2 * Author......: Jens Steube <jens.steube@gmail.com>
9 //#define NEW_SIMD_CODE
11 #include "inc_vendor.cl"
12 #include "inc_hash_constants.h"
13 #include "inc_hash_functions.cl"
14 #include "inc_types.cl"
15 #include "inc_common.cl"
18 #include "inc_simd.cl"
20 __constant u32 k_sha256[64] =
22 SHA256C00, SHA256C01, SHA256C02, SHA256C03,
23 SHA256C04, SHA256C05, SHA256C06, SHA256C07,
24 SHA256C08, SHA256C09, SHA256C0a, SHA256C0b,
25 SHA256C0c, SHA256C0d, SHA256C0e, SHA256C0f,
26 SHA256C10, SHA256C11, SHA256C12, SHA256C13,
27 SHA256C14, SHA256C15, SHA256C16, SHA256C17,
28 SHA256C18, SHA256C19, SHA256C1a, SHA256C1b,
29 SHA256C1c, SHA256C1d, SHA256C1e, SHA256C1f,
30 SHA256C20, SHA256C21, SHA256C22, SHA256C23,
31 SHA256C24, SHA256C25, SHA256C26, SHA256C27,
32 SHA256C28, SHA256C29, SHA256C2a, SHA256C2b,
33 SHA256C2c, SHA256C2d, SHA256C2e, SHA256C2f,
34 SHA256C30, SHA256C31, SHA256C32, SHA256C33,
35 SHA256C34, SHA256C35, SHA256C36, SHA256C37,
36 SHA256C38, SHA256C39, SHA256C3a, SHA256C3b,
37 SHA256C3c, SHA256C3d, SHA256C3e, SHA256C3f,
40 void sha256_transform (const u32x w[16], u32x digest[8])
68 #define ROUND_EXPAND() \
70 w0_t = SHA256_EXPAND (we_t, w9_t, w1_t, w0_t); \
71 w1_t = SHA256_EXPAND (wf_t, wa_t, w2_t, w1_t); \
72 w2_t = SHA256_EXPAND (w0_t, wb_t, w3_t, w2_t); \
73 w3_t = SHA256_EXPAND (w1_t, wc_t, w4_t, w3_t); \
74 w4_t = SHA256_EXPAND (w2_t, wd_t, w5_t, w4_t); \
75 w5_t = SHA256_EXPAND (w3_t, we_t, w6_t, w5_t); \
76 w6_t = SHA256_EXPAND (w4_t, wf_t, w7_t, w6_t); \
77 w7_t = SHA256_EXPAND (w5_t, w0_t, w8_t, w7_t); \
78 w8_t = SHA256_EXPAND (w6_t, w1_t, w9_t, w8_t); \
79 w9_t = SHA256_EXPAND (w7_t, w2_t, wa_t, w9_t); \
80 wa_t = SHA256_EXPAND (w8_t, w3_t, wb_t, wa_t); \
81 wb_t = SHA256_EXPAND (w9_t, w4_t, wc_t, wb_t); \
82 wc_t = SHA256_EXPAND (wa_t, w5_t, wd_t, wc_t); \
83 wd_t = SHA256_EXPAND (wb_t, w6_t, we_t, wd_t); \
84 we_t = SHA256_EXPAND (wc_t, w7_t, wf_t, we_t); \
85 wf_t = SHA256_EXPAND (wd_t, w8_t, w0_t, wf_t); \
88 #define ROUND_STEP(i) \
90 SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha256[i + 0]); \
91 SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha256[i + 1]); \
92 SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha256[i + 2]); \
93 SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha256[i + 3]); \
94 SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha256[i + 4]); \
95 SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha256[i + 5]); \
96 SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha256[i + 6]); \
97 SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha256[i + 7]); \
98 SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha256[i + 8]); \
99 SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha256[i + 9]); \
100 SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha256[i + 10]); \
101 SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha256[i + 11]); \
102 SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha256[i + 12]); \
103 SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha256[i + 13]); \
104 SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, we_t, k_sha256[i + 14]); \
105 SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha256[i + 15]); \
113 for (int i = 16; i < 64; i += 16)
115 ROUND_EXPAND (); ROUND_STEP (i);
128 void memcat64c_be (u32x block[16], const u32 offset, u32x carry[16])
130 const u32 mod = offset & 3;
131 const u32 div = offset / 4;
