1
0
mirror of https://github.com/hashcat/hashcat.git synced 2025-07-10 00:28:11 +00:00
Commit Graph

38 Commits

Author SHA1 Message Date
Jens Steube
ed10e6a913 Autotune and Benchmark refactoring
This change affects three key areas, each improving autotuning:

- Autotune refactoring itself

The main autotune algorithm had become too complex to maintain and has
now been rewritten from scratch. The engine is now closer to the old
v6.0.0 version, using a much more straightforward approach.

Additionally, the backend is now informed when the autotune engine runs
its operations and runs an extra invisible kernel invocation. This
significantly improves runtime accuracy because the same caching
mechanisms which kick in normal cracking sessions now also apply during
autotuning. This leads to more consistent and reliable automatic
workload tuning.

- Benchmarking and '--speed-only' accuracy bugs fixed

Benchmark runtimes had become too short, especially since the default
benchmark mask changed from '?b?b?b?b?b?b?b' to '?a?a?a?a?a?a?a?a'. For
very fast hashes like NTLM, benchmarks often stopped immediately when
base words needed to be regenerated, producing highly inaccurate
results.

This issue also misled users tuning '-n' values, as manually
oversubscribing kernels could mask the problem, creating the impression
that increasing '-n' had a larger impact on performance than it truly
does. While '-n' still has an effect, it’s not as significant. With this
fix, users achieve the same speed without needing to tune '-n' manually.

The bug was fixed by enforcing a minimum benchmark runtime of 4 seconds,
regardless of kernel runtime or kernel type. This ensures more stable
and realistic benchmark results, but typically increasing the benchmark
duration by up to 4 seconds.

- Kernel-Threads set to 32 and plugin configuration cleanup

Some plugin configurations existed solely to work around the old
benchmarking bug and can now be removed. For example,
'OPTS_TYPE_MAXIMUM_THREADS' is no longer required and has been removed
from all plugins, although the parameter itself remains to avoid
breaking custom plugins.

Because increasing threads beyond 32 no longer offers meaningful
performance gains, the default is now capped at 32 (unless overridden
with '-T'). This simplifies GPU memory management. Currently, work-item
counts are indirectly limited by buffer sizes (e.g., 'pws_buf[]'), which
must not exceed 4 GiB (a hard-coded limit). This buffer size depends on
the product of 'kernel-accel', 'kernel-threads', and the device’s
compute units. By reducing the default threads from 1024 to 32, there is
now more space available for base words.
2025-06-22 20:17:52 +02:00
Jens Steube
ceb5ff5641 The Assimilation Bridge (Framework) 2025-05-29 15:38:13 +02:00
Gabriele Gristina
003579d21b Modules: Updated module_unstable_warning 2025-05-11 15:40:52 +02:00
Rosen Penev
795674c939 fix some const-qual warnings
Signed-off-by: Rosen Penev <rosenp@gmail.com>
2023-08-20 20:55:54 -07:00
Gabriele Gristina
2adf735e51 Fixed bug in input_tokenizer when TOKEN_ATTR_FIXED_LENGTH is used and refactor modules 2023-04-11 19:34:01 +02:00
Jens Steube
8a82c75a99 macOS 13.0 fixed many issues in the OpenCL API. The module_unstable_warning() functions in many hash modes can then be removed 2022-11-20 12:21:41 +01:00
philsmd
d9749e8799
change interface, add module_benchmark_charset () 2022-07-15 17:17:57 +02:00
Jens Steube
6fce6fb3ff Update all existing modules to use the stock module marker 2022-04-08 14:11:50 +02:00
Gabriele Gristina
490702fcfa Backends: added Metal host-code 2022-02-05 22:48:16 +01:00
Jens Steube
5015bc0d2e Module Parser: Renamed struct token_t to hc_token_t to avoid naming conflict with token_t on MacOS 2021-12-20 13:19:40 +01:00
Jens Steube
5b4ac09e91 User Options: Add new module function module_hash_decode_postprocess() to override hash specific configurations from command line 2021-11-28 13:58:27 +01:00
Jens Steube
21f91c5bb8 Module Optimizaters: Added OPTS_TYPE_MAXIMUM_THREADS to deactivate the else branch route in the section to find -T before compilation
Set the new flag based on some testings with RX6900XT
2021-11-13 17:27:06 +01:00
Jens Steube
01738fafa0 Deprecated Plugins: Add new module function module_deprecated_notice() to mark a plugin as deprecated and to return a free text user notice
Added option --deprecated-check-disable to enable deprecated plugins
2021-08-10 17:59:52 +02:00
Jens Steube
20a7b9f992 Tuning-Database: Add new module function module_extra_tuningdb_block() to extend hashcat.hctune content from a plugin
See src/modules/module_08900.c as an example
2021-08-01 16:25:37 +02:00
Jens Steube
7c5fe5e233 Slow Kernels: Set some of the slowest kernels to OPTS_TYPE_MP_MULTI_DISABLE 2021-07-20 12:47:40 +02:00
Jens Steube
141b59b80e
Merge pull request #2883 from hashcat/HIP
Merge HIP branch into master
2021-07-16 19:54:00 +02:00
Jens Steube
11295e4679 Fix missing OPTI_TYPE_USES_BITS_64 in several modules 2021-07-14 17:01:46 +02:00
Jens Steube
7faf6859d6 Backport hand-optimized compiler settings in modules from ROCM to HIP
Backport DECLSPEC settings from ROCM to HIP
2021-07-13 20:45:01 +02:00
Jens Steube
282eb75fe9 Update module_unstable_warning for benchmark long selection on macOS for CPU and GPU; Allow use of GPU without --force testwise 2021-05-10 19:57:56 +02:00
Jens Steube
ff72a8ed21 Remove module_unstable_warning() entries for AMD (legacy) driver after workaround inside UTF16 conversion function is in use 2021-05-08 15:55:32 +00:00
Jens Steube
95489b0473 Update module_unstable_warning() for amdgpu-pro-20.50-1234664-ubuntu-20.04 (legacy) 2021-05-02 18:18:50 +00:00
Jens Steube
9c8509e101 Update module_unstable_warning() on latest Intel OpenCL runtime 2021-04-26 09:25:57 +02:00
Jens Steube
a834574daf Update module_unstable_warning() based on the latest available drivers:
- NVIDIA-Linux-x86_64-460.73.01
- amdgpu-pro-20.50-1234664-ubuntu-20.04 (rocr for new gpu, legacy for
old gpu)
- w_opencl_runtime_p_2021.2.0.616.exe

