From 4c7f61e473933804e85f84085908df6299d8cecd Mon Sep 17 00:00:00 2001 From: jsteube Date: Wed, 9 Aug 2017 14:26:39 +0200 Subject: [PATCH] Add pure kernels for sha1(CX) --- OpenCL/m14400_a0.cl | 548 +++++++++++++++++++++++++++++++++++++++++ OpenCL/m14400_a1.cl | 528 ++++++++++++++++++++++++++++++++++++++++ OpenCL/m14400_a3.cl | 582 ++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 1658 insertions(+) create mode 100644 OpenCL/m14400_a0.cl create mode 100644 OpenCL/m14400_a1.cl create mode 100644 OpenCL/m14400_a3.cl diff --git a/OpenCL/m14400_a0.cl b/OpenCL/m14400_a0.cl new file mode 100644 index 000000000..7188d535c --- /dev/null +++ b/OpenCL/m14400_a0.cl @@ -0,0 +1,548 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +//#define NEW_SIMD_CODE + +#include "inc_vendor.cl" +#include "inc_hash_constants.h" +#include "inc_hash_functions.cl" +#include "inc_types.cl" +#include "inc_common.cl" +#include "inc_rp.h" +#include "inc_rp.cl" +#include "inc_scalar.cl" +#include "inc_hash_sha1.cl" + +#if VECT_SIZE == 1 +#define uint_to_hex_lower8_le(i) (u32x) (l_bin2asc[(i)]) +#elif VECT_SIZE == 2 +#define uint_to_hex_lower8_le(i) (u32x) (l_bin2asc[(i).s0], l_bin2asc[(i).s1]) +#elif VECT_SIZE == 4 +#define uint_to_hex_lower8_le(i) (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_le(i) (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_le(i) (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 void m14400_mxx (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) +{ + /** + * modifier + */ + + const u32 gid = get_global_id (0); + const u32 lid = get_local_id (0); + const u32 lsz = get_local_size (0); + + /** + * bin2asc table + */ + + __local 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) << 0 + | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 8; + } + + barrier (CLK_LOCAL_MEM_FENCE); + + if (gid >= gid_max) return; + + /** + * base + */ + + const u32 pw_len = pws[gid].pw_len; + + const u32 pw_lenv = ceil ((float) pw_len / 4); + + u32 w[64] = { 0 }; + + for (int idx = 0; idx < pw_lenv; idx++) + { + w[idx] = pws[gid].i[idx]; + } + + sha1_ctx_t ctx0; + + sha1_init (&ctx0); + + u32 d20[4]; + u32 d21[4]; + u32 d22[4]; + u32 d23[4]; + + d20[0] = 0x2d2d0000; + d20[1] = 0; + d20[2] = 0; + d20[3] = 0; + d21[0] = 0; + d21[1] = 0; + d21[2] = 0; + d21[3] = 0; + d22[0] = 0; + d22[1] = 0; + d22[2] = 0; + d22[3] = 0; + d23[0] = 0; + d23[1] = 0; + d23[2] = 0; + d23[3] = 0; + + sha1_update_64 (&ctx0, d20, d21, d22, d23, 2); + + sha1_update_global_swap (&ctx0, salt_bufs[salt_pos].salt_buf, salt_bufs[salt_pos].salt_len); + + u32 d40[4]; + u32 d41[4]; + u32 d42[4]; + u32 d43[4]; + + d40[0] = 0x2d2d2d2d; + d40[1] = 0; + d40[2] = 0; + d40[3] = 0; + d41[0] = 0; + d41[1] = 0; + d41[2] = 0; + d41[3] = 0; + d42[0] = 0; + d42[1] = 0; + d42[2] = 0; + d42[3] = 0; + d43[0] = 0; + d43[1] = 0; + d43[2] = 0; + d43[3] = 0; + + sha1_update_64 (&ctx0, d20, d21, d22, d23, 2); + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos++) + { + // todo: add rules engine + + sha1_ctx_t ctx1 = ctx0; + + d20[0] = 0x2d2d0000; + d20[1] = 0; + d20[2] = 0; + d20[3] = 0; + d21[0] = 0; + d21[1] = 0; + d21[2] = 0; + d21[3] = 0; + d22[0] = 0; + d22[1] = 0; + d22[2] = 0; + d22[3] = 0; + d23[0] = 0; + d23[1] = 0; + d23[2] = 0; + d23[3] = 0; + + sha1_update_64 (&ctx1, d20, d21, d22, d23, 2); + + sha1_update_swap (&ctx1, w, pw_len); + + d40[0] = 0x2d2d2d2d; + d40[1] = 0; + d40[2] = 0; + d40[3] = 0; + d41[0] = 0; + d41[1] = 0; + d41[2] = 0; + d41[3] = 0; + d42[0] = 0; + d42[1] = 0; + d42[2] = 0; + d42[3] = 0; + d43[0] = 0; + d43[1] = 0; + d43[2] = 0; + d43[3] = 0; + + sha1_update_64 (&ctx1, d40, d41, d42, d43, 4); + + sha1_final (&ctx1); + + u32 a = ctx1.h[0]; + u32 b = ctx1.h[1]; + u32 c = ctx1.h[2]; + u32 d = ctx1.h[3]; + u32 e = ctx1.h[4]; + + sha1_ctx_t ctx; + + for (int i = 1; i < 10; i++) + { + ctx = ctx0; + + u32 w0[4]; + u32 w1[4]; + u32 w2[4]; + u32 w3[4]; + + w0[0] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((a >> 24) & 255) << 16; + w0[1] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((a >> 8) & 255) << 16; + w0[2] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((b >> 24) & 255) << 16; + w0[3] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((b >> 8) & 255) << 16; + w1[0] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((c >> 24) & 255) << 16; + w1[1] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((c >> 8) & 255) << 16; + w1[2] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((d >> 24) & 255) << 16; + w1[3] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((d >> 8) & 255) << 16; + w2[0] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((e >> 24) & 255) << 16; + w2[1] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((e >> 8) & 255) << 16; + w2[2] = 0; + w2[3] = 0; + w3[0] = 0; + w3[1] = 0; + w3[2] = 0; + w3[3] = 0; + + sha1_update_64 (&ctx, w0, w1, w2, w3, 40); + + d20[0] = 0x2d2d0000; + d20[1] = 0; + d20[2] = 0; + d20[3] = 0; + d21[0] = 0; + d21[1] = 0; + d21[2] = 0; + d21[3] = 0; + d22[0] = 0; + d22[1] = 0; + d22[2] = 0; + d22[3] = 0; + d23[0] = 0; + d23[1] = 0; + d23[2] = 0; + d23[3] = 0; + + sha1_update_64 (&ctx, d20, d21, d22, d23, 2); + + sha1_update_swap (&ctx, w, pw_len); + + d40[0] = 0x2d2d2d2d; + d40[1] = 0; + d40[2] = 0; + d40[3] = 0; + d41[0] = 0; + d41[1] = 0; + d41[2] = 0; + d41[3] = 0; + d42[0] = 0; + d42[1] = 0; + d42[2] = 0; + d42[3] = 0; + d43[0] = 0; + d43[1] = 0; + d43[2] = 0; + d43[3] = 0; + + sha1_update_64 (&ctx, d40, d41, d42, d43, 4); + + sha1_final (&ctx); + + a = ctx.h[0]; + b = ctx.h[1]; + c = ctx.h[2]; + d = ctx.h[3]; + e = ctx.h[4]; + } + + const u32 r0 = ctx.h[DGST_R0]; + const u32 r1 = ctx.h[DGST_R1]; + const u32 r2 = ctx.h[DGST_R2]; + const u32 r3 = ctx.h[DGST_R3]; + + COMPARE_M_SCALAR (r0, r1, r2, r3); + } +} + +__kernel void m14400_sxx (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) +{ + /** + * modifier + */ + + const u32 gid = get_global_id (0); + const u32 lid = get_local_id (0); + const u32 lsz = get_local_size (0); + + /** + * bin2asc table + */ + + __local 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) << 0 + | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 8; + } + + barrier (CLK_LOCAL_MEM_FENCE); + + if (gid >= gid_max) return; + + /** + * 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] + }; + + /** + * base + */ + + const u32 pw_len = pws[gid].pw_len; + + const u32 pw_lenv = ceil ((float) pw_len / 4); + + u32 w[64] = { 0 }; + + for (int idx = 0; idx < pw_lenv; idx++) + { + w[idx] = pws[gid].i[idx]; + } + + sha1_ctx_t ctx0; + + sha1_init (&ctx0); + + u32 d20[4]; + u32 d21[4]; + u32 d22[4]; + u32 d23[4]; + + d20[0] = 0x2d2d0000; + d20[1] = 0; + d20[2] = 0; + d20[3] = 0; + d21[0] = 0; + d21[1] = 0; + d21[2] = 0; + d21[3] = 0; + d22[0] = 0; + d22[1] = 0; + d22[2] = 0; + d22[3] = 0; + d23[0] = 0; + d23[1] = 0; + d23[2] = 0; + d23[3] = 0; + + sha1_update_64 (&ctx0, d20, d21, d22, d23, 2); + + sha1_update_global_swap (&ctx0, salt_bufs[salt_pos].salt_buf, salt_bufs[salt_pos].salt_len); + + u32 d40[4]; + u32 d41[4]; + u32 d42[4]; + u32 d43[4]; + + d40[0] = 0x2d2d2d2d; + d40[1] = 0; + d40[2] = 0; + d40[3] = 0; + d41[0] = 0; + d41[1] = 0; + d41[2] = 0; + d41[3] = 0; + d42[0] = 0; + d42[1] = 0; + d42[2] = 0; + d42[3] = 0; + d43[0] = 0; + d43[1] = 0; + d43[2] = 0; + d43[3] = 0; + + sha1_update_64 (&ctx0, d20, d21, d22, d23, 2); + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos++) + { + // todo: add rules engine + + sha1_ctx_t ctx1 = ctx0; + + d20[0] = 0x2d2d0000; + d20[1] = 0; + d20[2] = 0; + d20[3] = 0; + d21[0] = 0; + d21[1] = 0; + d21[2] = 0; + d21[3] = 0; + d22[0] = 0; + d22[1] = 0; + d22[2] = 0; + d22[3] = 0; + d23[0] = 