152 const int selector = (0x76543210 >> ((offset & 3) * 4)) & 0xffff;
154 tmp00 = __byte_perm (carry[ 0], 0, selector);
155 tmp01 = __byte_perm (carry[ 1], carry[ 0], selector);
156 tmp02 = __byte_perm (carry[ 2], carry[ 1], selector);
157 tmp03 = __byte_perm (carry[ 3], carry[ 2], selector);
158 tmp04 = __byte_perm (carry[ 4], carry[ 3], selector);
159 tmp05 = __byte_perm (carry[ 5], carry[ 4], selector);
160 tmp06 = __byte_perm (carry[ 6], carry[ 5], selector);
161 tmp07 = __byte_perm (carry[ 7], carry[ 6], selector);
162 tmp08 = __byte_perm (carry[ 8], carry[ 7], selector);
163 tmp09 = __byte_perm (carry[ 9], carry[ 8], selector);
164 tmp10 = __byte_perm (carry[10], carry[ 9], selector);
165 tmp11 = __byte_perm (carry[11], carry[10], selector);
166 tmp12 = __byte_perm (carry[12], carry[11], selector);
167 tmp13 = __byte_perm (carry[13], carry[12], selector);
168 tmp14 = __byte_perm (carry[14], carry[13], selector);
169 tmp15 = __byte_perm (carry[15], carry[14], selector);
170 tmp16 = __byte_perm ( 0, carry[15], selector);
173 #if defined IS_AMD || defined IS_GENERIC
174 tmp00 = amd_bytealign ( 0, carry[ 0], offset);
175 tmp01 = amd_bytealign (carry[ 0], carry[ 1], offset);
176 tmp02 = amd_bytealign (carry[ 1], carry[ 2], offset);
177 tmp03 = amd_bytealign (carry[ 2], carry[ 3], offset);
178 tmp04 = amd_bytealign (carry[ 3], carry[ 4], offset);
179 tmp05 = amd_bytealign (carry[ 4], carry[ 5], offset);
180 tmp06 = amd_bytealign (carry[ 5], carry[ 6], offset);
181 tmp07 = amd_bytealign (carry[ 6], carry[ 7], offset);
182 tmp08 = amd_bytealign (carry[ 7], carry[ 8], offset);
183 tmp09 = amd_bytealign (carry[ 8], carry[ 9], offset);
184 tmp10 = amd_bytealign (carry[ 9], carry[10], offset);
185 tmp11 = amd_bytealign (carry[10], carry[11], offset);
186 tmp12 = amd_bytealign (carry[11], carry[12], offset);
187 tmp13 = amd_bytealign (carry[12], carry[13], offset);
188 tmp14 = amd_bytealign (carry[13], carry[14], offset);
189 tmp15 = amd_bytealign (carry[14], carry[15], offset);
190 tmp16 = amd_bytealign (carry[15], 0, offset);
212 case 0: block[ 0] |= tmp00;
230 case 1: block[ 1] |= tmp00;
248 case 2: block[ 2] |= tmp00;
266 case 3: block[ 3] |= tmp00;
284 case 4: block[ 4] |= tmp00;
302 case 5: block[ 5] |= tmp00;
320 case 6: block[ 6] |= tmp00;
338 case 7: block[ 7] |= tmp00;
356 case 8: block[ 8] |= tmp00;
374 case 9: block[ 9] |= tmp00;
392 case 10: block[10] |= tmp00;
410 case 11: block[11] |= tmp00;
428 case 12: block[12] |= tmp00;
446 case 13: block[13] |= tmp00;
464 case 14: block[14] |= tmp00;
482 case 15: block[15] |= tmp00;
503 __kernel void m13800_m04 (__global pw_t *pws, __global kernel_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 win8phone_t *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)
509 const u32 gid = get_global_id (0);
510 const u32 lid = get_local_id (0);
511 const u32 lsz = get_local_size (0);
520 pw_buf0[0] = pws[gid].i[0];
521 pw_buf0[1] = pws[gid].i[1];
522 pw_buf0[2] = pws[gid].i[2];
523 pw_buf0[3] = pws[gid].i[3];
524 pw_buf1[0] = pws[gid].i[4];
525 pw_buf1[1] = pws[gid].i[5];
526 pw_buf1[2] = pws[gid].i[6];
527 pw_buf1[3] = pws[gid].i[7];
529 const u32 pw_len = pws[gid].pw_len;
535 __local u32 s_esalt[32];
537 for (u32 i = lid; i < 32; i += lsz)
539 s_esalt[i] = esalt_bufs[salt_pos].salt_buf[i];
542 barrier (CLK_LOCAL_MEM_FENCE);
544 if (gid >= gid_max) return;
