diff --git a/OpenCL/m21200_a0-optimized.cl b/OpenCL/m21200_a0-optimized.cl new file mode 100644 index 000000000..fe2cb311b --- /dev/null +++ b/OpenCL/m21200_a0-optimized.cl @@ -0,0 +1,1297 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +#define NEW_SIMD_CODE + +#ifdef KERNEL_STATIC +#include "inc_vendor.h" +#include "inc_types.h" +#include "inc_platform.cl" +#include "inc_common.cl" +#include "inc_rp_optimized.h" +#include "inc_rp_optimized.cl" +#include "inc_simd.cl" +#include "inc_hash_md5.cl" +#include "inc_hash_sha1.cl" +#endif + +#if VECT_SIZE == 1 +#define uint_to_hex_lower8(i) make_u32x (l_bin2asc[(i)]) +#elif VECT_SIZE == 2 +#define uint_to_hex_lower8(i) make_u32x (l_bin2asc[(i).s0], l_bin2asc[(i).s1]) +#elif VECT_SIZE == 4 +#define uint_to_hex_lower8(i) make_u32x (l_bin2asc[(i).s0], l_bin2asc[(i).s1], l_bin2asc[(i).s2], l_bin2asc[(i).s3]) +#elif VECT_SIZE == 8 +#define uint_to_hex_lower8(i) make_u32x (l_bin2asc[(i).s0], l_bin2asc[(i).s1], l_bin2asc[(i).s2], l_bin2asc[(i).s3], l_bin2asc[(i).s4], l_bin2asc[(i).s5], l_bin2asc[(i).s6], l_bin2asc[(i).s7]) +#elif VECT_SIZE == 16 +#define uint_to_hex_lower8(i) make_u32x (l_bin2asc[(i).s0], l_bin2asc[(i).s1], l_bin2asc[(i).s2], l_bin2asc[(i).s3], l_bin2asc[(i).s4], l_bin2asc[(i).s5], l_bin2asc[(i).s6], l_bin2asc[(i).s7], l_bin2asc[(i).s8], l_bin2asc[(i).s9], l_bin2asc[(i).sa], l_bin2asc[(i).sb], l_bin2asc[(i).sc], l_bin2asc[(i).sd], l_bin2asc[(i).se], l_bin2asc[(i).sf]) +#endif + +KERNEL_FQ void m21200_m04 (KERN_ATTR_RULES ()) +{ + /** + * modifier + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * bin2asc table + */ + + LOCAL_VK u32 l_bin2asc[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + const u32 i0 = (i >> 0) & 15; + const u32 i1 = (i >> 4) & 15; + + l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 8 + | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 0; + } + + SYNC_THREADS (); + + if (gid >= gid_max) return; + + /** + * base + */ + + u32 pw_buf0[4]; + u32 pw_buf1[4]; + + pw_buf0[0] = pws[gid].i[0]; + pw_buf0[1] = pws[gid].i[1]; + pw_buf0[2] = pws[gid].i[2]; + pw_buf0[3] = pws[gid].i[3]; + pw_buf1[0] = pws[gid].i[4]; + pw_buf1[1] = pws[gid].i[5]; + pw_buf1[2] = pws[gid].i[6]; + pw_buf1[3] = pws[gid].i[7]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * salt + */ + + u32 salt_buf0[4]; + u32 salt_buf1[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]; + salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4]; + salt_buf1[1] = 0; + salt_buf1[2] = 0; + salt_buf1[3] = 0; + + const u32 salt_len = salt_bufs[salt_pos].salt_len; + + append_0x80_2x4_S (salt_buf0, salt_buf1, salt_len); + + /** + * sha1(salt) + */ + + u32x w0_t = hc_swap32 (salt_buf0[0]); + u32x w1_t = hc_swap32 (salt_buf0[1]); + u32x w2_t = hc_swap32 (salt_buf0[2]); + u32x w3_t = hc_swap32 (salt_buf0[3]); + u32x w4_t = hc_swap32 (salt_buf1[0]); + u32x w5_t = 0; + u32x w6_t = 0; + u32x w7_t = 0; + u32x w8_t = 0; + u32x w9_t = 0; + u32x wa_t = 0; + u32x wb_t = 0; + u32x wc_t = 0; + u32x wd_t = 0; + u32x we_t = 0; + u32x wf_t = 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 = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t); + w3_t = hc_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 = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t); + w7_t = hc_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 = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t); + wb_t = hc_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 = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + const u32x a0 = a; + const u32x b0 = b; + const u32x c0 = c; + const u32x d0 = d; + const u32x e0 = e; + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + u32x w0[4] = { 0 }; + u32x w1[4] = { 0 }; + u32x w2[4] = { 0 }; + u32x w3[4] = { 0 }; + + const u32x out_len = apply_rules_vect_optimized (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1); + + append_0x80_2x4_VV (w0, w1, out_len); + + /** + * md5 + */ + + w0_t = w0[0]; + w1_t = w0[1]; + w2_t = w0[2]; + w3_t = w0[3]; + w4_t = w1[0]; + w5_t = w1[1]; + w6_t = w1[2]; + w7_t = w1[3]; + w8_t = w2[0]; + w9_t = w2[1]; + wa_t = w2[2]; + wb_t = w2[3]; + wc_t = w3[0]; + wd_t = w3[1]; + we_t = out_len * 8; + wf_t = 0; + + a = MD5M_A; + b = MD5M_B; + c = MD5M_C; + d = MD5M_D; + + MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03); + + MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13); + + u32x t; + + MD5_STEP (MD5_H1, a, b, c, d, w5_t, MD5C20, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w8_t, MD5C21, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wb_t, MD5C22, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, we_t, MD5C23, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w1_t, MD5C24, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w4_t, MD5C25, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w7_t, MD5C26, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, wa_t, MD5C27, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, wd_t, MD5C28, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w0_t, MD5C29, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w3_t, MD5C2a, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w6_t, MD5C2b, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w9_t, MD5C2c, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, wc_t, MD5C2d, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wf_t, MD5C2e, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w2_t, MD5C2f, MD5S23); + + MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); + + a += MD5M_A; + b += MD5M_B; + c += MD5M_C; + d += MD5M_D; + + const u32x a1 = hc_swap32 (a); + const u32x b1 = hc_swap32 (b); + const u32x c1 = hc_swap32 (c); + const u32x d1 = hc_swap32 (d); + + /** + * md5 + */ + + w0_t = uint_to_hex_lower8 ((a0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((a0 >> 16) & 255) << 16; + w1_t = uint_to_hex_lower8 ((a0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((a0 >> 0) & 255) << 16; + w2_t = uint_to_hex_lower8 ((b0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((b0 >> 16) & 255) << 16; + w3_t = uint_to_hex_lower8 ((b0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((b0 >> 0) & 255) << 16; + w4_t = uint_to_hex_lower8 ((c0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((c0 >> 16) & 255) << 16; + w5_t = uint_to_hex_lower8 ((c0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((c0 >> 0) & 255) << 16; + w6_t = uint_to_hex_lower8 ((d0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((d0 >> 16) & 255) << 16; + w7_t = uint_to_hex_lower8 ((d0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((d0 >> 0) & 255) << 16; + w8_t = uint_to_hex_lower8 ((e0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((e0 >> 16) & 255) << 16; + w9_t = uint_to_hex_lower8 ((e0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((e0 >> 0) & 255) << 16; + wa_t = 0; + wb_t = 0; + wc_t = 0; + wd_t = 0; + we_t = 0; + wf_t = 0; + + // ctx len 40, pos 40 + + w0[0] = uint_to_hex_lower8 ((a1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((a1 >> 16) & 255) << 16; + w0[1] = uint_to_hex_lower8 ((a1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((a1 >> 0) & 255) << 16; + w0[2] = uint_to_hex_lower8 ((b1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((b1 >> 16) & 255) << 16; + w0[3] = uint_to_hex_lower8 ((b1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((b1 >> 0) & 255) << 16; + w1[0] = uint_to_hex_lower8 ((c1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((c1 >> 16) & 255) << 16; + w1[1] = uint_to_hex_lower8 ((c1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((c1 >> 0) & 255) << 16; + w1[2] = uint_to_hex_lower8 ((d1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((d1 >> 16) & 255) << 16; + w1[3] = uint_to_hex_lower8 ((d1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((d1 >> 0) & 255) << 16; + + w2[0] = 0; + w2[1] = 0; + w2[2] = 0; + w2[3] = 0; + w3[0] = 0; + w3[1] = 0; + w3[2] = 0; + w3[3] = 0; + + u32x _w0[4] = { 0 }; + u32x _w1[4] = { 0 }; + u32x _w2[4] = { 0 }; + u32x _w3[4] = { 0 }; + + switch_buffer_by_offset_carry_le (w0, w1, w2, w3, _w0, _w1, _w2, _w3, 40); + + w0_t |= w0[0]; + w1_t |= w0[1]; + w2_t |= w0[2]; + w3_t |= w0[3]; + w4_t |= w1[0]; + w5_t |= w1[1]; + w6_t |= w1[2]; + w7_t |= w1[3]; + w8_t |= w2[0]; + w9_t |= w2[1]; + wa_t |= w2[2]; + wb_t |= w2[3]; + wc_t |= w3[0]; + wd_t |= w3[1]; + we_t |= w3[2]; + wf_t |= w3[3]; + + // md5 transform + + a = MD5M_A; + b = MD5M_B; + c = MD5M_C; + d = MD5M_D; + + u32x digest[4]; + + digest[0] = a; + digest[1] = b; + digest[2] = c; + digest[3] = d; + + MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03); + + MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13); + + MD5_STEP (MD5_H1, a, b, c, d, w5_t, MD5C20, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w8_t, MD5C21, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wb_t, MD5C22, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, we_t, MD5C23, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w1_t, MD5C24, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w4_t, MD5C25, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w7_t, MD5C26, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, wa_t, MD5C27, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, wd_t, MD5C28, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w0_t, MD5C29, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w3_t, MD5C2a, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w6_t, MD5C2b, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w9_t, MD5C2c, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, wc_t, MD5C2d, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wf_t, MD5C2e, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w2_t, MD5C2f, MD5S23); + + MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); + + digest[0] += a; + digest[1] += b; + digest[2] += c; + digest[3] += d; + + w0[0] = _w0[0]; + w0[1] = _w0[1]; + w0[2] = _w0[2]; + w0[3] = _w0[3]; + w1[0] = _w1[0]; + w1[1] = _w1[1]; + w1[2] = _w1[2]; + w1[3] = _w1[3]; + w2[0] = _w2[0]; + w2[1] = _w2[1]; + w2[2] = _w2[2]; + w2[3] = _w2[3]; + w3[0] = _w3[0]; + w3[1] = _w3[1]; + w3[2] = _w3[2]; + w3[3] = _w3[3]; + + // ctx