/** * Author......: Jens Steube * License.....: MIT */ #define _SHA1_ #include "include/constants.h" #include "include/kernel_vendor.h" #ifdef VLIW1 #define VECT_SIZE1 #endif #ifdef VLIW2 #define VECT_SIZE4 #endif #define DGST_R0 3 #define DGST_R1 4 #define DGST_R2 2 #define DGST_R3 1 #include "include/kernel_functions.c" #include "types_nv.c" #include "common_nv.c" #ifdef VECT_SIZE1 #define VECT_COMPARE_S "check_single_vect1_comp4.c" #define VECT_COMPARE_M "check_multi_vect1_comp4.c" #endif #ifdef VECT_SIZE2 #define VECT_COMPARE_S "check_single_vect2_comp4.c" #define VECT_COMPARE_M "check_multi_vect2_comp4.c" #endif #ifdef VECT_SIZE4 #define VECT_COMPARE_S "check_single_vect4_comp4.c" #define VECT_COMPARE_M "check_multi_vect4_comp4.c" #endif __device__ __constant__ comb_t c_combs[1024]; extern "C" __global__ void __launch_bounds__ (256, 1) m00110_m04 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, 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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) { /** * modifier */ const u32 lid = threadIdx.x; /** * base */ const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x; if (gid >= gid_max) return; u32x wordl0[4]; wordl0[0] = pws[gid].i[ 0]; wordl0[1] = pws[gid].i[ 1]; wordl0[2] = pws[gid].i[ 2]; wordl0[3] = pws[gid].i[ 3]; u32x wordl1[4]; wordl1[0] = pws[gid].i[ 4]; wordl1[1] = pws[gid].i[ 5]; wordl1[2] = pws[gid].i[ 6]; wordl1[3] = pws[gid].i[ 7]; u32x wordl2[4]; wordl2[0] = 0; wordl2[1] = 0; wordl2[2] = 0; wordl2[3] = 0; u32x wordl3[4]; wordl3[0] = 0; wordl3[1] = 0; wordl3[2] = 0; wordl3[3] = 0; const u32 pw_l_len = pws[gid].pw_len; if (combs_mode == COMBINATOR_MODE_BASE_RIGHT) { switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, c_combs[0].pw_len); } /** * 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]; const u32 salt_len = salt_bufs[salt_pos].salt_len; /** * loop */ for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++) { const u32 pw_r_len = c_combs[il_pos].pw_len; const u32 pw_len = pw_l_len + pw_r_len; u32 wordr0[4]; wordr0[0] = c_combs[il_pos].i[0]; wordr0[1] = c_combs[il_pos].i[1]; wordr0[2] = c_combs[il_pos].i[2]; wordr0[3] = c_combs[il_pos].i[3]; u32 wordr1[4]; wordr1[0] = c_combs[il_pos].i[4]; wordr1[1] = c_combs[il_pos].i[5]; wordr1[2] = c_combs[il_pos].i[6]; wordr1[3] = c_combs[il_pos].i[7]; u32 wordr2[4]; wordr2[0] = 0; wordr2[1] = 0; wordr2[2] = 0; wordr2[3] = 0; u32 wordr3[4]; wordr3[0] = 0; wordr3[1] = 0; wordr3[2] = 0; wordr3[3] = 0; if (combs_mode == COMBINATOR_MODE_BASE_LEFT) { switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len); } /** * append salt */ u32 s0[4]; s0[0] = salt_buf0[0]; s0[1] = salt_buf0[1]; s0[2] = salt_buf0[2]; s0[3] = salt_buf0[3]; u32 s1[4]; s1[0] = salt_buf1[0]; s1[1] = salt_buf1[1]; s1[2] = salt_buf1[2]; s1[3] = salt_buf1[3]; u32 s2[4]; s2[0] = 0; s2[1] = 0; s2[2] = 0; s2[3] = 0; u32 s3[4]; s3[0] = 0; s3[1] = 0; s3[2] = 0; s3[3] = 0; switch_buffer_by_offset (s0, s1, s2, s3, pw_len); const u32 pw_salt_len = pw_len + salt_len; u32x w0[4]; w0[0] = wordl0[0] | wordr0[0] | s0[0]; w0[1] = wordl0[1] | wordr0[1] | s0[1]; w0[2] = wordl0[2] | wordr0[2] | s0[2]; w0[3] = wordl0[3] | wordr0[3] | s0[3]; u32x w1[4]; w1[0] = wordl1[0] | wordr1[0] | s1[0]; w1[1] = wordl1[1] | wordr1[1] | s1[1]; w1[2] = wordl1[2] | wordr1[2] | s1[2]; w1[3] = wordl1[3] | wordr1[3] | s1[3]; u32x w2[4]; w2[0] = wordl2[0] | wordr2[0] | s2[0]; w2[1] = wordl2[1] | wordr2[1] | s2[1]; w2[2] = wordl2[2] | wordr2[2] | s2[2]; w2[3] = wordl2[3] | wordr2[3] | s2[3]; u32x w3[4]; w3[0] = wordl3[0] | wordr3[0] | s3[0]; w3[1] = wordl3[1] | wordr3[1] | s3[1]; w3[2] = 0; w3[3] = 0; append_0x80_4 (w0, w1, w2, w3, pw_salt_len); /** * sha1 */ u32x w0_t = swap_workaround (w0[0]); u32x w1_t = swap_workaround (w0[1]); u32x w2_t = swap_workaround (w0[2]); u32x w3_t = swap_workaround (w0[3]); u32x w4_t = swap_workaround (w1[0]); u32x w5_t = swap_workaround (w1[1]); u32x w6_t = swap_workaround (w1[2]); u32x w7_t = swap_workaround (w1[3]); u32x w8_t = swap_workaround (w2[0]); u32x w9_t = swap_workaround (w2[1]); u32x wa_t = swap_workaround (w2[2]); u32x wb_t = swap_workaround (w2[3]); u32x wc_t = swap_workaround (w3[0]); u32x wd_t = swap_workaround (w3[1]); u32x we_t = 0; u32x wf_t = pw_salt_len * 8; u32x a = SHA1M_A; u32x b = SHA1M_B; u32x c = SHA1M_C; u32x d = SHA1M_D; u32x e = SHA1M_E; #undef K #define K SHA1C00 SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t); SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t); SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t); SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t); SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t); SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t); SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t); SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t); w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t); w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t); w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t); w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t); #undef K #define K SHA1C01 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t); w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t); w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t); w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t); w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t); w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t); wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t); wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t); wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t); wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t); we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t); wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t); w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t); w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t); w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t); w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t); w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t); w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t); w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t); w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t); #undef