550 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
557 const u32x out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
559 const u32x out_len2 = out_len * 2;
561 make_unicode (w1, w2, w3);
562 make_unicode (w0, w0, w1);
566 w[ 0] = swap32 (w0[0]);
567 w[ 1] = swap32 (w0[1]);
568 w[ 2] = swap32 (w0[2]);
569 w[ 3] = swap32 (w0[3]);
570 w[ 4] = swap32 (w1[0]);
571 w[ 5] = swap32 (w1[1]);
572 w[ 6] = swap32 (w1[2]);
573 w[ 7] = swap32 (w1[3]);
574 w[ 8] = swap32 (w2[0]);
575 w[ 9] = swap32 (w2[1]);
576 w[10] = swap32 (w2[2]);
577 w[11] = swap32 (w2[3]);
578 w[12] = swap32 (w3[0]);
579 w[13] = swap32 (w3[1]);
580 w[14] = swap32 (w3[2]);
581 w[15] = swap32 (w3[3]);
585 carry[ 0] = s_esalt[ 0];
586 carry[ 1] = s_esalt[ 1];
587 carry[ 2] = s_esalt[ 2];
588 carry[ 3] = s_esalt[ 3];
589 carry[ 4] = s_esalt[ 4];
590 carry[ 5] = s_esalt[ 5];
591 carry[ 6] = s_esalt[ 6];
592 carry[ 7] = s_esalt[ 7];
593 carry[ 8] = s_esalt[ 8];
594 carry[ 9] = s_esalt[ 9];
595 carry[10] = s_esalt[10];
596 carry[11] = s_esalt[11];
597 carry[12] = s_esalt[12];
598 carry[13] = s_esalt[13];
599 carry[14] = s_esalt[14];
600 carry[15] = s_esalt[15];
602 // we can always use pw_len here, since we add exactly the hash buffer size
603 memcat64c_be (w, out_len2, carry);
607 digest[0] = SHA256M_A;
608 digest[1] = SHA256M_B;
609 digest[2] = SHA256M_C;
610 digest[3] = SHA256M_D;
611 digest[4] = SHA256M_E;
612 digest[5] = SHA256M_F;
613 digest[6] = SHA256M_G;
614 digest[7] = SHA256M_H;
616 sha256_transform (w, digest);
635 carry[ 0] = s_esalt[16];
636 carry[ 1] = s_esalt[17];
637 carry[ 2] = s_esalt[18];
638 carry[ 3] = s_esalt[19];
639 carry[ 4] = s_esalt[20];
640 carry[ 5] = s_esalt[21];
641 carry[ 6] = s_esalt[22];
642 carry[ 7] = s_esalt[23];
643 carry[ 8] = s_esalt[24];
644 carry[ 9] = s_esalt[25];
645 carry[10] = s_esalt[26];
646 carry[11] = s_esalt[27];
647 carry[12] = s_esalt[28];
648 carry[13] = s_esalt[29];
649 carry[14] = s_esalt[30];
650 carry[15] = s_esalt[31];
652 // we can always use pw_len here, since we add exactly the hash buffer size
653 memcat64c_be (w, out_len2, carry);
655 sha256_transform (w, digest);
674 // we can always use pw_len here, since we add exactly the hash buffer size
675 append_0x80_4x4 (w + 0, w + 4, w + 8, w + 12, out_len2 ^ 3);
678 w[15] = (out_len2 + 128) * 8;
680 sha256_transform (w, digest);
682 const u32x d = digest[DGST_R0];
683 const u32x h = digest[DGST_R1];
684 const u32x c = digest[DGST_R2];
685 const u32x g = digest[DGST_R3];
687 COMPARE_M_SIMD (d, h, c, g);
691 __kernel void m13800_m08 (__global pw_t *pws, __global kernel_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 win8phone_t *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)
695 __kernel void m13800_m16 (__global pw_t *pws, __global kernel_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 win8phone_t *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)
699 __kernel void m13800_s04 (__global pw_t *pws, __global kernel_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 win8phone_t *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)
705 const u32 gid = get_global_id (0);
706 const u32 lid = get_local_id (0);
707 const u32 lsz = get_local_size (0);
716 pw_buf0[0] = pws[gid].i[0];
717 pw_buf0[1] = pws[gid].i[1];
718 pw_buf0[2] = pws[gid].i[2];
719 pw_buf0[3] = pws[gid].i[3];
720 pw_buf1[0] = pws[gid].i[4];
721 pw_buf1[1] = pws[gid].i[5];
722 pw_buf1[2] = pws[gid].i[6];