0; + d23[1] = 0; + d23[2] = 0; + d23[3] = 0; + + sha1_update_64 (&ctx1, d20, d21, d22, d23, 2); + + sha1_update_swap (&ctx1, w, pw_len); + + d40[0] = 0x2d2d2d2d; + d40[1] = 0; + d40[2] = 0; + d40[3] = 0; + d41[0] = 0; + d41[1] = 0; + d41[2] = 0; + d41[3] = 0; + d42[0] = 0; + d42[1] = 0; + d42[2] = 0; + d42[3] = 0; + d43[0] = 0; + d43[1] = 0; + d43[2] = 0; + d43[3] = 0; + + sha1_update_64 (&ctx1, d40, d41, d42, d43, 4); + + sha1_final (&ctx1); + + u32 a = ctx1.h[0]; + u32 b = ctx1.h[1]; + u32 c = ctx1.h[2]; + u32 d = ctx1.h[3]; + u32 e = ctx1.h[4]; + + sha1_ctx_t ctx; + + for (int i = 1; i < 10; i++) + { + ctx = ctx0; + + u32 w0[4]; + u32 w1[4]; + u32 w2[4]; + u32 w3[4]; + + w0[0] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((a >> 24) & 255) << 16; + w0[1] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((a >> 8) & 255) << 16; + w0[2] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((b >> 24) & 255) << 16; + w0[3] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((b >> 8) & 255) << 16; + w1[0] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((c >> 24) & 255) << 16; + w1[1] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((c >> 8) & 255) << 16; + w1[2] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((d >> 24) & 255) << 16; + w1[3] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((d >> 8) & 255) << 16; + w2[0] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((e >> 24) & 255) << 16; + w2[1] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((e >> 8) & 255) << 16; + w2[2] = 0; + w2[3] = 0; + w3[0] = 0; + w3[1] = 0; + w3[2] = 0; + w3[3] = 0; + + sha1_update_64 (&ctx, w0, w1, w2, w3, 40); + + d20[0] = 0x2d2d0000; + d20[1] = 0; + d20[2] = 0; + d20[3] = 0; + d21[0] = 0; + d21[1] = 0; + d21[2] = 0; + d21[3] = 0; + d22[0] = 0; + d22[1] = 0; + d22[2] = 0; + d22[3] = 0; + d23[0] = 0; + d23[1] = 0; + d23[2] = 0; + d23[3] = 0; + + sha1_update_64 (&ctx, d20, d21, d22, d23, 2); + + sha1_update_swap (&ctx, w, pw_len); + + d40[0] = 0x2d2d2d2d; + d40[1] = 0; + d40[2] = 0; + d40[3] = 0; + d41[0] = 0; + d41[1] = 0; + d41[2] = 0; + d41[3] = 0; + d42[0] = 0; + d42[1] = 0; + d42[2] = 0; + d42[3] = 0; + d43[0] = 0; + d43[1] = 0; + d43[2] = 0; + d43[3] = 0; + + sha1_update_64 (&ctx, d40, d41, d42, d43, 4); + + sha1_final (&ctx); + + a = ctx.h[0]; + b = ctx.h[1]; + c = ctx.h[2]; + d = ctx.h[3]; + e = ctx.h[4]; + } + + const u32 r0 = ctx.h[DGST_R0]; + const u32 r1 = ctx.h[DGST_R1]; + const u32 r2 = ctx.h[DGST_R2]; + const u32 r3 = ctx.h[DGST_R3]; + + COMPARE_S_SCALAR (r0, r1, r2, r3); + } +} diff --git a/OpenCL/m14400_a1.cl b/OpenCL/m14400_a1.cl new file mode 100644 index 000000000..5a62f87f9 --- /dev/null +++ b/OpenCL/m14400_a1.cl @@ -0,0 +1,528 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +//#define NEW_SIMD_CODE + +#include "inc_vendor.cl" +#include "inc_hash_constants.h" +#include "inc_hash_functions.cl" +#include "inc_types.cl" +#include "inc_common.cl" +#include "inc_scalar.cl" +#include "inc_hash_sha1.cl" + +#if VECT_SIZE == 1 +#define uint_to_hex_lower8_le(i) (u32x) (l_bin2asc[(i)]) +#elif VECT_SIZE == 2 +#define uint_to_hex_lower8_le(i) (u32x) (l_bin2asc[(i).s0], l_bin2asc[(i).s1]) +#elif VECT_SIZE == 4 +#define uint_to_hex_lower8_le(i) (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_le(i) (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_le(i) (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 void m14400_mxx (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) +{ + /** + * modifier + */ + + const u32 gid = get_global_id (0); + const u32 lid = get_local_id (0); + const u32 lsz = get_local_size (0); + + /** + * bin2asc table + */ + + __local 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) << 0 + | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 8; + } + + barrier (CLK_LOCAL_MEM_FENCE); + + if (gid >= gid_max) return; + + /** + * base + */ + + sha1_ctx_t ctx0; + + sha1_init (&ctx0); + + u32 d20[4]; + u32 d21[4]; + u32 d22[4]; + u32 d23[4]; + + d20[0] = 0x2d2d0000; + d20[1] = 0; + d20[2] = 0; + d20[3] = 0; + d21[0] = 0; + d21[1] = 0; + d21[2] = 0; + d21[3] = 0; + d22[0] = 0; + d22[1] = 0; + d22[2] = 0; + d22[3] = 0; + d23[0] = 0; + d23[1] = 0; + d23[2] = 0; + d23[3] = 0; + + sha1_update_64 (&ctx0, d20, d21, d22, d23, 2); + + sha1_update_global_swap (&ctx0, salt_bufs[salt_pos].salt_buf, salt_bufs[salt_pos].salt_len); + + u32 d40[4]; + u32 d41[4]; + u32 d42[4]; + u32 d43[4]; + + d40[0] = 0x2d2d2d2d; + d40[1] = 0; + d40[2] = 0; + d40[3] = 0; + d41[0] = 0; + d41[1] = 0; + d41[2] = 0; + d41[3] = 0; + d42[0] = 0; + d42[1] = 0; + d42[2] = 0; + d42[3] = 0; + d43[0] = 0; + d43[1] = 0; + d43[2] = 0; + d43[3] = 0; + + sha1_update_64 (&ctx0, d20, d21, d22, d23, 2); + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos++) + { + sha1_ctx_t ctx1 = ctx0; + + d20[0] = 0x2d2d0000; + d20[1] = 0; + d20[2] = 0; + d20[3] = 0; + d21[0] = 0; + d21[1] = 0; + d21[2] = 0; + d21[3] = 0; + d22[0] = 0; + d22[1] = 0; + d22[2] = 0; + d22[3] = 0; + d23[0] = 0; + d23[1] = 0; + d23[2] = 0; + d23[3] = 0; + + sha1_update_64 (&ctx1, d20, d21, d22, d23, 2); + + sha1_update_global_swap (&ctx1, pws[gid].i, pws[gid].pw_len); + + sha1_update_global_swap (&ctx1, combs_buf[il_pos].i, combs_buf[il_pos].pw_len); + + d40[0] = 0x2d2d2d2d; + d40[1] = 0; + d40[2] = 0; + d40[3] = 0; + d41[0] = 0; + d41[1] = 0; + d41[2] = 0; + d41[3] = 0; + d42[0] = 0; + d42[1] = 0; + d42[2] = 0; + d42[3] = 0; + d43[0] = 0; + d43[1] = 0; + d43[2] = 0; + d43[3] = 0; + + sha1_update_64 (&ctx1, d40, d41, d42, d43, 4); + + sha1_final (&ctx1); + + u32 a = ctx1.h[0]; + u32 b = ctx1.h[1]; + u32 c = ctx1.h[2]; + u32 d = ctx1.h[3]; + u32 e = ctx1.h[4]; + + sha1_ctx_t ctx; + + for (int i = 1; i < 10; i++) + { + ctx = ctx0; + + u32 w0[4]; + u32 w1[4]; + u32 w2[4]; + u32 w3[4]; + + w0[0] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((a >> 24) & 255) << 16; + w0[1] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((a >> 8) & 255) << 16; + w0[2] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((b >> 24) & 255) << 16; + w0[3] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((b >> 8) & 255) << 16; + w1[0] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((c >> 24) & 255) << 16; + w1[1] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((c >> 8) & 255) << 16; + w1[2] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((d >> 24) & 255) << 16; + w1[3] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((d >> 8) & 255) << 16; + w2[0] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((e >> 24) & 255) << 16; + w2[1] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((e >> 8) & 255) << 16; + w2[2] = 0; + w2[3] = 0; + w3[0] = 0; + w3[1] = 0; + w3[2] = 0; + w3[3] = 0; + + sha1_update_64 (&ctx, w0, w1, w2, w3, 40); + + d20[0] = 0x2d2d0000; + d20[1] = 0; + d20[2] = 0; + d20[3] = 0; + d21[0] = 0; + d21[1] = 0; + d21[2] = 0; + d21[3] = 0; + d22[0] = 0; + d22[1] = 0; + d22[2] = 0; + d22[3] = 0; + d23[0] = 0; + d23[1] = 0; + d23[2] = 0; + d23[3] = 0; + + sha1_update_64 (&ctx, d20, d21, d22, d23, 2); + + sha1_update_global_swap (&ctx, pws[gid].i, pws[gid].pw_len); + + sha1_update_global_swap (&ctx, combs_buf[il_pos].i, combs_buf[il_pos].pw_len); + + d40[0] = 0x2d2d2d2d; + d40[1] = 0; + d40[2] = 0; + d40[3] = 0; + d41[0] = 0; + d41[1] = 0; + d41[2] = 0; + d41[3] = 0; + d42[0] = 0; + d42[1] = 0; + d42[2] = 0; + d42[3] = 0; + d43[0] = 0; + d43[1] = 0; + d43[2] = 0; + d43[3] = 0; + + sha1_update_64 (&ctx, d40, d41, d42, d43, 4); + + sha1_final (&ctx); + + a = ctx.h[0]; + b = ctx.h[1]; + c = ctx.h[2]; + d = ctx.h[3]; + e = ctx.h[4]; + } + + const u32 r0 = ctx.h[DGST_R0]; + const u32 r1 = ctx.h[DGST_R1]; + const u32 r2 = ctx.h[DGST_R2]; + const u32 r3 = ctx.h[DGST_R3]; + + COMPARE_M_SCALAR (r0, r1, r2, r3); + } +} + +__kernel void m14400_sxx (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) +{ + /** + * modifier + */ + + const u32 gid = get_global_id (0); + const u32 lid = get_local_id (0); + const u32 lsz = get_local_size (0); + + /** + * bin2asc table + */ + + __local 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) << 0 + | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 8; + } + + barrier (CLK_LOCAL_MEM_FENCE); + + if (gid >= gid_max) return; + + /** + * 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] + }; + + /** + * base + */ + + sha1_ctx_t ctx0; + + sha1_init (&ctx0); + + u32 d20[4]; + u32 d21[4]; + u32 d22[4]; + u32 d23[4]; + + d20[0] = 0x2d2d0000; + d20[1] = 0; + d20[2] = 0; + d20[3] = 0; + d21[0] = 0; + d21[1] = 0; + d21[2] = 0; + d21[3] = 0; + d22[0] = 0; + d22[1] = 0; + d22[2] = 0; + d22[3] = 0; + d23[0] = 0; + d23[1] = 0; + d23[2] = 0; + d23[3] = 0; + + sha1_update_64 (&ctx0, d20, d21, d22, d23, 2); + + sha1_update_global_swap (&ctx0, salt_bufs[salt_pos].salt_buf, salt_bufs[salt_pos].salt_len); + + u32 d40[4]; + u32 d41[4]; + u32 d42[4]; + u32 d43[4]; + + d40[0] = 0x2d2d2d2d; + d40[1] = 0; + d40[2] = 0; + d40[3] = 0; + d41[0] = 0; + d41[1] = 0; + d41[2] = 0; + d41[3] = 0; + d42[0] = 0; + d42[1] = 0; + d42[2] = 0; + d42[3] = 0; + d43[0] = 0; + d43[1] = 0; + d43[2] = 0; + d43[3] = 0; + + sha1_update_64 (&ctx0, d20, d21, d22, d23, 2); + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos++) + { + sha1_ctx_t ctx1 = ctx0; + + d20[0] = 0x2d2d0000; + d20[1] = 0; + d20[2] = 0; + d20[3] = 0; + d21[0] = 0; + d21[1] = 0; + d21[2] = 0; + d21[3] = 0; + d22[0] = 0; + d22[1] = 0; + d22[2] = 0; + d22[3] = 0; + d23[0] = 0; + d23[1] = 0; + d23[2] = 0; + d23[3] = 0; + + sha1_update_64 (&ctx1, d20, d21, d22, d23, 2); + + sha1_update_global_swap (&ctx1, pws[gid].i, pws[gid].pw_len); + + sha1_update_global_swap (&ctx1, combs_buf[il_pos].i, combs_buf[il_pos].pw_len); + + d40[0] = 0x2d2d2d2d; + d40[1] = 0; + d40[2] = 0; + d40[3] = 0; + d41[0] = 0; + d41[1] = 0; + d41[2] = 0; + d41[3] = 0; + d42[0] = 0; + d42[1] = 0; + d42[2] = 0; + d42[3] = 0; + d43[0] = 0; + d43[1] = 0; + d43[2] = 0; + d43[3] = 0; + + sha1_update_64 (&ctx1, d40, d41, d42, d43, 4); + + sha1_final (&ctx1); + + u32 a = ctx1.h[0]; + u32 b = ctx1.h[1]; + u32 c = ctx1.h[2]; + u32 d = ctx1.h[3]; + u32 e = ctx1.h[4]; + + sha1_ctx_t ctx; + + for (int i = 1; i < 10; i++) + { + ctx = ctx0; + + u32 w0[4]; + u32 w1[4]; + u32 w2[4]; + u32 w3[4]; + + w0[0] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((a >> 24) & 255) << 16; + w0[1] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((a >> 8) & 255) << 16; + w0[2] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((b >> 24) & 255) << 16; + w0[3] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((b >> 8) & 255) << 16; + w1[0] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((c >> 24) & 255) << 16; + w1[1] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((c >> 8) & 255) << 16; + w1[2] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((d >> 24) & 255) << 16; + w1[3] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((d >> 8) & 255) << 16; + w2[0] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((e >> 24) & 255) << 16; + w2[1] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((e >> 8) & 255) << 16; + w2[2] = 0; + w2[3] = 0; + w3[0] = 0; + w3[1] = 0; + w3[2] = 0; + w3[3] = 0; + + sha1_update_64 (&ctx, w0, w1, w2, w3, 40); + + d20[0] = 0x2d2d0000; + d20[1] = 0; + d20[2] = 0; + d20[3] = 0; + d21[0] = 0; + d21[1] = 0; + d21[2] = 0; + d21[3] = 0; + d22[0] = 0; + d22[1] = 0; + d22[2] = 0; + d22[3] = 0; + d23[0] = 0; + d23[1] = 0; + d23[2] = 0; + d23[3] = 0; + + sha1_update_64 (&ctx, d20, d21, d22, d23, 2); + + sha1_update_global_swap (&ctx, pws[gid].i, pws[gid].pw_len); + + sha1_update_global_swap (&ctx, combs_buf[il_pos].i, combs_buf[il_pos].pw_len); + + d40[0] = 0x2d2d2d2d; + d40[1] = 0; + d40[2] = 0; + d40[3] = 0; + d41[0] = 0; + d41[1] = 0; + d41[2] = 0; + d41[3] = 0; + d42[0] = 0; + d42[1] = 0; + d42[2] = 0; + d42[3] = 0; + d43[0] = 0; + d43[1] = 0; + d43[2] = 0; + d43[3] = 0; + + sha1_update_64 (&ctx, d40, d41, d42, d43, 4); + + sha1_final (&ctx); + + a = ctx.h[0]; + b = ctx.h[1]; + c = ctx.h[2]; + d = ctx.h[3]; + e = ctx.h[4]; + } + + const u32 r0 = ctx.h[DGST_R0]; + const u32 r1 = ctx.h[DGST_R1]; + const u32 r2 = ctx.h[DGST_R2]; + const u32 r3 = ctx.h[DGST_R3]; + + COMPARE_S_SCALAR (r0, r1, r2, r3); + } +} diff --git a/OpenCL/m14400_a3.cl b/OpenCL/m14400_a3.cl new file mode 100644 index 000000000..b6f30b739 --- /dev/null +++ b/OpenCL/m14400_a3.cl @@ -0,0 +1,582 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +#define NEW_SIMD_CODE + +#include "inc_vendor.cl" +#include "inc_hash_constants.h" +#include "inc_hash_functions.cl" +#include "inc_types.cl" +#include "inc_common.cl" +#include "inc_simd.cl" +#include "inc_hash_sha1.cl" + +#if VECT_SIZE == 1 +#define uint_to_hex_lower8_le(i) (u32x) (l_bin2asc[(i)]) +#elif VECT_SIZE == 2 +#define uint_to_hex_lower8_le(i) (u32x) (l_bin2asc[(i).s0], l_bin2asc[(i).s1]) +#elif VECT_SIZE == 4 +#define uint_to_hex_lower8_le(i) (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_le(i) (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_le(i) (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 void m14400_mxx (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __constant const u32x *words_buf_r, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) +{ + /** + * modifier + */ + + const u32 gid = get_global_id (0); + const u32 lid = get_local_id (0); + const u32 lsz = get_local_size (0); + + /** + * bin2asc table + */ + + __local 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) << 0 + | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 8; + } + + barrier (CLK_LOCAL_MEM_FENCE); + + if (gid >= gid_max) return; + + /** + * base + */ + + const u32 pw_len = pws[gid].pw_len; + + const u32 pw_lenv = ceil ((float) pw_len / 4); + + u32x w[64] = { 0 }; + + for (int idx = 0; idx < pw_lenv; idx++) + { + w[idx] = pws[gid].i[idx]; + } + + sha1_ctx_t ctx0; + + sha1_init (&ctx0); + + u32 d20[4]; + u32 d21[4]; + u32 d22[4]; + u32 d23[4]; + + d20[0] = 0x2d2d0000; + d20[1] = 0; + d20[2] = 0; + d20[3] = 0; + d21[0] = 0; + d21[1] = 0; + d21[2] = 0; + d21[3] = 0; + d22[0] = 0; + d22[1] = 0; + d22[2] = 0; + d22[3] = 0; + d23[0] = 0; + d23[1] = 0; + d23[2] = 0; + d23[3] = 0; + + sha1_update_64 (&ctx0, d20, d21, d22, d23, 2); + + sha1_update_global_swap (&ctx0, salt_bufs[salt_pos].salt_buf, salt_bufs[salt_pos].salt_len); + + u32 d40[4]; + u32 d41[4]; + u32 d42[4]; + u32 d43[4]; + + d40[0] = 0x2d2d2d2d; + d40[1] = 0; + d40[2] = 0; + d40[3] = 0; + d41[0] = 0; + d41[1] = 0; + d41[2] = 0; + d41[3] = 0; + d42[0] = 0; + d42[1] = 0; + d42[2] = 0; + d42[3] = 0; + d43[0] = 0; + d43[1] = 0; + d43[2] = 0; + d43[3] = 0; + + sha1_update_64 (&ctx0, d20, d21, d22, d23, 2); + + /** + * loop + */ + + u32x w0l = w[0]; + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x w0r = words_buf_r[il_pos / VECT_SIZE]; + + const u32x w0lr = w0l | w0r; + + w[0] = w0lr; + + sha1_ctx_vector_t ctx1; + + sha1_init_vector_from_scalar (&ctx1, &ctx0); + + u32x dx20[4]; + u32x dx21[4]; + u32x dx22[4]; + u32x dx23[4]; + + dx20[0] = 0x2d2d0000; + dx20[1] = 0; + dx20[2] = 0; + dx20[3] = 0; + dx21[0] = 0; + dx21[1] = 0; + dx21[2] = 0; + dx21[3] = 0; + dx22[0] = 0; + dx22[1] = 0; + dx22[2] = 0; + dx22[3] = 0; + dx23[0] = 0; + dx23[1] = 0; + dx23[2] = 0; + dx23[3] = 0; + + sha1_update_vector_64 (&ctx1, dx20, dx21, dx22, dx23, 2); + + sha1_update_vector_swap (&ctx1, w, pw_len); + + u32x dx40[4]; + u32x