jsteube
c4f30220a0
OpenCL Kernels: Refactored kernel thread management from native to maximum per kernel
2018-02-11 21:29:57 +01:00
Jens Steube
d38608b9bc
Add get_kernel_preferred_wgs_multiple() for future use
2018-02-11 16:22:24 +01:00
Jens Steube
1f442b5b0c
Add get_kernel_local_mem_size() for future use
2018-02-11 16:09:50 +01:00
Jens Steube
d656e9c3a4
OpenCL Kernels: Use the kernel local buffer size as additional reference in order to limit the thread-count
2018-02-11 10:56:08 +01:00
Jens Steube
9b1e66d87c
Fixed a race condition in combinator- and hybrid-mode where the same scratch buffer was used by multiple threads
2018-02-09 09:50:13 +01:00
jsteube
05a01d3843
fix some datatypes
2018-02-08 19:13:29 +01:00
jsteube
a3a16f676f
OpenCL Kernels: Add a decompressing kernel and a compressing host code in order to reduce PCIe transfer time
...
For details see https://hashcat.net/forum/thread-7267.html
2018-02-05 17:18:58 +01:00
jsteube
11b18512c7
Terminal: Add workitem settings to status display (can be handy for debugging)
2018-02-02 23:22:21 +01:00
Jens Steube
7062425d2b
OpenCL Kernels: Use a special kernel to initialize the password buffer used during autotune measurements to reduce startup time
2018-01-23 20:33:26 +01:00
jsteube
0796c074c3
Added -m 16500 Kernels
...
Also changed function declaration of parser function from const hashconfig_t to just hashconfig_t
2018-01-21 18:53:55 +01:00
Gabriele Gristina
03fab4a345
Fix clock_gettime() on OSX (v2)
2018-01-11 02:41:16 +01:00
Jens Steube
c41af948ba
Merge pull request #1471 from neheb/32to64
...
remove special time handling in hashcat.
2017-12-11 16:27:00 +01:00
jsteube
bbe9b723e1
Timer: Switch from gettimeofday() to clock_gettime() to workaround problems on cygwin
2017-12-10 14:02:43 +01:00
Rosen Penev
1c5915cae3
remove special time handling in hashcat.
...
There's a way to get 64-bit time_t and 64-bit stat in MinGW, so use that.
2017-12-09 16:40:45 -08:00
jsteube
494c2724f5
Add some types for later use
2017-12-09 21:39:10 +01:00
jsteube
ea795ef5ab
Move out code to check for hashfile or hash on commandline into dedicated function
2017-12-06 16:12:34 +01:00
jsteube
1184ae1cdd
Added option --benchmark-all to benchmark all hash-modes not just the default selection
2017-12-05 11:08:59 +01:00
jsteube
b864fa5ba7
Removed option --gpu-temp-retain that tried to retain GPU temperature at X degrees celsius, please use driver specific tools
...
Removed option --powertune-enable to enable power tuning, please use driver specific tools
2017-12-01 15:19:10 +01:00
Rosen Penev
353d3c3008
Fix a bunch of cast-qual warnings
2017-11-13 20:46:26 -08:00
Rosen Penev
8458f07a57
Replace char[] with *char
...
The latter is more correct. The side effect is that filesize becomes smaller.