723 pw_buf1[3] = pws[gid].i[7];
725 const u32 pw_len = pws[gid].pw_len;
731 __local u32 s_esalt[32];
733 for (u32 i = lid; i < 32; i += lsz)
735 s_esalt[i] = esalt_bufs[salt_pos].salt_buf[i];
738 barrier (CLK_LOCAL_MEM_FENCE);
740 if (gid >= gid_max) return;
746 const u32 search[4] =
748 digests_buf[digests_offset].digest_buf[DGST_R0],
749 digests_buf[digests_offset].digest_buf[DGST_R1],
750 digests_buf[digests_offset].digest_buf[DGST_R2],
751 digests_buf[digests_offset].digest_buf[DGST_R3]
758 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
765 const u32x out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
767 const u32x out_len2 = out_len * 2;
769 make_unicode (w1, w2, w3);
770 make_unicode (w0, w0, w1);
774 w[ 0] = swap32 (w0[0]);
775 w[ 1] = swap32 (w0[1]);
776 w[ 2] = swap32 (w0[2]);
777 w[ 3] = swap32 (w0[3]);
778 w[ 4] = swap32 (w1[0]);
779 w[ 5] = swap32 (w1[1]);
780 w[ 6] = swap32 (w1[2]);
781 w[ 7] = swap32 (w1[3]);
782 w[ 8] = swap32 (w2[0]);
783 w[ 9] = swap32 (w2[1]);
784 w[10] = swap32 (w2[2]);
785 w[11] = swap32 (w2[3]);
786 w[12] = swap32 (w3[0]);
787 w[13] = swap32 (w3[1]);
788 w[14] = swap32 (w3[2]);
789 w[15] = swap32 (w3[3]);
793 carry[ 0] = s_esalt[ 0];
794 carry[ 1] = s_esalt[ 1];
795 carry[ 2] = s_esalt[ 2];
796 carry[ 3] = s_esalt[ 3];
797 carry[ 4] = s_esalt[ 4];
798 carry[ 5] = s_esalt[ 5];
799 carry[ 6] = s_esalt[ 6];
800 carry[ 7] = s_esalt[ 7];
801 carry[ 8] = s_esalt[ 8];
802 carry[ 9] = s_esalt[ 9];
803 carry[10] = s_esalt[10];
804 carry[11] = s_esalt[11];
805 carry[12] = s_esalt[12];
806 carry[13] = s_esalt[13];
807 carry[14] = s_esalt[14];
808 carry[15] = s_esalt[15];
810 // we can always use pw_len here, since we add exactly the hash buffer size
811 memcat64c_be (w, out_len2, carry);
815 digest[0] = SHA256M_A;
816 digest[1] = SHA256M_B;
817 digest[2] = SHA256M_C;
818 digest[3] = SHA256M_D;
819 digest[4] = SHA256M_E;
820 digest[5] = SHA256M_F;
821 digest[6] = SHA256M_G;
822 digest[7] = SHA256M_H;
824 sha256_transform (w, digest);
843 carry[ 0] = s_esalt[16];
844 carry[ 1] = s_esalt[17];
845 carry[ 2] = s_esalt[18];
846 carry[ 3] = s_esalt[19];
847 carry[ 4] = s_esalt[20];
848 carry[ 5] = s_esalt[21];
849 carry[ 6] = s_esalt[22];
850 carry[ 7] = s_esalt[23];
851 carry[ 8] = s_esalt[24];
852 carry[ 9] = s_esalt[25];
853 carry[10] = s_esalt[26];
854 carry[11] = s_esalt[27];
855 carry[12] = s_esalt[28];
856 carry[13] = s_esalt[29];
857 carry[14] = s_esalt[30];
858 carry[15] = s_esalt[31];
860 // we can always use pw_len here, since we add exactly the hash buffer size
861 memcat64c_be (w, out_len2, carry);
863 sha256_transform (w, digest);
882 // we can always use pw_len here, since we add exactly the hash buffer size
883 append_0x80_4x4 (w + 0, w + 4, w + 8, w + 12, out_len2 ^ 3);
886 w[15] = (out_len2 + 128) * 8;
888 sha256_transform (w, digest);
890 const u32x d = digest[DGST_R0];
891 const u32x h = digest[DGST_R1];
892 const u32x c = digest[DGST_R2];
893 const u32x g = digest[DGST_R3];
895 COMPARE_S_SIMD (d, h, c, g);
899 __kernel void m13800_s08 (__global pw_t *pws, __global kernel_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 win8phone_t *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)
903 __kernel void m13800_s16 (__global pw_t *pws, __global kernel_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 win8phone_t *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)