X-Git-Url: https://www.flypig.org.uk/git/?a=blobdiff_plain;f=OpenCL%2Fm01100_a3.cl;h=cd23d404bb41e7efb34ab601c9cd71fccbe6ae6e;hb=8702d0e3e155e6a61ca1b0dcdc1ec924d92b8436;hp=1036de158c71bd5db3ce0bf1849a170a41c96eb2;hpb=6893a9ad857a1d4fe63f1a92365802f5d9d3b39e;p=hashcat.git diff --git a/OpenCL/m01100_a3.cl b/OpenCL/m01100_a3.cl index 1036de1..cd23d40 100644 --- a/OpenCL/m01100_a3.cl +++ b/OpenCL/m01100_a3.cl @@ -5,22 +5,16 @@ #define _MD4_ -#include "include/constants.h" -#include "include/kernel_vendor.h" +#define NEW_SIMD_CODE -#define DGST_R0 0 -#define DGST_R1 3 -#define DGST_R2 2 -#define DGST_R3 1 +#include "inc_vendor.cl" +#include "inc_hash_constants.h" +#include "inc_hash_functions.cl" +#include "inc_types.cl" +#include "inc_common.cl" +#include "inc_simd.cl" -#include "include/kernel_functions.c" -#include "types_ocl.c" -#include "common.c" - -#define COMPARE_S "check_single_comp4.c" -#define COMPARE_M "check_multi_comp4.c" - -static void m01100m (u32 w[16], const u32 pw_len, __global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32 * words_buf_r, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset) +void m01100m (__local salt_t s_salt_buf[1], u32 w[16], const u32 pw_len, __global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32x * words_buf_r, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset) { /** * modifier @@ -33,28 +27,17 @@ static void m01100m (u32 w[16], const u32 pw_len, __global pw_t *pws, __global g * salt */ - u32 salt_buf0[4]; - - salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0]; - salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1]; - salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2]; - salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3]; - - u32 salt_buf1[4]; - - salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4]; - salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5]; - salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6]; - salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7]; - - u32 salt_buf2[4]; - - salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8]; - salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9]; - salt_buf2[2] = 0; - salt_buf2[3] = 0; - - const u32 salt_len = salt_bufs[salt_pos].salt_len; + #define salt_buf00 s_salt_buf[0].salt_buf[ 0] + #define salt_buf01 s_salt_buf[0].salt_buf[ 1] + #define salt_buf02 s_salt_buf[0].salt_buf[ 2] + #define salt_buf03 s_salt_buf[0].salt_buf[ 3] + #define salt_buf04 s_salt_buf[0].salt_buf[ 4] + #define salt_buf05 s_salt_buf[0].salt_buf[ 5] + #define salt_buf06 s_salt_buf[0].salt_buf[ 6] + #define salt_buf07 s_salt_buf[0].salt_buf[ 7] + #define salt_buf08 s_salt_buf[0].salt_buf[ 8] + #define salt_buf09 s_salt_buf[0].salt_buf[ 9] + #define salt_buf10 s_salt_buf[0].salt_buf[10] /** * base @@ -117,16 +100,16 @@ static void m01100m (u32 w[16], const u32 pw_len, __global pw_t *pws, __global g u32 w0l = w[0]; - for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos++) + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) { - const u32 w0r = words_buf_r[il_pos]; + const u32x w0r = words_buf_r[il_pos / VECT_SIZE]; - const u32 w0 = w0l | w0r; + const u32x w0 = w0l | w0r; - u32 a = MD4M_A; - u32 b = MD4M_B; - u32 c = MD4M_C; - u32 d = MD4M_D; + u32x a = MD4M_A; + u32x b = MD4M_B; + u32x c = MD4M_C; + u32x d = MD4M_D; MD4_STEP (MD4_Fo, a, b, c, d, w0, F_w0c00, MD4S00); MD4_STEP0(MD4_Fo, d, a, b, c, F_w1c00, MD4S01); @@ -184,26 +167,26 @@ static void m01100m (u32 w[16], const u32 pw_len, __global pw_t *pws, __global