len 72, pos 8 + + append_0x80_4x4 (w0, w1, w2, w3, 8); + + w3[2] = 72 * 8; + w3[3] = 0; + + // md5 final transform + + a = digest[0]; + b = digest[1]; + c = digest[2]; + d = digest[3]; + + w0_t = w0[0]; + w1_t = w0[1]; + w2_t = w0[2]; + w3_t = w0[3]; + w4_t = w1[0]; + w5_t = w1[1]; + w6_t = w1[2]; + w7_t = w1[3]; + w8_t = w2[0]; + w9_t = w2[1]; + wa_t = w2[2]; + wb_t = w2[3]; + wc_t = w3[0]; + wd_t = w3[1]; + we_t = w3[2]; + wf_t = w3[3]; + + MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03); + + MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13); + + MD5_STEP (MD5_H1, a, b, c, d, w5_t, MD5C20, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w8_t, MD5C21, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wb_t, MD5C22, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, we_t, MD5C23, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w1_t, MD5C24, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w4_t, MD5C25, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w7_t, MD5C26, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, wa_t, MD5C27, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, wd_t, MD5C28, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w0_t, MD5C29, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w3_t, MD5C2a, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w6_t, MD5C2b, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w9_t, MD5C2c, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, wc_t, MD5C2d, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wf_t, MD5C2e, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w2_t, MD5C2f, MD5S23); + + MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); + + a += digest[0] - MD5M_A; + b += digest[1] - MD5M_B; + c += digest[2] - MD5M_C; + d += digest[3] - MD5M_D; + + COMPARE_M_SIMD (a, d, c, b); + } +} + +KERNEL_FQ void m21200_m08 (KERN_ATTR_RULES ()) +{ +} + +KERNEL_FQ void m21200_m16 (KERN_ATTR_RULES ()) +{ +} + +KERNEL_FQ void m21200_s04 (KERN_ATTR_RULES ()) +{ + /** + * modifier + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * bin2asc table + */ + + LOCAL_VK u32 l_bin2asc[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + const u32 i0 = (i >> 0) & 15; + const u32 i1 = (i >> 4) & 15; + + l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 8 + | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 0; + } + + SYNC_THREADS (); + + if (gid >= gid_max) return; + + /** + * base + */ + + u32 pw_buf0[4]; + u32 pw_buf1[4]; + + pw_buf0[0] = pws[gid].i[0]; + pw_buf0[1] = pws[gid].i[1]; + pw_buf0[2] = pws[gid].i[2]; + pw_buf0[3] = pws[gid].i[3]; + pw_buf1[0] = pws[gid].i[4]; + pw_buf1[1] = pws[gid].i[5]; + pw_buf1[2] = pws[gid].i[6]; + pw_buf1[3] = pws[gid].i[7]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * salt + */ + + u32 salt_buf0[4]; + u32 salt_buf1[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]; + salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4]; + salt_buf1[1] = 0; + salt_buf1[2] = 0; + salt_buf1[3] = 0; + + const u32 salt_len = salt_bufs[salt_pos].salt_len; + + append_0x80_2x4_S (salt_buf0, salt_buf1, salt_len); + + /** + * sha1(salt) + */ + + u32x w0_t = hc_swap32 (salt_buf0[0]); + u32x w1_t = hc_swap32 (salt_buf0[1]); + u32x w2_t = hc_swap32 (salt_buf0[2]); + u32x w3_t = hc_swap32 (salt_buf0[3]); + u32x w4_t = hc_swap32 (salt_buf1[0]); + u32x w5_t = 0; + u32x w6_t = 0; + u32x w7_t = 0; + u32x w8_t = 0; + u32x w9_t = 0; + u32x wa_t = 0; + u32x wb_t = 0; + u32x wc_t = 0; + u32x wd_t = 0; + u32x we_t = 0; + u32x wf_t = 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 = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t); + w3_t = hc_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 = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t); + w7_t = hc_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 = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t); + wb_t = hc_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 = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + const u32x a0 = a; + const u32x b0 = b; + const u32x c0 = c; + const u32x d0 = d; + const u32x e0 = e; + + /** + * 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] + }; + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + u32x w0[4] = { 0 }; + u32x w1[4] = { 0 }; + u32x w2[4] = { 0 }; + u32x w3[4] = { 0 }; + + const u32x out_len = apply_rules_vect_optimized (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1); + + append_0x80_2x4_VV (w0, w1, out_len); + + /** + * md5 + */ + + w0_t = w0[0]; + w1_t = w0[1]; + w2_t = w0[2]; + w3_t = w0[3]; + w4_t = w1[0]; + w5_t = w1[1]; + w6_t = w1[2]; + w7_t = w1[3]; + w8_t = w2[0]; + w9_t = w2[1]; + wa_t = w2[2]; + wb_t = w2[3]; + wc_t = w3[0]; + wd_t = w3[1]; + we_t = out_len * 8; + wf_t = 0; + + a = MD5M_A; + b = MD5M_B; + c = MD5M_C; + d = MD5M_D; + + MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03); + + MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13); + + u32x t; + + MD5_STEP (MD5_H1, a, b, c, d, w5_t, MD5C20, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w8_t, MD5C21, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wb_t, MD5C22, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, we_t, MD5C23, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w1_t, MD5C24, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w4_t, MD5C25, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w7_t, MD5C26, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, wa_t, MD5C27, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, wd_t, MD5C28, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w0_t, MD5C29, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w3_t, MD5C2a, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w6_t, MD5C2b, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w9_t, MD5C2c, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, wc_t, MD5C2d, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wf_t, MD5C2e, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w2_t, MD5C2f, MD5S23); + + MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); + + a += MD5M_A; + b += MD5M_B; + c += MD5M_C; + d += MD5M_D; + + const u32x a1 = hc_swap32 (a); + const u32x b1 = hc_swap32 (b); + const u32x c1 = hc_swap32 (c); + const u32x d1 = hc_swap32 (d); + + /** + * md5 + */ + + w0_t = uint_to_hex_lower8 ((a0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((a0 >> 16) & 255) << 16; + w1_t = uint_to_hex_lower8 ((a0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((a0 >> 0) & 255) << 16; + w2_t = uint_to_hex_lower8 ((b0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((b0 >> 16) & 255) << 16; + w3_t = uint_to_hex_lower8 ((b0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((b0 >> 0) & 255) << 16; + w4_t = uint_to_hex_lower8 ((c0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((c0 >> 16) & 255) << 16; + w5_t = uint_to_hex_lower8 ((c0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((c0 >> 0) & 255) << 16; + w6_t = uint_to_hex_lower8 ((d0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((d0 >> 16) & 255) << 16; + w7_t = uint_to_hex_lower8 ((d0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((d0 >> 0) & 255) << 16; + w8_t = uint_to_hex_lower8 ((e0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((e0 >> 16) & 255) << 16; + w9_t = uint_to_hex_lower8 ((e0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((e0 >> 0) & 255) << 16; + wa_t = 0; + wb_t = 0; + wc_t = 0; + wd_t = 0; + we_t = 0; + wf_t = 0; + + // ctx len 40, pos 40 + + w0[0] = uint_to_hex_lower8 ((a1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((a1 >> 16) & 255) << 16; + w0[1] = uint_to_hex_lower8 ((a1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((a1 >> 0) & 255) << 16; + w0[2] = uint_to_hex_lower8 ((b1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((b1 >> 16) & 255) << 16; + w0[3] = uint_to_hex_lower8 ((b1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((b1 >> 0) & 255) << 16; + w1[0] = uint_to_hex_lower8 ((c1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((c1 >> 16) & 255) << 16; + w1[1] = uint_to_hex_lower8 ((c1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((c1 >> 0) & 255) << 16; + w1[2] = uint_to_hex_lower8 ((d1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((d1 >> 16) & 255) << 16; + w1[3] = uint_to_hex_lower8 ((d1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((d1 >> 0) & 255) << 16; + + w2[0] = 0; + w2[1] = 0; + w2[2] = 0; + w2[3] = 0; + w3[0] = 0; + w3[1] = 0; + w3[2] = 0; + w3[3] = 0; + + u32x _w0[4] = { 0 }; + u32x _w1[4] = { 0 }; + u32x _w2[4] = { 0 }; + u32x _w3[4] = { 0 }; + + switch_buffer_by_offset_carry_le (w0, w1, w2, w3, _w0, _w1, _w2, _w3, 40); + + w0_t |= w0[0]; + w1_t |= w0[1]; + w2_t |= w0[2]; + w3_t |= w0[3]; + w4_t |= w1[0]; + w5_t |= w1[1]; + w6_t |= w1[2]; + w7_t |= w1[3]; + w8_t |= w2[0]; + w9_t |= w2[1]; + wa_t |= w2[2]; + wb_t |= w2[3]; + wc_t |= w3[0]; + wd_t |= w3[1]; + we_t |= w3[2]; + wf_t |= w3[3]; + + // md5 transform + + a = MD5M_A; + b = MD5M_B; + c = MD5M_C; + d = MD5M_D; + + u32x