/** * Author......: Jens Steube * License.....: MIT */ #define _SHA1_ #include "include/constants.h" #include "include/kernel_vendor.h" #ifdef VLIW1 #define VECT_SIZE1 #endif #ifdef VLIW4 #define VECT_SIZE4 #endif #ifdef VLIW5 #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_amd.c" #include "common_amd.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 static void overwrite_at (u32x sw[16], const u32x w0, const u32 salt_len) { switch (salt_len) { case 0: sw[0] = w0; break; case 1: sw[0] = (sw[0] & 0xff000000) | (w0 >> 8); sw[1] = (sw[1] & 0x00ffffff) | (w0 << 24); break; case 2: sw[0] = (sw[0] & 0xffff0000) | (w0 >> 16); sw[1] = (sw[1] & 0x0000ffff) | (w0 << 16); break; case 3: sw[0] = (sw[0] & 0xffffff00) | (w0 >> 24); sw[1] = (sw[1] & 0x000000ff) | (w0 << 8); break; case 4: sw[1] = w0; break; case 5: sw[1] = (sw[1] & 0xff000000) | (w0 >> 8); sw[2] = (sw[2] & 0x00ffffff) | (w0 << 24); break; case 6: sw[1] = (sw[1] & 0xffff0000) | (w0 >> 16); sw[2] = (sw[2] & 0x0000ffff) | (w0 << 16); break; case 7: sw[1] = (sw[1] & 0xffffff00) | (w0 >> 24); sw[2] = (sw[2] & 0x000000ff) | (w0 << 8); break; case 8: sw[2] = w0; break; case 9: sw[2] = (sw[2] & 0xff000000) | (w0 >> 8); sw[3] = (sw[3] & 0x00ffffff) | (w0 << 24); break; case 10: sw[2] = (sw[2] & 0xffff0000) | (w0 >> 16); sw[3] = (sw[3] & 0x0000ffff) | (w0 << 16); break; case 11: sw[2] = (sw[2] & 0xffffff00) | (w0 >> 24); sw[3] = (sw[3] & 0x000000ff) | (w0 << 8); break; case 12: sw[3] = w0; break; case 13: sw[3] = (sw[3] & 0xff000000) | (w0 >> 8); sw[4] = (sw[4] & 0x00ffffff) | (w0 << 24); break; case 14: sw[3] = (sw[3] & 0xffff0000) | (w0 >> 16); sw[4] = (sw[4] & 0x0000ffff) | (w0 << 16); break; case 15: sw[3] = (sw[3] & 0xffffff00) | (w0 >> 24); sw[4] = (sw[4] & 0x000000ff) | (w0 << 8); break; case 16: sw[4] = w0; break; case 17: sw[4] = (sw[4] & 0xff000000) | (w0 >> 8); sw[5] = (sw[5] & 0x00ffffff) | (w0 << 24); break; case 18: sw[4] = (sw[4] & 0xffff0000) | (w0 >> 16); sw[5] = (sw[5] & 0x0000ffff) | (w0 << 16); break; case 19: sw[4] = (sw[4] & 0xffffff00) | (w0 >> 24); sw[5] = (sw[5] & 0x000000ff) | (w0 << 8); break; case 20: sw[5] = w0; break; case 21: sw[5] = (sw[5] & 0xff000000) | (w0 >> 8); sw[6] = (sw[6] & 0x00ffffff) | (w0 << 24); break; case 22: sw[5] = (sw[5] & 0xffff0000) | (w0 >> 16); sw[6] = (sw[6] & 0x0000ffff) | (w0 << 16); break; case 23: sw[5] = (sw[5] & 0xffffff00) | (w0 >> 24); sw[6] = (sw[6] & 0x000000ff) | (w0 << 8); break; case 24: sw[6] = w0; break; case 25: sw[6] = (sw[6] & 0xff000000) | (w0 >> 8); sw[7] = (sw[7] & 0x00ffffff) | (w0 << 24); break; case 26: sw[6] = (sw[6] & 0xffff0000) | (w0 >> 16); sw[7] = (sw[7] & 0x0000ffff) | (w0 << 16); break; case 27: sw[6] = (sw[6] & 0xffffff00) | (w0 >> 24); sw[7] = (sw[7] & 0x000000ff) | (w0 << 8); break; case 28: sw[7] = w0; break; case 29: sw[7] = (sw[7] & 0xff000000) | (w0 >> 8); sw[8] = (sw[8] & 0x00ffffff) | (w0 << 24); break; case 30: sw[7] = (sw[7] & 0xffff0000) | (w0 >> 16); sw[8] = (sw[8] & 0x0000ffff) | (w0 << 16); break; case 31: sw[7] = (sw[7] & 0xffffff00) | (w0 >> 24); sw[8] = (sw[8] & 0x000000ff) | (w0 << 8); break; } } static void m00120m (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], const u32 pw_len, __global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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) { /** * modifier */ const u32 gid = get_global_id (0); const u32 lid = get_local_id (0); /** * 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] = 0; salt_buf2[1] = 0; salt_buf2[2] = 0; salt_buf2[3] = 0; u32 salt_buf3[4]; salt_buf3[0] = 0; salt_buf3[1] = 0; salt_buf3[2] = 0; salt_buf3[3] = 0; const u32 salt_len = salt_bufs[salt_pos].salt_len; const u32 pw_salt_len = pw_len + salt_len; /** * prepend salt */ u32x w0_t[4]; u32x w1_t[4]; u32x w2_t[4]; u32x w3_t[4]; w0_t[0] = swap_workaround (w0[0]); w0_t[1] = swap_workaround (w0[1]); w0_t[2] = swap_workaround (w0[2]); w0_t[3] = swap_workaround (w0[3]); w1_t[0] = swap_workaround (w1[0]); w1_t[1] = swap_workaround (w1[1]); w1_t[2] = swap_workaround (w1[2]); w1_t[3] = swap_workaround (w1[3]); w2_t[0] = swap_workaround (w2[0]); w2_t[1] = swap_workaround (w2[1]); w2_t[2] = swap_workaround (w2[2]); w2_t[3] = swap_workaround (w2[3]); w3_t[0] = swap_workaround (w3[0]); w3_t[1] = swap_workaround (w3[1]); w3_t[2] = swap_workaround (w3[2]); w3_t[3] = swap_workaround (w3[3]); switch_buffer_by_offset (w0_t, w1_t, w2_t, w3_t, salt_len); w0_t[0] |= salt_buf0[0]; w0_t[1] |= salt_buf0[1]; w0_t[2] |= salt_buf0[2]; w0_t[3] |= salt_buf0[3]; w1_t[0] |= salt_buf1[0]; w1_t[1] |= salt_buf1[1]; w1_t[2] |= salt_buf1[2]; w1_t[3] |= salt_buf1[3]; w2_t[0] |= salt_buf2[0]; w2_t[1] |= salt_buf2[1]; w2_t[2] |= salt_buf2[2]; w2_t[3] |= salt_buf2[3]; w3_t[0] |= salt_buf3[0]; w3_t[1] |= salt_buf3[1]; w3_t[2] |= salt_buf3[2]; w3_t[3] |= salt_buf3[3]; w0_t[0] = swap_workaround (w0_t[0]); w0_t[1] = swap_workaround (w0_t[1]); w0_t[2] = swap_workaround (w0_t[2]); w0_t[3] = swap_workaround (w0_t[3]); w1_t[0] = swap_workaround (w1_t[0]); w1_t[1] = swap_workaround (w1_t[1]); w1_t[2] = swap_workaround (w1_t[2]); w1_t[3] = swap_workaround (w1_t[3]); w2_t[0] = swap_workaround (w2_t[0]); w2_t[1] = swap_workaround (w2_t[1]); w2_t[2] = swap_workaround (w2_t[2]); w2_t[3] = swap_workaround (w2_t[3]); w3_t[0] = swap_workaround (w3_t[0]); w3_t[1] = swap_workaround (w3_t[1]); w3_t[2] = swap_workaround (w3_t[2]); w3_t[3] = swap_workaround (w3_t[3]); /** * loop */ u32x w0l = w0[0]; for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos++) { const u32 w0r = bfs_buf[il_pos].i; const u32x w0n = w0l | w0r; u32x wx[16]; wx[ 0] = w0_t[0]; wx[ 1] = w0_t[1]; wx[ 2] = w0_t[2]; wx[ 3] = w0_t[3]; wx[ 4] = w1_t[0]; wx[ 5] = w1_t[1]; wx[ 6] = w1_t[2]; wx[ 7] = w1_t[3]; wx[ 8] = w2_t[0]; wx[ 9] = w2_t[1]; wx[10] = w2_t[2]; wx[11] = w2_t[3]; wx[12] = w3_t[0]; wx[13] = w3_t[1]; wx[14] = w3_t[2]; wx[15] = w3_t[3]; overwrite_at (wx, w0n, salt_len); u32x w0_t[4]; u32x w1_t[4]; u32x w2_t[4]; u32x w3_t[4]; w0_t[0] = wx[ 0]; w0_t[1] = wx[ 1]; w0_t[2] = wx[ 2]; w0_t[3] = wx[ 3]; w1_t[0] = wx[ 4]; w1_t[1] = wx[ 5]; w1_t[2] = wx[ 6]; w1_t[3] = wx[ 7]; w2_t[0] = wx[ 8]; w2_t[1] = wx[ 9]; w2_t[2] = wx[10]; w2_t[3] = wx[11]; w3_t[0] = wx[12]; w3_t[1] = wx[13]; w3_t[2] = 0; w3_t[3] = pw_salt_len * 8; /** * sha1 */ 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[0]); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[1]); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[2]); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[3]); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w1_t[0]); SHA1_STEP (SHA1_F0o, a, b, c, d, e, w1_t[1]); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t[2]); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t[3]); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t[0]); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w2_t[1]); SHA1_STEP (SHA1_F0o, a, b, c, d, e, w2_t[2]); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w2_t[3]); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w3_t[0]); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t[1]); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t[2]); SHA1_STEP (SHA1_F0o, a, b, c, d, e, w3_t[3]); w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[0]); w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[1]); w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[2]); w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w0_t[3]); #undef K #define K SHA1C01 