Not tested:

- Apple *
- l_opencl_p_18.1.0.015.tgz
- Intel GPU
- POCL

Update docs/license.txt
2021-04-18 14:41:44 +02:00
Jens Steube
98aef2ae92 Module Structure: Add 3rd party library hook management functions. This also requires an update to all existing module_init() 2020-08-29 16:12:15 +02:00
Jens Steube
b9f6777f1b OpenCL Runtime: Add some unstable warnings for some SHA512 based algorithms on AMD GPU on macOS 2020-07-15 11:27:46 +02:00
Jens Steube
193aa77cdf Update unroll and unstable configuration for ROCM OpenCL runtime after unroll whitelisting change 2020-02-15 16:09:37 +01:00
Jens Steube
c90d83c3eb Prepare for UNROLL whitelisting 2020-02-15 12:44:12 +01:00
Jens Steube
8a905d19aa Updated optimized kernel unstable markers to amdgpu-pro-18.50-708488-ubuntu-18.04 2020-01-13 16:06:04 +01:00
Chick3nman
253db764b7 Fixed issue where multiple hashes with the same salt would fail to crack in module/kernel for 9500. Remove unused include in module for 9600. 2019-09-05 05:27:39 -05:00
Jens Steube
f1632b933e Add support to configure hash-mode specific range of number of hashes supported 2019-05-19 14:46:05 +02:00
jsteube
926e99811c Add some more NO_UNROLL to avoid module_unstable_warnings 2019-04-20 16:36:43 +02:00
jsteube
b8d609ba16 WPA/WPA2 cracking: In the potfile, replace password with PMK in order to detect already cracked networks across all WPA modes 2019-04-02 11:24:22 +02:00
jsteube
c0a31b3239 Prepare potfile specific module_hash_decode and module_hash_encode hooks 2019-04-01 12:32:11 +02:00
jsteube
7806be32c2 Add module_unstable_warning() for several modes used with AMDGPU drivers 2019-03-27 12:05:34 +01:00
Jens Steube
b585f25c8c Move more no unrolling exceptions from inc_vendor.cl to specific modules 2019-03-08 09:45:51 +01:00
jsteube
88a051629c Support module specific JiT compiler build options 2019-03-02 11:12:13 +01:00
jsteube
bab735b367 Get rid of hash_type variable. This hopefully reduces some confusion for new hashcat kernel developers 2019-02-12 16:02:27 +01:00
jsteube
c9e796fcf0 Add some more modules converted 2019-02-08 13:00:40 +01:00