K #define K SHA1C02 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t); w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t); wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t); wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t); wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t); wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t); we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t); wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t); w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t); w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t); w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t); w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t); w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t); w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t); w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t); w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t); w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t); w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t); wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t); wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t); #undef K #define K SHA1C03 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t); wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t); we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t); wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t); w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t); w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t); w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t); w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t); w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t); w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t); w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t); w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t); w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t); w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t); wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t); wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t); wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t); wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t); we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); const u32x r0 = d; const u32x r1 = e; const u32x r2 = c; const u32x r3 = b; #include VECT_COMPARE_M } } extern "C" __global__ void __launch_bounds__ (256, 1) m00110_m08 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, 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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) { } extern "C" __global__ void __launch_bounds__ (256, 1) m00110_m16 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, 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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) { } extern "C" __global__ void __launch_bounds__ (256, 1) m00110_s04 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, 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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) { /** * modifier */ const u32 lid = threadIdx.x; /** * base */ const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x; if (gid >= gid_max) return; u32x wordl0[4]; wordl0[0] = pws[gid].i[ 0]; wordl0[1] = pws[gid].i[ 1]; wordl0[2] = pws[gid].i[ 2]; wordl0[3] = pws[gid].i[ 3]; u32x wordl1[4]; wordl1[0] = pws[gid].i[ 4]; wordl1[1] = pws[gid].i[ 5]; wordl1[2] = pws[gid].i[ 6]; wordl1[3] = pws[gid].i[ 7]; u32x wordl2[4]; wordl2[0] = 0; wordl2[1] = 0; wordl2[2] = 0; wordl2[3] = 0; u32x wordl3[4]; wordl3[0] = 0; wordl3[1] = 0; wordl3[2] = 0; wordl3[3] = 0; const u32 pw_l_len = pws[gid].pw_len; if (combs_mode == COMBINATOR_MODE_BASE_RIGHT) { switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, c_combs[0].pw_len); } /** * 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]; const u32 salt_len = salt_bufs[salt_pos].salt_len; /** * digest */ const u32 search[4] = { digests_buf[digests_offset].digest_buf[DGST_R0], digests_buf[digests_offset].digest_buf[DGST_R1], digests_buf[digests_offset].digest_buf[DGST_R2], digests_buf[digests_offset].digest_buf[DGST_R3] }; /** * reverse */ const u32 e_rev = rotl32 (search[1], 2u); /** * loop */ for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++) { const u32 pw_r_len = c_combs[il_pos].pw_len; const u32 pw_len = pw_l_len + pw_r_len; u32 wordr0[4]; wordr0[0] = c_combs[il_pos].i[0]; wordr0[1] = c_combs[il_pos].i[1]; wordr0[2] = c_combs[il_pos].i[2]; wordr0[3] = c_combs[il_pos].i[3]; u32 wordr1[4]; wordr1[0] = c_combs[il_pos].i[4]; wordr1[1] = c_combs[il_pos].i[5]; wordr1[2] = c_combs[il_pos].i[6]; wordr1[3] = c_combs[il_pos].i[7]; u32 wordr2[4]; wordr2[0] = 0; wordr2[1] = 0; wordr2[2] = 0; wordr2[3] = 0; u32 wordr3[4]; wordr3[0] = 0; wordr3[1] = 0; wordr3[2] = 0; wordr3[3] = 0; if (combs_mode == COMBINATOR_MODE_BASE_LEFT) { switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len); } /** * append salt */ u32 s0[4]; s0[0] = salt_buf0[0]; s0[1] = salt_buf0[1]; s0[2] = salt_buf0[2]; s0[3] = salt_buf0[3]; u32 s1[4]; s1[0] = salt_buf1[0]; s1[1] = salt_buf1[1]; s1[2] = salt_buf1[2]; s1[3] = salt_buf1[3]; u32 s2[4]; s2[0] = 0; s2[1] = 0; s2[2] = 0; s2[3] = 0; u32 s3[4]; s3[0] = 0; s3[1] = 0; s3[2] = 0; s3[3] = 0; switch_buffer_by_offset (s0, s1, s2, s3, pw_len); const u32 pw_salt_len = pw_len + salt_len; u32x w0[4]; w0[0] = wordl0[0] | wordr0[0] | s0[0]; w0[1] = wordl0[1] | wordr0[1] | s0[1]; w0[2] = wordl0[2] | wordr0[2] | s0[2]; w0[3] = wordl0[3] | wordr0[3] | s0[3]; u32x