 +0000 (15:56 +0100)]
Merge pull request #237 from gpuhash/master
Fairly strict stamping of WPA salt components
Jens Steube [Wed, 24 Feb 2016 14:27:02 +0000 (15:27 +0100)]
Cleanup of all raw-SHA1 based algorithms
Jens Steube [Wed, 24 Feb 2016 12:40:38 +0000 (13:40 +0100)]
Cleanup of all raw-MD5 based algorithms; small change important for later changes
Jens Steube [Wed, 24 Feb 2016 10:35:13 +0000 (11:35 +0100)]
Cleanup of all raw-MD5 based algorithms; Should be done for all raw-SHA1, -SHA256 and -SHA512 based algorithms as well
Jens Steube [Tue, 23 Feb 2016 20:24:03 +0000 (21:24 +0100)]
Fix this evil bug, pw->i is u32, not u8
Jens Steube [Tue, 23 Feb 2016 19:53:55 +0000 (20:53 +0100)]
Add another example of an old budget GPU that benefits from higher vector width than returned by preferred
gpuhash [Tue, 23 Feb 2016 19:01:55 +0000 (22:01 +0300)]
Minor bugfix
gpuhash [Tue, 23 Feb 2016 18:31:18 +0000 (21:31 +0300)]
Fairly strict stamping of WPA salt components
Jens Steube [Tue, 23 Feb 2016 17:51:28 +0000 (18:51 +0100)]
Ensure a minimum length for password candidates in autotune
Jens Steube [Tue, 23 Feb 2016 14:36:06 +0000 (15:36 +0100)]
We still need on initial testrun in autotune
Jens Steube [Tue, 23 Feb 2016 14:00:56 +0000 (15:00 +0100)]
Do not use values that can actually crack a hash in autotune
Jens Steube [Tue, 23 Feb 2016 11:18:47 +0000 (12:18 +0100)]
Speed up -m 20 in -a 3 mode
Jens Steube [Tue, 23 Feb 2016 09:52:02 +0000 (10:52 +0100)]
Fix missing memset after previous fix
Jens Steube [Tue, 23 Feb 2016 09:36:43 +0000 (10:36 +0100)]
Evil Bug: Because of not-zeroing a buffer it's possible a hash was cracked but not reported as cracked because it was not in scope, but when it was in scope it was already marked as cracked and therefore not reported
Jens Steube [Tue, 23 Feb 2016 06:57:52 +0000 (07:57 +0100)]
Try to load libOpenCL.so.1 if libOpenCL.so was not found
Jens Steube [Mon, 22 Feb 2016 20:35:37 +0000 (21:35 +0100)]
Increase salt length for raw-md5 based algorithms
Jens Steube [Mon, 22 Feb 2016 20:33:33 +0000 (21:33 +0100)]
Zero pws_buf before reuse
Jens Steube [Mon, 22 Feb 2016 20:32:38 +0000 (21:32 +0100)]
Revert "Zero pws_buf before reuse"
This reverts commit
b409e5e9e162927b25af88e672326b246f17ec5f.
