From 5d137ba03682d010db0108b1a844ec3ec94e0f02 Mon Sep 17 00:00:00 2001 From: jsteube Date: Mon, 31 Jul 2017 09:36:57 +0200 Subject: [PATCH] Add pure kernels for IKE-PSK SHA1 --- OpenCL/m05300_a3.cl | 6 +- OpenCL/m05400_a0.cl | 194 ++++++++++++++++++++++++++++++++++++ OpenCL/m05400_a1.cl | 232 ++++++++++++++++++++++++++++++++++++++++++++ OpenCL/m05400_a3.cl | 204 ++++++++++++++++++++++++++++++++++++++ 4 files changed, 633 insertions(+), 3 deletions(-) create mode 100644 OpenCL/m05400_a0.cl create mode 100644 OpenCL/m05400_a1.cl create mode 100644 OpenCL/m05400_a3.cl diff --git a/OpenCL/m05300_a3.cl b/OpenCL/m05300_a3.cl index 55fc8c1fb..29e83a70d 100644 --- a/OpenCL/m05300_a3.cl +++ b/OpenCL/m05300_a3.cl @@ -10,7 +10,7 @@ #include "inc_hash_functions.cl" #include "inc_types.cl" #include "inc_common.cl" -#include "inc_simd.cl" +#include "inc_scalar.cl" #include "inc_hash_md5.cl" __kernel void m05300_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 ikepsk_t *ikepsk_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) @@ -98,7 +98,7 @@ __kernel void m05300_mxx (__global pw_t *pws, __global const kernel_rule_t *rule const u32 r2 = ctx.opad.h[DGST_R2]; const u32 r3 = ctx.opad.h[DGST_R3]; - COMPARE_M_SIMD (r0, r1, r2, r3); + COMPARE_M_SCALAR (r0, r1, r2, r3); } } @@ -199,6 +199,6 @@ __kernel void m05300_sxx (__global pw_t *pws, __global const kernel_rule_t *rule const u32 r2 = ctx.opad.h[DGST_R2]; const u32 r3 = ctx.opad.h[DGST_R3]; - COMPARE_S_SIMD (r0, r1, r2, r3); + COMPARE_S_SCALAR (r0, r1, r2, r3); } } diff --git a/OpenCL/m05400_a0.cl b/OpenCL/m05400_a0.cl new file mode 100644 index 000000000..ccafef2f5 --- /dev/null +++ b/OpenCL/m05400_a0.cl @@ -0,0 +1,194 @@ +/** + * 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" + +__kernel void m05400_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 ikepsk_t *ikepsk_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 lid = get_local_id (0); + const u32 gid = get_global_id (0); + + 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]; + + barrier (CLK_GLOBAL_MEM_FENCE); + } + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos++) + { + // todo: add rules engine + + sha1_hmac_ctx_t ctx0; + + sha1_hmac_init_swap (&ctx0, w, pw_len); + + sha1_hmac_update_global_swap (&ctx0, ikepsk_bufs[digests_offset].nr_buf, ikepsk_bufs[digests_offset].nr_len); + + sha1_hmac_final (&ctx0); + + u32 w0[4]; + u32 w1[4]; + u32 w2[4]; + u32 w3[4]; + + w0[0] = ctx0.opad.h[0]; + w0[1] = ctx0.opad.h[1]; + w0[2] = ctx0.opad.h[2]; + w0[3] = ctx0.opad.h[3]; + w1[0] = ctx0.opad.h[4]; + w1[1] = 0; + w1[2] = 0; + w1[3] = 0; + w2[0] = 0; + w2[1] = 0; + w2[2] = 0; + w2[3] = 0; + w3[0] = 0; + w3[1] = 0; + w3[2] = 0; + w3[3] = 0; + + sha1_hmac_ctx_t ctx; + + sha1_hmac_init_vector_64 (&ctx, w0, w1, w2, w3); + + sha1_hmac_update_global_swap (&ctx, ikepsk_bufs[digests_offset].msg_buf, ikepsk_bufs[digests_offset].msg_len); + + sha1_hmac_final (&ctx); + + const u32x r0 = ctx.opad.h[DGST_R0]; + const u32x r1 = ctx.opad.h[DGST_R1]; + const u32x r2 = ctx.opad.h[DGST_R2]; + const u32x r3 = ctx.opad.h[DGST_R3]; + + COMPARE_M_SCALAR (r0, r1, r2, r3); + } +} + +__kernel void m05400_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 ikepsk_t *ikepsk_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 lid = get_local_id (0); + const u32 gid = get_global_id (0); + + 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]; + + barrier (CLK_GLOBAL_MEM_FENCE); + } + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos++) + { + // todo: add rules engine + + sha1_hmac_ctx_t ctx0; + + sha1_hmac_init_swap (&ctx0, w, pw_len); + + sha1_hmac_update_global_swap (&ctx0, ikepsk_bufs[digests_offset].nr_buf, ikepsk_bufs[digests_offset].nr_len); + + sha1_hmac_final (&ctx0); + + u32 w0[4]; + u32 w1[4]; + u32 w2[4]; + u32 w3[4]; + + w0[0] = ctx0.opad.h[0]; + w0[1] = ctx0.opad.h[1]; + w0[2] = ctx0.opad.h[2]; + w0[3] = ctx0.opad.h[3]; + w1[0] = ctx0.opad.h[4]; + w1[1] = 0; + w1[2] = 0; + w1[3] = 0; + w2[0] = 0; + w2[1] = 0; + w2[2] = 0; + w2[3] = 0; + w3[0] = 0; + w3[1] = 0; + w3[2] = 0; + w3[3] = 0; + + sha1_hmac_ctx_t ctx; + + sha1_hmac_init_vector_64 (&ctx, w0, w1, w2, w3); + + sha1_hmac_update_global_swap (&ctx, ikepsk_bufs[digests_offset].msg_buf, ikepsk_bufs[digests_offset].msg_len); + + sha1_hmac_final (&ctx); + + const u32x r0 = ctx.opad.h[DGST_R0]; + const u32x r1 = ctx.opad.h[DGST_R1]; + const u32x r2 = ctx.opad.h[DGST_R2]; + const u32x r3 = ctx.opad.h[DGST_R3]; + + COMPARE_S_SCALAR (r0, r1, r2, r3); + } +} diff --git a/OpenCL/m05400_a1.cl b/OpenCL/m05400_a1.cl new file mode 100644 index 000000000..415d537d1 --- /dev/null +++ b/OpenCL/m05400_a1.cl @@ -0,0 +1,232 @@ +/** + * 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" + +__kernel void m05400_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 ikepsk_t *ikepsk_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 lid = get_local_id (0); + const u32 gid = get_global_id (0); + + 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] = swap32_S (pws[gid].i[idx]); + + barrier (CLK_GLOBAL_MEM_FENCE); + } + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos++) + { + const u32 comb_len = combs_buf[il_pos].pw_len; + + u32 c[64]; + + #ifdef _unroll + #pragma unroll + #endif + for (int idx = 0; idx < 64; idx++) + { + c[idx] = swap32_S (combs_buf[il_pos].i[idx]); + } + + switch_buffer_by_offset_1x64_be_S (c, pw_len); + + #ifdef _unroll + #pragma unroll + #endif + for (int i = 0; i < 64; i++) + { + c[i] |= w[i]; + } + + sha1_hmac_ctx_t ctx0; + + sha1_hmac_init (&ctx0, c, pw_len + comb_len); + + sha1_hmac_update_global_swap (&ctx0, ikepsk_bufs[digests_offset].nr_buf, ikepsk_bufs[digests_offset].nr_len); + + sha1_hmac_final (&ctx0); + + u32 w0[4]; + u32 w1[4]; + u32 w2[4]; + u32 w3[4]; + + w0[0] = ctx0.opad.h[0]; + w0[1] = ctx0.opad.h[1]; + w0[2] = ctx0.opad.h[2]; + w0[3] = ctx0.opad.h[3]; + w1[0] = ctx0.opad.h[4]; + w1[1] = 0; + w1[2] = 0; + w1[3] = 0; + w2[0] = 0; + w2[1] = 0; + w2[2] = 0; + w2[3] = 0; + w3[0] = 0; + w3[1] = 0; + w3[2] = 0; + w3[3] = 0; + + sha1_hmac_ctx_t ctx; + + sha1_hmac_init_vector_64 (&ctx, w0, w1, w2, w3); + + sha1_hmac_update_global_swap (&ctx, ikepsk_bufs[digests_offset].msg_buf, ikepsk_bufs[digests_offset].msg_len); + + sha1_hmac_final (&ctx); + + const u32 r0 = ctx.opad.h[DGST_R0]; + const u32 r1 = ctx.opad.h[DGST_R1]; + const u32 r2 = ctx.opad.h[DGST_R2]; + const u32 r3 = ctx.opad.h[DGST_R3]; + + COMPARE_M_SCALAR (r0, r1, r2, r3); + } +} + +__kernel void m05400_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 ikepsk_t *ikepsk_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 lid = get_local_id (0); + const u32 gid = get_global_id (0); + + 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] = swap32_S (pws[gid].i[idx]); + + barrier (CLK_GLOBAL_MEM_FENCE); + } + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos++) + { + const u32 comb_len = combs_buf[il_pos].pw_len; + + u32 c[64]; + + #ifdef _unroll + #pragma unroll + #endif + for (int idx = 0; idx < 64; idx++) + { + c[idx] = swap32_S (combs_buf[il_pos].i[idx]); + } + + switch_buffer_by_offset_1x64_be_S (c, pw_len); + + #ifdef _unroll + #pragma unroll + #endif + for (int i = 0; i < 64; i++) + { + c[i] |= w[i]; + } + + sha1_hmac_ctx_t ctx0; + + sha1_hmac_init (&ctx0, c, pw_len + comb_len); + + sha1_hmac_update_global_swap (&ctx0, ikepsk_bufs[digests_offset].nr_buf, ikepsk_bufs[digests_offset].nr_len); + + sha1_hmac_final (&ctx0); + + u32 w0[4]; + u32 w1[4]; + u32 w2[4]; + u32 w3[4]; + + w0[0] = ctx0.opad.h[0]; + w0[1] = ctx0.opad.h[1]; + w0[2] = ctx0.opad.h[2]; + w0[3] = ctx0.opad.h[3]; + w1[0] = ctx0.opad.h[4]; + w1[1] = 0; + w1[2] = 0; + w1[3] = 0; + w2[0] = 0; + w2[1] = 0; + w2[2] = 0; + w2[3] = 0; + w3[0] = 0; + w3[1] = 0; + w3[2] = 0; + w3[3] = 0; + + sha1_hmac_ctx_t ctx; + + sha1_hmac_init_vector_64 (&ctx, w0, w1, w2, w3); + + sha1_hmac_update_global_swap (&ctx, ikepsk_bufs[digests_offset].msg_buf, ikepsk_bufs[digests_offset].msg_len); + + sha1_hmac_final (&ctx); + + const u32 r0 = ctx.opad.h[DGST_R0]; + const u32 r1 = ctx.opad.h[DGST_R1]; + const u32 r2 = ctx.opad.h[DGST_R2]; + const u32 r3 = ctx.opad.h[DGST_R3]; + + COMPARE_S_SCALAR (r0, r1, r2, r3); + } +} diff --git a/OpenCL/m05400_a3.cl b/OpenCL/m05400_a3.cl new file mode 100644 index 000000000..ee4ab2b53 --- /dev/null +++ b/OpenCL/m05400_a3.cl @@ -0,0 +1,204 @@ +/** + * 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" + +__kernel void m05400_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 ikepsk_t *ikepsk_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 lid = get_local_id (0); + const u32 gid = get_global_id (0); + + 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]; + + barrier (CLK_GLOBAL_MEM_FENCE); + } + + /** + * 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_hmac_ctx_t ctx0; + + sha1_hmac_init (&ctx0, w, pw_len); + + sha1_hmac_update_global_swap (&ctx0, ikepsk_bufs[digests_offset].nr_buf, ikepsk_bufs[digests_offset].nr_len); + + sha1_hmac_final (&ctx0); + + u32 w0[4]; + u32 w1[4]; + u32 w2[4]; + u32 w3[4]; + + w0[0] = ctx0.opad.h[0]; + w0[1] = ctx0.opad.h[1]; + w0[2] = ctx0.opad.h[2]; + w0[3] = ctx0.opad.h[3]; + w1[0] = ctx0.opad.h[4]; + w1[1] = 0; + w1[2] = 0; + w1[3] = 0; + w2[0] = 0; + w2[1] = 0; + w2[2] = 0; + w2[3] = 0; + w3[0] = 0; + w3[1] = 0; + w3[2] = 0; + w3[3] = 0; + + sha1_hmac_ctx_t ctx; + + sha1_hmac_init_vector_64 (&ctx, w0, w1, w2, w3); + + sha1_hmac_update_global_swap (&ctx, ikepsk_bufs[digests_offset].msg_buf, ikepsk_bufs[digests_offset].msg_len); + + sha1_hmac_final (&ctx); + + const u32x r0 = ctx.opad.h[DGST_R0]; + const u32x r1 = ctx.opad.h[DGST_R1]; + const u32x r2 = ctx.opad.h[DGST_R2]; + const u32x r3 = ctx.opad.h[DGST_R3]; + + COMPARE_M_SCALAR (r0, r1, r2, r3); + } +} + +__kernel void m05400_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 ikepsk_t *ikepsk_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 lid = get_local_id (0); + const u32 gid = get_global_id (0); + + 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]; + + barrier (CLK_GLOBAL_MEM_FENCE); + } + + /** + * 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_hmac_ctx_t ctx0; + + sha1_hmac_init (&ctx0, w, pw_len); + + sha1_hmac_update_global_swap (&ctx0, ikepsk_bufs[digests_offset].nr_buf, ikepsk_bufs[digests_offset].nr_len); + + sha1_hmac_final (&ctx0); + + u32 w0[4]; + u32 w1[4]; + u32 w2[4]; + u32 w3[4]; + + w0[0] = ctx0.opad.h[0]; + w0[1] = ctx0.opad.h[1]; + w0[2] = ctx0.opad.h[2]; + w0[3] = ctx0.opad.h[3]; + w1[0] = ctx0.opad.h[4]; + w1[1] = 0; + w1[2] = 0; + w1[3] = 0; + w2[0] = 0; + w2[1] = 0; + w2[2] = 0; + w2[3] = 0; + w3[0] = 0; + w3[1] = 0; + w3[2] = 0; + w3[3] = 0; + + sha1_hmac_ctx_t ctx; + + sha1_hmac_init_vector_64 (&ctx, w0, w1, w2, w3); + + sha1_hmac_update_global_swap (&ctx, ikepsk_bufs[digests_offset].msg_buf, ikepsk_bufs[digests_offset].msg_len); + + sha1_hmac_final (&ctx); + + const u32x r0 = ctx.opad.h[DGST_R0]; + const u32x r1 = ctx.opad.h[DGST_R1]; + const u32x r2 = ctx.opad.h[DGST_R2]; + const u32x r3 = ctx.opad.h[DGST_R3]; + + COMPARE_S_SCALAR (r0, r1, r2, r3); + } +}