g c += MD4M_C; d += MD4M_D; - u32 w0_t[4]; - u32 w1_t[4]; - u32 w2_t[4]; - u32 w3_t[4]; + u32x w0_t[4]; + u32x w1_t[4]; + u32x w2_t[4]; + u32x w3_t[4]; w0_t[0] = a; w0_t[1] = b; w0_t[2] = c; w0_t[3] = d; - w1_t[0] = salt_buf0[0]; - w1_t[1] = salt_buf0[1]; - w1_t[2] = salt_buf0[2]; - w1_t[3] = salt_buf0[3]; - w2_t[0] = salt_buf1[0]; - w2_t[1] = salt_buf1[1]; - w2_t[2] = salt_buf1[2]; - w2_t[3] = salt_buf1[3]; - w3_t[0] = salt_buf2[0]; - w3_t[1] = salt_buf2[1]; - w3_t[2] = (16 + salt_len) * 8; + w1_t[0] = salt_buf00; + w1_t[1] = salt_buf01; + w1_t[2] = salt_buf02; + w1_t[3] = salt_buf03; + w2_t[0] = salt_buf04; + w2_t[1] = salt_buf05; + w2_t[2] = salt_buf06; + w2_t[3] = salt_buf07; + w3_t[0] = salt_buf08; + w3_t[1] = salt_buf09; + w3_t[2] = salt_buf10; w3_t[3] = 0; a = MD4M_A; @@ -262,16 +245,11 @@ static void m01100m (u32 w[16], const u32 pw_len, __global pw_t *pws, __global g MD4_STEP (MD4_H , c, d, a, b, w1_t[3], MD4C02, MD4S22); MD4_STEP (MD4_H , b, c, d, a, w3_t[3], MD4C02, MD4S23); - const u32 r0 = a; - const u32 r1 = d; - const u32 r2 = c; - const u32 r3 = b; - - #include COMPARE_M + COMPARE_M_SIMD (a, d, c, b); } } -static void m01100s (u32 w[16], const u32 pw_len, __global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32 * words_buf_r, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset) +void m01100s (__local salt_t s_salt_buf[1], u32 w[16], const u32 pw_len, __global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32x * words_buf_r, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset) { /** * modifier @@ -284,28 +262,17 @@ static void m01100s (u32 w[16], const u32 pw_len, __global pw_t *pws, __global g * salt */ - u32 salt_buf0[4]; - - salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0]; - salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1]; - salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2]; - salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3]; - - u32 salt_buf1[4]; - - salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4]; - salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5]; - salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6]; - salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7]; - - u32 salt_buf2[4]; - - salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8]; - salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9]; - salt_buf2[2] = 0; - salt_buf2[3] = 0; - - const u32 salt_len = salt_bufs[salt_pos].salt_len; + #define salt_buf00 s_salt_buf[0].salt_buf[ 0] + #define salt_buf01 s_salt_buf[0].salt_buf[ 1] + #define salt_buf02 s_salt_buf[0].salt_buf[ 2] + #define salt_buf03 s_salt_buf[0].salt_buf[ 3] + #define salt_buf04 s_salt_buf[0].salt_buf[ 4] + #define salt_buf05 s_salt_buf[0].salt_buf[ 5] + #define salt_buf06 s_salt_buf[0].salt_buf[ 6] + #define salt_buf07 s_salt_buf[0].salt_buf[ 7] + #define salt_buf08 s_salt_buf[0].salt_buf[ 8] + #define salt_buf09 s_salt_buf[0].salt_buf[ 9] + #define salt_buf10 s_salt_buf[0].salt_buf[10] /** * base @@ -380,16 +347,16 @@ static void m01100s (u32 w[16], const u32 pw_len, __global pw_t *pws, __global g u32 w0l = w[0]; - for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos++) + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) { - const u32 w0r = words_buf_r[il_pos]; + const u32x w0r = words_buf_r[il_pos / VECT_SIZE]; - const u32 w0 = w0l | w0r; + const u32x w0 = w0l | w0r; - u32 a = MD4M_A; - u32 b = MD4M_B; - u32 c = MD4M_C; - u32 d = MD4M_D; + u32x a = MD4M_A; + u32x b = MD4M_B; + u32x c = MD4M_C; + u32x d = MD4M_D; MD4_STEP (MD4_Fo, a, b, c, d, w0, F_w0c00, MD4S00); MD4_STEP0(MD4_Fo, d, a, b, c, F_w1c00, MD4S01); @@ -447,26 +414,26 @@ static void m01100s (u32 w[16], const u32 pw_len, __global pw_t *pws, __global g c += MD4M_C; d += MD4M_D; - u32 w0_t[4]; - u32 w1_t[4]; - u32 w2_t[4]; - u32 w3_t[4]; + u32x w0_t[4]; + u32x w1_t[4]; + u32x w2_t[4]; + u32x w3_t[4]; w0_t[0] = a; w0_t[1] = b; w0_t[2] = c; w0_t[3] = d; - w1_t[0] = salt_buf0[0]; - w1_t[1] = salt_buf0[1]; - w1_t[2] = salt_buf0[2]; - w1_t[3] = salt_buf0[3]; - w2_t[0] = salt_buf1[0]; - w2_t[1] = salt_buf1[1]; - w2_t[2] = salt_buf1[2]; - w2_t[3] = salt_buf1[3]; - w3_t[0] = salt_buf2[0]; - w3_t[1] = salt_buf2[1]; - w3_t[2] = (16 + salt_len) * 8; + w1_t[0] = salt_buf00; + w1_t[1] = salt_buf01; + w1_t[2] = salt_buf02; + w1_t[3] = salt_buf03; + w2_t[0] = salt_buf04; + w2_t[1] = salt_buf05; + w2_t[2] = salt_buf06; + w2_t[3] = salt_buf07; + w3_t[0] = salt_buf08; + w3_t[1] = salt_buf09; + w3_t[2] = salt_buf10; w3_t[3] = 0; a = MD4M_A; @@ -522,32 +489,24 @@ static void m01100s (u32 w[16], const u32 pw_len, __global pw_t *pws, __global g MD4_STEP (MD4_H , b, c, d, a, w3_t[1], MD4C02, MD4S23); MD4_STEP (MD4_H , a, b, c, d, w0_t[3], MD4C02, MD4S20); - bool q_cond = allx (search[0] != a); - - if (q_cond) continue; + if (MATCHES_NONE_VS (a, search[0])) continue; MD4_STEP (MD4_H , d, a, b, c, w2_t[3], MD4C02, MD4S21); MD4_STEP (MD4_H , c, d, a, b, w1_t[3], MD4C02, MD4S22); MD4_STEP (MD4_H , b, c, d, a, w3_t[3], MD4C02, MD4S23); - const u32 r0 = a; - const u32 r1 = d; - const u32 r2 = c; - const u32 r3 = b; - - #include COMPARE_S + COMPARE_S_SIMD (a, d, c, b); } } -__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01100_m04 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32 * words_buf_r, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) +__kernel void m01100_m04 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32x * words_buf_r, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) { /** * base */ const u32 gid = get_global_id (0); - - if (gid >= gid_max) return; + const u32 lid = get_local_id (0); u32 w[16]; @@ -570,22 +529,38 @@ __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01100_m04 (__glo const u32 pw_len = pws[gid].pw_len; + /** + * salt + */ + + __local salt_t s_salt_buf[1]; + + if (lid == 0) + { + s_salt_buf[0] = salt_bufs[salt_pos]; + + s_salt_buf[0].salt_buf[10] = (16 + s_salt_buf[0].salt_len) * 8; + } + + barrier (CLK_LOCAL_MEM_FENCE); + + if (gid >= gid_max) return; + /** * main */ - m01100m (w, pw_len, pws, rules_buf, combs_buf, words_buf_r, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset); + m01100m (s_salt_buf, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_scryptV0_buf, d_scryptV1_buf, d_scryptV2_buf, d_scryptV3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset); } -__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01100_m08 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32 * words_buf_r, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) +__kernel void m01100_m08 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32x * words_buf_r, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) { /** * base */ const u32 gid = get_global_id (0); - - if (gid >= gid_max) return; + const u32 lid = get_local_id (0); u32 w[16]; @@ -608,26 +583,42 @@ __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01100_m08 (__glo const u32 pw_len = pws[gid].pw_len; + /** + * salt + */ + + __local salt_t s_salt_buf[1]; + + if (lid == 0) + { + s_salt_buf[0] = salt_bufs[salt_pos]; + + s_salt_buf[0].salt_buf[10] = (16 + s_salt_buf[0].salt_len) * 8; + } + + barrier (CLK_LOCAL_MEM_FENCE); + + if (gid >= gid_max) return; + /** * main */ - m01100m (w, pw_len, pws, rules_buf, combs_buf, words_buf_r, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset); + m01100m (s_salt_buf, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_scryptV0_buf, d_scryptV1_buf, d_scryptV2_buf, d_scryptV3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset); } -__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01100_m16 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32 * words_buf_r, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) +__kernel void m01100_m16 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32x * words_buf_r, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) { } -__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01100_s04 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32 * words_buf_r, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) +__kernel void m01100_s04 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32x * words_buf_r, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) { /** * base */ const u32 gid = get_global_id (0); - - if (gid >= gid_max) return; + const u32 lid = get_local_id (0); u32 w[16]; @@ -650,22 +641,38 @@ __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01100_s04 (__glo const u32 pw_len = pws[gid].pw_len; + /** + * salt + */ + + __local salt_t s_salt_buf[1]; + + if (lid == 0) + { + s_salt_buf[0] = salt_bufs[salt_pos]; + + s_salt_buf[0].salt_buf[10] = (16 + s_salt_buf[0].salt_len) * 8; + } + + barrier (CLK_LOCAL_MEM_FENCE); + + if (gid >= gid_max) return; + /** * main */ - m01100s (w, pw_len, pws, rules_buf, combs_buf, words_buf_r, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset); + m01100s (s_salt_buf, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_scryptV0_buf, d_scryptV1_buf, d_scryptV2_buf, d_scryptV3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset); } -__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01100_s08 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32 * words_buf_r, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) +__kernel void m01100_s08 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32x * words_buf_r, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) { /** * base */ const u32 gid = get_global_id (0); - - if (gid >= gid_max) return; + const u32 lid = get_local_id (0); u32 w[16]; @@ -688,13 +695,30 @@ __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01100_s08 (__glo const u32 pw_len = pws[gid].pw_len; + /** + * salt + */ + + __local salt_t s_salt_buf[1]; + + if (lid == 0) + { + s_salt_buf[0] = salt_bufs[salt_pos]; + + s_salt_buf[0].salt_buf[10] = (16 + s_salt_buf[0].salt_len) * 8; + } + + barrier (CLK_LOCAL_MEM_FENCE); + + if (gid >= gid_max) return; + /** * main */ - m01100s (w, pw_len, pws, rules_buf, combs_buf, words_buf_r, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset); + m01100s (s_salt_buf, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_scryptV0_buf, d_scryptV1_buf, d_scryptV2_buf, d_scryptV3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset); } -__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01100_s16 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32 * words_buf_r, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) +__kernel void m01100_s16 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32x * words_buf_r, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) { }