Jens Steube
f3bd936971
Add hardware monitor mapping for HIP devices
2021-07-14 08:23:39 +02:00
Jens Steube
219bed457f
Fix use of --gpu-max-threads-per-block
2021-07-12 14:02:43 +02:00
Jens Steube
674ca7d88f
Add GPU threads to kernel cache checksum because it has an influence on HIP offline compile options
...
Add V_ALIGNBIT_B32 inline assembly wrapper because HIP does not provide amd_bitalign()
2021-07-12 11:27:05 +02:00
Jens Steube
23c3c178bf
Limit max threads per block to 64 to enable offline compiler to make better use if available registers
...
Fix double free() for hip_event1/hip_event2 and hip_stream causes segfault
Replace hc_cuCtxSetCurrent() with hc_cuCtxPushCurrent() in order to align changes with HIP
Add vector datatype operators (if we decide to use them - currently unused)
2021-07-12 09:28:26 +02:00
Jens Steube
ca3beacd93
Disable dynamic shared memory on HIP, because hipFuncSetAttribute() maps to cudaFuncSetAttribute() and not to cuFuncSetAttribute()
2021-07-11 14:30:49 +02:00
Jens Steube
1b84a9e53b
Add missing backports from code base v6.2.2
...
Fix context to thread management
Fix missing code in selftest.c, autotune.c, hashes.c, dispatch.c and backend.c
Use IS_HIP depending code makes it easier for future optimization related to inline assembly calls - instead of using IS_CUDA || IS_HIP
See TODO markers for more optimizations / next steps
2021-07-11 12:38:59 +02:00
Jens Steube
bfe83ec138
Added temperature watchdog for CPU on linux using sysfs
2021-07-10 08:43:15 +02:00
Jens Steube
979f9e9868
Rename hardware monitor interface sysfs to sysfs_amdgpu
2021-07-09 20:48:10 +02:00
Jukka Ojanen
a0eaefa0c2
Missing whitespaces
2021-07-05 20:20:51 +03:00
Jukka Ojanen
e133bd4ec4
Change rc_board_name_amd initial value to CL_INVALID_VALUE. If CHECK_BOARD_NAME_AMD is changed to 0, there is a problem with CL_SUCCESS which equals to 0, device will be skipped.
2021-07-05 19:05:10 +03:00
Jukka Ojanen
b3d18f86e2
Fix early return leaks in backend_session_begin
2021-07-05 19:03:56 +03:00
Jukka Ojanen
b976e52bc7
Fix early return leaks in load_kernel; nvrtc_options, nvrtc_options_string, build_log. Ensure build log NULL termination.
2021-07-05 19:00:35 +03:00
Jukka Ojanen
2f7eec2fd7
Fix early return leaks in backend_ctx_init and backend_ctx_devices_init
2021-07-05 15:52:48 +03:00
Jukka Ojanen
bcbb9b0d2c
Fix skipped device param leak in backend_ctx_devices_destroy
2021-07-05 15:38:07 +03:00
Jukka Ojanen
5f109b5862
Fix iconv_ctx and iconv_tmp leaks in backend.c
2021-07-03 12:51:37 +03:00
Jens Steube
8a3eee3fe5
OpenCL Runtime: Workaround JiT crash (SC failed. No reason given.) on macOS by limiting local memory allocations to 32k
2021-06-27 10:18:38 +02:00
Jens Steube
909d5e64a5
Added hash-mode: AES-128/192/256-ECB NOKDF
...
This mode is probably very rare in real-life scenarios,
but it is a nice template for kernels which do
not use a KDF,
or use AES,
or simple fast hashes with lookup tables
or simple optimized kernels in general
2021-06-26 17:12:10 +02:00
Jens Steube
28c08de9ef
Folders: Do not escape the variable cpath_real to prevent certain OpenCL runtimes from running into an error which do not support escape characters
2021-06-24 09:24:02 +02:00
Jens Steube
90f0e78b5b
Add event about autodetection process start
...
Remove "..." substring from final messages
Add patient message on "..." startup messages
Add missing docs/changes.txt entry
2021-06-22 11:37:59 +02:00
Jens Steube
7e267b9b37
Merge pull request #2825 from matrix/hwmon_osx_v2
...