Jens Steube [Mon, 22 Feb 2016 20:20:16 +0000 (21:20 +0100)]
Zero pws_buf before reuse
Jens Steube [Mon, 22 Feb 2016 11:50:04 +0000 (12:50 +0100)]
Fix use of find_kernel_power_div()
Jens Steube [Mon, 22 Feb 2016 10:57:37 +0000 (11:57 +0100)]
Get rid of old pw_cache mechanism to control host-based vector data-types
Jens Steube [Mon, 22 Feb 2016 09:36:04 +0000 (10:36 +0100)]
Add auto-append for 0x01 and 0x80 in ATTACK_MODE_HYBRID2
Jens Steube [Mon, 22 Feb 2016 09:09:16 +0000 (10:09 +0100)]
New SIMD code for -a 1 -m 100
Jens Steube [Mon, 22 Feb 2016 09:02:23 +0000 (10:02 +0100)]
New SIMD code for -a 1 -m 60
Jens Steube [Mon, 22 Feb 2016 09:01:55 +0000 (10:01 +0100)]
New SIMD code for -a 1 -m 40
Jens Steube [Mon, 22 Feb 2016 09:01:53 +0000 (10:01 +0100)]
New SIMD code for -a 1 -m 40
Jens Steube [Mon, 22 Feb 2016 09:01:49 +0000 (10:01 +0100)]
New SIMD code for -a 1 -m 30
Jens Steube [Sun, 21 Feb 2016 17:40:10 +0000 (18:40 +0100)]
New SIMD code for -a 1 -m 20
Jens Steube [Sun, 21 Feb 2016 17:40:06 +0000 (18:40 +0100)]
New SIMD code for -a 1 -m 10
Jens Steube [Sun, 21 Feb 2016 17:40:01 +0000 (18:40 +0100)]
New SIMD code for -a 1 -m 0
Jens Steube [Sun, 21 Feb 2016 17:39:44 +0000 (18:39 +0100)]
New SIMD code for -a 1 prepare
Jens Steube [Sun, 21 Feb 2016 17:23:04 +0000 (18:23 +0100)]
Autotune should respect --quiet parameter
Jens Steube [Sun, 21 Feb 2016 12:30:28 +0000 (13:30 +0100)]
Fix force_jit_compilation usage
Jens Steube [Sat, 20 Feb 2016 15:13:06 +0000 (16:13 +0100)]
Prepare NEW_SIMD_MODE for -a 1 kernels
Jens Steube [Sat, 20 Feb 2016 11:47:01 +0000 (12:47 +0100)]
Merge pull request #233 from Fist0urs/format_krb5tgs
Fix test.pl verify -m 13100
Fist0urs [Sat, 20 Feb 2016 00:38:00 +0000 (01:38 +0100)]
Fix test.pl verify
Jens Steube [Fri, 19 Feb 2016 23:58:22 +0000 (00:58 +0100)]
Add hard-wired Device-Name for Tuning-Database which matches all Device-Types:
- DEVICE_TYPE_CPU
- DEVICE_TYPE_GPU
- DEVICE_TYPE_ACCELERATOR
Optimized IRIS GPU workaround settings
Jens Steube [Fri, 19 Feb 2016 23:05:30 +0000 (00:05 +0100)]
Merge pull request #232 from Fist0urs/format_krb5tgs
Add verification of checksum for -m 13100
Fist0urs [Fri, 19 Feb 2016 22:12:46 +0000 (23:12 +0100)]
Add verification of checksum for -m 13100
This avoid collisions by arc4'ing all data then hmac-md5
when valid ASN1 structures headers are found.
Performance should not be impacted.
Jens Steube [Fri, 19 Feb 2016 17:58:03 +0000 (18:58 +0100)]
Allow mixed settings auto and fixed values for parameter and inside tuningdb for loops and accel
Jens Steube [Thu, 18 Feb 2016 13:44:44 +0000 (14:44 +0100)]
Simplify steps for autotune
Jens Steube [Thu, 18 Feb 2016 12:04:25 +0000 (13:04 +0100)]
Fix final autotune balancing
Jens Steube [Thu, 18 Feb 2016 09:49:23 +0000 (10:49 +0100)]
Move fake word generation to try_run() to avoid unwanted caching
Jens Steube [Thu, 18 Feb 2016 08:02:01 +0000 (09:02 +0100)]
Merge pull request #230 from magnumripper/master
Support vector width 16
magnum [Tue, 16 Feb 2016 15:42:08 +0000 (16:42 +0100)]
Allow and support vector-width 16, which is current maximum for
OpenCL. Closes #226.
magnum [Tue, 16 Feb 2016 14:14:15 +0000 (15:14 +0100)]
Simplify code. No change in functionality at all.
