/** * Author......: See docs/credits.txt * License.....: MIT */ #include "common.h" #include "types.h" #include "timer.h" #include "memory.h" #include "logging.h" #include "locking.h" #include "ext_ADL.h" #include "ext_nvapi.h" #include "ext_nvml.h" #include "ext_xnvctrl.h" #include "ext_OpenCL.h" #include "cpu_md5.h" #include "interface.h" #include "tuningdb.h" #include "thread.h" #include "opencl.h" #include "hwmon.h" #include "restore.h" #include "hash_management.h" #include "thread.h" #include "status.h" #include "stdout.h" #include "mpsp.h" #include "rp_cpu.h" #include "outfile.h" #include "potfile.h" #include "debugfile.h" #include "loopback.h" #include "filenames.h" #include "data.h" #include "shared.h" #include "filehandling.h" #include "convert.h" #include "dictstat.h" #include "wordlist.h" extern hc_global_data_t data; extern hc_thread_mutex_t mux_counter; extern hc_thread_mutex_t mux_hwmon; extern const int comptime; char *strstatus (const uint devices_status) { switch (devices_status) { case STATUS_INIT: return ((char *) ST_0000); case STATUS_AUTOTUNE: return ((char *) ST_0001); case STATUS_RUNNING: return ((char *) ST_0002); case STATUS_PAUSED: return ((char *) ST_0003); case STATUS_EXHAUSTED: return ((char *) ST_0004); case STATUS_CRACKED: return ((char *) ST_0005); case STATUS_ABORTED: return ((char *) ST_0006); case STATUS_QUIT: return ((char *) ST_0007); case STATUS_BYPASS: return ((char *) ST_0008); } return ((char *) "Uninitialized! Bug!"); } static uint setup_opencl_platforms_filter (const char *opencl_platforms) { uint opencl_platforms_filter = 0; if (opencl_platforms) { char *platforms = mystrdup (opencl_platforms); char *next = strtok (platforms, ","); do { int platform = atoi (next); if (platform < 1 || platform > 32) { log_error ("ERROR: Invalid OpenCL platform %u specified", platform); exit (-1); } opencl_platforms_filter |= 1u << (platform - 1); } while ((next = strtok (NULL, ",")) != NULL); myfree (platforms); } else { opencl_platforms_filter = -1u; } return opencl_platforms_filter; } static u32 setup_devices_filter (const char *opencl_devices) { u32 devices_filter = 0; if (opencl_devices) { char *devices = mystrdup (opencl_devices); char *next = strtok (devices, ","); do { int device_id = atoi (next); if (device_id < 1 || device_id > 32) { log_error ("ERROR: Invalid device_id %u specified", device_id); exit (-1); } devices_filter |= 1u << (device_id - 1); } while ((next = strtok (NULL, ",")) != NULL); myfree (devices); } else { devices_filter = -1u; } return devices_filter; } static cl_device_type setup_device_types_filter (const char *opencl_device_types) { cl_device_type device_types_filter = 0; if (opencl_device_types) { char *device_types = mystrdup (opencl_device_types); char *next = strtok (device_types, ","); do { int device_type = atoi (next); if (device_type < 1 || device_type > 3) { log_error ("ERROR: Invalid device_type %u specified", device_type); exit (-1); } device_types_filter |= 1u << device_type; } while ((next = strtok (NULL, ",")) != NULL); myfree (device_types); } else { // Do not use CPU by default, this often reduces GPU performance because // the CPU is too busy to handle GPU synchronization device_types_filter = CL_DEVICE_TYPE_ALL & ~CL_DEVICE_TYPE_CPU; } return device_types_filter; } void load_kernel (const char *kernel_file, int num_devices, size_t *kernel_lengths, const u8 **kernel_sources) { FILE *fp = fopen (kernel_file, "rb"); if (fp != NULL) { struct stat st; memset (&st, 0, sizeof (st)); stat (kernel_file, &st); u8 *buf = (u8 *) mymalloc (st.st_size + 1); size_t num_read = fread (buf, sizeof (u8), st.st_size, fp); if (num_read != (size_t) st.st_size) { log_error ("ERROR: %s: %s", kernel_file, strerror (errno)); exit (-1); } fclose (fp); buf[st.st_size] = 0; for (int i = 0; i < num_devices; i++) { kernel_lengths[i] = (size_t) st.st_size; kernel_sources[i] = buf; } } else { log_error ("ERROR: %s: %s", kernel_file, strerror (errno)); exit (-1); } return; } void writeProgramBin (char *dst, u8 *binary, size_t binary_size) { if (binary_size > 0) { FILE *fp = fopen (dst, "wb"); lock_file (fp); fwrite (binary, sizeof (u8), binary_size, fp); fflush (fp); fclose (fp); } } int gidd_to_pw_t (opencl_ctx_t *opencl_ctx, hc_device_param_t *device_param, const u64 gidd, pw_t *pw) { cl_int CL_err = hc_clEnqueueReadBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_pws_buf, CL_TRUE, gidd * sizeof (pw_t), sizeof (pw_t), pw, 0, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueReadBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } return 0; } int choose_kernel (opencl_ctx_t *opencl_ctx, hc_device_param_t *device_param, hashconfig_t *hashconfig, const uint attack_exec, const uint attack_mode, const uint opts_type, const salt_t *salt_buf, const uint highest_pw_len, const uint pws_cnt, const uint fast_iteration) { cl_int CL_err = CL_SUCCESS; if (hashconfig->hash_mode == 2000) { process_stdout (opencl_ctx, device_param, pws_cnt); return 0; } if (attack_exec == ATTACK_EXEC_INSIDE_KERNEL) { if (attack_mode == ATTACK_MODE_BF) { if (opts_type & OPTS_TYPE_PT_BITSLICE) { const uint size_tm = 32 * sizeof (bs_word_t); run_kernel_bzero (opencl_ctx, device_param, device_param->d_tm_c, size_tm); run_kernel_tm (opencl_ctx, device_param); CL_err = hc_clEnqueueCopyBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_tm_c, device_param->d_bfs_c, 0, 0, size_tm, 0, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueCopyBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } } } if (highest_pw_len < 16) { run_kernel (KERN_RUN_1, opencl_ctx, device_param, pws_cnt, true, fast_iteration, hashconfig); } else if (highest_pw_len < 32) { run_kernel (KERN_RUN_2, opencl_ctx, device_param, pws_cnt, true, fast_iteration, hashconfig); } else { run_kernel (KERN_RUN_3, opencl_ctx, device_param, pws_cnt, true, fast_iteration, hashconfig); } } else { run_kernel_amp (opencl_ctx, device_param, pws_cnt); run_kernel (KERN_RUN_1, opencl_ctx, device_param, pws_cnt, false, 0, hashconfig); if (opts_type & OPTS_TYPE_HOOK12) { run_kernel (KERN_RUN_12, opencl_ctx, device_param, pws_cnt, false, 0, hashconfig); CL_err = hc_clEnqueueReadBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_hooks, CL_TRUE, 0, device_param->size_hooks, device_param->hooks_buf, 0, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueReadBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } // do something with data CL_err = hc_clEnqueueWriteBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_hooks, CL_TRUE, 0, device_param->size_hooks, device_param->hooks_buf, 0, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueWriteBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } } uint iter = salt_buf->salt_iter; uint loop_step = device_param->kernel_loops; for (uint loop_pos = 0, slow_iteration = 0; loop_pos < iter; loop_pos += loop_step, slow_iteration++) { uint loop_left = iter - loop_pos; loop_left = MIN (loop_left, loop_step); device_param->kernel_params_buf32[28] = loop_pos; device_param->kernel_params_buf32[29] = loop_left; run_kernel (KERN_RUN_2, opencl_ctx, device_param, pws_cnt, true, slow_iteration, hashconfig); while (opencl_ctx->run_thread_level2 == false) break; /** * speed */ const float iter_part = (float) (loop_pos + loop_left) / iter; const u64 perf_sum_all = (u64) (pws_cnt * iter_part); double speed_ms; hc_timer_get (device_param->timer_speed, speed_ms); const u32 speed_pos = device_param->speed_pos; device_param->speed_cnt[speed_pos] = perf_sum_all; device_param->speed_ms[speed_pos] = speed_ms; if (data.benchmark == 1) { if (speed_ms > 4096) myabort (opencl_ctx); } } if (opts_type & OPTS_TYPE_HOOK23) { run_kernel (KERN_RUN_23, opencl_ctx, device_param, pws_cnt, false, 0, hashconfig); CL_err = hc_clEnqueueReadBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_hooks, CL_TRUE, 0, device_param->size_hooks, device_param->hooks_buf, 0, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueReadBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } // do something with data CL_err = hc_clEnqueueWriteBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_hooks, CL_TRUE, 0, device_param->size_hooks, device_param->hooks_buf, 0, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueWriteBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } } run_kernel (KERN_RUN_3, opencl_ctx, device_param, pws_cnt, false, 0, hashconfig); } return 0; } int run_kernel (const uint kern_run, opencl_ctx_t *opencl_ctx, hc_device_param_t *device_param, const uint num, const uint event_update, const uint iteration, hashconfig_t *hashconfig) { cl_int CL_err = CL_SUCCESS; uint num_elements = num; device_param->kernel_params_buf32[33] = data.combs_mode; device_param->kernel_params_buf32[34] = num; uint kernel_threads = device_param->kernel_threads; while (num_elements % kernel_threads) num_elements++; cl_kernel kernel = NULL; switch (kern_run) { case KERN_RUN_1: kernel = device_param->kernel1; break; case KERN_RUN_12: kernel = device_param->kernel12; break; case KERN_RUN_2: kernel = device_param->kernel2; break; case KERN_RUN_23: kernel = device_param->kernel23; break; case KERN_RUN_3: kernel = device_param->kernel3; break; } CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 24, sizeof (cl_uint), device_param->kernel_params[24]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 25, sizeof (cl_uint), device_param->kernel_params[25]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 26, sizeof (cl_uint), device_param->kernel_params[26]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 27, sizeof (cl_uint), device_param->kernel_params[27]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 28, sizeof (cl_uint), device_param->kernel_params[28]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 29, sizeof (cl_uint), device_param->kernel_params[29]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 30, sizeof (cl_uint), device_param->kernel_params[30]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 31, sizeof (cl_uint), device_param->kernel_params[31]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 32, sizeof (cl_uint), device_param->kernel_params[32]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 33, sizeof (cl_uint), device_param->kernel_params[33]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 34, sizeof (cl_uint), device_param->kernel_params[34]); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clSetKernelArg(): %s\n", val2cstr_cl (CL_err)); return -1; } cl_event event; if ((hashconfig->opts_type & OPTS_TYPE_PT_BITSLICE) && (data.attack_mode == ATTACK_MODE_BF)) { const size_t global_work_size[3] = { num_elements, 32, 1 }; const size_t local_work_size[3] = { kernel_threads / 32, 32, 1 }; CL_err = hc_clEnqueueNDRangeKernel (opencl_ctx->ocl, device_param->command_queue, kernel, 2, NULL, global_work_size, local_work_size, 0, NULL, &event); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueNDRangeKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } } else { if (kern_run == KERN_RUN_2) { if (hashconfig->opti_type & OPTI_TYPE_SLOW_HASH_SIMD) { num_elements = CEIL (num_elements / device_param->vector_width); } } while (num_elements % kernel_threads) num_elements++; const size_t global_work_size[3] = { num_elements, 1, 1 }; const size_t local_work_size[3] = { kernel_threads, 1, 1 }; CL_err = hc_clEnqueueNDRangeKernel (opencl_ctx->ocl, device_param->command_queue, kernel, 1, NULL, global_work_size, local_work_size, 0, NULL, &event); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueNDRangeKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } } CL_err = hc_clFlush (opencl_ctx->ocl, device_param->command_queue); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clFlush(): %s\n", val2cstr_cl (CL_err)); return -1; } if (device_param->nvidia_spin_damp > 0) { if (opencl_ctx->devices_status == STATUS_RUNNING) { if (iteration < EXPECTED_ITERATIONS) { switch (kern_run) { case KERN_RUN_1: if (device_param->exec_us_prev1[iteration] > 0) usleep ((useconds_t)(device_param->exec_us_prev1[iteration] * device_param->nvidia_spin_damp)); break; case KERN_RUN_2: if (device_param->exec_us_prev2[iteration] > 0) usleep ((useconds_t)(device_param->exec_us_prev2[iteration] * device_param->nvidia_spin_damp)); break; case KERN_RUN_3: if (device_param->exec_us_prev3[iteration] > 0) usleep ((useconds_t)(device_param->exec_us_prev3[iteration] * device_param->nvidia_spin_damp)); break; } } } } CL_err = hc_clWaitForEvents (opencl_ctx->ocl, 1, &event); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clWaitForEvents(): %s\n", val2cstr_cl (CL_err)); return -1; } cl_ulong time_start; cl_ulong time_end; CL_err |= hc_clGetEventProfilingInfo (opencl_ctx->ocl, event, CL_PROFILING_COMMAND_START, sizeof (time_start), &time_start, NULL); CL_err |= hc_clGetEventProfilingInfo (opencl_ctx->ocl, event, CL_PROFILING_COMMAND_END, sizeof (time_end), &time_end, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetEventProfilingInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } const double exec_us = (double) (time_end - time_start) / 1000; if (opencl_ctx->devices_status == STATUS_RUNNING) { if (iteration < EXPECTED_ITERATIONS) { switch (kern_run) { case KERN_RUN_1: device_param->exec_us_prev1[iteration] = exec_us; break; case KERN_RUN_2: device_param->exec_us_prev2[iteration] = exec_us; break; case KERN_RUN_3: device_param->exec_us_prev3[iteration] = exec_us; break; } } } if (event_update) { uint exec_pos = device_param->exec_pos; device_param->exec_ms[exec_pos] = exec_us / 1000; exec_pos++; if (exec_pos == EXEC_CACHE) { exec_pos = 0; } device_param->exec_pos = exec_pos; } CL_err = hc_clReleaseEvent (opencl_ctx->ocl, event); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clReleaseEvent(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clFinish (opencl_ctx->ocl, device_param->command_queue); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clFinish(): %s\n", val2cstr_cl (CL_err)); return -1; } return 0; } int run_kernel_mp (const uint kern_run, opencl_ctx_t *opencl_ctx, hc_device_param_t *device_param, const uint num) { cl_int CL_err = CL_SUCCESS; uint num_elements = num; switch (kern_run) { case KERN_RUN_MP: device_param->kernel_params_mp_buf32[8] = num; break; case KERN_RUN_MP_R: device_param->kernel_params_mp_r_buf32[8] = num; break; case KERN_RUN_MP_L: device_param->kernel_params_mp_l_buf32[9] = num; break; } // causes problems with special threads like in bcrypt // const uint kernel_threads = device_param->kernel_threads; uint kernel_threads = device_param->kernel_threads; while (num_elements % kernel_threads) num_elements++; cl_kernel kernel = NULL; switch (kern_run) { case KERN_RUN_MP: kernel = device_param->kernel_mp; break; case KERN_RUN_MP_R: kernel = device_param->kernel_mp_r; break; case KERN_RUN_MP_L: kernel = device_param->kernel_mp_l; break; } switch (kern_run) { case KERN_RUN_MP: CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 3, sizeof (cl_ulong), device_param->kernel_params_mp[3]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 4, sizeof (cl_uint), device_param->kernel_params_mp[4]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 5, sizeof (cl_uint), device_param->kernel_params_mp[5]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 6, sizeof (cl_uint), device_param->kernel_params_mp[6]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 7, sizeof (cl_uint), device_param->kernel_params_mp[7]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 8, sizeof (cl_uint), device_param->kernel_params_mp[8]); break; case KERN_RUN_MP_R: CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 3, sizeof (cl_ulong), device_param->kernel_params_mp_r[3]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 4, sizeof (cl_uint), device_param->kernel_params_mp_r[4]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 5, sizeof (cl_uint), device_param->kernel_params_mp_r[5]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 6, sizeof (cl_uint), device_param->kernel_params_mp_r[6]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 7, sizeof (cl_uint), device_param->kernel_params_mp_r[7]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 8, sizeof (cl_uint), device_param->kernel_params_mp_r[8]); break; case KERN_RUN_MP_L: CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 3, sizeof (cl_ulong), device_param->kernel_params_mp_l[3]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 4, sizeof (cl_uint), device_param->kernel_params_mp_l[4]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 5, sizeof (cl_uint), device_param->kernel_params_mp_l[5]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 6, sizeof (cl_uint), device_param->kernel_params_mp_l[6]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 7, sizeof (cl_uint), device_param->kernel_params_mp_l[7]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 8, sizeof (cl_uint), device_param->kernel_params_mp_l[8]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 9, sizeof (cl_uint), device_param->kernel_params_mp_l[9]); break; } if (CL_err != CL_SUCCESS) { log_error ("ERROR: clSetKernelArg(): %s\n", val2cstr_cl (CL_err)); return -1; } const size_t global_work_size[3] = { num_elements, 1, 1 }; const size_t local_work_size[3] = { kernel_threads, 1, 1 }; CL_err = hc_clEnqueueNDRangeKernel (opencl_ctx->ocl, device_param->command_queue, kernel, 1, NULL, global_work_size, local_work_size, 0, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueNDRangeKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clFlush (opencl_ctx->ocl, device_param->command_queue); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clFlush(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clFinish (opencl_ctx->ocl, device_param->command_queue); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clFinish(): %s\n", val2cstr_cl (CL_err)); return -1; } return 0; } int run_kernel_tm (opencl_ctx_t *opencl_ctx, hc_device_param_t *device_param) { cl_int CL_err = CL_SUCCESS; const uint num_elements = 1024; // fixed uint kernel_threads = 32; cl_kernel kernel = device_param->kernel_tm; const size_t global_work_size[3] = { num_elements, 1, 1 }; const size_t local_work_size[3] = { kernel_threads, 1, 1 }; CL_err = hc_clEnqueueNDRangeKernel (opencl_ctx->ocl, device_param->command_queue, kernel, 1, NULL, global_work_size, local_work_size, 0, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueNDRangeKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clFlush (opencl_ctx->ocl, device_param->command_queue); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clFlush(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clFinish (opencl_ctx->ocl, device_param->command_queue); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clFinish(): %s\n", val2cstr_cl (CL_err)); return -1; } return 0; } int run_kernel_amp (opencl_ctx_t *opencl_ctx, hc_device_param_t *device_param, const uint num) { cl_int CL_err = CL_SUCCESS; uint num_elements = num; device_param->kernel_params_amp_buf32[5] = data.combs_mode; device_param->kernel_params_amp_buf32[6] = num_elements; // causes problems with special threads like in bcrypt // const uint kernel_threads = device_param->kernel_threads; uint kernel_threads = device_param->kernel_threads; while (num_elements % kernel_threads) num_elements++; cl_kernel kernel = device_param->kernel_amp; CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 5, sizeof (cl_uint), device_param->kernel_params_amp[5]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 6, sizeof (cl_uint), device_param->kernel_params_amp[6]); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clSetKernelArg(): %s\n", val2cstr_cl (CL_err)); return -1; } const size_t global_work_size[3] = { num_elements, 1, 1 }; const size_t local_work_size[3] = { kernel_threads, 1, 1 }; CL_err = hc_clEnqueueNDRangeKernel (opencl_ctx->ocl, device_param->command_queue, kernel, 1, NULL, global_work_size, local_work_size, 0, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueNDRangeKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clFlush (opencl_ctx->ocl, device_param->command_queue); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clFlush(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clFinish (opencl_ctx->ocl, device_param->command_queue); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clFinish(): %s\n", val2cstr_cl (CL_err)); return -1; } return 0; } int run_kernel_memset (opencl_ctx_t *opencl_ctx, hc_device_param_t *device_param, cl_mem buf, const uint value, const uint num) { cl_int CL_err = CL_SUCCESS; const u32 num16d = num / 16; const u32 num16m = num % 16; if (num16d) { device_param->kernel_params_memset_buf32[1] = value; device_param->kernel_params_memset_buf32[2] = num16d; uint kernel_threads = device_param->kernel_threads; uint num_elements = num16d; while (num_elements % kernel_threads) num_elements++; cl_kernel kernel = device_param->kernel_memset; CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 0, sizeof (cl_mem), (void *) &buf); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 1, sizeof (cl_uint), device_param->kernel_params_memset[1]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, kernel, 2, sizeof (cl_uint), device_param->kernel_params_memset[2]); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clSetKernelArg(): %s\n", val2cstr_cl (CL_err)); return -1; } const size_t global_work_size[3] = { num_elements, 1, 1 }; const size_t local_work_size[3] = { kernel_threads, 1, 1 }; CL_err = hc_clEnqueueNDRangeKernel (opencl_ctx->ocl, device_param->command_queue, kernel, 1, NULL, global_work_size, local_work_size, 0, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueNDRangeKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clFlush (opencl_ctx->ocl, device_param->command_queue); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clFlush(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clFinish (opencl_ctx->ocl, device_param->command_queue); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clFinish(): %s\n", val2cstr_cl (CL_err)); return -1; } } if (num16m) { u32 tmp[4]; tmp[0] = value; tmp[1] = value; tmp[2] = value; tmp[3] = value; CL_err = hc_clEnqueueWriteBuffer (opencl_ctx->ocl, device_param->command_queue, buf, CL_TRUE, num16d * 16, num16m, tmp, 0, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueWriteBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } } return 0; } int run_kernel_bzero (opencl_ctx_t *opencl_ctx, hc_device_param_t *device_param, cl_mem buf, const size_t size) { return run_kernel_memset (opencl_ctx, device_param, buf, 0, size); } int run_copy (opencl_ctx_t *opencl_ctx, hc_device_param_t *device_param, hashconfig_t *hashconfig, const uint pws_cnt) { cl_int CL_err = CL_SUCCESS; if (data.attack_kern == ATTACK_KERN_STRAIGHT) { CL_err = hc_clEnqueueWriteBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_pws_buf, CL_TRUE, 0, pws_cnt * sizeof (pw_t), device_param->pws_buf, 0, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueWriteBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } } else if (data.attack_kern == ATTACK_KERN_COMBI) { if (data.attack_mode == ATTACK_MODE_COMBI) { if (data.combs_mode == COMBINATOR_MODE_BASE_RIGHT) { if (hashconfig->opts_type & OPTS_TYPE_PT_ADD01) { for (u32 i = 0; i < pws_cnt; i++) { const u32 pw_len = device_param->pws_buf[i].pw_len; u8 *ptr = (u8 *) device_param->pws_buf[i].i; ptr[pw_len] = 0x01; } } else if (hashconfig->opts_type & OPTS_TYPE_PT_ADD80) { for (u32 i = 0; i < pws_cnt; i++) { const u32 pw_len = device_param->pws_buf[i].pw_len; u8 *ptr = (u8 *) device_param->pws_buf[i].i; ptr[pw_len] = 0x80; } } } } else if (data.attack_mode == ATTACK_MODE_HYBRID2) { if (hashconfig->opts_type & OPTS_TYPE_PT_ADD01) { for (u32 i = 0; i < pws_cnt; i++) { const u32 pw_len = device_param->pws_buf[i].pw_len; u8 *ptr = (u8 *) device_param->pws_buf[i].i; ptr[pw_len] = 0x01; } } else if (hashconfig->opts_type & OPTS_TYPE_PT_ADD80) { for (u32 i = 0; i < pws_cnt; i++) { const u32 pw_len = device_param->pws_buf[i].pw_len; u8 *ptr = (u8 *) device_param->pws_buf[i].i; ptr[pw_len] = 0x80; } } } CL_err = hc_clEnqueueWriteBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_pws_buf, CL_TRUE, 0, pws_cnt * sizeof (pw_t), device_param->pws_buf, 0, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueWriteBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } } else if (data.attack_kern == ATTACK_KERN_BF) { const u64 off = device_param->words_off; device_param->kernel_params_mp_l_buf64[3] = off; run_kernel_mp (KERN_RUN_MP_L, opencl_ctx, device_param, pws_cnt); } return 0; } int run_cracker (opencl_ctx_t *opencl_ctx, hc_device_param_t *device_param, hashconfig_t *hashconfig, hashes_t *hashes, const uint pws_cnt) { char *line_buf = (char *) mymalloc (HCBUFSIZ_LARGE); // init speed timer uint speed_pos = device_param->speed_pos; #if defined (_POSIX) if (device_param->timer_speed.tv_sec == 0) { hc_timer_set (&device_param->timer_speed); } #endif #if defined (_WIN) if (device_param->timer_speed.QuadPart == 0) { hc_timer_set (&device_param->timer_speed); } #endif // find higest password length, this is for optimization stuff uint highest_pw_len = 0; if (data.attack_kern == ATTACK_KERN_STRAIGHT) { } else if (data.attack_kern == ATTACK_KERN_COMBI) { } else if (data.attack_kern == ATTACK_KERN_BF) { highest_pw_len = device_param->kernel_params_mp_l_buf32[4] + device_param->kernel_params_mp_l_buf32[5]; } // loop start: most outer loop = salt iteration, then innerloops (if multi) for (uint salt_pos = 0; salt_pos < hashes->salts_cnt; salt_pos++) { while (opencl_ctx->devices_status == STATUS_PAUSED) hc_sleep (1); salt_t *salt_buf = &hashes->salts_buf[salt_pos]; device_param->kernel_params_buf32[27] = salt_pos; device_param->kernel_params_buf32[31] = salt_buf->digests_cnt; device_param->kernel_params_buf32[32] = salt_buf->digests_offset; FILE *combs_fp = device_param->combs_fp; if (data.attack_mode == ATTACK_MODE_COMBI) { rewind (combs_fp); } // iteration type uint innerloop_step = 0; uint innerloop_cnt = 0; if (hashconfig->attack_exec == ATTACK_EXEC_INSIDE_KERNEL) innerloop_step = device_param->kernel_loops; else innerloop_step = 1; if (data.attack_kern == ATTACK_KERN_STRAIGHT) innerloop_cnt = data.kernel_rules_cnt; else if (data.attack_kern == ATTACK_KERN_COMBI) innerloop_cnt = data.combs_cnt; else if (data.attack_kern == ATTACK_KERN_BF) innerloop_cnt = data.bfs_cnt; // innerloops for (uint innerloop_pos = 0; innerloop_pos < innerloop_cnt; innerloop_pos += innerloop_step) { while (opencl_ctx->devices_status == STATUS_PAUSED) hc_sleep (1); uint fast_iteration = 0; uint innerloop_left = innerloop_cnt - innerloop_pos; if (innerloop_left > innerloop_step) { innerloop_left = innerloop_step; fast_iteration = 1; } device_param->innerloop_pos = innerloop_pos; device_param->innerloop_left = innerloop_left; device_param->kernel_params_buf32[30] = innerloop_left; // i think we can get rid of this if (innerloop_left == 0) { puts ("bug, how should this happen????\n"); continue; } if (hashes->salts_shown[salt_pos] == 1) { data.words_progress_done[salt_pos] += (u64) pws_cnt * (u64) innerloop_left; continue; } // initialize amplifiers if (data.attack_mode == ATTACK_MODE_COMBI) { uint i = 0; while (i < innerloop_left) { if (feof (combs_fp)) break; int line_len = fgetl (combs_fp, line_buf); if (line_len >= PW_MAX1) continue; line_len = convert_from_hex (line_buf, line_len); char *line_buf_new = line_buf; if (run_rule_engine (data.rule_len_r, data.rule_buf_r)) { char rule_buf_out[BLOCK_SIZE] = { 0 }; int rule_len_out = _old_apply_rule (data.rule_buf_r, data.rule_len_r, line_buf, line_len, rule_buf_out); if (rule_len_out < 0) { data.words_progress_rejected[salt_pos] += pws_cnt; continue; } line_len = rule_len_out; line_buf_new = rule_buf_out; } line_len = MIN (line_len, PW_DICTMAX); u8 *ptr = (u8 *) device_param->combs_buf[i].i; memcpy (ptr, line_buf_new, line_len); memset (ptr + line_len, 0, PW_DICTMAX1 - line_len); if (hashconfig->opts_type & OPTS_TYPE_PT_UPPER) { uppercase (ptr, line_len); } if (data.combs_mode == COMBINATOR_MODE_BASE_LEFT) { if (hashconfig->opts_type & OPTS_TYPE_PT_ADD80) { ptr[line_len] = 0x80; } if (hashconfig->opts_type & OPTS_TYPE_PT_ADD01) { ptr[line_len] = 0x01; } } device_param->combs_buf[i].pw_len = line_len; i++; } for (uint j = i; j < innerloop_left; j++) { device_param->combs_buf[j].i[0] = 0; device_param->combs_buf[j].i[1] = 0; device_param->combs_buf[j].i[2] = 0; device_param->combs_buf[j].i[3] = 0; device_param->combs_buf[j].i[4] = 0; device_param->combs_buf[j].i[5] = 0; device_param->combs_buf[j].i[6] = 0; device_param->combs_buf[j].i[7] = 0; device_param->combs_buf[j].pw_len = 0; } innerloop_left = i; } else if (data.attack_mode == ATTACK_MODE_BF) { u64 off = innerloop_pos; device_param->kernel_params_mp_r_buf64[3] = off; run_kernel_mp (KERN_RUN_MP_R, opencl_ctx, device_param, innerloop_left); } else if (data.attack_mode == ATTACK_MODE_HYBRID1) { u64 off = innerloop_pos; device_param->kernel_params_mp_buf64[3] = off; run_kernel_mp (KERN_RUN_MP, opencl_ctx, device_param, innerloop_left); } else if (data.attack_mode == ATTACK_MODE_HYBRID2) { u64 off = innerloop_pos; device_param->kernel_params_mp_buf64[3] = off; run_kernel_mp (KERN_RUN_MP, opencl_ctx, device_param, innerloop_left); } // copy amplifiers if (data.attack_mode == ATTACK_MODE_STRAIGHT) { cl_int CL_err = hc_clEnqueueCopyBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_rules, device_param->d_rules_c, innerloop_pos * sizeof (kernel_rule_t), 0, innerloop_left * sizeof (kernel_rule_t), 0, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueCopyBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } } else if (data.attack_mode == ATTACK_MODE_COMBI) { cl_int CL_err = hc_clEnqueueWriteBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_combs_c, CL_TRUE, 0, innerloop_left * sizeof (comb_t), device_param->combs_buf, 0, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueWriteBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } } else if (data.attack_mode == ATTACK_MODE_BF) { cl_int CL_err = hc_clEnqueueCopyBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_bfs, device_param->d_bfs_c, 0, 0, innerloop_left * sizeof (bf_t), 0, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueCopyBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } } else if (data.attack_mode == ATTACK_MODE_HYBRID1) { cl_int CL_err = hc_clEnqueueCopyBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_combs, device_param->d_combs_c, 0, 0, innerloop_left * sizeof (comb_t), 0, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueCopyBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } } else if (data.attack_mode == ATTACK_MODE_HYBRID2) { cl_int CL_err = hc_clEnqueueCopyBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_combs, device_param->d_combs_c, 0, 0, innerloop_left * sizeof (comb_t), 0, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueCopyBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } } if (data.benchmark == 1) { hc_timer_set (&device_param->timer_speed); } int rc = choose_kernel (opencl_ctx, device_param, hashconfig, hashconfig->attack_exec, data.attack_mode, hashconfig->opts_type, salt_buf, highest_pw_len, pws_cnt, fast_iteration); if (rc == -1) return -1; /** * result */ if (data.benchmark == 0) { check_cracked (opencl_ctx, device_param, hashconfig, hashes, salt_pos); } /** * progress */ u64 perf_sum_all = (u64) pws_cnt * (u64) innerloop_left; hc_thread_mutex_lock (mux_counter); data.words_progress_done[salt_pos] += perf_sum_all; hc_thread_mutex_unlock (mux_counter); /** * speed */ double speed_ms; hc_timer_get (device_param->timer_speed, speed_ms); hc_timer_set (&device_param->timer_speed); // current speed //hc_thread_mutex_lock (mux_display); device_param->speed_cnt[speed_pos] = perf_sum_all; device_param->speed_ms[speed_pos] = speed_ms; //hc_thread_mutex_unlock (mux_display); speed_pos++; if (speed_pos == SPEED_CACHE) { speed_pos = 0; } /** * benchmark */ if (data.benchmark == 1) break; if (opencl_ctx->run_thread_level2 == false) break; } if (opencl_ctx->run_thread_level2 == false) break; } device_param->speed_pos = speed_pos; myfree (line_buf); return 0; } int opencl_ctx_init (opencl_ctx_t *opencl_ctx, const char *opencl_platforms, const char *opencl_devices, const char *opencl_device_types, const uint opencl_vector_width, const uint opencl_vector_width_chgd, const uint nvidia_spin_damp, const uint nvidia_spin_damp_chgd, const uint workload_profile, const uint kernel_accel, const uint kernel_accel_chgd, const uint kernel_loops, const uint kernel_loops_chgd, const uint keyspace, const uint stdout_flag) { if (keyspace == 1) { opencl_ctx->disable = 1; return 0; } opencl_ctx->devices_status = STATUS_INIT; opencl_ctx->run_main_level1 = true; opencl_ctx->run_main_level2 = true; opencl_ctx->run_main_level3 = true; opencl_ctx->run_thread_level1 = true; opencl_ctx->run_thread_level2 = true; opencl_ctx->opencl_vector_width_chgd = opencl_vector_width_chgd; opencl_ctx->opencl_vector_width = opencl_vector_width; opencl_ctx->nvidia_spin_damp_chgd = nvidia_spin_damp_chgd; opencl_ctx->nvidia_spin_damp = nvidia_spin_damp; opencl_ctx->kernel_accel_chgd = kernel_accel_chgd; opencl_ctx->kernel_accel = kernel_accel; opencl_ctx->kernel_loops_chgd = kernel_loops_chgd; opencl_ctx->kernel_loops = kernel_loops; opencl_ctx->workload_profile = workload_profile; opencl_ctx->ocl = (OCL_PTR *) mymalloc (sizeof (OCL_PTR)); hc_device_param_t *devices_param = (hc_device_param_t *) mycalloc (DEVICES_MAX, sizeof (hc_device_param_t)); opencl_ctx->devices_param = devices_param; /** * Load and map OpenCL library calls * TODO: remove exit() calls in there */ ocl_init (opencl_ctx->ocl); /** * OpenCL platform selection */ u32 opencl_platforms_filter = setup_opencl_platforms_filter (opencl_platforms); opencl_ctx->opencl_platforms_filter = opencl_platforms_filter; /** * OpenCL device selection */ u32 devices_filter = setup_devices_filter (opencl_devices); opencl_ctx->devices_filter = devices_filter; /** * OpenCL device type selection */ cl_device_type device_types_filter = setup_device_types_filter (opencl_device_types); opencl_ctx->device_types_filter = device_types_filter; /** * OpenCL platforms: detect */ cl_uint platforms_cnt = 0; cl_platform_id *platforms = (cl_platform_id *) mycalloc (CL_PLATFORMS_MAX, sizeof (cl_platform_id)); cl_uint platform_devices_cnt = 0; cl_device_id *platform_devices = (cl_device_id *) mycalloc (DEVICES_MAX, sizeof (cl_device_id)); cl_int CL_err = hc_clGetPlatformIDs (opencl_ctx->ocl, CL_PLATFORMS_MAX, platforms, &platforms_cnt); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetPlatformIDs(): %s\n", val2cstr_cl (CL_err)); return -1; } if (platforms_cnt == 0) { log_info (""); log_info ("ATTENTION! No OpenCL compatible platform found"); log_info (""); log_info ("You're probably missing the OpenCL runtime installation"); log_info (" AMD users require AMD drivers 14.9 or later (recommended 15.12 or later)"); log_info (" Intel users require Intel OpenCL Runtime 14.2 or later (recommended 15.1 or later)"); log_info (" NVidia users require NVidia drivers 346.59 or later (recommended 361.x or later)"); log_info (""); return -1; } if (opencl_platforms_filter != (uint) -1) { uint platform_cnt_mask = ~(((uint) -1 >> platforms_cnt) << platforms_cnt); if (opencl_platforms_filter > platform_cnt_mask) { log_error ("ERROR: The platform selected by the --opencl-platforms parameter is larger than the number of available platforms (%d)", platforms_cnt); return -1; } } if (opencl_device_types == NULL) { /** * OpenCL device types: * In case the user did not specify --opencl-device-types and the user runs hashcat in a system with only a CPU only he probably want to use that CPU. */ cl_device_type device_types_all = 0; for (uint platform_id = 0; platform_id < platforms_cnt; platform_id++) { if ((opencl_platforms_filter & (1u << platform_id)) == 0) continue; cl_platform_id platform = platforms[platform_id]; cl_int CL_err = hc_clGetDeviceIDs (opencl_ctx->ocl, platform, CL_DEVICE_TYPE_ALL, DEVICES_MAX, platform_devices, &platform_devices_cnt); if (CL_err != CL_SUCCESS) { //log_error ("ERROR: clGetDeviceIDs(): %s\n", val2cstr_cl (CL_err)); //return -1; // Silently ignore at this point, it will be reused later and create a note for the user at that point continue; } for (uint platform_devices_id = 0; platform_devices_id < platform_devices_cnt; platform_devices_id++) { cl_device_id device = platform_devices[platform_devices_id]; cl_device_type device_type; cl_int CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device, CL_DEVICE_TYPE, sizeof (device_type), &device_type, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } device_types_all |= device_type; } } // In such a case, automatically enable CPU device type support, since it's disabled by default. if ((device_types_all & (CL_DEVICE_TYPE_GPU | CL_DEVICE_TYPE_ACCELERATOR)) == 0) { device_types_filter |= CL_DEVICE_TYPE_CPU; } // In another case, when the user uses --stdout, using CPU devices is much faster to setup // If we have a CPU device, force it to be used if (stdout_flag == 1) { if (device_types_all & CL_DEVICE_TYPE_CPU) { device_types_filter = CL_DEVICE_TYPE_CPU; } } } opencl_ctx->platforms_cnt = platforms_cnt; opencl_ctx->platforms = platforms; opencl_ctx->platform_devices_cnt = platform_devices_cnt; opencl_ctx->platform_devices = platform_devices; return 0; } void opencl_ctx_destroy (opencl_ctx_t *opencl_ctx) { if (opencl_ctx->disable == 1) return; myfree (opencl_ctx->devices_param); ocl_close (opencl_ctx->ocl); myfree (opencl_ctx->ocl); myfree (opencl_ctx->platforms); myfree (opencl_ctx->platform_devices); myfree (opencl_ctx); } int opencl_ctx_devices_init (opencl_ctx_t *opencl_ctx, const hashconfig_t *hashconfig, const tuning_db_t *tuning_db, const uint attack_mode, const bool quiet, const bool force, const bool benchmark, const bool opencl_info, const bool machine_readable, const uint algorithm_pos) { if (opencl_ctx->disable == 1) return 0; /** * OpenCL devices: simply push all devices from all platforms into the same device array */ cl_uint platforms_cnt = opencl_ctx->platforms_cnt; cl_platform_id *platforms = opencl_ctx->platforms; cl_uint platform_devices_cnt = opencl_ctx->platform_devices_cnt; cl_device_id *platform_devices = opencl_ctx->platform_devices; int need_adl = 0; int need_nvml = 0; int need_nvapi = 0; int need_xnvctrl = 0; u32 devices_cnt = 0; u32 devices_active = 0; if (opencl_info) { fprintf (stdout, "OpenCL Info:\n"); } for (uint platform_id = 0; platform_id < platforms_cnt; platform_id++) { cl_int CL_err = CL_SUCCESS; cl_platform_id platform = platforms[platform_id]; char platform_vendor[HCBUFSIZ_TINY] = { 0 }; CL_err = hc_clGetPlatformInfo (opencl_ctx->ocl, platform, CL_PLATFORM_VENDOR, sizeof (platform_vendor), platform_vendor, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetPlatformInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } // find our own platform vendor because pocl and mesa are pushing original vendor_id through opencl // this causes trouble with vendor id based macros // we'll assign generic to those without special optimization available cl_uint platform_vendor_id = 0; if (strcmp (platform_vendor, CL_VENDOR_AMD) == 0) { platform_vendor_id = VENDOR_ID_AMD; } else if (strcmp (platform_vendor, CL_VENDOR_AMD_USE_INTEL) == 0) { platform_vendor_id = VENDOR_ID_AMD_USE_INTEL; } else if (strcmp (platform_vendor, CL_VENDOR_APPLE) == 0) { platform_vendor_id = VENDOR_ID_APPLE; } else if (strcmp (platform_vendor, CL_VENDOR_INTEL_BEIGNET) == 0) { platform_vendor_id = VENDOR_ID_INTEL_BEIGNET; } else if (strcmp (platform_vendor, CL_VENDOR_INTEL_SDK) == 0) { platform_vendor_id = VENDOR_ID_INTEL_SDK; } else if (strcmp (platform_vendor, CL_VENDOR_MESA) == 0) { platform_vendor_id = VENDOR_ID_MESA; } else if (strcmp (platform_vendor, CL_VENDOR_NV) == 0) { platform_vendor_id = VENDOR_ID_NV; } else if (strcmp (platform_vendor, CL_VENDOR_POCL) == 0) { platform_vendor_id = VENDOR_ID_POCL; } else { platform_vendor_id = VENDOR_ID_GENERIC; } uint platform_skipped = ((opencl_ctx->opencl_platforms_filter & (1u << platform_id)) == 0); CL_err = hc_clGetDeviceIDs (opencl_ctx->ocl, platform, CL_DEVICE_TYPE_ALL, DEVICES_MAX, platform_devices, &platform_devices_cnt); if (CL_err != CL_SUCCESS) { //log_error ("ERROR: clGetDeviceIDs(): %s\n", val2cstr_cl (CL_err)); //return -1; platform_skipped = 2; } if (opencl_info) { char platform_name[HCBUFSIZ_TINY] = { 0 }; CL_err = hc_clGetPlatformInfo (opencl_ctx->ocl, platform, CL_PLATFORM_NAME, HCBUFSIZ_TINY, platform_name, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetPlatformInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } char platform_version[HCBUFSIZ_TINY] = { 0 }; CL_err = hc_clGetPlatformInfo (opencl_ctx->ocl, platform, CL_PLATFORM_VERSION, sizeof (platform_version), platform_version, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetPlatformInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } fprintf (stdout, "\nPlatform ID #%u\n Vendor : %s\n Name : %s\n Version : %s\n\n", platform_id, platform_vendor, platform_name, platform_version); } if ((benchmark == 1 || quiet == 0) && (algorithm_pos == 0)) { if (machine_readable == 0) { if (platform_skipped == 0) { const int len = log_info ("OpenCL Platform #%u: %s", platform_id + 1, platform_vendor); char line[256] = { 0 }; for (int i = 0; i < len; i++) line[i] = '='; log_info (line); } else if (platform_skipped == 1) { log_info ("OpenCL Platform #%u: %s, skipped", platform_id + 1, platform_vendor); log_info (""); } else if (platform_skipped == 2) { log_info ("OpenCL Platform #%u: %s, skipped! No OpenCL compatible devices found", platform_id + 1, platform_vendor); log_info (""); } } } if (platform_skipped == 1) continue; if (platform_skipped == 2) continue; hc_device_param_t *devices_param = opencl_ctx->devices_param; for (uint platform_devices_id = 0; platform_devices_id < platform_devices_cnt; platform_devices_id++) { size_t param_value_size = 0; const uint device_id = devices_cnt; hc_device_param_t *device_param = &devices_param[device_id]; device_param->platform_vendor_id = platform_vendor_id; device_param->device = platform_devices[platform_devices_id]; device_param->device_id = device_id; device_param->platform_devices_id = platform_devices_id; device_param->platform = platform; // device_type cl_device_type device_type; CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_TYPE, sizeof (device_type), &device_type, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } device_type &= ~CL_DEVICE_TYPE_DEFAULT; device_param->device_type = device_type; // device_name CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_NAME, 0, NULL, ¶m_value_size); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } char *device_name = (char *) mymalloc (param_value_size); CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_NAME, param_value_size, device_name, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } device_param->device_name = device_name; // device_vendor CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_VENDOR, 0, NULL, ¶m_value_size); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } char *device_vendor = (char *) mymalloc (param_value_size); CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_VENDOR, param_value_size, device_vendor, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } device_param->device_vendor = device_vendor; cl_uint device_vendor_id = 0; if (strcmp (device_vendor, CL_VENDOR_AMD) == 0) { device_vendor_id = VENDOR_ID_AMD; } else if (strcmp (device_vendor, CL_VENDOR_AMD_USE_INTEL) == 0) { device_vendor_id = VENDOR_ID_AMD_USE_INTEL; } else if (strcmp (device_vendor, CL_VENDOR_APPLE) == 0) { device_vendor_id = VENDOR_ID_APPLE; } else if (strcmp (device_vendor, CL_VENDOR_INTEL_BEIGNET) == 0) { device_vendor_id = VENDOR_ID_INTEL_BEIGNET; } else if (strcmp (device_vendor, CL_VENDOR_INTEL_SDK) == 0) { device_vendor_id = VENDOR_ID_INTEL_SDK; } else if (strcmp (device_vendor, CL_VENDOR_MESA) == 0) { device_vendor_id = VENDOR_ID_MESA; } else if (strcmp (device_vendor, CL_VENDOR_NV) == 0) { device_vendor_id = VENDOR_ID_NV; } else if (strcmp (device_vendor, CL_VENDOR_POCL) == 0) { device_vendor_id = VENDOR_ID_POCL; } else { device_vendor_id = VENDOR_ID_GENERIC; } device_param->device_vendor_id = device_vendor_id; // device_version CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_VERSION, 0, NULL, ¶m_value_size); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } char *device_version = (char *) mymalloc (param_value_size); CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_VERSION, param_value_size, device_version, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } device_param->device_version = device_version; // device_opencl_version CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_OPENCL_C_VERSION, 0, NULL, ¶m_value_size); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } char *device_opencl_version = (char *) mymalloc (param_value_size); CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_OPENCL_C_VERSION, param_value_size, device_opencl_version, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } device_param->opencl_v12 = device_opencl_version[9] > '1' || device_opencl_version[11] >= '2'; // vector_width cl_uint vector_width; if (opencl_ctx->opencl_vector_width_chgd == 0) { // tuning db tuning_db_entry_t *tuningdb_entry = tuning_db_search (tuning_db, device_param->device_name, device_param->device_type, attack_mode, hashconfig->hash_mode); if (tuningdb_entry == NULL || tuningdb_entry->vector_width == -1) { if (hashconfig->opti_type & OPTI_TYPE_USES_BITS_64) { CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_NATIVE_VECTOR_WIDTH_LONG, sizeof (vector_width), &vector_width, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } } else { CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_NATIVE_VECTOR_WIDTH_INT, sizeof (vector_width), &vector_width, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } } } else { vector_width = (cl_uint) tuningdb_entry->vector_width; } } else { vector_width = opencl_ctx->opencl_vector_width; } if (vector_width > 16) vector_width = 16; device_param->vector_width = vector_width; // max_compute_units cl_uint device_processors; CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_MAX_COMPUTE_UNITS, sizeof (device_processors), &device_processors, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } device_param->device_processors = device_processors; // device_maxmem_alloc // note we'll limit to 2gb, otherwise this causes all kinds of weird errors because of possible integer overflows in opencl runtimes cl_ulong device_maxmem_alloc; CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_MAX_MEM_ALLOC_SIZE, sizeof (device_maxmem_alloc), &device_maxmem_alloc, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } device_param->device_maxmem_alloc = MIN (device_maxmem_alloc, 0x7fffffff); // device_global_mem cl_ulong device_global_mem; CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_GLOBAL_MEM_SIZE, sizeof (device_global_mem), &device_global_mem, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } device_param->device_global_mem = device_global_mem; // max_work_group_size size_t device_maxworkgroup_size; CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_MAX_WORK_GROUP_SIZE, sizeof (device_maxworkgroup_size), &device_maxworkgroup_size, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } device_param->device_maxworkgroup_size = device_maxworkgroup_size; // max_clock_frequency cl_uint device_maxclock_frequency; CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_MAX_CLOCK_FREQUENCY, sizeof (device_maxclock_frequency), &device_maxclock_frequency, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } device_param->device_maxclock_frequency = device_maxclock_frequency; // device_endian_little cl_bool device_endian_little; CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_ENDIAN_LITTLE, sizeof (device_endian_little), &device_endian_little, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } if (device_endian_little == CL_FALSE) { log_info ("- Device #%u: WARNING: Not a little endian device", device_id + 1); device_param->skipped = 