Add OSX HW Monitor initial support (2021)
2021-06-20 11:09:35 +02:00
Jens Steube
2d7f67fd23
Alias Devices: Show a warning in case the user specifically listed a device to use which in a later step is skipped because it is an alias of another active device
2021-06-15 14:06:24 +02:00
Gabriele Gristina
bc4ce4cbeb
Add support for CPU/GPU device temperature and fanspeed using iokit (Apple)
2021-06-12 20:13:31 +02:00
Jens Steube
ee7fca82f5
Scrypt Kernels: Re-enable scrypt based kernels to use kernel cache
2021-06-12 11:42:19 +02:00
Jens Steube
7fc0ac4ef1
Do not initialize backend devices in case --id is used and some bug fixes
2021-06-12 10:47:48 +02:00
Jens Steube
65d81c0f7b
Hashrate: Innerloop hashrate prediction requires update because of the new salt_repeats feature and also respect _loop2 kernel runtime
2021-06-10 22:43:06 +02:00
Jens Steube
fdccc8287d
Fixed free memory size output for skipped GPU (both automatic and manual) of --backend-info information screen
2021-06-07 12:12:01 +02:00
Royce Williams
7bea7ca177
tighten output to be physical-terminal friendly
2021-06-06 10:47:18 -08:00
nycex
470e844e5d
use XDG_CACHE_HOME for kernels
2021-06-05 00:38:43 +02:00
Jens Steube
ce8c121b50
BCRYPT Kernels: Improved bcrypt performance by 6.5% for high-end NVIDIA GPU devices using CUDA backend
2021-06-01 22:52:07 +02:00
Jens Steube
9f5a22a3ab
OpenCL Backend: Use CL_DEVICE_BOARD_NAME_AMD instead of CL_DEVICE_NAME for device name in case OpenCL runtime supports this query
2021-05-26 19:09:50 +02:00
Jens Steube
5c6501444a
Kernels: Add standalone true UTF8 to UTF16 converter kernel that runs after amplifier. Use OPTS_TYPE_POST_AMP_UTF16LE from plugin
2021-05-20 14:34:24 +02:00
Jens Steube
fe91f6276d
CUDA Backend: Do not warn about missing CUDA SDK installation if --stdout is used
2021-05-16 18:31:43 +02:00
Jens Steube
a2fcb03fe3
Update AMD GPU check on macOS warning message
2021-05-11 10:54:39 +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
bbd6e55968
Add missing null pointer to cuda_module_shared and opencl_program_shared in backend cleanup function
2021-05-09 14:55:52 +02:00
Jens Steube
0c2afde83b
Add support for clUnloadPlatformCompiler()
2021-05-02 08:15:25 +00:00
Jens Steube
59459d0e5b
Fixed memory leak causing problems in sessions with many iterations. for instance, --benchmark-all or large mask files
2021-05-01 17:27:33 +02:00
Jens Steube
9a87d5aa01
Fixed out-of-boundary reads in case user activates -S for fast but pure hashes in -a 1 or -a 3 mode
2021-04-27 19:55:30 +02:00
Jens Steube
0ba77fe761
Kernel Development: Kernel cache is disabled automatically in casehashcat is compiled with DEBUG=1
...
See https://github.com/hashcat/hashcat/issues/2750
2021-04-26 09:51:50 +02:00
Jens Steube
1dac869cb7
Removed unnecessary swaps in SCRYPT based algorithms
2021-04-23 20:55:13 +02:00
Jens Steube
15f35fa68c
Scrypt Kernels: Reduced kernel wait times by making it a true split kernel where iteration count = N value
2021-04-21 15:59:14 +02:00
Jens Steube
8e47fdf8f5
Add 4 times single workitem extra buffer size to total extra buffer size to workaround single workitem buffer overflows
2021-04-19 10:27:51 +02:00
Jens Steube
57a8923b81
Update complete SCRYPT workload tuning logic.
...
A detailed description will follow.
Set -m 8900 defaults to 16k:8:1 (default scrypt settings).
2021-04-16 20:17:53 +02:00
Jens Steube
ff96015f53
Add OPTS_TYPE_NATIVE_THREADS for use by plugin developer to enforce native thread count (useful for scrypt)
2021-04-14 15:22:30 +02:00
Jens Steube
67d189e10a
Update calculation of EXTRA_SPACE in backend.c and add upper and lower hard limit
2021-04-13 12:02:52 +02:00
Jens Steube
51e8661070
Update calculation of EXTRA_SPACE in backend.c to make it depending from kernel-accel
2021-04-13 11:47:37 +02:00
Jens Steube
9033975efd
Allow plugins to disable the multiplication of the kernel-accel value with the multiprocessor count of the compute device. Will be used later.
2021-04-13 11:26:17 +02:00
Jens Steube
a0eae9050c
OpenCL Runtime: Workaround JiT compiler deadlock on NVIDIA driver >= 465.89
2021-04-11 13:35:40 +02:00
Jens Steube
3c199bfa1b
Merge pull request #2693 from matrix/out_of_host_memory
...
[backend.c] skipping devices instead of stop with error
2021-04-05 12:58:47 +02:00
Jens Steube
d53913f444
Merge pull request #2672 from matrix/example2info
...
Replaced --example-hashes with --hash-info
2021-04-03 13:25:51 +02:00