digest[4]; + + digest[0] = a; + digest[1] = b; + digest[2] = c; + digest[3] = d; + + MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03); + + MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13); + + MD5_STEP (MD5_H1, a, b, c, d, w5_t, MD5C20, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w8_t, MD5C21, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wb_t, MD5C22, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, we_t, MD5C23, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w1_t, MD5C24, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w4_t, MD5C25, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w7_t, MD5C26, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, wa_t, MD5C27, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, wd_t, MD5C28, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w0_t, MD5C29, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w3_t, MD5C2a, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w6_t, MD5C2b, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w9_t, MD5C2c, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, wc_t, MD5C2d, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wf_t, MD5C2e, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w2_t, MD5C2f, MD5S23); + + MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); + + digest[0] += a; + digest[1] += b; + digest[2] += c; + digest[3] += d; + + w0[0] = _w0[0]; + w0[1] = _w0[1]; + w0[2] = _w0[2]; + w0[3] = _w0[3]; + w1[0] = _w1[0]; + w1[1] = _w1[1]; + w1[2] = _w1[2]; + w1[3] = _w1[3]; + w2[0] = _w2[0]; + w2[1] = _w2[1]; + w2[2] = _w2[2]; + w2[3] = _w2[3]; + w3[0] = _w3[0]; + w3[1] = _w3[1]; + w3[2] = _w3[2]; + w3[3] = _w3[3]; + + // ctx len 72, pos 8 + + append_0x80_4x4 (w0, w1, w2, w3, 8); + + w3[2] = 72 * 8; + w3[3] = 0; + + // md5 final transform + + a = digest[0]; + b = digest[1]; + c = digest[2]; + d = digest[3]; + + w0_t = w0[0]; + w1_t = w0[1]; + w2_t = w0[2]; + w3_t = w0[3]; + w4_t = w1[0]; + w5_t = w1[1]; + w6_t = w1[2]; + w7_t = w1[3]; + w8_t = w2[0]; + w9_t = w2[1]; + wa_t = w2[2]; + wb_t = w2[3]; + wc_t = w3[0]; + wd_t = w3[1]; + we_t = w3[2]; + wf_t = w3[3]; + + MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03); + + MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13); + + MD5_STEP (MD5_H1, a, b, c, d, w5_t, MD5C20, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w8_t, MD5C21, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wb_t, MD5C22, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, we_t, MD5C23, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w1_t, MD5C24, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w4_t, MD5C25, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w7_t, MD5C26, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, wa_t, MD5C27, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, wd_t, MD5C28, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w0_t, MD5C29, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w3_t, MD5C2a, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w6_t, MD5C2b, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w9_t, MD5C2c, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, wc_t, MD5C2d, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wf_t, MD5C2e, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w2_t, MD5C2f, MD5S23); + + MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30); + + if (MATCHES_NONE_VS ((a+digest[0]-MD5M_A), search[0])) continue; + + MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); + + a += digest[0] - MD5M_A; + b += digest[1] - MD5M_B; + c += digest[2] - MD5M_C; + d += digest[3] - MD5M_D; + + COMPARE_S_SIMD (a, d, c, b); + } +} + +KERNEL_FQ void m21200_s08 (KERN_ATTR_RULES ()) +{ +} + +KERNEL_FQ void m21200_s16 (KERN_ATTR_RULES ()) +{ +} diff --git a/OpenCL/m21200_a1-optimized.cl b/OpenCL/m21200_a1-optimized.cl new file mode 100644 index 000000000..4f835c979 --- /dev/null +++ b/OpenCL/m21200_a1-optimized.cl @@ -0,0 +1,1411 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +#define NEW_SIMD_CODE + +#ifdef KERNEL_STATIC +#include "inc_vendor.h" +#include "inc_types.h" +#include "inc_platform.cl" +#include "inc_common.cl" +#include "inc_simd.cl" +#include "inc_hash_md5.cl" +#include "inc_hash_sha1.cl" +#endif + +#if VECT_SIZE == 1 +#define uint_to_hex_lower8(i) make_u32x (l_bin2asc[(i)]) +#elif VECT_SIZE == 2 +#define uint_to_hex_lower8(i) make_u32x (l_bin2asc[(i).s0], l_bin2asc[(i).s1]) +#elif VECT_SIZE == 4 +#define uint_to_hex_lower8(i) make_u32x (l_bin2asc[(i).s0], l_bin2asc[(i).s1], l_bin2asc[(i).s2], l_bin2asc[(i).s3]) +#elif VECT_SIZE == 8 +#define uint_to_hex_lower8(i) make_u32x (l_bin2asc[(i).s0], l_bin2asc[(i).s1], l_bin2asc[(i).s2], l_bin2asc[(i).s3], l_bin2asc[(i).s4], l_bin2asc[(i).s5], l_bin2asc[(i).s6], l_bin2asc[(i).s7]) +#elif VECT_SIZE == 16 +#define uint_to_hex_lower8(i) make_u32x (l_bin2asc[(i).s0], l_bin2asc[(i).s1], l_bin2asc[(i).s2], l_bin2asc[(i).s3], l_bin2asc[(i).s4], l_bin2asc[(i).s5], l_bin2asc[(i).s6], l_bin2asc[(i).s7], l_bin2asc[(i).s8], l_bin2asc[(i).s9], l_bin2asc[(i).sa], l_bin2asc[(i).sb], l_bin2asc[(i).sc], l_bin2asc[(i).sd], l_bin2asc[(i).se], l_bin2asc[(i).sf]) +#endif + +KERNEL_FQ void m21200_m04 (KERN_ATTR_BASIC ()) +{ + /** + * modifier + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * bin2asc table + */ + + LOCAL_VK u32 l_bin2asc[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + const u32 i0 = (i >> 0) & 15; + const u32 i1 = (i >> 4) & 15; + + l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 8 + | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 0; + } + + SYNC_THREADS (); + + if (gid >= gid_max) return; + + /** + * base + */ + + u32 pw_buf0[4]; + u32 pw_buf1[4]; + + pw_buf0[0] = pws[gid].i[0]; + pw_buf0[1] = pws[gid].i[1]; + pw_buf0[2] = pws[gid].i[2]; + pw_buf0[3] = pws[gid].i[3]; + pw_buf1[0] = pws[gid].i[4]; + pw_buf1[1] = pws[gid].i[5]; + pw_buf1[2] = pws[gid].i[6]; + pw_buf1[3] = pws[gid].i[7]; + + const u32 pw_l_len = pws[gid].pw_len & 63; + + /** + * salt + */ + + u32 salt_buf0[4]; + u32 salt_buf1[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]; + salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4]; + salt_buf1[1] = 0; + salt_buf1[2] = 0; + salt_buf1[3] = 0; + + const u32 salt_len = salt_bufs[salt_pos].salt_len; + + append_0x80_2x4_S (salt_buf0, salt_buf1, salt_len); + + /** + * sha1(salt) + */ + + u32x w0_t = hc_swap32 (salt_buf0[0]); + u32x w1_t = hc_swap32 (salt_buf0[1]); + u32x w2_t = hc_swap32 (salt_buf0[2]); + u32x w3_t = hc_swap32 (salt_buf0[3]); + u32x w4_t = hc_swap32 (salt_buf1[0]); + u32x w5_t = 0; + u32x w6_t = 0; + u32x w7_t = 0; + u32x w8_t = 0; + u32x w9_t = 0; + u32x wa_t = 0; + u32x wb_t = 0; + u32x wc_t = 0; + u32x wd_t = 0; + u32x we_t = 0; + u32x wf_t = 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 = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t); + w3_t = hc_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 = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t); + w7_t = hc_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 = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t); + wb_t = hc_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 = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + const u32x a0 = a; + const u32x b0 = b; + const u32x c0 = c; + const u32x d0 = d; + const u32x e0 = e; + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos) & 63; + + const u32x pw_len = (pw_l_len + pw_r_len) & 63; + + /** + * concat password candidate + */ + + u32x wordl0[4] = { 0 }; + u32x wordl1[4] = { 0 }; + u32x wordl2[4] = { 0 }; + u32x wordl3[4] = { 0 }; + + wordl0[0] = pw_buf0[0]; + wordl0[1] = pw_buf0[1]; + wordl0[2] = pw_buf0[2]; + wordl0[3] = pw_buf0[3]; + wordl1[0] = pw_buf1[0]; + wordl1[1] = pw_buf1[1]; + wordl1[2] = pw_buf1[2]; + wordl1[3] = pw_buf1[3]; + + u32x wordr0[4] = { 0 }; + u32x wordr1[4] = { 0 }; + u32x wordr2[4] = { 0 }; + u32x wordr3[4] = { 0 }; + + wordr0[0] = ix_create_combt (combs_buf, il_pos, 0); + wordr0[1] = ix_create_combt (combs_buf, il_pos, 1); + wordr0[2] = ix_create_combt (combs_buf, il_pos, 2); + wordr0[3] = ix_create_combt (combs_buf, il_pos, 3); + wordr1[0] = ix_create_combt (combs_buf, il_pos, 4); + wordr1[1] = ix_create_combt (combs_buf, il_pos, 5); + wordr1[2] = ix_create_combt (combs_buf, il_pos, 6); + wordr1[3] = ix_create_combt (combs_buf, il_pos, 7); + + if (combs_mode == COMBINATOR_MODE_BASE_LEFT) + { + switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len); + } + else + { + switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len); + } + + u32x w0[4]; + u32x w1[4]; + u32x w2[4]; + u32x w3[4]; + + w0[0] = wordl0[0] | wordr0[0]; + w0[1] = wordl0[1] | wordr0[1]; + w0[2] = wordl0[2] | wordr0[2]; + w0[3] = wordl0[3] | wordr0[3]; + w1[0] = wordl1[0] | wordr1[0]; + w1[1] = wordl1[1] | wordr1[1]; + w1[2] = wordl1[2] | wordr1[2]; + w1[3] = wordl1[3] | wordr1[3]; + w2[0] = wordl2[0] | wordr2[0]; + w2[1] = wordl2[1] | wordr2[1]; + w2[2] = wordl2[2] | wordr2[2]; + w2[3] = wordl2[3] | wordr2[3]; + w3[0] = wordl3[0] | wordr3[0]; + w3[1] = wordl3[1] | wordr3[1]; + w3[2] = wordl3[2] | wordr3[2]; + w3[3] = wordl3[3] | wordr3[3]; + + /** + * md5 + */ + + w0_t = w0[0]; + w1_t = w0[1]; + w2_t = w0[2]; + w3_t = w0[3]; + w4_t = w1[0]; + w5_t = w1[1]; + w6_t = w1[2]; + w7_t = w1[3]; + w8_t = w2[0]; + w9_t = w2[1]; + wa_t = w2[2]; + wb_t = w2[3]; + wc_t = w3[0]; + wd_t = w3[1]; + we_t = pw_len * 8; + wf_t = 0; + + a = MD5M_A; + b = MD5M_B; + c = MD5M_C; + d = MD5M_D; + + MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03); + + MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13); + + u32x