w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[0]); w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[1]); w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[2]); w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[3]); w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[0]); w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[1]); w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t[2]); w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[3]); w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[0]); w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[1]); w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[2]); w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[3]); w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[0]); w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w0_t[1]); w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[2]); w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[3]); w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[0]); w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[1]); w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[2]); w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[3]); #undef K #define K SHA1C02 w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t[0]); w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[1]); w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[2]); w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[3]); w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w3_t[0]); w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w3_t[1]); w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t[2]); w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w3_t[3]); w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t[0]); w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w0_t[1]); w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w0_t[2]); w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w0_t[3]); w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w1_t[0]); w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w1_t[1]); w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t[2]); w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w1_t[3]); w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[0]); w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[1]); w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[2]); w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w2_t[3]); #undef K #define K SHA1C03 w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[0]); w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[1]); w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[2]); w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[3]); w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[0]); w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[1]); w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w0_t[2]); w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[3]); w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[0]); w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[1]); w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[2]); w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[3]); w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[0]); w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w2_t[1]); w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[2]); w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[3]); w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[0]); w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[1]); w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[2]); w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[3]); const u32x r0 = d; const u32x r1 = e; const u32x r2 = c; const u32x r3 = b; #include VECT_COMPARE_M } } static void m00120s (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], const u32 pw_len, __global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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) { /** * modifier */ const u32 gid = get_global_id (0); const u32 lid = get_local_id (0); /** * 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); /** * 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] = 0; salt_buf2[1] = 0; salt_buf2[2] = 0; salt_buf2[3] = 0; u32 salt_buf3[4]; salt_buf3[0] = 0; salt_buf3[1] = 0; salt_buf3[2] = 0; salt_buf3[3] = 0; const u32 salt_len = salt_bufs[salt_pos].salt_len; const u32 pw_salt_len = pw_len + salt_len; /** * prepend salt */ u32x w0_t[4]; u32x w1_t[4]; u32x w2_t[4]; u32x w3_t[4]; w0_t[0] = swap_workaround (w0[0]); w0_t[1] = swap_workaround (w0[1]); w0_t[2] = swap_workaround (w0[2]); w0_t[3] = swap_workaround (w0[3]); w1_t[0] = swap_workaround (w1[0]); w1_t[1] = swap_workaround (w1[1]); w1_t[2] = swap_workaround (w1[2]); w1_t[3] = swap_workaround (w1[3]); w2_t[0] = swap_workaround (w2[0]); w2_t[1] = swap_workaround (w2[1]); w2_t[2] = swap_workaround (w2[2]); w2_t[3] = swap_workaround (w2[3]); w3_t[0] = swap_workaround (w3[0]); w3_t[1] = swap_workaround (w3[1]); w3_t[2] = swap_workaround (w3[2]); w3_t[3] = swap_workaround (w3[3]); switch_buffer_by_offset (w0_t, w1_t, w2_t, w3_t, salt_len); w0_t[0] |= salt_buf0[0]; w0_t[1] |= salt_buf0[1]; w0_t[2] |= salt_buf0[2]; w0_t[3] |= salt_buf0[3]; w1_t[0] |= salt_buf1[0]; w1_t[1] |= salt_buf1[1]; w1_t[2] |= salt_buf1[2]; w1_t[3] |= salt_buf1[3]; w2_t[0] |= salt_buf2[0]; w2_t[1] |= salt_buf2[1]; w2_t[2] |= salt_buf2[2]; w2_t[3] |= salt_buf2[3]; w3_t[0] |= salt_buf3[0]; w3_t[1] |= salt_buf3[1]; w3_t[2] |= salt_buf3[2]; w3_t[3] |= salt_buf3[3]; w0_t[0] = swap_workaround (w0_t[0]); w0_t[1] = swap_workaround (w0_t[1]); w0_t[2] = swap_workaround (w0_t[2]); w0_t[3] = swap_workaround (w0_t[3]); w1_t[0] = swap_workaround (w1_t[0]); w1_t[1] = swap_workaround (w1_t[1]); w1_t[2] = swap_workaround (w1_t[2]); w1_t[3] = swap_workaround (w1_t[3]); w2_t[0] = swap_workaround (w2_t[0]); w2_t[1] = swap_workaround (w2_t[1]); w2_t[2] = swap_workaround (w2_t[2]); w2_t[3] = swap_workaround (w2_t[3]); w3_t[0] = swap_workaround (w3_t[0]); w3_t[1] = swap_workaround (w3_t[1]); w3_t[2] = swap_workaround (w3_t[2]); w3_t[3] = swap_workaround (w3_t[3]); /** * loop */ u32x w0l = w0[0]; for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos++) { const u32 w0r = bfs_buf[il_pos].i; const u32x w0n = w0l | w0r; u32x wx[16]; wx[ 0] = w0_t[0]; wx[ 1] = w0_t[1]; wx[ 2] = w0_t[2]; wx[ 3] = w0_t[3]; wx[ 4] = w1_t[0]; wx[ 5] = w1_t[1]; wx[ 6] = w1_t[2]; wx[ 7] = w1_t[3]; wx[ 8] = w2_t[0]; wx[ 9] = w2_t[1]; wx[10] = w2_t[2]; wx[11] = w2_t[3]; wx[12] = w3_t[0]; wx[13] = w3_t[1]; wx[14] = w3_t[2]; wx[15] = w3_t[3]; overwrite_at (wx, w0n, salt_len); u32x w0_t[4]; u32x w1_t[4]; u32x w2_t[4]; u32x w3_t[4]; w0_t[0] = wx[ 0]; w0_t[1] = wx[ 1]; w0_t[2] = wx[ 2]; w0_t[3] = wx[ 3]; w1_t[0] = wx[ 4]; w1_t[1] = wx[ 5]; w1_t[2] = wx[ 6]; w1_t[3] = wx[ 7]; w2_t[0] = wx[ 8]; w2_t[1] = wx[ 9]; w2_t[2] = wx[10]; w2_t[3] = wx[11]; w3_t[0] = wx[12]; w3_t[1] = wx[13]; w3_t[2] = 0; w3_t[3] = pw_salt_len * 8; /** * sha1 */ 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[0]); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[1]); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[2]); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[3]); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w1_t[0]); SHA1_STEP (SHA1_F0o, a, b, c, d, e, w1_t[1]); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t[2]); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t[3]); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t[0]); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w2_t[1]); SHA1_STEP (SHA1_F0o, a, b, c, d, e, w2_t[2]); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w2_t[3]); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w3_t[0]); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t[1]); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t[2]); SHA1_STEP (SHA1_F0o, a, b, c, d, e, w3_t[3]); w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[0]); w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[1]); w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[2]); w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w0_t[3]); #undef K #define K SHA1C01 