dx41[4]; + u32x dx42[4]; + u32x dx43[4]; + + dx40[0] = 0x2d2d2d2d; + dx40[1] = 0; + dx40[2] = 0; + dx40[3] = 0; + dx41[0] = 0; + dx41[1] = 0; + dx41[2] = 0; + dx41[3] = 0; + dx42[0] = 0; + dx42[1] = 0; + dx42[2] = 0; + dx42[3] = 0; + dx43[0] = 0; + dx43[1] = 0; + dx43[2] = 0; + dx43[3] = 0; + + sha1_update_vector_64 (&ctx1, dx40, dx41, dx42, dx43, 4); + + sha1_final_vector (&ctx1); + + u32x a = ctx1.h[0]; + u32x b = ctx1.h[1]; + u32x c = ctx1.h[2]; + u32x d = ctx1.h[3]; + u32x e = ctx1.h[4]; + + sha1_ctx_vector_t ctx; + + for (int i = 1; i < 10; i++) + { + sha1_init_vector_from_scalar (&ctx, &ctx0); + + u32x w0[4]; + u32x w1[4]; + u32x w2[4]; + u32x w3[4]; + + w0[0] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((a >> 24) & 255) << 16; + w0[1] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((a >> 8) & 255) << 16; + w0[2] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((b >> 24) & 255) << 16; + w0[3] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((b >> 8) & 255) << 16; + w1[0] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((c >> 24) & 255) << 16; + w1[1] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((c >> 8) & 255) << 16; + w1[2] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((d >> 24) & 255) << 16; + w1[3] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((d >> 8) & 255) << 16; + w2[0] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((e >> 24) & 255) << 16; + w2[1] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((e >> 8) & 255) << 16; + w2[2] = 0; + w2[3] = 0; + w3[0] = 0; + w3[1] = 0; + w3[2] = 0; + w3[3] = 0; + + sha1_update_vector_64 (&ctx, w0, w1, w2, w3, 40); + + dx20[0] = 0x2d2d0000; + dx20[1] = 0; + dx20[2] = 0; + dx20[3] = 0; + dx21[0] = 0; + dx21[1] = 0; + dx21[2] = 0; + dx21[3] = 0; + dx22[0] = 0; + dx22[1] = 0; + dx22[2] = 0; + dx22[3] = 0; + dx23[0] = 0; + dx23[1] = 0; + dx23[2] = 0; + dx23[3] = 0; + + sha1_update_vector_64 (&ctx, dx20, dx21, dx22, dx23, 2); + + sha1_update_vector_swap (&ctx, w, pw_len); + + dx40[0] = 0x2d2d2d2d; + dx40[1] = 0; + dx40[2] = 0; + dx40[3] = 0; + dx41[0] = 0; + dx41[1] = 0; + dx41[2] = 0; + dx41[3] = 0; + dx42[0] = 0; + dx42[1] = 0; + dx42[2] = 0; + dx42[3] = 0; + dx43[0] = 0; + dx43[1] = 0; + dx43[2] = 0; + dx43[3] = 0; + + sha1_update_vector_64 (&ctx, dx40, dx41, dx42, dx43, 4); + + sha1_final_vector (&ctx); + + a = ctx.h[0]; + b = ctx.h[1]; + c = ctx.h[2]; + d = ctx.h[3]; + e = ctx.h[4]; + } + + const u32x r0 = ctx.h[DGST_R0]; + const u32x r1 = ctx.h[DGST_R1]; + const u32x r2 = ctx.h[DGST_R2]; + const u32x r3 = ctx.h[DGST_R3]; + + COMPARE_M_SIMD (r0, r1, r2, r3); + } +} + +__kernel void m14400_sxx (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __constant const u32x *words_buf_r, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) +{ + /** + * modifier + */ + + const u32 gid = get_global_id (0); + const u32 lid = get_local_id (0); + const u32 lsz = get_local_size (0); + + /** + * bin2asc table + */ + + __local 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) << 0 + | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 8; + } + + barrier (CLK_LOCAL_MEM_FENCE); + + if (gid >= gid_max) return; + + /** + * 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] + }; + + /** + * base + */ + + const u32 pw_len = pws[gid].pw_len; + + const u32 pw_lenv = ceil ((float) pw_len / 4); + + u32x w[64] = { 0 }; + + for (int idx = 0; idx < pw_lenv; idx++) + { + w[idx] = pws[gid].i[idx]; + } + + sha1_ctx_t ctx0; + + sha1_init (&ctx0); + + u32 d20[4]; + u32 d21[4]; + u32 d22[4]; + u32 d23[4]; + + d20[0] = 0x2d2d0000; + d20[1] = 0; + d20[2] = 0; + d20[3] = 0; + d21[0] = 0; + d21[1] = 0; + d21[2] = 0; + d21[3] = 0; + d22[0] = 0; + d22[1] = 0; + d22[2] = 0; + d22[3] = 0; + d23[0] = 0; + d23[1] = 0; + d23[2] = 0; + d23[3] = 0; + + sha1_update_64 (&ctx0, d20, d21, d22, d23, 