Jens Steube [Wed, 17 Feb 2016 14:11:43 +0000 (15:11 +0100)]
Merge pull request #229 from Fist0urs/format_krb5tgs
Cosmetic change -m 13100
Fist0urs [Wed, 17 Feb 2016 12:32:07 +0000 (13:32 +0100)]
Cosmetic change -m 13100
Removed commented part (the commented part was there to implement full last arc4'ing + hmac-md5)
We will see if some people find collision. In this case we will add this last check
Jens Steube [Wed, 17 Feb 2016 10:33:40 +0000 (11:33 +0100)]
Merge pull request #228 from philsmd/master
minor: some formatting fixes re: #225 (-m 13100 = Kerberos 5 TGS-REP etype 23)
philsmd [Wed, 17 Feb 2016 10:02:40 +0000 (11:02 +0100)]
minor: some formatting fixes re: #225 (-m 13100 = Kerberos 5 TGS-REP etype 23)
Jens Steube [Wed, 17 Feb 2016 07:50:12 +0000 (08:50 +0100)]
Add missing changes entry for -m 13100
Jens Steube [Wed, 17 Feb 2016 07:23:27 +0000 (08:23 +0100)]
Merge pull request #225 from Fist0urs/format_krb5tgs
Initial commit
Fist0urs [Tue, 16 Feb 2016 15:34:46 +0000 (16:34 +0100)]
Initial commit
jsteube [Tue, 16 Feb 2016 12:24:36 +0000 (13:24 +0100)]
Fix warp bug on -m 8500
Jens Steube [Tue, 16 Feb 2016 09:47:40 +0000 (10:47 +0100)]
Fix Truecrypt Whirlpool speed
Jens Steube [Tue, 16 Feb 2016 09:34:28 +0000 (10:34 +0100)]
Fix old NV Titan OpenCL device name
Jens Steube [Mon, 15 Feb 2016 21:27:30 +0000 (22:27 +0100)]
Replace OpenCL event's with own timers, this saves us a thread-pair spawned on nvidia's OpenCL runtime and doesn't affect others
Jens Steube [Mon, 15 Feb 2016 20:19:41 +0000 (21:19 +0100)]
Use CL_DEVICE_NATIVE_VECTOR_WIDTH_* instead of CL_DEVICE_PREFERRED_VECTOR_WIDTH_*
Jens Steube [Mon, 15 Feb 2016 20:08:16 +0000 (21:08 +0100)]
Fix autotune in -i mode
jsteube [Mon, 15 Feb 2016 19:32:01 +0000 (20:32 +0100)]
Fix another source of 0H/s in benchmark mode
Jens Steube [Mon, 15 Feb 2016 17:56:03 +0000 (18:56 +0100)]
Merge pull request #223 from gm4tr1x/clEnqueueNDRangeKernelMod
Get rid of hc_clEnqueueNDRangeKernel double call
Jens Steube [Mon, 15 Feb 2016 16:42:09 +0000 (17:42 +0100)]
Add option to repeat the kernel on the device NUM times to increase benchmark accuracy
Jens Steube [Mon, 15 Feb 2016 14:09:22 +0000 (15:09 +0100)]
Add autotune engine to changes
Jens Steube [Mon, 15 Feb 2016 13:57:03 +0000 (14:57 +0100)]
Final balance meassurements for autotune
Jens Steube [Mon, 15 Feb 2016 11:38:54 +0000 (12:38 +0100)]
In benchmark-mode, do not depend any longer on a fixed time, better use a single iteration instead
Jens Steube [Mon, 15 Feb 2016 10:11:20 +0000 (11:11 +0100)]
Remove measurement section from tuning database and create a forum-thread or wiki-entry for it
Jens Steube [Sun, 14 Feb 2016 19:15:14 +0000 (20:15 +0100)]
Disable GPU-Temp monitoring during benchmark
jsteube [Sun, 14 Feb 2016 19:07:16 +0000 (20:07 +0100)]
Autotune: 200 loops should be fine for a start
Jens Steube [Sun, 14 Feb 2016 17:23:21 +0000 (18:23 +0100)]
Add dedicated steps for loops and accel
Jens Steube [Sun, 14 Feb 2016 14:50:29 +0000 (15:50 +0100)]
Fixed a missing condition
Jens Steube [Sun, 14 Feb 2016 14:45:52 +0000 (15:45 +0100)]
Autotuning engine prototype
Jens Steube [Sat, 13 Feb 2016 15:07:58 +0000 (16:07 +0100)]
Prepare for better autotuning capabilities
Gabriele 'matrix' Gristina [Fri, 12 Feb 2016 16:11:23 +0000 (17:11 +0100)]
Get rid of hc_clEnqueueNDRangeKernel double call