w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[0]); w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[1]); w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[2]); w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[3]); w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[0]); w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[1]); w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t[2]); w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[3]); w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[0]); w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[1]); w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[2]); w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[3]); w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[0]); w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w0_t[1]); w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[2]); w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[3]); w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[0]); w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[1]); w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[2]); w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[3]); #undef K #define K SHA1C02 w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t[0]); w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[1]); w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[2]); w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[3]); w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w3_t[0]); w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w3_t[1]); w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t[2]); w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w3_t[3]); w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t[0]); w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w0_t[1]); w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w0_t[2]); w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w0_t[3]); w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w1_t[0]); w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w1_t[1]); w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t[2]); w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w1_t[3]); w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[0]); w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[1]); w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[2]); w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w2_t[3]); #undef K #define K SHA1C03 w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[0]); w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[1]); w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[2]); w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[3]); w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[0]); w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[1]); w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w0_t[2]); w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[3]); w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[0]); w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[1]); w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[2]); w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[3]); w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[0]); w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w2_t[1]); w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[2]); w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[3]); w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[0]); if (allx (e != e_rev)) continue; w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[1]); w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[2]); w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[3]); const u32x r0 = d; const u32x r1 = e; const u32x r2 = c; const u32x r3 = b; #include VECT_COMPARE_S } } __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m00120_m04 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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) { /** * base */ const u32 gid = get_global_id (0); if (gid >= gid_max) return; u32x w0[4]; w0[0] = pws[gid].i[ 0]; w0[1] = pws[gid].i[ 1]; w0[2] = pws[gid].i[ 2]; w0[3] = pws[gid].i[ 3]; u32x w1[4]; w1[0] = 0; w1[1] = 0; w1[2] = 0; w1[3] = 0; u32x w2[4]; w2[0] = 0; w2[1] = 0; w2[2] = 0; w2[3] = 0; u32x w3[4]; w3[0] = 0; w3[1] = 0; w3[2] = 0; w3[3] = 0; const u32 pw_len = pws[gid].pw_len; /** * main */ m00120m (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, 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); } __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m00120_m08 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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) { /** * base */ const u32 gid = get_global_id (0); if (gid >= gid_max) return; u32x w0[4]; w0[0] = pws[gid].i[ 0]; w0[1] = pws[gid].i[ 1]; w0[2] = pws[gid].i[ 2]; w0[3] = pws[gid].i[ 3]; u32x w1[4]; w1[0] = pws[gid].i[ 4]; w1[1] = pws[gid].i[ 5]; w1[2] = pws[gid].i[ 6]; w1[3] = pws[gid].i[ 7]; u32x w2[4]; w2[0] = 0; w2[1] = 0; w2[2] = 0; w2[3] = 0; u32x w3[4]; w3[0] = 0; w3[1] = 0; w3[2] = 0; w3[3] = 0; const u32 pw_len = pws[gid].pw_len; /** * main */ m00120m (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, 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); } __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m00120_m16 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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) { /** * base */ const u32 gid = get_global_id (0); if (gid >= gid_max) return; u32x w0[4]; w0[0] = pws[gid].i[ 0]; w0[1] = pws[gid].i[ 1]; w0[2] = pws[gid].i[ 2]; w0[3] = pws[gid].i[ 3]; u32x w1[4]; w1[0] = pws[gid].i[ 4]; w1[1] = pws[gid].i[ 5]; w1[2] = pws[gid].i[ 6]; w1[3] = pws[gid].i[ 7]; u32x w2[4]; w2[0] = pws[gid].i[ 8]; w2[1] = pws[gid].i[ 9]; w2[2] = pws[gid].i[10]; w2[3] = pws[gid].i[11]; u32x w3[4]; w3[0] = pws[gid].i[12]; w3[1] = pws[gid].i[13]; w3[2] = 0; w3[3] = 0; const u32 pw_len = pws[gid].pw_len; /** * main */ m00120m (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, 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); } __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m00120_s04 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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) { /** * base */ const u32 gid = get_global_id (0); if (gid >= gid_max) return; u32x w0[4]; w0[0] = pws[gid].i[ 0]; w0[1] = pws[gid].i[ 1]; w0[2] = pws[gid].i[ 2]; w0[3] = pws[gid].i[ 3]; u32x w1[4]; w1[0] = 0; w1[1] = 0; w1[2] = 0; w1[3] = 0; u32x w2[4]; w2[0] = 0; w2[1] = 0; w2[2] = 0; w2[3] = 0; u32x w3[4]; w3[0] = 0; w3[1] = 0; w3[2] = 0; w3[3] = 0; const u32 pw_len = pws[gid].pw_len; /** * main */ m00120s (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, 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); } __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m00120_s08 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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) { /** * base */ const u32 gid = get_global_id (0); if (gid >= gid_max) return; u32x w0[4]; w0[0] = pws[gid].i[ 0]; w0[1] = pws[gid].i[ 1]; w0[2] = pws[gid].i[ 2]; w0[3] = pws[gid].i[ 3]; u32x w1[4]; w1[0] = pws[gid].i[ 4]; w1[1] = pws[gid].i[ 5]; w1[2] = pws[gid].i[ 6]; w1[3] = pws[gid].i[ 7]; u32x w2[4]; w2[0] = 0; w2[1] = 0; w2[2] = 0; w2[3] = 0; u32x w3[4]; w3[0] = 0; w3[1] = 0; w3[2] = 0; w3[3] = 0; const u32 pw_len = pws[gid].pw_len; /** * main */ m00120s (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, 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); } __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m00120_s16 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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) { /** * base */ const u32 gid = get_global_id (0); if (gid >= gid_max) return; u32x w0[4]; w0[0] = pws[gid].i[ 0]; w0[1] = pws[gid].i[ 1]; w0[2] = pws[gid].i[ 2]; w0[3] = pws[gid].i[ 3]; u32x w1[4]; w1[0] = pws[gid].i[ 4]; w1[1] = pws[gid].i[ 5]; w1[2] = pws[gid].i[ 6]; w1[3] = pws[gid].i[ 7]; u32x w2[4]; w2[0] = pws[gid].i[ 8]; w2[1] = pws[gid].i[ 9]; w2[2] = pws[gid].i[10]; w2[3] = pws[gid].i[11]; u32x w3[4]; w3[0] = pws[gid].i[12]; w3[1] = pws[gid].i[13]; w3[2] = 0; w3[3] = 0; const u32 pw_len = pws[gid].pw_len; /** * main */ m00120s (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, 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); }