2); + + sha1_update_global_swap (&ctx0, salt_bufs[salt_pos].salt_buf, salt_bufs[salt_pos].salt_len); + + u32 d40[4]; + u32 d41[4]; + u32 d42[4]; + u32 d43[4]; + + d40[0] = 0x2d2d2d2d; + d40[1] = 0; + d40[2] = 0; + d40[3] = 0; + d41[0] = 0; + d41[1] = 0; + d41[2] = 0; + d41[3] = 0; + d42[0] = 0; + d42[1] = 0; + d42[2] = 0; + d42[3] = 0; + d43[0] = 0; + d43[1] = 0; + d43[2] = 0; + d43[3] = 0; + + sha1_update_64 (&ctx0, d20, d21, d22, d23, 2); + + /** + * loop + */ + + u32x w0l = w[0]; + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x w0r = words_buf_r[il_pos / VECT_SIZE]; + + const u32x w0lr = w0l | w0r; + + w[0] = w0lr; + + sha1_ctx_vector_t ctx1; + + sha1_init_vector_from_scalar (&ctx1, &ctx0); + + u32x dx20[4]; + u32x dx21[4]; + u32x dx22[4]; + u32x dx23[4]; + + dx20[0] = 0x2d2d0000; + dx20[1] = 0; + dx20[2] = 0; + dx20[3] = 0; + dx21[0] = 0; + dx21[1] = 0; + dx21[2] = 0; + dx21[3] = 0; + dx22[0] = 0; + dx22[1] = 0; + dx22[2] = 0; + dx22[3] = 0; + dx23[0] = 0; + dx23[1] = 0; + dx23[2] = 0; + dx23[3] = 0; + + sha1_update_vector_64 (&ctx1, dx20, dx21, dx22, dx23, 2); + + sha1_update_vector_swap (&ctx1, w, pw_len); + + u32x dx40[4]; + u32x dx41[4]; + u32x dx42[4]; + u32x dx43[4]; + + dx40[0] = 0x2d2d2d2d; + dx40[1] = 0; + dx40[2] = 0; + dx40[3] = 0; + dx41[0] = 0; + dx41[1] = 0; + dx41[2] = 0; + dx41[3] = 0; + dx42[0] = 0; + dx42[1] = 0; + dx42[2] = 0; + dx42[3] = 0; + dx43[0] = 0; + dx43[1] = 0; + dx43[2] = 0; + dx43[3] = 0; + + sha1_update_vector_64 (&ctx1, dx40, dx41, dx42, dx43, 4); + + sha1_final_vector (&ctx1); + + u32x a = ctx1.h[0]; + u32x b = ctx1.h[1]; + u32x c = ctx1.h[2]; + u32x d = ctx1.h[3]; + u32x e = ctx1.h[4]; + + sha1_ctx_vector_t ctx; + + for (int i = 1; i < 10; i++) + { + sha1_init_vector_from_scalar (&ctx, &ctx0); + + u32x w0[4]; + u32x w1[4]; + u32x w2[4]; + u32x w3[4]; + + w0[0] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((a >> 24) & 255) << 16; + w0[1] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((a >> 8) & 255) << 16; + w0[2] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((b >> 24) & 255) << 16; + w0[3] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((b >> 8) & 255) << 16; + w1[0] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((c >> 24) & 255) << 16; + w1[1] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((c >> 8) & 255) << 16; + w1[2] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((d >> 24) & 255) << 16; + w1[3] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((d >> 8) & 255) << 16; + w2[0] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0 + | uint_to_hex_lower8_le ((e >> 24) & 255) << 16; + w2[1] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0 + | uint_to_hex_lower8_le ((e >> 8) & 255) << 16; + w2[2] = 0; + w2[3] = 0; + w3[0] = 0; + w3[1] = 0; + w3[2] = 0; + w3[3] = 0; + + sha1_update_vector_64 (&ctx, w0, w1, w2, w3, 40); + + dx20[0] = 0x2d2d0000; + dx20[1] = 0; + dx20[2] = 0; + dx20[3] = 0; + dx21[0] = 0; + dx21[1] = 0; + dx21[2] = 0; + dx21[3] = 0; + dx22[0] = 0; + dx22[1] = 0; + dx22[2] = 0; + dx22[3] = 0; + dx23[0] = 0; + dx23[1] = 0; + dx23[2] = 0; + dx23[3] = 0; + + sha1_update_vector_64 (&ctx, dx20, dx21, dx22, dx23, 2); + + sha1_update_vector_swap (&ctx, w, pw_len); + + dx40[0] = 0x2d2d2d2d; + dx40[1] = 0; + dx40[2] = 0; + dx40[3] = 0; + dx41[0] = 0; + dx41[1] = 0; + dx41[2] = 0; + dx41[3] = 0; + dx42[0] = 0; + dx42[1] = 0; + dx42[2] = 0; + dx42[3] = 0; + dx43[0] = 0; + dx43[1] = 0; + dx43[2] = 0; + dx43[3] = 0; + + sha1_update_vector_64 (&ctx, dx40, dx41, dx42, dx43, 4); + + sha1_final_vector (&ctx); + + a = ctx.h[0]; + b = ctx.h[1]; + c = ctx.h[2]; + d = ctx.h[3]; + e = ctx.h[4]; + } + + const u32x r0 = ctx.h[DGST_R0]; + const u32x r1 = ctx.h[DGST_R1]; + const u32x r2 = ctx.h[DGST_R2]; + const u32x r3 = ctx.h[DGST_R3]; + + COMPARE_S_SIMD (r0, r1, r2, r3); + } +}