2 * Author......: Jens Steube <jens.steube@gmail.com>
8 #include "include/constants.h"
9 #include "include/kernel_vendor.h"
24 #include "include/kernel_functions.c"
26 #include "common_nv.c"
29 #define VECT_COMPARE_M "check_multi_vect1_comp4.c"
33 #define VECT_COMPARE_M "check_multi_vect2_comp4.c"
37 #define VECT_COMPARE_M "check_multi_vect4_comp4.c"
40 __device__ static void md5_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[4])
66 MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00);
67 MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01);
68 MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02);
69 MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03);
70 MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00);
71 MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01);
72 MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02);
73 MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03);
74 MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00);
75 MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01);
76 MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02);
77 MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03);
78 MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00);
79 MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01);
80 MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02);
81 MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03);
83 MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10);
84 MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11);
85 MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12);
86 MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13);
87 MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10);
88 MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11);
89 MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12);
90 MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13);
91 MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10);
92 MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11);
93 MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12);
94 MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13);
95 MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10);
96 MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11);
97 MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12);
98 MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13);
100 MD5_STEP (MD5_H1, a, b, c, d, w5_t, MD5C20, MD5S20);
101 MD5_STEP (MD5_H2, d, a, b, c, w8_t, MD5C21, MD5S21);
102 MD5_STEP (MD5_H1, c, d, a, b, wb_t, MD5C22, MD5S22);
103 MD5_STEP (MD5_H2, b, c, d, a, we_t, MD5C23, MD5S23);
104 MD5_STEP (MD5_H1, a, b, c, d, w1_t, MD5C24, MD5S20);
105 MD5_STEP (MD5_H2, d, a, b, c, w4_t, MD5C25, MD5S21);
106 MD5_STEP (MD5_H1, c, d, a, b, w7_t, MD5C26, MD5S22);
107 MD5_STEP (MD5_H2, b, c, d, a, wa_t, MD5C27, MD5S23);
108 MD5_STEP (MD5_H1, a, b, c, d, wd_t, MD5C28, MD5S20);
109 MD5_STEP (MD5_H2, d, a, b, c, w0_t, MD5C29, MD5S21);
110 MD5_STEP (MD5_H1, c, d, a, b, w3_t, MD5C2a, MD5S22);
111 MD5_STEP (MD5_H2, b, c, d, a, w6_t, MD5C2b, MD5S23);
112 MD5_STEP (MD5_H1, a, b, c, d, w9_t, MD5C2c, MD5S20);
113 MD5_STEP (MD5_H2, d, a, b, c, wc_t, MD5C2d, MD5S21);
114 MD5_STEP (MD5_H1, c, d, a, b, wf_t, MD5C2e, MD5S22);
115 MD5_STEP (MD5_H2, b, c, d, a, w2_t, MD5C2f, MD5S23);
117 MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30);
118 MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31);
119 MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32);
120 MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33);
121 MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30);
122 MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31);
123 MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32);
124 MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33);
125 MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30);
126 MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31);
127 MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32);
128 MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33);
129 MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30);
130 MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31);
131 MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32);
132 MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33);
140 extern "C" __global__ void __launch_bounds__ (256, 1) m00400_init (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, phpass_tmp_t *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)
142 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
144 if (gid >= gid_max) return;
148 w0[0] = pws[gid].i[ 0];
149 w0[1] = pws[gid].i[ 1];
150 w0[2] = pws[gid].i[ 2];
151 w0[3] = pws[gid].i[ 3];
155 w1[0] = pws[gid].i[ 4];
156 w1[1] = pws[gid].i[ 5];
157 w1[2] = pws[gid].i[ 6];
158 w1[3] = pws[gid].i[ 7];
162 w2[0] = pws[gid].i[ 8];
163 w2[1] = pws[gid].i[ 9];
167 const u32 pw_len = pws[gid].pw_len;
175 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
176 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
182 u32 block_len = 8 + pw_len;
186 block0[0] = salt_buf[0];
187 block0[1] = salt_buf[1];
209 block3[2] = block_len * 8;
212 append_0x80_4 (block0, block1, block2, block3, block_len);
225 md5_transform (block0, block1, block2, block3, digest);
227 tmps[gid].digest_buf[0] = digest[0];
228 tmps[gid].digest_buf[1] = digest[1];
229 tmps[gid].digest_buf[2] = digest[2];
230 tmps[gid].digest_buf[3] = digest[3];
233 extern "C" __global__ void __launch_bounds__ (256, 1) m00400_loop (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, phpass_tmp_t *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)
235 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
237 if (gid >= gid_max) return;
241 w0[0] = pws[gid].i[ 0];
242 w0[1] = pws[gid].i[ 1];
243 w0[2] = pws[gid].i[ 2];
244 w0[3] = pws[gid].i[ 3];
248 w1[0] = pws[gid].i[ 4];
249 w1[1] = pws[gid].i[ 5];
250 w1[2] = pws[gid].i[ 6];
251 w1[3] = pws[gid].i[ 7];
255 w2[0] = pws[gid].i[ 8];
256 w2[1] = pws[gid].i[ 9];
260 const u32 pw_len = pws[gid].pw_len;
268 digest[0] = tmps[gid].digest_buf[0];
269 digest[1] = tmps[gid].digest_buf[1];
270 digest[2] = tmps[gid].digest_buf[2];
271 digest[3] = tmps[gid].digest_buf[3];
277 u32 block_len = (16 + pw_len);
304 block3[2] = block_len * 8;
307 append_0x80_4 (block0, block1, block2, block3, block_len);
313 for (u32 i = 0; i < loop_cnt; i++)
315 block0[0] = digest[0];
316 block0[1] = digest[1];
317 block0[2] = digest[2];
318 block0[3] = digest[3];
325 md5_transform (block0, block1, block2, block3, digest);
328 tmps[gid].digest_buf[0] = digest[0];
329 tmps[gid].digest_buf[1] = digest[1];
330 tmps[gid].digest_buf[2] = digest[2];
331 tmps[gid].digest_buf[3] = digest[3];
334 extern "C" __global__ void __launch_bounds__ (256, 1) m00400_comp (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, phpass_tmp_t *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)
340 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
342 if (gid >= gid_max) return;
344 const u32 lid = threadIdx.x;
350 const u32x r0 = tmps[gid].digest_buf[DGST_R0];
351 const u32x r1 = tmps[gid].digest_buf[DGST_R1];
352 const u32x r2 = tmps[gid].digest_buf[DGST_R2];
353 const u32x r3 = tmps[gid].digest_buf[DGST_R3];
357 #include VECT_COMPARE_M