1; } // device_available cl_bool device_available; CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_AVAILABLE, sizeof (device_available), &device_available, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } if (device_available == CL_FALSE) { log_info ("- Device #%u: WARNING: Device not available", device_id + 1); device_param->skipped = 1; } // device_compiler_available cl_bool device_compiler_available; CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_COMPILER_AVAILABLE, sizeof (device_compiler_available), &device_compiler_available, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } if (device_compiler_available == CL_FALSE) { log_info ("- Device #%u: WARNING: No compiler available for device", device_id + 1); device_param->skipped = 1; } // device_execution_capabilities cl_device_exec_capabilities device_execution_capabilities; CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_EXECUTION_CAPABILITIES, sizeof (device_execution_capabilities), &device_execution_capabilities, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } if ((device_execution_capabilities & CL_EXEC_KERNEL) == 0) { log_info ("- Device #%u: WARNING: Device does not support executing kernels", device_id + 1); device_param->skipped = 1; } // device_extensions size_t device_extensions_size; CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_EXTENSIONS, 0, NULL, &device_extensions_size); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } char *device_extensions = mymalloc (device_extensions_size + 1); CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_EXTENSIONS, device_extensions_size, device_extensions, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } if (strstr (device_extensions, "base_atomics") == 0) { log_info ("- Device #%u: WARNING: Device does not support base atomics", device_id + 1); device_param->skipped = 1; } if (strstr (device_extensions, "byte_addressable_store") == 0) { log_info ("- Device #%u: WARNING: Device does not support byte addressable store", device_id + 1); device_param->skipped = 1; } myfree (device_extensions); // device_local_mem_size cl_ulong device_local_mem_size; CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_LOCAL_MEM_SIZE, sizeof (device_local_mem_size), &device_local_mem_size, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } if (device_local_mem_size < 32768) { log_info ("- Device #%u: WARNING: Device local mem size is too small", device_id + 1); device_param->skipped = 1; } // If there's both an Intel CPU and an AMD OpenCL runtime it's a tricky situation // Both platforms support CPU device types and therefore both will try to use 100% of the physical resources // This results in both utilizing it for 50% // However, Intel has much better SIMD control over their own hardware // It makes sense to give them full control over their own hardware if (device_type & CL_DEVICE_TYPE_CPU) { if (device_param->device_vendor_id == VENDOR_ID_AMD_USE_INTEL) { if (force == 0) { if (algorithm_pos == 0) { log_info ("- Device #%u: WARNING: Not a native Intel OpenCL runtime, expect massive speed loss", device_id + 1); log_info (" You can use --force to override this but do not post error reports if you do so"); } device_param->skipped = 1; } } } // skipped device_param->skipped |= ((opencl_ctx->devices_filter & (1u << device_id)) == 0); device_param->skipped |= ((opencl_ctx->device_types_filter & (device_type)) == 0); // driver_version CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DRIVER_VERSION, 0, NULL, ¶m_value_size); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } char *driver_version = (char *) mymalloc (param_value_size); CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DRIVER_VERSION, param_value_size, driver_version, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } device_param->driver_version = driver_version; // device_name_chksum char *device_name_chksum = (char *) mymalloc (HCBUFSIZ_TINY); #if defined (__x86_64__) snprintf (device_name_chksum, HCBUFSIZ_TINY - 1, "%u-%u-%u-%s-%s-%s-%u", 64, device_param->platform_vendor_id, device_param->vector_width, device_param->device_name, device_param->device_version, device_param->driver_version, comptime); #else snprintf (device_name_chksum, HCBUFSIZ_TINY - 1, "%u-%u-%u-%s-%s-%s-%u", 32, device_param->platform_vendor_id, device_param->vector_width, device_param->device_name, device_param->device_version, device_param->driver_version, comptime); #endif uint device_name_digest[4] = { 0 }; md5_64 ((uint *) device_name_chksum, device_name_digest); snprintf (device_name_chksum, HCBUFSIZ_TINY - 1, "%08x", device_name_digest[0]); device_param->device_name_chksum = device_name_chksum; // vendor specific if (device_param->device_type & CL_DEVICE_TYPE_GPU) { if ((device_param->platform_vendor_id == VENDOR_ID_AMD) && (device_param->device_vendor_id == VENDOR_ID_AMD)) { need_adl = 1; } if ((device_param->platform_vendor_id == VENDOR_ID_NV) && (device_param->device_vendor_id == VENDOR_ID_NV)) { need_nvml = 1; #if defined (__linux__) need_xnvctrl = 1; #endif #if defined (_WIN) need_nvapi = 1; #endif } } if (device_type & CL_DEVICE_TYPE_GPU) { if (device_vendor_id == VENDOR_ID_NV) { cl_uint kernel_exec_timeout = 0; #define CL_DEVICE_KERNEL_EXEC_TIMEOUT_NV 0x4005 CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_KERNEL_EXEC_TIMEOUT_NV, sizeof (kernel_exec_timeout), &kernel_exec_timeout, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } device_param->kernel_exec_timeout = kernel_exec_timeout; cl_uint sm_minor = 0; cl_uint sm_major = 0; #define CL_DEVICE_COMPUTE_CAPABILITY_MAJOR_NV 0x4000 #define CL_DEVICE_COMPUTE_CAPABILITY_MINOR_NV 0x4001 CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_COMPUTE_CAPABILITY_MINOR_NV, sizeof (sm_minor), &sm_minor, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clGetDeviceInfo (opencl_ctx->ocl, device_param->device, CL_DEVICE_COMPUTE_CAPABILITY_MAJOR_NV, sizeof (sm_major), &sm_major, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetDeviceInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } device_param->sm_minor = sm_minor; device_param->sm_major = sm_major; // CPU burning loop damper // Value is given as number between 0-100 // By default 100% device_param->nvidia_spin_damp = (double) opencl_ctx->nvidia_spin_damp; if (opencl_ctx->nvidia_spin_damp_chgd == 0) { if (attack_mode == ATTACK_MODE_STRAIGHT) { /** * the workaround is not a friend of rule based attacks * the words from the wordlist combined with fast and slow rules cause * fluctuations which cause inaccurate wait time estimations * using a reduced damping percentage almost compensates this */ device_param->nvidia_spin_damp = 64; } } device_param->nvidia_spin_damp /= 100; } } // display results if (opencl_info) { char *format = " Device ID #%u\n Type : %s\n Vendor ID : %u\n Vendor : %s\n Name : %s\n Processor(s) : %u\n Clock : %u\n Memory : %lu/%lu MB allocatable\n OpenCL Version : %s\n\n"; fprintf(stdout, format, device_id, ((device_type & CL_DEVICE_TYPE_CPU) ? "Cpu" : ((device_type & CL_DEVICE_TYPE_GPU) ? "Gpu" : "Accelerator")), device_vendor_id, device_vendor, device_name, device_processors, device_maxclock_frequency, device_maxmem_alloc/1024/1024, device_global_mem/1024/1024, device_opencl_version); } myfree (device_opencl_version); if ((benchmark == 1 || quiet == 0) && (algorithm_pos == 0)) { if (machine_readable == 0) { if (device_param->skipped == 0) { log_info ("- Device #%u: %s, %lu/%lu MB allocatable, %uMCU", device_id + 1, device_name, (unsigned int) (device_maxmem_alloc / 1024 / 1024), (unsigned int) (device_global_mem / 1024 / 1024), (unsigned int) device_processors); } else { log_info ("- Device #%u: %s, skipped", device_id + 1, device_name); } } } // common driver check if (device_param->skipped == 0) { if (device_type & CL_DEVICE_TYPE_GPU) { if (platform_vendor_id == VENDOR_ID_AMD) { int catalyst_check = (force == 1) ? 0 : 1; int catalyst_warn = 0; int catalyst_broken = 0; if (catalyst_check == 1) { catalyst_warn = 1; // v14.9 and higher if (atoi (device_param->driver_version) >= 1573) { catalyst_warn = 0; } catalyst_check = 0; } if (catalyst_broken == 1) { log_info (""); log_info ("ATTENTION! The Catalyst driver installed on your system is known to be broken!"); log_info ("It passes over cracked hashes and will not report them as cracked"); log_info ("You are STRONGLY encouraged not to use it"); log_info ("You can use --force to override this but do not post error reports if you do so"); log_info (""); return -1; } if (catalyst_warn == 1) { log_info (""); log_info ("ATTENTION! Unsupported or incorrectly installed Catalyst driver detected!"); log_info ("You are STRONGLY encouraged to use the official supported catalyst driver"); log_info ("See hashcat's homepage for official supported catalyst drivers"); #if defined (_WIN) log_info ("Also see: http://hashcat.net/wiki/doku.php?id=upgrading_amd_drivers_how_to"); #endif log_info ("You can use --force to override this but do not post error reports if you do so"); log_info (""); return -1; } } else if (platform_vendor_id == VENDOR_ID_NV) { if (device_param->kernel_exec_timeout != 0) { if (quiet == 0) log_info ("- Device #%u: WARNING! Kernel exec timeout is not disabled, it might cause you errors of code 702", device_id + 1); if (quiet == 0) log_info (" See the wiki on how to disable it: https://hashcat.net/wiki/doku.php?id=timeout_patch"); } } } /* turns out pocl still creates segfaults (because of llvm) if (device_type & CL_DEVICE_TYPE_CPU) { if (platform_vendor_id == VENDOR_ID_AMD) { if (force == 0) { log_info (""); log_info ("ATTENTION! OpenCL support for CPU of catalyst driver is not reliable."); log_info ("You are STRONGLY encouraged not to use it"); log_info ("You can use --force to override this but do not post error reports if you do so"); log_info ("A good alternative is the free pocl >= v0.13, but make sure to use a LLVM >= v3.8"); log_info (""); return -1; } } } */ /** * kernel accel and loops tuning db adjustment */ device_param->kernel_accel_min = 1; device_param->kernel_accel_max = 1024; device_param->kernel_loops_min = 1; device_param->kernel_loops_max = 1024; tuning_db_entry_t *tuningdb_entry = tuning_db_search (tuning_db, device_param->device_name, device_param->device_type, attack_mode, hashconfig->hash_mode); if (tuningdb_entry != NULL) { u32 _kernel_accel = tuningdb_entry->kernel_accel; u32 _kernel_loops = tuningdb_entry->kernel_loops; if (_kernel_accel) { device_param->kernel_accel_min = _kernel_accel; device_param->kernel_accel_max = _kernel_accel; } if (_kernel_loops) { if (opencl_ctx->workload_profile == 1) { _kernel_loops = (_kernel_loops > 8) ? _kernel_loops / 8 : 1; } else if (opencl_ctx->workload_profile == 2) { _kernel_loops = (_kernel_loops > 4) ? _kernel_loops / 4 : 1; } device_param->kernel_loops_min = _kernel_loops; device_param->kernel_loops_max = _kernel_loops; } } // commandline parameters overwrite tuningdb entries if (opencl_ctx->kernel_accel_chgd == 1) { device_param->kernel_accel_min = opencl_ctx->kernel_accel; device_param->kernel_accel_max = opencl_ctx->kernel_accel; } if (opencl_ctx->kernel_loops_chgd == 1) { device_param->kernel_loops_min = opencl_ctx->kernel_loops; device_param->kernel_loops_max = opencl_ctx->kernel_loops; } /** * activate device */ devices_active++; } // next please devices_cnt++; } if ((benchmark == 1 || quiet == 0) && (algorithm_pos == 0)) { if (machine_readable == 0) { log_info (""); } } } if (opencl_info) { exit(0); } if (devices_active == 0) { log_error ("ERROR: No devices found/left"); return -1; } // additional check to see if the user has chosen a device that is not within the range of available devices (i.e. larger than devices_cnt) if (opencl_ctx->devices_filter != (uint) -1) { const uint devices_cnt_mask = ~(((uint) -1 >> devices_cnt) << devices_cnt); if (opencl_ctx->devices_filter > devices_cnt_mask) { log_error ("ERROR: The device specified by the --opencl-devices parameter is larger than the number of available devices (%d)", devices_cnt); return -1; } } opencl_ctx->devices_cnt = devices_cnt; opencl_ctx->devices_active = devices_active; opencl_ctx->need_adl = need_adl; opencl_ctx->need_nvml = need_nvml; opencl_ctx->need_nvapi = need_nvapi; opencl_ctx->need_xnvctrl = need_xnvctrl; return 0; } void opencl_ctx_devices_destroy (opencl_ctx_t *opencl_ctx) { for (uint device_id = 0; device_id < opencl_ctx->devices_cnt; device_id++) { hc_device_param_t *device_param = &opencl_ctx->devices_param[device_id]; if (device_param->skipped) continue; myfree (device_param->device_name); myfree (device_param->device_name_chksum); myfree (device_param->device_version); myfree (device_param->driver_version); } opencl_ctx->devices_cnt = 0; opencl_ctx->devices_active = 0; opencl_ctx->need_adl = 0; opencl_ctx->need_nvml = 0; opencl_ctx->need_nvapi = 0; opencl_ctx->need_xnvctrl = 0; } int opencl_session_begin (opencl_ctx_t *opencl_ctx, const hashconfig_t *hashconfig, const hashes_t *hashes, const session_ctx_t *session_ctx) { for (uint device_id = 0; device_id < opencl_ctx->devices_cnt; device_id++) { cl_int CL_err = CL_SUCCESS; /** * host buffer */ hc_device_param_t *device_param = &opencl_ctx->devices_param[device_id]; if (device_param->skipped) continue; /** * device properties */ const char *device_name_chksum = device_param->device_name_chksum; const u32 device_processors = device_param->device_processors; /** * create context for each device */ cl_context_properties properties[3]; properties[0] = CL_CONTEXT_PLATFORM; properties[1] = (cl_context_properties) device_param->platform; properties[2] = 0; CL_err = hc_clCreateContext (opencl_ctx->ocl, properties, 1, &device_param->device, NULL, NULL, &device_param->context); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateContext(): %s\n", val2cstr_cl (CL_err)); return -1; } /** * create command-queue */ // not supported with NV // device_param->command_queue = hc_clCreateCommandQueueWithProperties (device_param->context, device_param->device, NULL); CL_err = hc_clCreateCommandQueue (opencl_ctx->ocl, device_param->context, device_param->device, CL_QUEUE_PROFILING_ENABLE, &device_param->command_queue); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateCommandQueue(): %s\n", val2cstr_cl (CL_err)); return -1; } /** * kernel threads: some algorithms need a fixed kernel-threads count * because of shared memory usage or bitslice * there needs to be some upper limit, otherwise there's too much overhead */ uint kernel_threads = MIN (KERNEL_THREADS_MAX, device_param->device_maxworkgroup_size); if (hashconfig->hash_mode == 8900) kernel_threads = 64; // Scrypt if (hashconfig->hash_mode == 9300) kernel_threads = 64; // Scrypt if (device_param->device_type & CL_DEVICE_TYPE_CPU) { kernel_threads = KERNEL_THREADS_MAX_CPU; } if (hashconfig->hash_mode == 1500) kernel_threads = 64; // DES if (hashconfig->hash_mode == 3000) kernel_threads = 64; // DES if (hashconfig->hash_mode == 3100) kernel_threads = 64; // DES if (hashconfig->hash_mode == 3200) kernel_threads = 8; // Blowfish if (hashconfig->hash_mode == 7500) kernel_threads = 64; // RC4 if (hashconfig->hash_mode == 8500) kernel_threads = 64; // DES if (hashconfig->hash_mode == 9000) kernel_threads = 8; // Blowfish if (hashconfig->hash_mode == 9700) kernel_threads = 64; // RC4 if (hashconfig->hash_mode == 9710) kernel_threads = 64; // RC4 if (hashconfig->hash_mode == 9800) kernel_threads = 64; // RC4 if (hashconfig->hash_mode == 9810) kernel_threads = 64; // RC4 if (hashconfig->hash_mode == 10400) kernel_threads = 64; // RC4 if (hashconfig->hash_mode == 10410) kernel_threads = 64; // RC4 if (hashconfig->hash_mode == 10500) kernel_threads = 64; // RC4 if (hashconfig->hash_mode == 13100) kernel_threads = 64; // RC4 if (hashconfig->hash_mode == 14000) kernel_threads = 64; // DES if (hashconfig->hash_mode == 14100) kernel_threads = 64; // DES device_param->kernel_threads = kernel_threads; device_param->hardware_power = device_processors * kernel_threads; /** * create input buffers on device : calculate size of fixed memory buffers */ size_t size_root_css = SP_PW_MAX * sizeof (cs_t); size_t size_markov_css = SP_PW_MAX * CHARSIZ * sizeof (cs_t); device_param->size_root_css = size_root_css; device_param->size_markov_css = size_markov_css; size_t size_results = sizeof (uint); device_param->size_results = size_results; size_t size_rules = session_ctx->kernel_rules_cnt * sizeof (kernel_rule_t); size_t size_rules_c = KERNEL_RULES * sizeof (kernel_rule_t); size_t size_plains = hashes->digests_cnt * sizeof (plain_t); size_t size_salts = hashes->salts_cnt * sizeof (salt_t); size_t size_esalts = hashes->salts_cnt * hashconfig->esalt_size; size_t size_shown = hashes->digests_cnt * sizeof (uint); size_t size_digests = hashes->digests_cnt * hashconfig->dgst_size; device_param->size_plains = size_plains; device_param->size_digests = size_digests; device_param->size_shown = size_shown; device_param->size_salts = size_salts; size_t size_combs = KERNEL_COMBS * sizeof (comb_t); size_t size_bfs = KERNEL_BFS * sizeof (bf_t); size_t size_tm = 32 * sizeof (bs_word_t); // scryptV stuff u32 scrypt_tmp_size = 0; u32 scrypt_tmto_final = 0; size_t size_scrypt = 4; if ((hashconfig->hash_mode == 8900) || (hashconfig->hash_mode == 9300)) { // we need to check that all hashes have the same scrypt settings const u32 scrypt_N = hashes->salts_buf[0].scrypt_N; const u32 scrypt_r = hashes->salts_buf[0].scrypt_r; const u32 scrypt_p = hashes->salts_buf[0].scrypt_p; for (uint i = 1; i < hashes->salts_cnt; i++) { if ((hashes->salts_buf[i].scrypt_N != scrypt_N) || (hashes->salts_buf[i].scrypt_r != scrypt_r) || (hashes->salts_buf[i].scrypt_p != scrypt_p)) { log_error ("ERROR: Mixed scrypt settings not supported"); return -1; } } scrypt_tmp_size = (128 * scrypt_r * scrypt_p); uint tmto_start = 0; uint tmto_stop = 10; if (session_ctx->scrypt_tmto) { tmto_start = session_ctx->scrypt_tmto; } else { // in case the user did not specify the tmto manually // use some values known to run best (tested on 290x for AMD and GTX1080 for NV) if (hashconfig->hash_mode == 8900) { if (device_param->device_vendor_id == VENDOR_ID_AMD) { tmto_start = 3; } else if (device_param->device_vendor_id == VENDOR_ID_NV) { tmto_start = 2; } } else if (hashconfig->hash_mode == 9300) { if (device_param->device_vendor_id == VENDOR_ID_AMD) { tmto_start = 2; } else if (device_param->device_vendor_id == VENDOR_ID_NV) { tmto_start = 4; } } } device_param->kernel_accel_min = 1; device_param->kernel_accel_max = 8; uint tmto; for (tmto = tmto_start; tmto < tmto_stop; tmto++) { size_scrypt = (128 * scrypt_r) * scrypt_N; size_scrypt /= 1u << tmto; size_scrypt *= device_param->device_processors * device_param->kernel_threads * device_param->kernel_accel_max; if ((size_scrypt / 4) > device_param->device_maxmem_alloc) { if (session_ctx->quiet == 0) log_info ("WARNING: Not enough single-block device memory allocatable to use --scrypt-tmto %d, increasing...", tmto); continue; } if (size_scrypt > device_param->device_global_mem) { if (session_ctx->quiet == 0) log_info ("WARNING: Not enough total device memory allocatable to use --scrypt-tmto %d, increasing...", tmto); continue; } for (uint salts_pos = 0; salts_pos < hashes->salts_cnt; salts_pos++) { scrypt_tmto_final = tmto; } break; } if (tmto == tmto_stop) { log_error ("ERROR: Can't allocate enough device memory"); return -1; } if (session_ctx->quiet == 0) log_info ("SCRYPT tmto optimizer value set to: %u, mem: %" PRIu64 "\n", scrypt_tmto_final, size_scrypt); } size_t size_scrypt4 = size_scrypt / 4; /** * some algorithms need a fixed kernel-loops count */ if (hashconfig->hash_mode == 1500 && session_ctx->attack_mode == ATTACK_MODE_BF) { const u32 kernel_loops_fixed = 1024; device_param->kernel_loops_min = kernel_loops_fixed; device_param->kernel_loops_max = kernel_loops_fixed; } if (hashconfig->hash_mode == 3000 && session_ctx->attack_mode == ATTACK_MODE_BF) { const u32 kernel_loops_fixed = 1024; device_param->kernel_loops_min = kernel_loops_fixed; device_param->kernel_loops_max = kernel_loops_fixed; } if (hashconfig->hash_mode == 8900) { const u32 kernel_loops_fixed = 1; device_param->kernel_loops_min = kernel_loops_fixed; device_param->kernel_loops_max = kernel_loops_fixed; } if (hashconfig->hash_mode == 9300) { const u32 kernel_loops_fixed = 1; device_param->kernel_loops_min = kernel_loops_fixed; device_param->kernel_loops_max = kernel_loops_fixed; } if (hashconfig->hash_mode == 12500) { const u32 kernel_loops_fixed = ROUNDS_RAR3 / 16; device_param->kernel_loops_min = kernel_loops_fixed; device_param->kernel_loops_max = kernel_loops_fixed; } if (hashconfig->hash_mode == 14000 && session_ctx->attack_mode == ATTACK_MODE_BF) { const u32 kernel_loops_fixed = 1024; device_param->kernel_loops_min = kernel_loops_fixed; device_param->kernel_loops_max = kernel_loops_fixed; } if (hashconfig->hash_mode == 14100 && session_ctx->attack_mode == ATTACK_MODE_BF) { const u32 kernel_loops_fixed = 1024; device_param->kernel_loops_min = kernel_loops_fixed; device_param->kernel_loops_max = kernel_loops_fixed; } u32 kernel_accel_min = device_param->kernel_accel_min; u32 kernel_accel_max = device_param->kernel_accel_max; // find out if we would request too much memory on memory blocks which are based on kernel_accel size_t size_pws = 4; size_t size_tmps = 4; size_t size_hooks = 4; while (kernel_accel_max >= kernel_accel_min) { const u32 kernel_power_max = device_processors * kernel_threads * kernel_accel_max; // size_pws size_pws = kernel_power_max * sizeof (pw_t); // size_tmps switch (hashconfig->hash_mode) { case 400: size_tmps = kernel_power_max * sizeof (phpass_tmp_t); break; case 500: size_tmps = kernel_power_max * sizeof (md5crypt_tmp_t); break; case 501: size_tmps = kernel_power_max * sizeof (md5crypt_tmp_t); break; case 1600: size_tmps = kernel_power_max * sizeof (md5crypt_tmp_t); break; case 1800: size_tmps = kernel_power_max * sizeof (sha512crypt_tmp_t); break; case 2100: size_tmps = kernel_power_max * sizeof (dcc2_tmp_t); break; case 2500: size_tmps = kernel_power_max * sizeof (wpa_tmp_t); break; case 3200: size_tmps = kernel_power_max * sizeof (bcrypt_tmp_t); break; case 5200: size_tmps = kernel_power_max * sizeof (pwsafe3_tmp_t); break; case 5800: size_tmps = kernel_power_max * sizeof (androidpin_tmp_t); break; case 6211: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 6212: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 6213: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 6221: size_tmps = kernel_power_max * sizeof (tc64_tmp_t); break; case 6222: size_tmps = kernel_power_max * sizeof (tc64_tmp_t); break; case 6223: size_tmps = kernel_power_max * sizeof (tc64_tmp_t); break; case 6231: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 6232: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 6233: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 6241: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 6242: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 6243: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 6300: size_tmps = kernel_power_max * sizeof (md5crypt_tmp_t); break; case 6400: size_tmps = kernel_power_max * sizeof (sha256aix_tmp_t); break; case 6500: size_tmps = kernel_power_max * sizeof (sha512aix_tmp_t); break; case 6600: size_tmps = kernel_power_max * sizeof (agilekey_tmp_t); break; case 6700: size_tmps = kernel_power_max * sizeof (sha1aix_tmp_t); break; case 6800: size_tmps = kernel_power_max * sizeof (lastpass_tmp_t); break; case 7100: size_tmps = kernel_power_max * sizeof (pbkdf2_sha512_tmp_t); break; case 7200: size_tmps = kernel_power_max * sizeof (pbkdf2_sha512_tmp_t); break; case 7400: size_tmps = kernel_power_max * sizeof (sha256crypt_tmp_t); break; case 7900: size_tmps = kernel_power_max * sizeof (drupal7_tmp_t); break; case 8200: size_tmps = kernel_power_max * sizeof (pbkdf2_sha512_tmp_t); break; case 8800: size_tmps = kernel_power_max * sizeof (androidfde_tmp_t); break; case 8900: size_tmps = kernel_power_max * scrypt_tmp_size; break; case 9000: size_tmps = kernel_power_max * sizeof (pwsafe2_tmp_t); break; case 9100: size_tmps = kernel_power_max * sizeof (lotus8_tmp_t); break; case 9200: size_tmps = kernel_power_max * sizeof (pbkdf2_sha256_tmp_t); break; case 9300: size_tmps = kernel_power_max * scrypt_tmp_size; break; case 9400: size_tmps = kernel_power_max * sizeof (office2007_tmp_t); break; case 9500: size_tmps = kernel_power_max * sizeof (office2010_tmp_t); break; case 9600: size_tmps = kernel_power_max * sizeof (office2013_tmp_t); break; case 10000: size_tmps = kernel_power_max * sizeof (pbkdf2_sha256_tmp_t); break; case 10200: size_tmps = kernel_power_max * sizeof (cram_md5_t); break; case 10300: size_tmps = kernel_power_max * sizeof (saph_sha1_tmp_t); break; case 10500: size_tmps = kernel_power_max * sizeof (pdf14_tmp_t); break; case 10700: size_tmps = kernel_power_max * sizeof (pdf17l8_tmp_t); break; case 10900: size_tmps = kernel_power_max * sizeof (pbkdf2_sha256_tmp_t); break; case 11300: size_tmps = kernel_power_max * sizeof (bitcoin_wallet_tmp_t); break; case 11600: size_tmps = kernel_power_max * sizeof (seven_zip_tmp_t); break; case 11900: size_tmps = kernel_power_max * sizeof (pbkdf2_md5_tmp_t); break; case 12000: size_tmps = kernel_power_max * sizeof (pbkdf2_sha1_tmp_t); break; case 12100: size_tmps = kernel_power_max * sizeof (pbkdf2_sha512_tmp_t); break; case 12200: size_tmps = kernel_power_max * sizeof (ecryptfs_tmp_t); break; case 12300: size_tmps = kernel_power_max * sizeof (oraclet_tmp_t); break; case 12400: size_tmps = kernel_power_max * sizeof (bsdicrypt_tmp_t); break; case 12500: size_tmps = kernel_power_max * sizeof (rar3_tmp_t); break; case 12700: size_tmps = kernel_power_max * sizeof (mywallet_tmp_t); break; case 12800: size_tmps = kernel_power_max * sizeof (pbkdf2_sha256_tmp_t); break; case 12900: size_tmps = kernel_power_max * sizeof (pbkdf2_sha256_tmp_t); break; case 13000: size_tmps = kernel_power_max * sizeof (pbkdf2_sha256_tmp_t); break; case 13200: size_tmps = kernel_power_max * sizeof (axcrypt_tmp_t); break; case 13400: size_tmps = kernel_power_max * sizeof (keepass_tmp_t); break; case 13600: size_tmps = kernel_power_max * sizeof (pbkdf2_sha1_tmp_t); break; case 13711: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 13712: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 13713: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 13721: size_tmps = kernel_power_max * sizeof (tc64_tmp_t); break; case 13722: size_tmps = kernel_power_max * sizeof (tc64_tmp_t); break; case 13723: size_tmps = kernel_power_max * sizeof (tc64_tmp_t); break; case 13731: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 13732: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 13733: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 13741: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 13742: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 13743: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 13751: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 13752: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 13753: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 13761: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 13762: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; case 13763: size_tmps = kernel_power_max * sizeof (tc_tmp_t); break; }; // size_hooks if ((hashconfig->opts_type & OPTS_TYPE_HOOK12) || (hashconfig->opts_type & OPTS_TYPE_HOOK23)) { switch (hashconfig->hash_mode) { } } // now check if all device-memory sizes which depend on the kernel_accel_max amplifier are within its boundaries // if not, decrease amplifier and try again int memory_limit_hit = 0; if (size_pws > device_param->device_maxmem_alloc) memory_limit_hit = 1; if (size_tmps > device_param->device_maxmem_alloc) memory_limit_hit = 1; if (size_hooks > device_param->device_maxmem_alloc) memory_limit_hit = 1; const u64 size_total = session_ctx->bitmap_size + session_ctx->bitmap_size + session_ctx->bitmap_size + session_ctx->bitmap_size + session_ctx->bitmap_size + session_ctx->bitmap_size + session_ctx->bitmap_size + session_ctx->bitmap_size + size_bfs + size_combs + size_digests + size_esalts + size_hooks + size_markov_css + size_plains + size_pws + size_pws // not a bug + size_results + size_root_css + size_rules + size_rules_c + size_salts + size_scrypt4 + size_scrypt4 + size_scrypt4 + size_scrypt4 + size_shown + size_tm + size_tmps; if (size_total > device_param->device_global_mem) memory_limit_hit = 1; if (memory_limit_hit == 1) { kernel_accel_max--; continue; } break; } if (kernel_accel_max < kernel_accel_min) { log_error ("- Device #%u: Device does not provide enough allocatable device-memory to handle this attack", device_id + 1); return -1; } device_param->kernel_accel_min = kernel_accel_min; device_param->kernel_accel_max = kernel_accel_max; /* if (kernel_accel_max < kernel_accel) { if (session_ctx->quiet == 0) log_info ("- Device #%u: Reduced maximum kernel-accel to %u", device_id + 1, kernel_accel_max); device_param->kernel_accel = kernel_accel_max; } */ device_param->size_bfs = size_bfs; device_param->size_combs = size_combs; device_param->size_rules = size_rules; device_param->size_rules_c = size_rules_c; device_param->size_pws = size_pws; device_param->size_tmps = size_tmps; device_param->size_hooks = size_hooks; /** * default building options */ if (chdir (session_ctx->cpath_real) == -1) { log_error ("ERROR: %s: %s", session_ctx->cpath_real, strerror (errno)); return -1; } char build_opts[1024] = { 0 }; #if defined (_WIN) snprintf (build_opts, sizeof (build_opts) - 1, "-I \"%s\"", session_ctx->cpath_real); #else snprintf (build_opts, sizeof (build_opts) - 1, "-I %s", session_ctx->cpath_real); #endif // include check // this test needs to be done manually because of osx opencl runtime // if there's a problem with permission, its not reporting back and erroring out silently #define files_cnt 15 const char *files_names[files_cnt] = { "inc_cipher_aes256.cl", "inc_cipher_serpent256.cl", "inc_cipher_twofish256.cl", "inc_common.cl", "inc_comp_multi_bs.cl", "inc_comp_multi.cl", "inc_comp_single_bs.cl", "inc_comp_single.cl", "inc_hash_constants.h", "inc_hash_functions.cl", "inc_rp.cl", "inc_rp.h", "inc_simd.cl", "inc_types.cl", "inc_vendor.cl", }; for (int i = 0; i < files_cnt; i++) { FILE *fd = fopen (files_names[i], "r"); if (fd == NULL) { log_error ("ERROR: %s: fopen(): %s", files_names[i], strerror (errno)); return -1; } char buf[1]; size_t n = fread (buf, 1, 1, fd); if (n != 1) { log_error ("ERROR: %s: fread(): %s", files_names[i], strerror (errno)); return -1; } fclose (fd); } // we don't have sm_* on vendors not NV but it doesn't matter char build_opts_new[1024] = { 0 }; #if defined (DEBUG) snprintf (build_opts_new, sizeof (build_opts_new) - 1, "%s -D VENDOR_ID=%u -D CUDA_ARCH=%d -D VECT_SIZE=%u -D DEVICE_TYPE=%u -D DGST_R0=%u -D DGST_R1=%u -D DGST_R2=%u -D DGST_R3=%u -D DGST_ELEM=%u -D KERN_TYPE=%u -D _unroll -cl-std=CL1.1", build_opts, device_param->device_vendor_id, (device_param->sm_major * 100) + device_param->sm_minor, device_param->vector_width, (u32) device_param->device_type, hashconfig->dgst_pos0, hashconfig->dgst_pos1, hashconfig->dgst_pos2, hashconfig->dgst_pos3, hashconfig->dgst_size / 4, hashconfig->kern_type); #else snprintf (build_opts_new, sizeof (build_opts_new) - 1, "%s -D VENDOR_ID=%u -D CUDA_ARCH=%d -D VECT_SIZE=%u -D DEVICE_TYPE=%u -D DGST_R0=%u -D DGST_R1=%u -D DGST_R2=%u -D DGST_R3=%u -D DGST_ELEM=%u -D KERN_TYPE=%u -D _unroll -cl-std=CL1.1 -w", build_opts, device_param->device_vendor_id, (device_param->sm_major * 100) + device_param->sm_minor, device_param->vector_width, (u32) device_param->device_type, hashconfig->dgst_pos0, hashconfig->dgst_pos1, hashconfig->dgst_pos2, hashconfig->dgst_pos3, hashconfig->dgst_size / 4, hashconfig->kern_type); #endif strncpy (build_opts, build_opts_new, sizeof (build_opts)); #if defined (DEBUG) log_info ("- Device #%u: build_opts '%s'\n", device_id + 1, build_opts); #endif /** * main kernel */ { /** * kernel source filename */ char source_file[256] = { 0 }; generate_source_kernel_filename (hashconfig->attack_exec, session_ctx->attack_kern, hashconfig->kern_type, session_ctx->shared_dir, source_file); struct stat sst; if (stat (source_file, &sst) == -1) { log_error ("ERROR: %s: %s", source_file, strerror (errno)); return -1; } /** * kernel cached filename */ char cached_file[256] = { 0 }; generate_cached_kernel_filename (hashconfig->attack_exec, session_ctx->attack_kern, hashconfig->kern_type, session_ctx->profile_dir, device_name_chksum, cached_file); int cached = 1; struct stat cst; if ((stat (cached_file, &cst) == -1) || cst.st_size == 0) { cached = 0; } /** * kernel compile or load */ size_t *kernel_lengths = (size_t *) mymalloc (sizeof (size_t)); const u8 **kernel_sources = (const u8 **) mymalloc (sizeof (u8 *)); if (opencl_ctx->force_jit_compilation == -1) { if (cached == 0) { if (session_ctx->quiet == 0) log_info ("- Device #%u: Kernel %s not found in cache! Building may take a while...", device_id + 1, filename_from_filepath (cached_file)); load_kernel (source_file, 1, kernel_lengths, kernel_sources); CL_err = hc_clCreateProgramWithSource (opencl_ctx->ocl, device_param->context, 1, (const char **) kernel_sources, NULL, &device_param->program); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateProgramWithSource(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clBuildProgram (opencl_ctx->ocl, device_param->program, 1, &device_param->device, build_opts, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clBuildProgram(): %s\n", val2cstr_cl (CL_err)); //return -1; } size_t build_log_size = 0; /* CL_err = hc_clGetProgramBuildInfo (opencl_ctx->ocl, device_param->program, device_param->device, CL_PROGRAM_BUILD_LOG, 0, NULL, &build_log_size); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetProgramBuildInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } */ hc_clGetProgramBuildInfo (opencl_ctx->ocl, device_param->program, device_param->device, CL_PROGRAM_BUILD_LOG, 0, NULL, &build_log_size); #if defined (DEBUG) if ((build_log_size != 0) || (CL_err != CL_SUCCESS)) #else if (CL_err != CL_SUCCESS) #endif { char *build_log = (char *) mymalloc (build_log_size + 1); CL_err = hc_clGetProgramBuildInfo (opencl_ctx->ocl, device_param->program, device_param->device, CL_PROGRAM_BUILD_LOG, build_log_size, build_log, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetProgramBuildInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } puts (build_log); myfree (build_log); } if (CL_err != CL_SUCCESS) { device_param->skipped = true; log_info ("- Device #%u: Kernel %s build failure. Proceeding without this device.", device_id + 1, source_file); continue; } size_t binary_size; CL_err = hc_clGetProgramInfo (opencl_ctx->ocl, device_param->program, CL_PROGRAM_BINARY_SIZES, sizeof (size_t), &binary_size, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetProgramInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } u8 *binary = (u8 *) mymalloc (binary_size); CL_err = hc_clGetProgramInfo (opencl_ctx->ocl, device_param->program, CL_PROGRAM_BINARIES, sizeof (binary), &binary, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetProgramInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } writeProgramBin (cached_file, binary, binary_size); local_free (binary); } else { #if defined (DEBUG) log_info ("- Device #%u: Kernel %s (%ld bytes)", device_id + 1, cached_file, cst.st_size); #endif load_kernel (cached_file, 1, kernel_lengths, kernel_sources); CL_err = hc_clCreateProgramWithBinary (opencl_ctx->ocl, device_param->context, 1, &device_param->device, kernel_lengths, (const u8 **) kernel_sources, NULL, &device_param->program); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateProgramWithBinary(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clBuildProgram (opencl_ctx->ocl, device_param->program, 1, &device_param->device, build_opts, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clBuildProgram(): %s\n", val2cstr_cl (CL_err)); return -1; } } } else { #if defined (DEBUG) log_info ("- Device #%u: Kernel %s (%ld bytes)", device_id + 1, source_file, sst.st_size); #endif load_kernel (source_file, 1, kernel_lengths, kernel_sources); CL_err = hc_clCreateProgramWithSource (opencl_ctx->ocl, device_param->context, 1, (const char **) kernel_sources, NULL, &device_param->program); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateProgramWithSource(): %s\n", val2cstr_cl (CL_err)); return -1; } char build_opts_update[1024] = { 0 }; if (opencl_ctx->force_jit_compilation == 1500) { snprintf (build_opts_update, sizeof (build_opts_update) - 1, "%s -DDESCRYPT_SALT=%u", build_opts, hashes->salts_buf[0].salt_buf[0]); } else if (opencl_ctx->force_jit_compilation == 8900) { snprintf (build_opts_update, sizeof (build_opts_update) - 1, "%s -DSCRYPT_N=%u -DSCRYPT_R=%u -DSCRYPT_P=%u -DSCRYPT_TMTO=%u -DSCRYPT_TMP_ELEM=%u", build_opts, hashes->salts_buf[0].scrypt_N, hashes->salts_buf[0].scrypt_r, hashes->salts_buf[0].scrypt_p, 1 << scrypt_tmto_final, scrypt_tmp_size / 16); } else { snprintf (build_opts_update, sizeof (build_opts_update) - 1, "%s", build_opts); } CL_err = hc_clBuildProgram (opencl_ctx->ocl, device_param->program, 1, &device_param->device, build_opts_update, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clBuildProgram(): %s\n", val2cstr_cl (CL_err)); //return -1; } size_t build_log_size = 0; /* CL_err = hc_clGetProgramBuildInfo (opencl_ctx->ocl, device_param->program, device_param->device, CL_PROGRAM_BUILD_LOG, 0, NULL, &build_log_size); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetProgramBuildInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } */ hc_clGetProgramBuildInfo (opencl_ctx->ocl, device_param->program, device_param->device, CL_PROGRAM_BUILD_LOG, 0, NULL, &build_log_size); #if defined (DEBUG) if ((build_log_size != 0) || (CL_err != CL_SUCCESS)) #else if (CL_err != CL_SUCCESS) #endif { char *build_log = (char *) mymalloc (build_log_size + 1); CL_err = hc_clGetProgramBuildInfo (opencl_ctx->ocl, device_param->program, device_param->device, CL_PROGRAM_BUILD_LOG, build_log_size, build_log, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetProgramBuildInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } puts (build_log); myfree (build_log); } if (CL_err != CL_SUCCESS) { device_param->skipped = true; log_info ("- Device #%u: Kernel %s build failure. Proceeding without this device.", device_id + 1, source_file); } } local_free (kernel_lengths); local_free (kernel_sources[0]); local_free (kernel_sources); } /** * word generator kernel */ if (session_ctx->attack_mode != ATTACK_MODE_STRAIGHT) { /** * kernel mp source filename */ char source_file[256] = { 0 }; generate_source_kernel_mp_filename (hashconfig->opti_type, hashconfig->opts_type, session_ctx->shared_dir, source_file); struct stat sst; if (stat (source_file, &sst) == -1) { log_error ("ERROR: %s: %s", source_file, strerror (errno)); return -1; } /** * kernel mp cached filename */ char cached_file[256] = { 0 }; generate_cached_kernel_mp_filename (hashconfig->opti_type, hashconfig->opts_type, session_ctx->profile_dir, device_name_chksum, cached_file); int cached = 1; struct stat cst; if (stat (cached_file, &cst) == -1) { cached = 0; } /** * kernel compile or load */ size_t *kernel_lengths = (size_t *) mymalloc (sizeof (size_t)); const u8 **kernel_sources = (const u8 **) mymalloc (sizeof (u8 *)); if (cached == 0) { if (session_ctx->quiet == 0) log_info ("- Device #%u: Kernel %s not found in cache! Building may take a while...", device_id + 1, filename_from_filepath (cached_file)); if (session_ctx->quiet == 0) log_info (""); load_kernel (source_file, 1, kernel_lengths, kernel_sources); CL_err = hc_clCreateProgramWithSource (opencl_ctx->ocl, device_param->context, 1, (const char **) kernel_sources, NULL, &device_param->program_mp); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateProgramWithSource(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clBuildProgram (opencl_ctx->ocl, device_param->program_mp, 1, &device_param->device, build_opts, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clBuildProgram(): %s\n", val2cstr_cl (CL_err)); //return -1; } size_t build_log_size = 0; /* CL_err = hc_clGetProgramBuildInfo (opencl_ctx->ocl, device_param->program_mp, device_param->device, CL_PROGRAM_BUILD_LOG, 0, NULL, &build_log_size); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetProgramBuildInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } */ hc_clGetProgramBuildInfo (opencl_ctx->ocl, device_param->program_mp, device_param->device, CL_PROGRAM_BUILD_LOG, 0, NULL, &build_log_size); #if defined (DEBUG) if ((build_log_size != 0) || (CL_err != CL_SUCCESS)) #else if (CL_err != CL_SUCCESS) #endif { char *build_log = (char *) mymalloc (build_log_size + 1); CL_err = hc_clGetProgramBuildInfo (opencl_ctx->ocl, device_param->program_mp, device_param->device, CL_PROGRAM_BUILD_LOG, build_log_size, build_log, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetProgramBuildInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } puts (build_log); myfree (build_log); } if (CL_err != CL_SUCCESS) { device_param->skipped = true; log_info ("- Device #%u: Kernel %s build failure. Proceeding without this device.", device_id + 1, source_file); continue; } size_t binary_size; CL_err = hc_clGetProgramInfo (opencl_ctx->ocl, device_param->program_mp, CL_PROGRAM_BINARY_SIZES, sizeof (size_t), &binary_size, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetProgramInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } u8 *binary = (u8 *) mymalloc (binary_size); CL_err = hc_clGetProgramInfo (opencl_ctx->ocl, device_param->program_mp, CL_PROGRAM_BINARIES, sizeof (binary), &binary, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetProgramInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } writeProgramBin (cached_file, binary, binary_size); local_free (binary); } else { #if defined (DEBUG) log_info ("- Device #%u: Kernel %s (%ld bytes)", device_id + 1, cached_file, cst.st_size); #endif load_kernel (cached_file, 1, kernel_lengths, kernel_sources); CL_err = hc_clCreateProgramWithBinary (opencl_ctx->ocl, device_param->context, 1, &device_param->device, kernel_lengths, (const u8 **) kernel_sources, NULL, &device_param->program_mp); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateProgramWithBinary(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clBuildProgram (opencl_ctx->ocl, device_param->program_mp, 1, &device_param->device, build_opts, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clBuildProgram(): %s\n", val2cstr_cl (CL_err)); return -1; } } local_free (kernel_lengths); local_free (kernel_sources[0]); local_free (kernel_sources); } /** * amplifier kernel */ if (hashconfig->attack_exec == ATTACK_EXEC_INSIDE_KERNEL) { } else { /** * kernel amp source filename */ char source_file[256] = { 0 }; generate_source_kernel_amp_filename (session_ctx->attack_kern, session_ctx->shared_dir, source_file); struct stat sst; if (stat (source_file, &sst) == -1) { log_error ("ERROR: %s: %s", source_file, strerror (errno)); return -1; } /** * kernel amp cached filename */ char cached_file[256] = { 0 }; generate_cached_kernel_amp_filename (session_ctx->attack_kern, session_ctx->profile_dir, device_name_chksum, cached_file); int cached = 1; struct stat cst; if (stat (cached_file, &cst) == -1) { cached = 0; } /** * kernel compile or load */ size_t *kernel_lengths = (size_t *) mymalloc (sizeof (size_t)); const u8 **kernel_sources = (const u8 **) mymalloc (sizeof (u8 *)); if (cached == 0) { if (session_ctx->quiet == 0) log_info ("- Device #%u: Kernel %s not found in cache! Building may take a while...", device_id + 1, filename_from_filepath (cached_file)); if (session_ctx->quiet == 0) log_info (""); load_kernel (source_file, 1, kernel_lengths, kernel_sources); CL_err = hc_clCreateProgramWithSource (opencl_ctx->ocl, device_param->context, 1, (const char **) kernel_sources, NULL, &device_param->program_amp); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateProgramWithSource(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clBuildProgram (opencl_ctx->ocl, device_param->program_amp, 1, &device_param->device, build_opts, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clBuildProgram(): %s\n", val2cstr_cl (CL_err)); //return -1; } size_t build_log_size = 0; /* CL_err = hc_clGetProgramBuildInfo (opencl_ctx->ocl, device_param->program_amp, device_param->device, CL_PROGRAM_BUILD_LOG, 0, NULL, &build_log_size); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetProgramBuildInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } */ hc_clGetProgramBuildInfo (opencl_ctx->ocl, device_param->program_amp, device_param->device, CL_PROGRAM_BUILD_LOG, 0, NULL, &build_log_size); #if defined (DEBUG) if ((build_log_size != 0) || (CL_err != CL_SUCCESS)) #else if (CL_err != CL_SUCCESS) #endif { char *build_log = (char *) mymalloc (build_log_size + 1); CL_err = hc_clGetProgramBuildInfo (opencl_ctx->ocl, device_param->program_amp, device_param->device, CL_PROGRAM_BUILD_LOG, build_log_size, build_log, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetProgramBuildInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } puts (build_log); myfree (build_log); } if (CL_err != CL_SUCCESS) { device_param->skipped = true; log_info ("- Device #%u: Kernel %s build failure. Proceed without this device.", device_id + 1, source_file); continue; } size_t binary_size; CL_err = hc_clGetProgramInfo (opencl_ctx->ocl, device_param->program_amp, CL_PROGRAM_BINARY_SIZES, sizeof (size_t), &binary_size, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetProgramInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } u8 *binary = (u8 *) mymalloc (binary_size); CL_err = hc_clGetProgramInfo (opencl_ctx->ocl, device_param->program_amp, CL_PROGRAM_BINARIES, sizeof (binary), &binary, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetProgramInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } writeProgramBin (cached_file, binary, binary_size); local_free (binary); } else { #if defined (DEBUG) if (session_ctx->quiet == 0) log_info ("- Device #%u: Kernel %s (%ld bytes)", device_id + 1, cached_file, cst.st_size); #endif load_kernel (cached_file, 1, kernel_lengths, kernel_sources); CL_err = hc_clCreateProgramWithBinary (opencl_ctx->ocl, device_param->context, 1, &device_param->device, kernel_lengths, (const u8 **) kernel_sources, NULL, &device_param->program_amp); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateProgramWithBinary(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clBuildProgram (opencl_ctx->ocl, device_param->program_amp, 1, &device_param->device, build_opts, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clBuildProgram(): %s\n", val2cstr_cl (CL_err)); return -1; } } local_free (kernel_lengths); local_free (kernel_sources[0]); local_free (kernel_sources); } // return back to the folder we came from initially (workaround) if (chdir (session_ctx->cwd) == -1) { log_error ("ERROR: %s: %s", session_ctx->cwd, strerror (errno)); return -1; } // some algorithm collide too fast, make that impossible if (session_ctx->benchmark == 1) { ((uint *) hashes->digests_buf)[0] = -1u; ((uint *) hashes->digests_buf)[1] = -1u; ((uint *) hashes->digests_buf)[2] = -1u; ((uint *) hashes->digests_buf)[3] = -1u; } /** * global buffers */ CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, size_pws, NULL, &device_param->d_pws_buf); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, size_pws, NULL, &device_param->d_pws_amp_buf); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_WRITE, size_tmps, NULL, &device_param->d_tmps); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_WRITE, size_hooks, NULL, &device_param->d_hooks); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, session_ctx->bitmap_size, NULL, &device_param->d_bitmap_s1_a); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, session_ctx->bitmap_size, NULL, &device_param->d_bitmap_s1_b); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, session_ctx->bitmap_size, NULL, &device_param->d_bitmap_s1_c); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, session_ctx->bitmap_size, NULL, &device_param->d_bitmap_s1_d); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, session_ctx->bitmap_size, NULL, &device_param->d_bitmap_s2_a); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, session_ctx->bitmap_size, NULL, &device_param->d_bitmap_s2_b); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, session_ctx->bitmap_size, NULL, &device_param->d_bitmap_s2_c); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, session_ctx->bitmap_size, NULL, &device_param->d_bitmap_s2_d); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_WRITE, size_plains, NULL, &device_param->d_plain_bufs); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, size_digests, NULL, &device_param->d_digests_buf); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_WRITE, size_shown, NULL, &device_param->d_digests_shown); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, size_salts, NULL, &device_param->d_salt_bufs); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_WRITE, size_results, NULL, &device_param->d_result); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_WRITE, size_scrypt4, NULL, &device_param->d_scryptV0_buf); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_WRITE, size_scrypt4, NULL, &device_param->d_scryptV1_buf); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_WRITE, size_scrypt4, NULL, &device_param->d_scryptV2_buf); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_WRITE, size_scrypt4, NULL, &device_param->d_scryptV3_buf); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err |= hc_clEnqueueWriteBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_bitmap_s1_a, CL_TRUE, 0, session_ctx->bitmap_size, session_ctx->bitmap_s1_a, 0, NULL, NULL); CL_err |= hc_clEnqueueWriteBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_bitmap_s1_b, CL_TRUE, 0, session_ctx->bitmap_size, session_ctx->bitmap_s1_b, 0, NULL, NULL); CL_err |= hc_clEnqueueWriteBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_bitmap_s1_c, CL_TRUE, 0, session_ctx->bitmap_size, session_ctx->bitmap_s1_c, 0, NULL, NULL); CL_err |= hc_clEnqueueWriteBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_bitmap_s1_d, CL_TRUE, 0, session_ctx->bitmap_size, session_ctx->bitmap_s1_d, 0, NULL, NULL); CL_err |= hc_clEnqueueWriteBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_bitmap_s2_a, CL_TRUE, 0, session_ctx->bitmap_size, session_ctx->bitmap_s2_a, 0, NULL, NULL); CL_err |= hc_clEnqueueWriteBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_bitmap_s2_b, CL_TRUE, 0, session_ctx->bitmap_size, session_ctx->bitmap_s2_b, 0, NULL, NULL); CL_err |= hc_clEnqueueWriteBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_bitmap_s2_c, CL_TRUE, 0, session_ctx->bitmap_size, session_ctx->bitmap_s2_c, 0, NULL, NULL); CL_err |= hc_clEnqueueWriteBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_bitmap_s2_d, CL_TRUE, 0, session_ctx->bitmap_size, session_ctx->bitmap_s2_d, 0, NULL, NULL); CL_err |= hc_clEnqueueWriteBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_digests_buf, CL_TRUE, 0, size_digests, hashes->digests_buf, 0, NULL, NULL); CL_err |= hc_clEnqueueWriteBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_digests_shown, CL_TRUE, 0, size_shown, hashes->digests_shown, 0, NULL, NULL); CL_err |= hc_clEnqueueWriteBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_salt_bufs, CL_TRUE, 0, size_salts, hashes->salts_buf, 0, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueWriteBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } /** * special buffers */ if (session_ctx->attack_kern == ATTACK_KERN_STRAIGHT) { CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, size_rules, NULL, &device_param->d_rules); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, size_rules_c, NULL, &device_param->d_rules_c); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clEnqueueWriteBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_rules, CL_TRUE, 0, size_rules, session_ctx->kernel_rules_buf, 0, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueWriteBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } } else if (session_ctx->attack_kern == ATTACK_KERN_COMBI) { CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, size_combs, NULL, &device_param->d_combs); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, size_combs, NULL, &device_param->d_combs_c); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, size_root_css, NULL, &device_param->d_root_css_buf); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, size_markov_css, NULL, &device_param->d_markov_css_buf); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } } else if (session_ctx->attack_kern == ATTACK_KERN_BF) { CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, size_bfs, NULL, &device_param->d_bfs); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, size_bfs, NULL, &device_param->d_bfs_c); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, size_tm, NULL, &device_param->d_tm_c); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, size_root_css, NULL, &device_param->d_root_css_buf); CL_err |= hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, size_markov_css, NULL, &device_param->d_markov_css_buf); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } } if (size_esalts) { CL_err = hc_clCreateBuffer (opencl_ctx->ocl, device_param->context, CL_MEM_READ_ONLY, size_esalts, NULL, &device_param->d_esalt_bufs); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clEnqueueWriteBuffer (opencl_ctx->ocl, device_param->command_queue, device_param->d_esalt_bufs, CL_TRUE, 0, size_esalts, hashes->esalts_buf, 0, NULL, NULL); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clEnqueueWriteBuffer(): %s\n", val2cstr_cl (CL_err)); return -1; } } /** * main host data */ pw_t *pws_buf = (pw_t *) mymalloc (size_pws); device_param->pws_buf = pws_buf; comb_t *combs_buf = (comb_t *) mycalloc (KERNEL_COMBS, sizeof (comb_t)); device_param->combs_buf = combs_buf; void *hooks_buf = mymalloc (size_hooks); device_param->hooks_buf = hooks_buf; /** * kernel args */ device_param->kernel_params_buf32[24] = session_ctx->bitmap_mask; device_param->kernel_params_buf32[25] = session_ctx->bitmap_shift1; device_param->kernel_params_buf32[26] = session_ctx->bitmap_shift2; device_param->kernel_params_buf32[27] = 0; // salt_pos device_param->kernel_params_buf32[28] = 0; // loop_pos device_param->kernel_params_buf32[29] = 0; // loop_cnt device_param->kernel_params_buf32[30] = 0; // kernel_rules_cnt device_param->kernel_params_buf32[31] = 0; // digests_cnt device_param->kernel_params_buf32[32] = 0; // digests_offset device_param->kernel_params_buf32[33] = 0; // combs_mode device_param->kernel_params_buf32[34] = 0; // gid_max device_param->kernel_params[ 0] = (hashconfig->attack_exec == ATTACK_EXEC_INSIDE_KERNEL) ? &device_param->d_pws_buf : &device_param->d_pws_amp_buf; device_param->kernel_params[ 1] = &device_param->d_rules_c; device_param->kernel_params[ 2] = &device_param->d_combs_c; device_param->kernel_params[ 3] = &device_param->d_bfs_c; device_param->kernel_params[ 4] = &device_param->d_tmps; device_param->kernel_params[ 5] = &device_param->d_hooks; device_param->kernel_params[ 6] = &device_param->d_bitmap_s1_a; device_param->kernel_params[ 7] = &device_param->d_bitmap_s1_b; device_param->kernel_params[ 8] = &device_param->d_bitmap_s1_c; device_param->kernel_params[ 9] = &device_param->d_bitmap_s1_d; device_param->kernel_params[10] = &device_param->d_bitmap_s2_a; device_param->kernel_params[11] = &device_param->d_bitmap_s2_b; device_param->kernel_params[12] = &device_param->d_bitmap_s2_c; device_param->kernel_params[13] = &device_param->d_bitmap_s2_d; device_param->kernel_params[14] = &device_param->d_plain_bufs; device_param->kernel_params[15] = &device_param->d_digests_buf; device_param->kernel_params[16] = &device_param->d_digests_shown; device_param->kernel_params[17] = &device_param->d_salt_bufs; device_param->kernel_params[18] = &device_param->d_esalt_bufs; device_param->kernel_params[19] = &device_param->d_result; device_param->kernel_params[20] = &device_param->d_scryptV0_buf; device_param->kernel_params[21] = &device_param->d_scryptV1_buf; device_param->kernel_params[22] = &device_param->d_scryptV2_buf; device_param->kernel_params[23] = &device_param->d_scryptV3_buf; device_param->kernel_params[24] = &device_param->kernel_params_buf32[24]; device_param->kernel_params[25] = &device_param->kernel_params_buf32[25]; device_param->kernel_params[26] = &device_param->kernel_params_buf32[26]; device_param->kernel_params[27] = &device_param->kernel_params_buf32[27]; device_param->kernel_params[28] = &device_param->kernel_params_buf32[28]; device_param->kernel_params[29] = &device_param->kernel_params_buf32[29]; device_param->kernel_params[30] = &device_param->kernel_params_buf32[30]; device_param->kernel_params[31] = &device_param->kernel_params_buf32[31]; device_param->kernel_params[32] = &device_param->kernel_params_buf32[32]; device_param->kernel_params[33] = &device_param->kernel_params_buf32[33]; device_param->kernel_params[34] = &device_param->kernel_params_buf32[34]; device_param->kernel_params_mp_buf64[3] = 0; device_param->kernel_params_mp_buf32[4] = 0; device_param->kernel_params_mp_buf32[5] = 0; device_param->kernel_params_mp_buf32[6] = 0; device_param->kernel_params_mp_buf32[7] = 0; device_param->kernel_params_mp_buf32[8] = 0; device_param->kernel_params_mp[0] = NULL; device_param->kernel_params_mp[1] = NULL; device_param->kernel_params_mp[2] = NULL; device_param->kernel_params_mp[3] = &device_param->kernel_params_mp_buf64[3]; device_param->kernel_params_mp[4] = &device_param->kernel_params_mp_buf32[4]; device_param->kernel_params_mp[5] = &device_param->kernel_params_mp_buf32[5]; device_param->kernel_params_mp[6] = &device_param->kernel_params_mp_buf32[6]; device_param->kernel_params_mp[7] = &device_param->kernel_params_mp_buf32[7]; device_param->kernel_params_mp[8] = &device_param->kernel_params_mp_buf32[8]; device_param->kernel_params_mp_l_buf64[3] = 0; device_param->kernel_params_mp_l_buf32[4] = 0; device_param->kernel_params_mp_l_buf32[5] = 0; device_param->kernel_params_mp_l_buf32[6] = 0; device_param->kernel_params_mp_l_buf32[7] = 0; device_param->kernel_params_mp_l_buf32[8] = 0; device_param->kernel_params_mp_l_buf32[9] = 0; device_param->kernel_params_mp_l[0] = NULL; device_param->kernel_params_mp_l[1] = NULL; device_param->kernel_params_mp_l[2] = NULL; device_param->kernel_params_mp_l[3] = &device_param->kernel_params_mp_l_buf64[3]; device_param->kernel_params_mp_l[4] = &device_param->kernel_params_mp_l_buf32[4]; device_param->kernel_params_mp_l[5] = &device_param->kernel_params_mp_l_buf32[5]; device_param->kernel_params_mp_l[6] = &device_param->kernel_params_mp_l_buf32[6]; device_param->kernel_params_mp_l[7] = &device_param->kernel_params_mp_l_buf32[7]; device_param->kernel_params_mp_l[8] = &device_param->kernel_params_mp_l_buf32[8]; device_param->kernel_params_mp_l[9] = &device_param->kernel_params_mp_l_buf32[9]; device_param->kernel_params_mp_r_buf64[3] = 0; device_param->kernel_params_mp_r_buf32[4] = 0; device_param->kernel_params_mp_r_buf32[5] = 0; device_param->kernel_params_mp_r_buf32[6] = 0; device_param->kernel_params_mp_r_buf32[7] = 0; device_param->kernel_params_mp_r_buf32[8] = 0; device_param->kernel_params_mp_r[0] = NULL; device_param->kernel_params_mp_r[1] = NULL; device_param->kernel_params_mp_r[2] = NULL; device_param->kernel_params_mp_r[3] = &device_param->kernel_params_mp_r_buf64[3]; device_param->kernel_params_mp_r[4] = &device_param->kernel_params_mp_r_buf32[4]; device_param->kernel_params_mp_r[5] = &device_param->kernel_params_mp_r_buf32[5]; device_param->kernel_params_mp_r[6] = &device_param->kernel_params_mp_r_buf32[6]; device_param->kernel_params_mp_r[7] = &device_param->kernel_params_mp_r_buf32[7]; device_param->kernel_params_mp_r[8] = &device_param->kernel_params_mp_r_buf32[8]; device_param->kernel_params_amp_buf32[5] = 0; // combs_mode device_param->kernel_params_amp_buf32[6] = 0; // gid_max device_param->kernel_params_amp[0] = &device_param->d_pws_buf; device_param->kernel_params_amp[1] = &device_param->d_pws_amp_buf; device_param->kernel_params_amp[2] = &device_param->d_rules_c; device_param->kernel_params_amp[3] = &device_param->d_combs_c; device_param->kernel_params_amp[4] = &device_param->d_bfs_c; device_param->kernel_params_amp[5] = &device_param->kernel_params_amp_buf32[5]; device_param->kernel_params_amp[6] = &device_param->kernel_params_amp_buf32[6]; device_param->kernel_params_tm[0] = &device_param->d_bfs_c; device_param->kernel_params_tm[1] = &device_param->d_tm_c; device_param->kernel_params_memset_buf32[1] = 0; // value device_param->kernel_params_memset_buf32[2] = 0; // gid_max device_param->kernel_params_memset[0] = NULL; device_param->kernel_params_memset[1] = &device_param->kernel_params_memset_buf32[1]; device_param->kernel_params_memset[2] = &device_param->kernel_params_memset_buf32[2]; /** * kernel name */ size_t kernel_wgs_tmp; char kernel_name[64] = { 0 }; if (hashconfig->attack_exec == ATTACK_EXEC_INSIDE_KERNEL) { if (hashconfig->opti_type & OPTI_TYPE_SINGLE_HASH) { snprintf (kernel_name, sizeof (kernel_name) - 1, "m%05d_s%02d", hashconfig->kern_type, 4); CL_err = hc_clCreateKernel (opencl_ctx->ocl, device_param->program, kernel_name, &device_param->kernel1); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } snprintf (kernel_name, sizeof (kernel_name) - 1, "m%05d_s%02d", hashconfig->kern_type, 8); CL_err = hc_clCreateKernel (opencl_ctx->ocl, device_param->program, kernel_name, &device_param->kernel2); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } snprintf (kernel_name, sizeof (kernel_name) - 1, "m%05d_s%02d", hashconfig->kern_type, 16); CL_err = hc_clCreateKernel (opencl_ctx->ocl, device_param->program, kernel_name, &device_param->kernel3); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } } else { snprintf (kernel_name, sizeof (kernel_name) - 1, "m%05d_m%02d", hashconfig->kern_type, 4); CL_err = hc_clCreateKernel (opencl_ctx->ocl, device_param->program, kernel_name, &device_param->kernel1); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } snprintf (kernel_name, sizeof (kernel_name) - 1, "m%05d_m%02d", hashconfig->kern_type, 8); CL_err = hc_clCreateKernel (opencl_ctx->ocl, device_param->program, kernel_name, &device_param->kernel2); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } snprintf (kernel_name, sizeof (kernel_name) - 1, "m%05d_m%02d", hashconfig->kern_type, 16); CL_err = hc_clCreateKernel (opencl_ctx->ocl, device_param->program, kernel_name, &device_param->kernel3); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } } if (session_ctx->attack_mode == ATTACK_MODE_BF) { if (hashconfig->opts_type & OPTS_TYPE_PT_BITSLICE) { snprintf (kernel_name, sizeof (kernel_name) - 1, "m%05d_tm", hashconfig->kern_type); CL_err = hc_clCreateKernel (opencl_ctx->ocl, device_param->program, kernel_name, &device_param->kernel_tm); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clGetKernelWorkGroupInfo (opencl_ctx->ocl, device_param->kernel_tm, device_param->device, CL_KERNEL_WORK_GROUP_SIZE, sizeof (size_t), &kernel_wgs_tmp, NULL); kernel_threads = MIN (kernel_threads, kernel_wgs_tmp); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetKernelWorkGroupInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } } } } else { snprintf (kernel_name, sizeof (kernel_name) - 1, "m%05d_init", hashconfig->kern_type); CL_err = hc_clCreateKernel (opencl_ctx->ocl, device_param->program, kernel_name, &device_param->kernel1); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } snprintf (kernel_name, sizeof (kernel_name) - 1, "m%05d_loop", hashconfig->kern_type); CL_err = hc_clCreateKernel (opencl_ctx->ocl, device_param->program, kernel_name, &device_param->kernel2); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } snprintf (kernel_name, sizeof (kernel_name) - 1, "m%05d_comp", hashconfig->kern_type); CL_err = hc_clCreateKernel (opencl_ctx->ocl, device_param->program, kernel_name, &device_param->kernel3); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } if (hashconfig->opts_type & OPTS_TYPE_HOOK12) { snprintf (kernel_name, sizeof (kernel_name) - 1, "m%05d_hook12", hashconfig->kern_type); CL_err = hc_clCreateKernel (opencl_ctx->ocl, device_param->program, kernel_name, &device_param->kernel12); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clGetKernelWorkGroupInfo (opencl_ctx->ocl, device_param->kernel12, device_param->device, CL_KERNEL_WORK_GROUP_SIZE, sizeof (size_t), &kernel_wgs_tmp, NULL); kernel_threads = MIN (kernel_threads, kernel_wgs_tmp); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetKernelWorkGroupInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } } if (hashconfig->opts_type & OPTS_TYPE_HOOK23) { snprintf (kernel_name, sizeof (kernel_name) - 1, "m%05d_hook23", hashconfig->kern_type); CL_err = hc_clCreateKernel (opencl_ctx->ocl, device_param->program, kernel_name, &device_param->kernel23); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clGetKernelWorkGroupInfo (opencl_ctx->ocl, device_param->kernel23, device_param->device, CL_KERNEL_WORK_GROUP_SIZE, sizeof (size_t), &kernel_wgs_tmp, NULL); kernel_threads = MIN (kernel_threads, kernel_wgs_tmp); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetKernelWorkGroupInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } } } CL_err |= hc_clGetKernelWorkGroupInfo (opencl_ctx->ocl, device_param->kernel1, device_param->device, CL_KERNEL_WORK_GROUP_SIZE, sizeof (size_t), &kernel_wgs_tmp, NULL); kernel_threads = MIN (kernel_threads, kernel_wgs_tmp); CL_err |= hc_clGetKernelWorkGroupInfo (opencl_ctx->ocl, device_param->kernel2, device_param->device, CL_KERNEL_WORK_GROUP_SIZE, sizeof (size_t), &kernel_wgs_tmp, NULL); kernel_threads = MIN (kernel_threads, kernel_wgs_tmp); CL_err |= hc_clGetKernelWorkGroupInfo (opencl_ctx->ocl, device_param->kernel3, device_param->device, CL_KERNEL_WORK_GROUP_SIZE, sizeof (size_t), &kernel_wgs_tmp, NULL); kernel_threads = MIN (kernel_threads, kernel_wgs_tmp); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetKernelWorkGroupInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } for (uint i = 0; i <= 23; i++) { CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, device_param->kernel1, i, sizeof (cl_mem), device_param->kernel_params[i]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, device_param->kernel2, i, sizeof (cl_mem), device_param->kernel_params[i]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, device_param->kernel3, i, sizeof (cl_mem), device_param->kernel_params[i]); if (hashconfig->opts_type & OPTS_TYPE_HOOK12) CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, device_param->kernel12, i, sizeof (cl_mem), device_param->kernel_params[i]); if (hashconfig->opts_type & OPTS_TYPE_HOOK23) CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, device_param->kernel23, i, sizeof (cl_mem), device_param->kernel_params[i]); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clSetKernelArg(): %s\n", val2cstr_cl (CL_err)); return -1; } } for (uint i = 24; i <= 34; i++) { CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, device_param->kernel1, i, sizeof (cl_uint), device_param->kernel_params[i]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, device_param->kernel2, i, sizeof (cl_uint), device_param->kernel_params[i]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, device_param->kernel3, i, sizeof (cl_uint), device_param->kernel_params[i]); if (hashconfig->opts_type & OPTS_TYPE_HOOK12) CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, device_param->kernel12, i, sizeof (cl_uint), device_param->kernel_params[i]); if (hashconfig->opts_type & OPTS_TYPE_HOOK23) CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, device_param->kernel23, i, sizeof (cl_uint), device_param->kernel_params[i]); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clSetKernelArg(): %s\n", val2cstr_cl (CL_err)); return -1; } } // GPU memset CL_err = hc_clCreateKernel (opencl_ctx->ocl, device_param->program, "gpu_memset", &device_param->kernel_memset); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clGetKernelWorkGroupInfo (opencl_ctx->ocl, device_param->kernel_memset, device_param->device, CL_KERNEL_WORK_GROUP_SIZE, sizeof (size_t), &kernel_wgs_tmp, NULL); kernel_threads = MIN (kernel_threads, kernel_wgs_tmp); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetKernelWorkGroupInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, device_param->kernel_memset, 0, sizeof (cl_mem), device_param->kernel_params_memset[0]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, device_param->kernel_memset, 1, sizeof (cl_uint), device_param->kernel_params_memset[1]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, device_param->kernel_memset, 2, sizeof (cl_uint), device_param->kernel_params_memset[2]); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clSetKernelArg(): %s\n", val2cstr_cl (CL_err)); return -1; } // MP start if (session_ctx->attack_mode == ATTACK_MODE_BF) { CL_err |= hc_clCreateKernel (opencl_ctx->ocl, device_param->program_mp, "l_markov", &device_param->kernel_mp_l); CL_err |= hc_clCreateKernel (opencl_ctx->ocl, device_param->program_mp, "r_markov", &device_param->kernel_mp_r); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err |= hc_clGetKernelWorkGroupInfo (opencl_ctx->ocl, device_param->kernel_mp_l, device_param->device, CL_KERNEL_WORK_GROUP_SIZE, sizeof (size_t), &kernel_wgs_tmp, NULL); kernel_threads = MIN (kernel_threads, kernel_wgs_tmp); CL_err |= hc_clGetKernelWorkGroupInfo (opencl_ctx->ocl, device_param->kernel_mp_r, device_param->device, CL_KERNEL_WORK_GROUP_SIZE, sizeof (size_t), &kernel_wgs_tmp, NULL); kernel_threads = MIN (kernel_threads, kernel_wgs_tmp); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetKernelWorkGroupInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } if (hashconfig->opts_type & OPTS_TYPE_PT_BITSLICE) { CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, device_param->kernel_tm, 0, sizeof (cl_mem), device_param->kernel_params_tm[0]); CL_err |= hc_clSetKernelArg (opencl_ctx->ocl, device_param->kernel_tm, 1, sizeof (cl_mem), device_param->kernel_params_tm[1]); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clSetKernelArg(): %s\n", val2cstr_cl (CL_err)); return -1; } } } else if (session_ctx->attack_mode == ATTACK_MODE_HYBRID1) { CL_err = hc_clCreateKernel (opencl_ctx->ocl, device_param->program_mp, "C_markov", &device_param->kernel_mp); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clGetKernelWorkGroupInfo (opencl_ctx->ocl, device_param->kernel_mp, device_param->device, CL_KERNEL_WORK_GROUP_SIZE, sizeof (size_t), &kernel_wgs_tmp, NULL); kernel_threads = MIN (kernel_threads, kernel_wgs_tmp); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetKernelWorkGroupInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } } else if (session_ctx->attack_mode == ATTACK_MODE_HYBRID2) { CL_err = hc_clCreateKernel (opencl_ctx->ocl, device_param->program_mp, "C_markov", &device_param->kernel_mp); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clGetKernelWorkGroupInfo (opencl_ctx->ocl, device_param->kernel_mp, device_param->device, CL_KERNEL_WORK_GROUP_SIZE, sizeof (size_t), &kernel_wgs_tmp, NULL); kernel_threads = MIN (kernel_threads, kernel_wgs_tmp); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetKernelWorkGroupInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } } if (hashconfig->attack_exec == ATTACK_EXEC_INSIDE_KERNEL) { // nothing to do } else { CL_err = hc_clCreateKernel (opencl_ctx->ocl, device_param->program_amp, "amp", &device_param->kernel_amp); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clCreateKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } CL_err = hc_clGetKernelWorkGroupInfo (opencl_ctx->ocl, device_param->kernel_amp, device_param->device, CL_KERNEL_WORK_GROUP_SIZE, sizeof (size_t), &kernel_wgs_tmp, NULL); kernel_threads = MIN (kernel_threads, kernel_wgs_tmp); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clGetKernelWorkGroupInfo(): %s\n", val2cstr_cl (CL_err)); return -1; } } if (hashconfig->attack_exec == ATTACK_EXEC_INSIDE_KERNEL) { // nothing to do } else { for (uint i = 0; i < 5; i++) { CL_err = hc_clSetKernelArg (opencl_ctx->ocl, device_param->kernel_amp, i, sizeof (cl_mem), device_param->kernel_params_amp[i]); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clSetKernelArg(): %s\n", val2cstr_cl (CL_err)); return -1; } } for (uint i = 5; i < 7; i++) { CL_err = hc_clSetKernelArg (opencl_ctx->ocl, device_param->kernel_amp, i, sizeof (cl_uint), device_param->kernel_params_amp[i]); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clSetKernelArg(): %s\n", val2cstr_cl (CL_err)); return -1; } } } // maybe this has been updated by clGetKernelWorkGroupInfo() // value can only be decreased, so we don't need to reallocate buffers device_param->kernel_threads = kernel_threads; // zero some data buffers run_kernel_bzero (opencl_ctx, device_param, device_param->d_pws_buf, size_pws); run_kernel_bzero (opencl_ctx, device_param, device_param->d_pws_amp_buf, size_pws); run_kernel_bzero (opencl_ctx, device_param, device_param->d_tmps, size_tmps); run_kernel_bzero (opencl_ctx, device_param, device_param->d_hooks, size_hooks); run_kernel_bzero (opencl_ctx, device_param, device_param->d_plain_bufs, size_plains); run_kernel_bzero (opencl_ctx, device_param, device_param->d_result, size_results); /** * special buffers */ if (session_ctx->attack_kern == ATTACK_KERN_STRAIGHT) { run_kernel_bzero (opencl_ctx, device_param, device_param->d_rules_c, size_rules_c); } else if (session_ctx->attack_kern == ATTACK_KERN_COMBI) { run_kernel_bzero (opencl_ctx, device_param, device_param->d_combs, size_combs); run_kernel_bzero (opencl_ctx, device_param, device_param->d_combs_c, size_combs); run_kernel_bzero (opencl_ctx, device_param, device_param->d_root_css_buf, size_root_css); run_kernel_bzero (opencl_ctx, device_param, device_param->d_markov_css_buf, size_markov_css); } else if (session_ctx->attack_kern == ATTACK_KERN_BF) { run_kernel_bzero (opencl_ctx, device_param, device_param->d_bfs, size_bfs); run_kernel_bzero (opencl_ctx, device_param, device_param->d_bfs_c, size_bfs); run_kernel_bzero (opencl_ctx, device_param, device_param->d_tm_c, size_tm); run_kernel_bzero (opencl_ctx, device_param, device_param->d_root_css_buf, size_root_css); run_kernel_bzero (opencl_ctx, device_param, device_param->d_markov_css_buf, size_markov_css); } } return 0; } int opencl_session_destroy (opencl_ctx_t *opencl_ctx) { for (uint device_id = 0; device_id < opencl_ctx->devices_cnt; device_id++) { hc_device_param_t *device_param = &opencl_ctx->devices_param[device_id]; if (device_param->skipped) continue; cl_int CL_err = CL_SUCCESS; myfree (device_param->pws_buf); myfree (device_param->combs_buf); myfree (device_param->hooks_buf); if (device_param->d_pws_buf) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_pws_buf); if (device_param->d_pws_amp_buf) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_pws_amp_buf); if (device_param->d_rules) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_rules); if (device_param->d_rules_c) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_rules_c); if (device_param->d_combs) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_combs); if (device_param->d_combs_c) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_combs_c); if (device_param->d_bfs) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_bfs); if (device_param->d_bfs_c) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_bfs_c); if (device_param->d_bitmap_s1_a) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_bitmap_s1_a); if (device_param->d_bitmap_s1_b) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_bitmap_s1_b); if (device_param->d_bitmap_s1_c) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_bitmap_s1_c); if (device_param->d_bitmap_s1_d) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_bitmap_s1_d); if (device_param->d_bitmap_s2_a) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_bitmap_s2_a); if (device_param->d_bitmap_s2_b) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_bitmap_s2_b); if (device_param->d_bitmap_s2_c) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_bitmap_s2_c); if (device_param->d_bitmap_s2_d) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_bitmap_s2_d); if (device_param->d_plain_bufs) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_plain_bufs); if (device_param->d_digests_buf) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_digests_buf); if (device_param->d_digests_shown) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_digests_shown); if (device_param->d_salt_bufs) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_salt_bufs); if (device_param->d_esalt_bufs) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_esalt_bufs); if (device_param->d_tmps) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_tmps); if (device_param->d_hooks) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_hooks); if (device_param->d_result) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_result); if (device_param->d_scryptV0_buf) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_scryptV0_buf); if (device_param->d_scryptV1_buf) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_scryptV1_buf); if (device_param->d_scryptV2_buf) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_scryptV2_buf); if (device_param->d_scryptV3_buf) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_scryptV3_buf); if (device_param->d_root_css_buf) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_root_css_buf); if (device_param->d_markov_css_buf) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_markov_css_buf); if (device_param->d_tm_c) CL_err |= hc_clReleaseMemObject (opencl_ctx->ocl, device_param->d_tm_c); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clReleaseMemObject(): %s\n", val2cstr_cl (CL_err)); return -1; } if (device_param->kernel1) CL_err |= hc_clReleaseKernel (opencl_ctx->ocl, device_param->kernel1); if (device_param->kernel12) CL_err |= hc_clReleaseKernel (opencl_ctx->ocl, device_param->kernel12); if (device_param->kernel2) CL_err |= hc_clReleaseKernel (opencl_ctx->ocl, device_param->kernel2); if (device_param->kernel23) CL_err |= hc_clReleaseKernel (opencl_ctx->ocl, device_param->kernel23); if (device_param->kernel3) CL_err |= hc_clReleaseKernel (opencl_ctx->ocl, device_param->kernel3); if (device_param->kernel_mp) CL_err |= hc_clReleaseKernel (opencl_ctx->ocl, device_param->kernel_mp); if (device_param->kernel_mp_l) CL_err |= hc_clReleaseKernel (opencl_ctx->ocl, device_param->kernel_mp_l); if (device_param->kernel_mp_r) CL_err |= hc_clReleaseKernel (opencl_ctx->ocl, device_param->kernel_mp_r); if (device_param->kernel_tm) CL_err |= hc_clReleaseKernel (opencl_ctx->ocl, device_param->kernel_tm); if (device_param->kernel_amp) CL_err |= hc_clReleaseKernel (opencl_ctx->ocl, device_param->kernel_amp); if (device_param->kernel_memset) CL_err |= hc_clReleaseKernel (opencl_ctx->ocl, device_param->kernel_memset); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clReleaseKernel(): %s\n", val2cstr_cl (CL_err)); return -1; } if (device_param->program) CL_err |= hc_clReleaseProgram (opencl_ctx->ocl, device_param->program); if (device_param->program_mp) CL_err |= hc_clReleaseProgram (opencl_ctx->ocl, device_param->program_mp); if (device_param->program_amp) CL_err |= hc_clReleaseProgram (opencl_ctx->ocl, device_param->program_amp); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clReleaseProgram(): %s\n", val2cstr_cl (CL_err)); return -1; } if (device_param->command_queue) CL_err |= hc_clReleaseCommandQueue (opencl_ctx->ocl, device_param->command_queue); if (CL_err != CL_SUCCESS) { log_error ("ERROR: clReleaseCommandQueue(): %s\n", val2cstr_cl (CL_err)); return -1; } if (device_param->context) CL_err |= hc_clReleaseContext (opencl_ctx->ocl, device_param->context); if (CL_err != CL_SUCCESS) { log_error ("ERROR: hc_clReleaseContext(): %s\n", val2cstr_cl (CL_err)); return -1; } } return 0; }