t; + + MD5_STEP (MD5_H1, a, b, c, d, w5_t, MD5C20, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w8_t, MD5C21, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wb_t, MD5C22, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, we_t, MD5C23, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w1_t, MD5C24, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w4_t, MD5C25, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w7_t, MD5C26, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, wa_t, MD5C27, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, wd_t, MD5C28, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w0_t, MD5C29, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w3_t, MD5C2a, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w6_t, MD5C2b, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w9_t, MD5C2c, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, wc_t, MD5C2d, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wf_t, MD5C2e, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w2_t, MD5C2f, MD5S23); + + MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); + + a += MD5M_A; + b += MD5M_B; + c += MD5M_C; + d += MD5M_D; + + const u32x a1 = hc_swap32 (a); + const u32x b1 = hc_swap32 (b); + const u32x c1 = hc_swap32 (c); + const u32x d1 = hc_swap32 (d); + + /** + * md5 + */ + + w0_t = uint_to_hex_lower8 ((a0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((a0 >> 16) & 255) << 16; + w1_t = uint_to_hex_lower8 ((a0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((a0 >> 0) & 255) << 16; + w2_t = uint_to_hex_lower8 ((b0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((b0 >> 16) & 255) << 16; + w3_t = uint_to_hex_lower8 ((b0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((b0 >> 0) & 255) << 16; + w4_t = uint_to_hex_lower8 ((c0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((c0 >> 16) & 255) << 16; + w5_t = uint_to_hex_lower8 ((c0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((c0 >> 0) & 255) << 16; + w6_t = uint_to_hex_lower8 ((d0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((d0 >> 16) & 255) << 16; + w7_t = uint_to_hex_lower8 ((d0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((d0 >> 0) & 255) << 16; + w8_t = uint_to_hex_lower8 ((e0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((e0 >> 16) & 255) << 16; + w9_t = uint_to_hex_lower8 ((e0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((e0 >> 0) & 255) << 16; + wa_t = 0; + wb_t = 0; + wc_t = 0; + wd_t = 0; + we_t = 0; + wf_t = 0; + + // ctx len 40, pos 40 + + w0[0] = uint_to_hex_lower8 ((a1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((a1 >> 16) & 255) << 16; + w0[1] = uint_to_hex_lower8 ((a1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((a1 >> 0) & 255) << 16; + w0[2] = uint_to_hex_lower8 ((b1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((b1 >> 16) & 255) << 16; + w0[3] = uint_to_hex_lower8 ((b1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((b1 >> 0) & 255) << 16; + w1[0] = uint_to_hex_lower8 ((c1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((c1 >> 16) & 255) << 16; + w1[1] = uint_to_hex_lower8 ((c1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((c1 >> 0) & 255) << 16; + w1[2] = uint_to_hex_lower8 ((d1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((d1 >> 16) & 255) << 16; + w1[3] = uint_to_hex_lower8 ((d1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((d1 >> 0) & 255) << 16; + + w2[0] = 0; + w2[1] = 0; + w2[2] = 0; + w2[3] = 0; + w3[0] = 0; + w3[1] = 0; + w3[2] = 0; + w3[3] = 0; + + u32x _w0[4] = { 0 }; + u32x _w1[4] = { 0 }; + u32x _w2[4] = { 0 }; + u32x _w3[4] = { 0 }; + + switch_buffer_by_offset_carry_le (w0, w1, w2, w3, _w0, _w1, _w2, _w3, 40); + + w0_t |= w0[0]; + w1_t |= w0[1]; + w2_t |= w0[2]; + w3_t |= w0[3]; + w4_t |= w1[0]; + w5_t |= w1[1]; + w6_t |= w1[2]; + w7_t |= w1[3]; + w8_t |= w2[0]; + w9_t |= w2[1]; + wa_t |= w2[2]; + wb_t |= w2[3]; + wc_t |= w3[0]; + wd_t |= w3[1]; + we_t |= w3[2]; + wf_t |= w3[3]; + + // md5 transform + + a = MD5M_A; + b = MD5M_B; + c = MD5M_C; + d = MD5M_D; + + u32x digest[4]; + + digest[0] = a; + digest[1] = b; + digest[2] = c; + digest[3] = d; + + MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03); + + MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13); + + MD5_STEP (MD5_H1, a, b, c, d, w5_t, MD5C20, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w8_t, MD5C21, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wb_t, MD5C22, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, we_t, MD5C23, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w1_t, MD5C24, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w4_t, MD5C25, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w7_t, MD5C26, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, wa_t, MD5C27, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, wd_t, MD5C28, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w0_t, MD5C29, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w3_t, MD5C2a, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w6_t, MD5C2b, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w9_t, MD5C2c, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, wc_t, MD5C2d, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wf_t, MD5C2e, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w2_t, MD5C2f, MD5S23); + + MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); + + digest[0] += a; + digest[1] += b; + digest[2] += c; + digest[3] += d; + + w0[0] = _w0[0]; + w0[1] = _w0[1]; + w0[2] = _w0[2]; + w0[3] = _w0[3]; + w1[0] = _w1[0]; + w1[1] = _w1[1]; + w1[2] = _w1[2]; + w1[3] = _w1[3]; + w2[0] = _w2[0]; + w2[1] = _w2[1]; + w2[2] = _w2[2]; + w2[3] = _w2[3]; + w3[0] = _w3[0]; + w3[1] = _w3[1]; + w3[2] = _w3[2]; + w3[3] = _w3[3]; + + // ctx len 72, pos 8 + + append_0x80_4x4 (w0, w1, w2, w3, 8); + + w3[2] = 72 * 8; + w3[3] = 0; + + // md5 final transform + + a = digest[0]; + b = digest[1]; + c = digest[2]; + d = digest[3]; + + w0_t = w0[0]; + w1_t = w0[1]; + w2_t = w0[2]; + w3_t = w0[3]; + w4_t = w1[0]; + w5_t = w1[1]; + w6_t = w1[2]; + w7_t = w1[3]; + w8_t = w2[0]; + w9_t = w2[1]; + wa_t = w2[2]; + wb_t = w2[3]; + wc_t = w3[0]; + wd_t = w3[1]; + we_t = w3[2]; + wf_t = w3[3]; + + MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03); + + MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13); + + MD5_STEP (MD5_H1, a, b, c, d, w5_t, MD5C20, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w8_t, MD5C21, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wb_t, MD5C22, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, we_t, MD5C23, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w1_t, MD5C24, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w4_t, MD5C25, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w7_t, MD5C26, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, wa_t, MD5C27, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, wd_t, MD5C28, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w0_t, MD5C29, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w3_t, MD5C2a, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w6_t, MD5C2b, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w9_t, MD5C2c, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, wc_t, MD5C2d, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wf_t, MD5C2e, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w2_t, MD5C2f, MD5S23); + + MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); + + a += digest[0] - MD5M_A; + b += digest[1] - MD5M_B; + c += digest[2] - MD5M_C; + d += digest[3] - MD5M_D; + + COMPARE_M_SIMD (a, d, c, b); + } +} + +KERNEL_FQ void m21200_m08 (KERN_ATTR_BASIC ()) +{ +} + +KERNEL_FQ void m21200_m16 (KERN_ATTR_BASIC ()) +{ +} + +KERNEL_FQ void m21200_s04 (KERN_ATTR_BASIC ()) +{ + /** + * modifier + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * bin2asc table + */ + + LOCAL_VK u32 l_bin2asc[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + const u32 i0 = (i >> 0) & 15; + const u32 i1 = (i >> 4) & 15; + + l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 8 + | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 0; + } + + SYNC_THREADS (); + + if (gid >= gid_max) return; + + /** + * base + */ + + u32 pw_buf0[4]; + u32 pw_buf1[4]; + + pw_buf0[0] = pws[gid].i[0]; + pw_buf0[1] = pws[gid].i[1]; + pw_buf0[2] = pws[gid].i[2]; + pw_buf0[3] = pws[gid].i[3]; + pw_buf1[0] = pws[gid].i[4]; + pw_buf1[1] = pws[gid].i[5]; + pw_buf1[2] = pws[gid].i[6]; + pw_buf1[3] = pws[gid].i[7]; + + const u32 pw_l_len = pws[gid].pw_len & 63; + + /** + * salt + */ + + u32 salt_buf0[4]; + u32 salt_buf1[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]; + salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4]; + salt_buf1[1] = 0; + salt_buf1[2] = 0; + salt_buf1[3] = 0; + + const u32 salt_len = salt_bufs[salt_pos].salt_len; + + append_0x80_2x4_S (salt_buf0, salt_buf1, salt_len); + + /** + * sha1(salt) + */ + + u32x w0_t = hc_swap32 (salt_buf0[0]); + u32x w1_t = hc_swap32 (salt_buf0[1]); + u32x w2_t = hc_swap32 (salt_buf0[2]); + u32x w3_t = hc_swap32 (salt_buf0[3]); + u32x w4_t = hc_swap32 (salt_buf1[0]); + u32x w5_t = 0; + u32x w6_t = 0; + u32x w7_t = 0; + u32x w8_t = 0; + u32x w9_t = 0; + u32x wa_t = 0; + u32x wb_t = 0; + u32x wc_t = 0; + u32x wd_t = 0; + u32x we_t = 0; + u32x wf_t = 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 = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t); + w3_t = hc_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 = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t); + w7_t = hc_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 = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t); + wb_t = hc_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 = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + const u32x a0 = a; + const u32x b0 = b; + const u32x c0 = c; + const u32x d0 = d; + const u32x e0 = e; + + /** + * 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] + }; + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos) & 63; + + const u32x pw_len = (pw_l_len + pw_r_len) & 63; + + /** + * concat password candidate + */ + + u32x wordl0[4] = { 0 }; + u32x wordl1[4] = { 0 }; + u32x wordl2[4] = { 0 }; + u32x wordl3[4] = { 0 }; + + wordl0[0] = pw_buf0[0]; + wordl0[1] = pw_buf0[1]; + wordl0[2] = pw_buf0[2]; + wordl0[3] = pw_buf0[3]; + wordl1[0] = pw_buf1[0]; + wordl1[1] = pw_buf1[1]; + wordl1[2] = pw_buf1[2]; + wordl1[3] = pw_buf1[3]; + + u32x wordr0[4] = { 0 }; + u32x wordr1[4] = { 0 }; + u32x wordr2[4] = { 0 }; + u32x wordr3[4] = { 0 }; + + wordr0[0] = ix_create_combt (combs_buf, il_pos, 0); + wordr0[1] = ix_create_combt (combs_buf, il_pos, 1); + wordr0[2] = ix_create_combt (combs_buf, il_pos, 2); + wordr0[3] = ix_create_combt (combs_buf, il_pos, 3); + wordr1[0] = ix_create_combt (combs_buf, il_pos, 4); + wordr1[1] = ix_create_combt (combs_buf, il_pos, 5); + wordr1[2] = ix_create_combt (combs_buf, il_pos, 6); + wordr1[3] = ix_create_combt (combs_buf, il_pos, 7); + + if (combs_mode == COMBINATOR_MODE_BASE_LEFT) + { + switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len); + } + else + { + switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len); + } + + u32x w0[4]; + u32x w1[4]; + u32x w2[4]; + u32x w3[4]; + + w0[0] = wordl0[0] | wordr0[0]; + w0[1] = wordl0[1] | wordr0[1]; + w0[2] = wordl0[2] | wordr0[2]; + w0[3] = wordl0[3] | wordr0[3]; + w1[0] = wordl1[0] | wordr1[0]; + w1[1] = wordl1[1] | wordr1[1]; + w1[2] = wordl1[2] | wordr1[2]; + w1[3] = wordl1[3] | wordr1[3]; + w2[0] = wordl2[0] | wordr2[0]; + w2[1] = wordl2[1] | wordr2[1]; + w2[2] = wordl2[2] | wordr2[2]; + w2[3] = wordl2[3] | wordr2[3]; + w3[0] = wordl3[0] | wordr3[0]; + w3[1] = wordl3[1] | wordr3[1]; + w3[2] = wordl3[2] | wordr3[2]; + w3[3] = wordl3[3] | wordr3[3]; + + /** + * md5 + */ + + w0_t = w0[0]; + w1_t = w0[1]; + w2_t = w0[2]; + w3_t = w0[3]; + w4_t = w1[0]; + w5_t = w1[1]; + w6_t = w1[2]; + w7_t = w1[3]; + w8_t = w2[0]; + w9_t = w2[1]; + wa_t = w2[2]; + wb_t = w2[3]; + wc_t = w3[0]; + wd_t = w3[1]; + we_t = pw_len * 8; + wf_t = 0; + + a = MD5M_A; + b = MD5M_B; + c = MD5M_C; + d = MD5M_D; + + MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03); + + MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13); + + u32x t; + + MD5_STEP (MD5_H1, a, b, c, d, w5_t, MD5C20, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w8_t, MD5C21, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wb_t, MD5C22, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, we_t, MD5C23, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w1_t, MD5C24, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w4_t, MD5C25, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w7_t, MD5C26, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, wa_t, MD5C27, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, wd_t, MD5C28, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w0_t, MD5C29, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w3_t, MD5C2a, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w6_t, MD5C2b, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w9_t, MD5C2c, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, wc_t, MD5C2d, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wf_t, MD5C2e, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w2_t, MD5C2f, MD5S23); + + MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); + + a += MD5M_A; + b += MD5M_B; + c += MD5M_C; + d += MD5M_D; + + const u32x a1 = hc_swap32 (a); + const u32x b1 = hc_swap32 (b); + const u32x c1 = hc_swap32 (c); + const u32x d1 = hc_swap32 (d); + + /** + * md5 + */ + + w0_t = uint_to_hex_lower8 ((a0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((a0 >> 16) & 255) << 16; + w1_t = uint_to_hex_lower8 ((a0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((a0 >> 0) & 255) << 16; + w2_t = uint_to_hex_lower8 ((b0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((b0 >> 16) & 255) << 16; + w3_t = uint_to_hex_lower8 ((b0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((b0 >> 0) & 255) << 16; + w4_t = uint_to_hex_lower8 ((c0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((c0 >> 16) & 255) << 16; + w5_t = uint_to_hex_lower8 ((c0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((c0 >> 0) & 255) << 16; + w6_t = uint_to_hex_lower8 ((d0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((d0 >> 16) & 255) << 16; + w7_t = uint_to_hex_lower8 ((d0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((d0 >> 0) & 255) << 16; + w8_t = uint_to_hex_lower8 ((e0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((e0 >> 16) & 255) << 16; + w9_t = uint_to_hex_lower8 ((e0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((e0 >> 0) & 255) << 16; + wa_t = 0; + wb_t = 0; + wc_t = 0; + wd_t = 0; + we_t = 0; + wf_t = 0; + + // ctx len 40, pos 40 + + w0[0] = uint_to_hex_lower8 ((a1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((a1 >> 16) & 255) << 16; + w0[1] = uint_to_hex_lower8 ((a1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((a1 >> 0) & 255) << 16; + w0[2] = uint_to_hex_lower8 ((b1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((b1 >> 16) & 255) << 16; + w0[3] = uint_to_hex_lower8 ((b1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((b1 >> 0) & 255) << 16; + w1[0] = uint_to_hex_lower8 ((c1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((c1 >> 16) & 255) << 16; + w1[1] = uint_to_hex_lower8 ((c1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((c1 >> 0) & 255) << 16; + w1[2] = uint_to_hex_lower8 ((d1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((d1 >> 16) & 255) << 16; + w1[3] = uint_to_hex_lower8 ((d1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((d1 >> 0) & 255) << 16; + + w2[0] = 0; + w2[1] = 0; + w2[2] = 0; + w2[3] = 0; + w3[0] = 0; + w3[1] = 0; + w3[2] = 0; + w3[3] = 0; + + u32x _w0[4] = { 0 }; + u32x _w1[4] = { 0 }; + u32x _w2[4] = { 0 }; + u32x _w3[4] = { 0 }; + + switch_buffer_by_offset_carry_le (w0, w1, w2, w3, _w0, _w1, _w2, _w3, 40); + + w0_t |= w0[0]; + w1_t |= w0[1]; + w2_t |= w0[2]; + w3_t |= w0[3]; + w4_t |= w1[0]; + w5_t |= w1[1]; + w6_t |= w1[2]; + w7_t |= w1[3]; + w8_t |= w2[0]; + w9_t |= w2[1]; + wa_t |= w2[2]; + wb_t |= w2[3]; + wc_t |= w3[0]; + wd_t |= w3[1]; + we_t |= w3[2]; + wf_t |= w3[3]; + + // md5 transform + + a = MD5M_A; + b = MD5M_B; + c = MD5M_C; + d = MD5M_D; + + u32x digest[4]; + + digest[0] = a; + digest[1] = b; + digest[2] = c; + digest[3] = d; + + MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03); + + MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13); + + MD5_STEP (MD5_H1, a, b, c, d, w5_t, MD5C20, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w8_t, MD5C21, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wb_t, MD5C22, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, we_t, MD5C23, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w1_t, MD5C24, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w4_t, MD5C25, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w7_t, MD5C26, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, wa_t, MD5C27, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, wd_t, MD5C28, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w0_t, MD5C29, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w3_t, MD5C2a, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w6_t, MD5C2b, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w9_t, MD5C2c, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, wc_t, MD5C2d, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wf_t, MD5C2e, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w2_t, MD5C2f, MD5S23); + + MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); + + digest[0] += a; + digest[1] += b; + digest[2] += c; + digest[3] += d; + + w0[0] = _w0[0]; + w0[1] = _w0[1]; + w0[2] = _w0[2]; + w0[3] = _w0[3]; + w1[0] = _w1[0]; + w1[1] = _w1[1]; + w1[2] = _w1[2]; + w1[3] = _w1[3]; + w2[0] = _w2[0]; + w2[1] = _w2[1]; + w2[2] = _w2[2]; + w2[3] = _w2[3]; + w3[0] = _w3[0]; + w3[1] = _w3[1]; + w3[2] = _w3[2]; + w3[3] = _w3[3]; + + // ctx len 72, pos 8 + + append_0x80_4x4 (w0, w1, w2, w3, 8); + + w3[2] = 72 * 8; + w3[3] = 0; + + // md5 final transform + + a = digest[0]; + b = digest[1]; + c = digest[2]; + d = digest[3]; + + w0_t = w0[0]; + w1_t = w0[1]; + w2_t = w0[2]; + w3_t = w0[3]; + w4_t = w1[0]; + w5_t = w1[1]; + w6_t = w1[2]; + w7_t = w1[3]; + w8_t = w2[0]; + w9_t = w2[1]; + wa_t = w2[2]; + wb_t = w2[3]; + wc_t = w3[0]; + wd_t = w3[1]; + we_t = w3[2]; + wf_t = w3[3]; + + MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03); + + MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13); + + MD5_STEP (MD5_H1, a, b, c, d, w5_t, MD5C20, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w8_t, MD5C21, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wb_t, MD5C22, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, we_t, MD5C23, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w1_t, MD5C24, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w4_t, MD5C25, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w7_t, MD5C26, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, wa_t, MD5C27, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, wd_t, MD5C28, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w0_t, MD5C29, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w3_t, MD5C2a, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w6_t, MD5C2b, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w9_t, MD5C2c, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, wc_t, MD5C2d, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wf_t, MD5C2e, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w2_t, MD5C2f, MD5S23); + + MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30); + + if (MATCHES_NONE_VS ((a+digest[0]-MD5M_A), search[0])) continue; + + MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); + + a += digest[0] - MD5M_A; + b += digest[1] - MD5M_B; + c += digest[2] - MD5M_C; + d += digest[3] - MD5M_D; + + COMPARE_S_SIMD (a, d, c, b); + } +} + +KERNEL_FQ void m21200_s08 (KERN_ATTR_BASIC ()) +{ +} + +KERNEL_FQ void m21200_s16 (KERN_ATTR_BASIC ()) +{ +} diff --git a/OpenCL/m21200_a3-optimized.cl b/OpenCL/m21200_a3-optimized.cl new file mode 100644 index 000000000..84f09e7d0 --- /dev/null +++ b/OpenCL/m21200_a3-optimized.cl @@ -0,0 +1,1629 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +#define NEW_SIMD_CODE + +#ifdef KERNEL_STATIC +#include "inc_vendor.h" +#include "inc_types.h" +#include "inc_platform.cl" +#include "inc_common.cl" +#include "inc_simd.cl" +#include "inc_hash_md5.cl" +#include "inc_hash_sha1.cl" +#endif + +#if VECT_SIZE == 1 +#define uint_to_hex_lower8(i) make_u32x (l_bin2asc[(i)]) +#elif VECT_SIZE == 2 +#define uint_to_hex_lower8(i) make_u32x (l_bin2asc[(i).s0], l_bin2asc[(i).s1]) +#elif VECT_SIZE == 4 +#define uint_to_hex_lower8(i) make_u32x (l_bin2asc[(i).s0], l_bin2asc[(i).s1], l_bin2asc[(i).s2], l_bin2asc[(i).s3]) +#elif VECT_SIZE == 8 +#define uint_to_hex_lower8(i) make_u32x (l_bin2asc[(i).s0], l_bin2asc[(i).s1], l_bin2asc[(i).s2], l_bin2asc[(i).s3], l_bin2asc[(i).s4], l_bin2asc[(i).s5], l_bin2asc[(i).s6], l_bin2asc[(i).s7]) +#elif VECT_SIZE == 16 +#define uint_to_hex_lower8(i) make_u32x (l_bin2asc[(i).s0], l_bin2asc[(i).s1], l_bin2asc[(i).s2], l_bin2asc[(i).s3], l_bin2asc[(i).s4], l_bin2asc[(i).s5], l_bin2asc[(i).s6], l_bin2asc[(i).s7], l_bin2asc[(i).s8], l_bin2asc[(i).s9], l_bin2asc[(i).sa], l_bin2asc[(i).sb], l_bin2asc[(i).sc], l_bin2asc[(i).sd], l_bin2asc[(i).se], l_bin2asc[(i).sf]) +#endif + +DECLSPEC void m21200m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_BASIC (), LOCAL_AS u32 *l_bin2asc) +{ + /** + * modifier + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + + /** + * salt + */ + + u32 salt_buf0[4]; + u32 salt_buf1[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]; + salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4]; + salt_buf1[1] = 0; + salt_buf1[2] = 0; + salt_buf1[3] = 0; + + const u32 salt_len = salt_bufs[salt_pos].salt_len; + + append_0x80_2x4_S (salt_buf0, salt_buf1, salt_len); + + /** + * sha1(salt) + */ + + u32x w0_t = hc_swap32 (salt_buf0[0]); + u32x w1_t = hc_swap32 (salt_buf0[1]); + u32x w2_t = hc_swap32 (salt_buf0[2]); + u32x w3_t = hc_swap32 (salt_buf0[3]); + u32x w4_t = hc_swap32 (salt_buf1[0]); + u32x w5_t = 0; + u32x w6_t = 0; + u32x w7_t = 0; + u32x w8_t = 0; + u32x w9_t = 0; + u32x wa_t = 0; + u32x wb_t = 0; + u32x wc_t = 0; + u32x wd_t = 0; + u32x we_t = 0; + u32x wf_t = 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 = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t); + w3_t = hc_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 = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t); + w7_t = hc_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 = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t); + wb_t = hc_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 = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + const u32x a0 = a; + const u32x b0 = b; + const u32x c0 = c; + const u32x d0 = d; + const u32x e0 = e; + + /** + * loop + */ + + u32 w0l = w0[0]; + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x w0r = ix_create_bft (bfs_buf, il_pos); + + const u32x w0lr = w0l | w0r; + + /** + * md5 + */ + + w0_t = w0lr; + w1_t = w0[1]; + w2_t = w0[2]; + w3_t = w0[3]; + w4_t = w1[0]; + w5_t = w1[1]; + w6_t = w1[2]; + w7_t = w1[3]; + w8_t = w2[0]; + w9_t = w2[1]; + wa_t = w2[2]; + wb_t = w2[3]; + wc_t = w3[0]; + wd_t = w3[1]; + we_t = pw_len * 8; + wf_t = 0; + + a = MD5M_A; + b = MD5M_B; + c = MD5M_C; + d = MD5M_D; + + MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03); + + MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13); + + u32x t; + + MD5_STEP (MD5_H1, a, b, c, d, w5_t, MD5C20, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w8_t, MD5C21, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wb_t, MD5C22, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, we_t, MD5C23, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w1_t, MD5C24, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w4_t, MD5C25, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w7_t, MD5C26, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, wa_t, MD5C27, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, wd_t, MD5C28, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w0_t, MD5C29, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w3_t, MD5C2a, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w6_t, MD5C2b, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w9_t, MD5C2c, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, wc_t, MD5C2d, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wf_t, MD5C2e, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w2_t, MD5C2f, MD5S23); + + MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); + + a += MD5M_A; + b += MD5M_B; + c += MD5M_C; + d += MD5M_D; + + const u32x a1 = hc_swap32 (a); + const u32x b1 = hc_swap32 (b); + const u32x c1 = hc_swap32 (c); + const u32x d1 = hc_swap32 (d); + + /** + * md5 + */ + + w0_t = uint_to_hex_lower8 ((a0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((a0 >> 16) & 255) << 16; + w1_t = uint_to_hex_lower8 ((a0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((a0 >> 0) & 255) << 16; + w2_t = uint_to_hex_lower8 ((b0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((b0 >> 16) & 255) << 16; + w3_t = uint_to_hex_lower8 ((b0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((b0 >> 0) & 255) << 16; + w4_t = uint_to_hex_lower8 ((c0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((c0 >> 16) & 255) << 16; + w5_t = uint_to_hex_lower8 ((c0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((c0 >> 0) & 255) << 16; + w6_t = uint_to_hex_lower8 ((d0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((d0 >> 16) & 255) << 16; + w7_t = uint_to_hex_lower8 ((d0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((d0 >> 0) & 255) << 16; + w8_t = uint_to_hex_lower8 ((e0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((e0 >> 16) & 255) << 16; + w9_t = uint_to_hex_lower8 ((e0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((e0 >> 0) & 255) << 16; + wa_t = 0; + wb_t = 0; + wc_t = 0; + wd_t = 0; + we_t = 0; + wf_t = 0; + + // ctx len 40, pos 40 + + u32x _w0[4] = { 0 }; + u32x _w1[4] = { 0 }; + u32x _w2[4] = { 0 }; + u32x _w3[4] = { 0 }; + + _w0[0] = uint_to_hex_lower8 ((a1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((a1 >> 16) & 255) << 16; + _w0[1] = uint_to_hex_lower8 ((a1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((a1 >> 0) & 255) << 16; + _w0[2] = uint_to_hex_lower8 ((b1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((b1 >> 16) & 255) << 16; + _w0[3] = uint_to_hex_lower8 ((b1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((b1 >> 0) & 255) << 16; + _w1[0] = uint_to_hex_lower8 ((c1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((c1 >> 16) & 255) << 16; + _w1[1] = uint_to_hex_lower8 ((c1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((c1 >> 0) & 255) << 16; + _w1[2] = uint_to_hex_lower8 ((d1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((d1 >> 16) & 255) << 16; + _w1[3] = uint_to_hex_lower8 ((d1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((d1 >> 0) & 255) << 16; + + _w2[0] = 0; + _w2[1] = 0; + _w2[2] = 0; + _w2[3] = 0; + _w3[0] = 0; + _w3[1] = 0; + _w3[2] = 0; + _w3[3] = 0; + + u32x _c0[4] = { 0 }; + u32x _c1[4] = { 0 }; + u32x _c2[4] = { 0 }; + u32x _c3[4] = { 0 }; + + switch_buffer_by_offset_carry_le (_w0, _w1, _w2, _w3, _c0, _c1, _c2, _c3, 40); + + w0_t |= _w0[0]; + w1_t |= _w0[1]; + w2_t |= _w0[2]; + w3_t |= _w0[3]; + w4_t |= _w1[0]; + w5_t |= _w1[1]; + w6_t |= _w1[2]; + w7_t |= _w1[3]; + w8_t |= _w2[0]; + w9_t |= _w2[1]; + wa_t |= _w2[2]; + wb_t |= _w2[3]; + wc_t |= _w3[0]; + wd_t |= _w3[1]; + we_t |= _w3[2]; + wf_t |= _w3[3]; + + // md5 transform + + a = MD5M_A; + b = MD5M_B; + c = MD5M_C; + d = MD5M_D; + + u32x digest[4]; + + digest[0] = a; + digest[1] = b; + digest[2] = c; + digest[3] = d; + + MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03); + + MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13); + + MD5_STEP (MD5_H1, a, b, c, d, w5_t, MD5C20, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w8_t, MD5C21, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wb_t, MD5C22, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, we_t, MD5C23, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w1_t, MD5C24, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w4_t, MD5C25, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w7_t, MD5C26, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, wa_t, MD5C27, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, wd_t, MD5C28, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w0_t, MD5C29, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w3_t, MD5C2a, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w6_t, MD5C2b, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w9_t, MD5C2c, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, wc_t, MD5C2d, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wf_t, MD5C2e, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w2_t, MD5C2f, MD5S23); + + MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); + + digest[0] += a; + digest[1] += b; + digest[2] += c; + digest[3] += d; + + _w0[0] = _c0[0]; + _w0[1] = _c0[1]; + _w0[2] = _c0[2]; + _w0[3] = _c0[3]; + _w1[0] = _c1[0]; + _w1[1] = _c1[1]; + _w1[2] = _c1[2]; + _w1[3] = _c1[3]; + _w2[0] = _c2[0]; + _w2[1] = _c2[1]; + _w2[2] = _c2[2]; + _w2[3] = _c2[3]; + _w3[0] = _c3[0]; + _w3[1] = _c3[1]; + _w3[2] = _c3[2]; + _w3[3] = _c3[3]; + + // ctx len 72, pos 8 + + append_0x80_4x4 (_w0, _w1, _w2, _w3, 8); + + _w3[2] = 72 * 8; + _w3[3] = 0; + + // md5 final transform + + a = digest[0]; + b = digest[1]; + c = digest[2]; + d = digest[3]; + + w0_t = _w0[0]; + w1_t = _w0[1]; + w2_t = _w0[2]; + w3_t = _w0[3]; + w4_t = _w1[0]; + w5_t = _w1[1]; + w6_t = _w1[2]; + w7_t = _w1[3]; + w8_t = _w2[0]; + w9_t = _w2[1]; + wa_t = _w2[2]; + wb_t = _w2[3]; + wc_t = _w3[0]; + wd_t = _w3[1]; + we_t = _w3[2]; + wf_t = _w3[3]; + + MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03); + + MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13); + + MD5_STEP (MD5_H1, a, b, c, d, w5_t, MD5C20, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w8_t, MD5C21, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wb_t, MD5C22, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, we_t, MD5C23, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w1_t, MD5C24, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w4_t, MD5C25, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w7_t, MD5C26, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, wa_t, MD5C27, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, wd_t, MD5C28, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w0_t, MD5C29, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w3_t, MD5C2a, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w6_t, MD5C2b, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w9_t, MD5C2c, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, wc_t, MD5C2d, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wf_t, MD5C2e, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w2_t, MD5C2f, MD5S23); + + MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); + + a += digest[0] - MD5M_A; + b += digest[1] - MD5M_B; + c += digest[2] - MD5M_C; + d += digest[3] - MD5M_D; + + COMPARE_M_SIMD (a, d, c, b); + } +} + +DECLSPEC void m21200s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_BASIC (), LOCAL_AS u32 *l_bin2asc) +{ + /** + * modifier + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + + /** + * salt + */ + + u32 salt_buf0[4]; + u32 salt_buf1[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]; + salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4]; + salt_buf1[1] = 0; + salt_buf1[2] = 0; + salt_buf1[3] = 0; + + const u32 salt_len = salt_bufs[salt_pos].salt_len; + + append_0x80_2x4_S (salt_buf0, salt_buf1, salt_len); + + /** + * sha1(salt) + */ + + u32x w0_t = hc_swap32 (salt_buf0[0]); + u32x w1_t = hc_swap32 (salt_buf0[1]); + u32x w2_t = hc_swap32 (salt_buf0[2]); + u32x w3_t = hc_swap32 (salt_buf0[3]); + u32x w4_t = hc_swap32 (salt_buf1[0]); + u32x w5_t = 0; + u32x w6_t = 0; + u32x w7_t = 0; + u32x w8_t = 0; + u32x w9_t = 0; + u32x wa_t = 0; + u32x wb_t = 0; + u32x wc_t = 0; + u32x wd_t = 0; + u32x we_t = 0; + u32x wf_t = 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 = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t); + w3_t = hc_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 = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t); + w7_t = hc_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 = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t); + wb_t = hc_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 = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + const u32x a0 = a; + const u32x b0 = b; + const u32x c0 = c; + const u32x d0 = d; + const u32x e0 = e; + +if (gid == 0) printf("sha1(salt): %08x %08x %08x %08x %08x\n", a0, b0, c0, d0, e0); + + /** + * 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] + }; + + /** + * loop + */ + + u32 w0l = w0[0]; + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x w0r = ix_create_bft (bfs_buf, il_pos); + + const u32x w0lr = w0l | w0r; + + /** + * md5 + */ + + w0_t = w0lr; + w1_t = w0[1]; + w2_t = w0[2]; + w3_t = w0[3]; + w4_t = w1[0]; + w5_t = w1[1]; + w6_t = w1[2]; + w7_t = w1[3]; + w8_t = w2[0]; + w9_t = w2[1]; + wa_t = w2[2]; + wb_t = w2[3]; + wc_t = w3[0]; + wd_t = w3[1]; + we_t = pw_len * 8; + wf_t = 0; + + a = MD5M_A; + b = MD5M_B; + c = MD5M_C; + d = MD5M_D; + + MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03); + + MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13); + + u32x t; + + MD5_STEP (MD5_H1, a, b, c, d, w5_t, MD5C20, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w8_t, MD5C21, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wb_t, MD5C22, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, we_t, MD5C23, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w1_t, MD5C24, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w4_t, MD5C25, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w7_t, MD5C26, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, wa_t, MD5C27, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, wd_t, MD5C28, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w0_t, MD5C29, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w3_t, MD5C2a, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w6_t, MD5C2b, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w9_t, MD5C2c, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, wc_t, MD5C2d, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wf_t, MD5C2e, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w2_t, MD5C2f, MD5S23); + + MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); + + a += MD5M_A; + b += MD5M_B; + c += MD5M_C; + d += MD5M_D; + + const u32x a1 = hc_swap32 (a); + const u32x b1 = hc_swap32 (b); + const u32x c1 = hc_swap32 (c); + const u32x d1 = hc_swap32 (d); + + /** + * md5 + */ + + w0_t = uint_to_hex_lower8 ((a0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((a0 >> 16) & 255) << 16; + w1_t = uint_to_hex_lower8 ((a0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((a0 >> 0) & 255) << 16; + w2_t = uint_to_hex_lower8 ((b0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((b0 >> 16) & 255) << 16; + w3_t = uint_to_hex_lower8 ((b0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((b0 >> 0) & 255) << 16; + w4_t = uint_to_hex_lower8 ((c0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((c0 >> 16) & 255) << 16; + w5_t = uint_to_hex_lower8 ((c0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((c0 >> 0) & 255) << 16; + w6_t = uint_to_hex_lower8 ((d0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((d0 >> 16) & 255) << 16; + w7_t = uint_to_hex_lower8 ((d0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((d0 >> 0) & 255) << 16; + w8_t = uint_to_hex_lower8 ((e0 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((e0 >> 16) & 255) << 16; + w9_t = uint_to_hex_lower8 ((e0 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((e0 >> 0) & 255) << 16; + wa_t = 0; + wb_t = 0; + wc_t = 0; + wd_t = 0; + we_t = 0; + wf_t = 0; + + // ctx len 40, pos 40 + + u32x _w0[4] = { 0 }; + u32x _w1[4] = { 0 }; + u32x _w2[4] = { 0 }; + u32x _w3[4] = { 0 }; + + _w0[0] = uint_to_hex_lower8 ((a1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((a1 >> 16) & 255) << 16; + _w0[1] = uint_to_hex_lower8 ((a1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((a1 >> 0) & 255) << 16; + _w0[2] = uint_to_hex_lower8 ((b1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((b1 >> 16) & 255) << 16; + _w0[3] = uint_to_hex_lower8 ((b1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((b1 >> 0) & 255) << 16; + _w1[0] = uint_to_hex_lower8 ((c1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((c1 >> 16) & 255) << 16; + _w1[1] = uint_to_hex_lower8 ((c1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((c1 >> 0) & 255) << 16; + _w1[2] = uint_to_hex_lower8 ((d1 >> 24) & 255) << 0 + | uint_to_hex_lower8 ((d1 >> 16) & 255) << 16; + _w1[3] = uint_to_hex_lower8 ((d1 >> 8) & 255) << 0 + | uint_to_hex_lower8 ((d1 >> 0) & 255) << 16; + + _w2[0] = 0; + _w2[1] = 0; + _w2[2] = 0; + _w2[3] = 0; + _w3[0] = 0; + _w3[1] = 0; + _w3[2] = 0; + _w3[3] = 0; + + u32x _c0[4] = { 0 }; + u32x _c1[4] = { 0 }; + u32x _c2[4] = { 0 }; + u32x _c3[4] = { 0 }; + + switch_buffer_by_offset_carry_le (_w0, _w1, _w2, _w3, _c0, _c1, _c2, _c3, 40); + + w0_t |= _w0[0]; + w1_t |= _w0[1]; + w2_t |= _w0[2]; + w3_t |= _w0[3]; + w4_t |= _w1[0]; + w5_t |= _w1[1]; + w6_t |= _w1[2]; + w7_t |= _w1[3]; + w8_t |= _w2[0]; + w9_t |= _w2[1]; + wa_t |= _w2[2]; + wb_t |= _w2[3]; + wc_t |= _w3[0]; + wd_t |= _w3[1]; + we_t |= _w3[2]; + wf_t |= _w3[3]; + + // md5 transform + + a = MD5M_A; + b = MD5M_B; + c = MD5M_C; + d = MD5M_D; + + u32x digest[4]; + + digest[0] = a; + digest[1] = b; + digest[2] = c; + digest[3] = d; + + MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03); + + MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13); + + MD5_STEP (MD5_H1, a, b, c, d, w5_t, MD5C20, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w8_t, MD5C21, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wb_t, MD5C22, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, we_t, MD5C23, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w1_t, MD5C24, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w4_t, MD5C25, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w7_t, MD5C26, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, wa_t, MD5C27, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, wd_t, MD5C28, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w0_t, MD5C29, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w3_t, MD5C2a, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w6_t, MD5C2b, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w9_t, MD5C2c, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, wc_t, MD5C2d, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wf_t, MD5C2e, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w2_t, MD5C2f, MD5S23); + + MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); + + digest[0] += a; + digest[1] += b; + digest[2] += c; + digest[3] += d; + + _w0[0] = _c0[0]; + _w0[1] = _c0[1]; + _w0[2] = _c0[2]; + _w0[3] = _c0[3]; + _w1[0] = _c1[0]; + _w1[1] = _c1[1]; + _w1[2] = _c1[2]; + _w1[3] = _c1[3]; + _w2[0] = _c2[0]; + _w2[1] = _c2[1]; + _w2[2] = _c2[2]; + _w2[3] = _c2[3]; + _w3[0] = _c3[0]; + _w3[1] = _c3[1]; + _w3[2] = _c3[2]; + _w3[3] = _c3[3]; + + // ctx len 72, pos 8 + + append_0x80_4x4 (_w0, _w1, _w2, _w3, 8); + + _w3[2] = 72 * 8; + _w3[3] = 0; + + // md5 final transform + + a = digest[0]; + b = digest[1]; + c = digest[2]; + d = digest[3]; + + w0_t = _w0[0]; + w1_t = _w0[1]; + w2_t = _w0[2]; + w3_t = _w0[3]; + w4_t = _w1[0]; + w5_t = _w1[1]; + w6_t = _w1[2]; + w7_t = _w1[3]; + w8_t = _w2[0]; + w9_t = _w2[1]; + wa_t = _w2[2]; + wb_t = _w2[3]; + wc_t = _w3[0]; + wd_t = _w3[1]; + we_t = _w3[2]; + wf_t = _w3[3]; + + MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03); + MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00); + MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01); + MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02); + MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03); + + MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13); + MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10); + MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11); + MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12); + MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13); + + MD5_STEP (MD5_H1, a, b, c, d, w5_t, MD5C20, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w8_t, MD5C21, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wb_t, MD5C22, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, we_t, MD5C23, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w1_t, MD5C24, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w4_t, MD5C25, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w7_t, MD5C26, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, wa_t, MD5C27, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, wd_t, MD5C28, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, w0_t, MD5C29, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, w3_t, MD5C2a, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w6_t, MD5C2b, MD5S23); + MD5_STEP (MD5_H1, a, b, c, d, w9_t, MD5C2c, MD5S20); + MD5_STEP (MD5_H2, d, a, b, c, wc_t, MD5C2d, MD5S21); + MD5_STEP (MD5_H1, c, d, a, b, wf_t, MD5C2e, MD5S22); + MD5_STEP (MD5_H2, b, c, d, a, w2_t, MD5C2f, MD5S23); + + MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30); + MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33); + MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30); + + if (MATCHES_NONE_VS ((a+digest[0]-MD5M_A), search[0])) continue; + + MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31); + MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); + MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); + + a += digest[0] - MD5M_A; + b += digest[1] - MD5M_B; + c += digest[2] - MD5M_C; + d += digest[3] - MD5M_D; + + COMPARE_S_SIMD (a, d, c, b); + } +} + +KERNEL_FQ void m21200_m04 (KERN_ATTR_BASIC ()) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * bin2asc table + */ + + LOCAL_VK u32 l_bin2asc[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + const u32 i0 = (i >> 0) & 15; + const u32 i1 = (i >> 4) & 15; + + l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 8 + | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 0; + } + + SYNC_THREADS (); + + if (gid >= gid_max) return; + + /** + * modifier + */ + + u32 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]; + + u32 w1[4]; + + w1[0] = 0; + w1[1] = 0; + w1[2] = 0; + w1[3] = 0; + + u32 w2[4]; + + w2[0] = 0; + w2[1] = 0; + w2[2] = 0; + w2[3] = 0; + + u32 w3[4]; + + w3[0] = 0; + w3[1] = 0; + w3[2] = 0; + w3[3] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m21200m (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_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max, l_bin2asc); +} + +KERNEL_FQ void m21200_m08 (KERN_ATTR_BASIC ()) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * bin2asc table + */ + + LOCAL_VK u32 l_bin2asc[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + const u32 i0 = (i >> 0) & 15; + const u32 i1 = (i >> 4) & 15; + + l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 8 + | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 0; + } + + SYNC_THREADS (); + + if (gid >= gid_max) return; + + /** + * modifier + */ + + u32 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]; + + u32 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]; + + u32 w2[4]; + + w2[0] = 0; + w2[1] = 0; + w2[2] = 0; + w2[3] = 0; + + u32 w3[4]; + + w3[0] = 0; + w3[1] = 0; + w3[2] = 0; + w3[3] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m21200m (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_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max, l_bin2asc); +} + +KERNEL_FQ void m21200_m16 (KERN_ATTR_BASIC ()) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * bin2asc table + */ + + LOCAL_VK u32 l_bin2asc[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + const u32 i0 = (i >> 0) & 15; + const u32 i1 = (i >> 4) & 15; + + l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 8 + | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 0; + } + + SYNC_THREADS (); + + if (gid >= gid_max) return; + + /** + * modifier + */ + + u32 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]; + + u32 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]; + + u32 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]; + + u32 w3[4]; + + w3[0] = pws[gid].i[12]; + w3[1] = pws[gid].i[13]; + w3[2] = pws[gid].i[14]; + w3[3] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m21200m (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_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max, l_bin2asc); +} + +KERNEL_FQ void m21200_s04 (KERN_ATTR_BASIC ()) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * bin2asc table + */ + + LOCAL_VK u32 l_bin2asc[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + const u32 i0 = (i >> 0) & 15; + const u32 i1 = (i >> 4) & 15; + + l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 8 + | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 0; + } + + SYNC_THREADS (); + + if (gid >= gid_max) return; + + /** + * modifier + */ + + u32 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]; + + u32 w1[4]; + + w1[0] = 0; + w1[1] = 0; + w1[2] = 0; + w1[3] = 0; + + u32 w2[4]; + + w2[0] = 0; + w2[1] = 0; + w2[2] = 0; + w2[3] = 0; + + u32 w3[4]; + + w3[0] = 0; + w3[1] = 0; + w3[2] = 0; + w3[3] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m21200s (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_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max, l_bin2asc); +} + +KERNEL_FQ void m21200_s08 (KERN_ATTR_BASIC ()) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * bin2asc table + */ + + LOCAL_VK u32 l_bin2asc[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + const u32 i0 = (i >> 0) & 15; + const u32 i1 = (i >> 4) & 15; + + l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 8 + | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 0; + } + + SYNC_THREADS (); + + if (gid >= gid_max) return; + + /** + * modifier + */ + + u32 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]; + + u32 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]; + + u32 w2[4]; + + w2[0] = 0; + w2[1] = 0; + w2[2] = 0; + w2[3] = 0; + + u32 w3[4]; + + w3[0] = 0; + w3[1] = 0; + w3[2] = 0; + w3[3] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m21200s (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_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max, l_bin2asc); +} + +KERNEL_FQ void m21200_s16 (KERN_ATTR_BASIC ()) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * bin2asc table + */ + + LOCAL_VK u32 l_bin2asc[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + const u32 i0 = (i >> 0) & 15; + const u32 i1 = (i >> 4) & 15; + + l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 8 + | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 0; + } + + SYNC_THREADS (); + + if (gid >= gid_max) return; + + /** + * modifier + */ + + u32 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]; + + u32 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]; + + u32 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]; + + u32 w3[4]; + + w3[0] = pws[gid].i[12]; + w3[1] = pws[gid].i[13]; + w3[2] = pws[gid].i[14]; + w3[3] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m21200s (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_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max, l_bin2asc); +}