w1[4]; w1[0] = wordl1[0] | wordr1[0] | s1[0]; w1[1] = wordl1[1] | wordr1[1] | s1[1]; w1[2] = wordl1[2] | wordr1[2] | s1[2]; w1[3] = wordl1[3] | wordr1[3] | s1[3]; u32x w2[4]; w2[0] = wordl2[0] | wordr2[0] | s2[0]; w2[1] = wordl2[1] | wordr2[1] | s2[1]; w2[2] = wordl2[2] | wordr2[2] | s2[2]; w2[3] = wordl2[3] | wordr2[3] | s2[3]; u32x w3[4]; w3[0] = wordl3[0] | wordr3[0] | s3[0]; w3[1] = wordl3[1] | wordr3[1] | s3[1]; w3[2] = 0; w3[3] = 0; append_0x80_4 (w0, w1, w2, w3, pw_salt_len); /** * sha1 */ u32x w0_t = swap_workaround (w0[0]); u32x w1_t = swap_workaround (w0[1]); u32x w2_t = swap_workaround (w0[2]); u32x w3_t = swap_workaround (w0[3]); u32x w4_t = swap_workaround (w1[0]); u32x w5_t = swap_workaround (w1[1]); u32x w6_t = swap_workaround (w1[2]); u32x w7_t = swap_workaround (w1[3]); u32x w8_t = swap_workaround (w2[0]); u32x w9_t = swap_workaround (w2[1]); u32x wa_t = swap_workaround (w2[2]); u32x wb_t = swap_workaround (w2[3]); u32x wc_t = swap_workaround (w3[0]); u32x wd_t = swap_workaround (w3[1]); u32x we_t = 0; u32x wf_t = pw_salt_len * 8; u32x a = SHA1M_A; u32x b = SHA1M_B; u32x c = SHA1M_C; u32x d = SHA1M_D; u32x e = SHA1M_E; #undef K #define K SHA1C00 SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t); SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t); SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t); SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t); SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t); SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t); SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t); SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t); w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t); w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t); w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t); w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t); #undef K #define K SHA1C01 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t); w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t); w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t); w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t); w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t); w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t); wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t); wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t); wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t); wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t); we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t); wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t); w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t); w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t); w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t); w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t); w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t); w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t); w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t); w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t); #undef K #define K SHA1C02 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t); w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t); wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t); wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t); wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t); wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t); we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t); wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t); w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t); w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t); w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t); w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t); w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t); w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t); w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t); w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t); w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t); w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t); wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t); wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t); #undef K #define K SHA1C03 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t); wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t); we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t); wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t); w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t); w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t); w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t); w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t); w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t); w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t); w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t); w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t); w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t); w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t); wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t); wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t); if (e != e_rev) continue; wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t); wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t); we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); const u32x r0 = d; const u32x r1 = e; const u32x r2 = c; const u32x r3 = b; #include VECT_COMPARE_S } } extern "C" __global__ void __launch_bounds__ (256, 1) m00110_s08 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, 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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) { } extern "C" __global__ void __launch_bounds__ (256, 1) m00110_s16 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, 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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) { }