gcc clang
before
O2
756112 762120
Os
674192 680216
after
O2
748456 760728
Os
666536 674728
2017-11-13 12:12:44 -08:00
jsteube
635b2bfd50
Do initial OpenCL device thread management a bit different
2017-10-19 14:54:03 +02:00
philsmd
276879b1ea
show/left: improve speed of --show/--left again by using a single malloc/free for all nodes and reduce RAM by using the first nodes hash information
2017-10-18 12:09:32 +02:00
philsmd
433513a47f
get rid of last_node in pot_tree_entry, we update the first node of the list instead of the last one
2017-10-18 08:54:21 +02:00
philsmd
6542331101
show/left: improved the performance by using a tree and linked node structure
2017-10-17 12:08:17 +02:00
jsteube
2517292ac1
Removed option --weak-hash-check (zero-length password check) to increase startup time, it also causes many Trap 6 error on OSX
2017-09-19 12:04:05 +02:00
jsteube
7a278ef035
Fix overflow in out_push()
2017-09-17 12:14:52 +02:00
jsteube
5551337cf6
Show advice to use optimized kernels only if there's actually one for this hash-mode
2017-09-16 22:33:04 +02:00
Jens Steube
fcd20fbacb
OpenCL Runtime: Fall back to 64 threads default (from 256) on AMD GPU to prevent creating too many workitems
2017-08-29 17:36:50 +02:00
jsteube
688a580d6c
Fixed an integer overflow in innerloop_step and innerloop_cnt variables
2017-08-25 20:52:13 +02:00
jsteube
1d04de3a8e
Limit kernel-loops in straight-mode to 256, therefore allow rules to be stored in constant memory
2017-08-23 12:43:59 +02:00
jsteube
51372438fe
Allow OpenCL kernel inline assembly if ROCm drivers was detected
2017-08-22 18:47:53 +02:00
jsteube
12d95fd22c
Added option --example-hashes to show an example hash for each hash-mode
2017-08-22 11:09:46 +02:00
jsteube
58d1dedd1e
OpenCL Buffers: Do not allocate memory for amplifiers for fast hashes, it's simply not needed
2017-08-19 19:25:37 +02:00
jsteube
319799bbbf
Switch the datatypes of the variables responsible for work-item count and work-item size from u32 to u64
2017-08-19 16:39:22 +02:00
philsmd
e33e828bc6
replace all time_t/__time64_t with hc_time_t and all related functions
2017-08-16 19:43:41 +02:00
philsmd
1644ac30b5
use an overflow check instead of a hard coded value for ETA
2017-08-16 17:51:35 +02:00
jsteube
ac9bc55c9e
Fall back to 64 threads default (from 256) because modern GPU create too many workitems which then results in too less GPU memory available
2017-08-16 16:13:28 +02:00
jsteube
1cf6a078b0
Make tmp_size a real size_t as with latest GPU number can exceed 2^32
2017-08-16 13:03:52 +02:00
philsmd
bc1a101d2b
fixes #1321 : new option --wordlist-autohex-disable
2017-08-16 12:42:28 +02:00
jsteube
b847bbb274
Fix calculation of device_name_chksum; should be done for each iteration
2017-07-22 18:46:12 +02:00
jsteube
beb6ee2061
Add OPTI_TYPE_OPTIMIZED_KERNEL
...
Rename unconverted fast hash kernels to optimized kernels
Finalize some converted fast hashes to default kernels
2017-07-18 14:45:15 +02:00
jsteube
fbea72ebd6
Renamed default kernels to optimized kernels
...
Renamed pure kernels to default kernels
Replaced long option --length-limit-disable with --optimized-kernel-enable
Replaced short option -L with -O
Set --optimized-kernel-enable to unset by default
2017-07-18 13:23:42 +02:00
jsteube
967d7b9323
Rename SALT_TYPE_INTERN to SALT_TYPE_GENERIC
2017-07-14 14:57:31 +02:00
jsteube
f70da8a04c
Increase salt length for salt of generic hash types
...
Remove --hex-salt support for non-generic hash types
2017-07-13 11:03:57 +02:00
jsteube
f7a8e7c54b
Multiple changes:
...
* Added more preparations to support to crack passwords and salts up to length 256
* Added option --length-limit-disable to disable optimization based on password- and salt-length
* Added option --self-test-disable to disable self-test functionality on startup
2017-06-29 12:19:05 +02:00
jsteube
c918173fcf
Get rid of comb_t which can be safely replace with pw_t now
2017-06-25 00:56:25 +02:00
jsteube
4174f06008
PoC using a length-independant MD4 hash processing in -m 2100
2017-06-18 23:31:40 +02:00
jsteube
a673aee037
Very hot commit, continue reading here:
...
This is a test commit using buffers large enough to handle both passwords and salts up to length 256.
It requires changes to the kernel code, which is not included in here.
It also requires some of the host code to be modified. Before we're going to modify kernel code to support the larger lengths I want to be
sure of:
1. Host code modification is ok (no overflows or underflows)
2. Passwords and Salts are printed correctly to status, outfile, show, left, etc.
3. Performance does not change (or only very minimal)
This is not a patch that supports actual cracking both passwords and salts up to length 256, but it can not fail anyway.
If if it does, there's no reason to continue to add support for both passwords and salts up to length 256.
2017-06-17 17:57:30 +02:00
jsteube
94a35ae50a
Add support for hooks in selftest function
2017-06-14 12:07:33 +02:00
Jens Steube
08fc0ec1fb
Added self-test funcionality for OpenCL kernels on startup
...
Fixed a parser error for mode -m 9820 = MS Office <= 2003 $3, SHA1 + RC4, collider #2
2017-06-13 19:07:08 +02:00