Jens Steube [Fri, 12 Feb 2016 15:35:16 +0000 (16:35 +0100)]
Merge pull request #221 from gm4tr1x/8900tuning
Updated 8900 osx kernel accel
Jens Steube [Fri, 12 Feb 2016 15:35:09 +0000 (16:35 +0100)]
Merge pull request #220 from gm4tr1x/weak_hash_check
Fix bug in weak_hash_check(), wrong kernel_loops assigned
Gabriele 'matrix' Gristina [Fri, 12 Feb 2016 14:23:35 +0000 (15:23 +0100)]
Updated 8900 osx kernel accel
Gabriele 'matrix' Gristina [Fri, 12 Feb 2016 13:04:13 +0000 (14:04 +0100)]
Fix bug in weak_hash_check(), wrong kernel_loops assigned
Jens Steube [Thu, 11 Feb 2016 13:26:18 +0000 (14:26 +0100)]
Update tuningdb help text
Jens Steube [Thu, 11 Feb 2016 13:16:08 +0000 (14:16 +0100)]
Remove Wordload-Profiles from tuningdb, we can simply calculate it
Jens Steube [Thu, 11 Feb 2016 12:38:59 +0000 (13:38 +0100)]
Updated Workload finding strategy, should be easy to automate now
Jens Steube [Thu, 11 Feb 2016 12:31:50 +0000 (13:31 +0100)]
Updated Workload finding strategy, should be easy to automate now
Jens Steube [Thu, 11 Feb 2016 12:01:45 +0000 (13:01 +0100)]
Merge pull request #219 from gm4tr1x/updateTuningDB
Added two aliases for osx
Gabriele 'matrix' Gristina [Thu, 11 Feb 2016 10:51:57 +0000 (11:51 +0100)]
Added two aliases for osx
Jens Steube [Thu, 11 Feb 2016 10:36:33 +0000 (11:36 +0100)]
Replace tab with spaces in .hctab so it's easier to view on github
Jens Steube [Thu, 11 Feb 2016 10:34:26 +0000 (11:34 +0100)]
Merge branch 'master' of https://github.com/hashcat/oclHashcat
Jens Steube [Thu, 11 Feb 2016 10:34:09 +0000 (11:34 +0100)]
Update strategy to make it easier to follow
Jens Steube [Thu, 11 Feb 2016 09:08:57 +0000 (10:08 +0100)]
Merge pull request #217 from magnumripper/master
Use device's preferred vector width, not its native one.
jsteube [Thu, 11 Feb 2016 08:54:50 +0000 (09:54 +0100)]
Update event handling to workaround event handling error in nvidia opencl runtime
magnum [Wed, 10 Feb 2016 22:42:04 +0000 (23:42 +0100)]
Use device's preferred vector width, not the native one. Also, don't
assume vector width for 'long' is half of that for 'int'.
Jens Steube [Wed, 10 Feb 2016 22:19:11 +0000 (23:19 +0100)]
Merge pull request #216 from gm4tr1x/revert-214-timerResolution
Revert "Use per-device timer resolution in total_time calculations"
Gabriele 'Matrix' Gristina [Wed, 10 Feb 2016 21:31:43 +0000 (22:31 +0100)]
Revert "Use per-device timer resolution in total_time calculations"
Jens Steube [Wed, 10 Feb 2016 19:40:21 +0000 (20:40 +0100)]
Added new concept of a tuning database; tryout phase
Jens Steube [Wed, 10 Feb 2016 15:56:51 +0000 (16:56 +0100)]
Fixed a bug in line counter: Conditional jump or move depends on uninitialised value
Jens Steube [Wed, 10 Feb 2016 12:31:04 +0000 (13:31 +0100)]
Check for allocatable device-memory depending on kernel_accel amplifier before trying to allocate
Jens Steube [Tue, 9 Feb 2016 23:14:15 +0000 (00:14 +0100)]
Merge pull request #214 from gm4tr1x/timerResolution
Use per-device timer resolution in total_time calculations
Gabriele 'matrix' Gristina [Tue, 9 Feb 2016 20:48:18 +0000 (21:48 +0100)]
Use CL_DEVICE_PROFILING_TIMER_RESOLUTION per-device value in total_time calculations
Jens Steube [Tue, 9 Feb 2016 19:01:50 +0000 (20:01 +0100)]
Added the execution time of the running kernel to the status display
Jens Steube [Mon, 8 Feb 2016 18:49:28 +0000 (19:49 +0100)]
Prepare SIMD for slow hashes macro