arno/hashcat
1
0
Fork 0
mirror of https://github.com/hashcat/hashcat.git synced 2024-12-04 22:08:40 +00:00
Code Issues Releases Wiki Activity
a7ab9f6ca9
hashcat/OpenCL/m12600_a3.cl

1045 lines
52 KiB
Common Lisp
Raw Normal View History

Initial commit
2015-12-04 14:47:52 +00:00
/**
Added docs/credits.txt Added docs/team.txt
2016-09-11 20:20:15 +00:00
* Author......: See docs/credits.txt
Initial commit
2015-12-04 14:47:52 +00:00
* License.....: MIT
*/
#define _SHA256_SHA1_
Converted _a3 kernels, use SIMD for CPU and GPU
2016-01-23 14:32:31 +00:00
#define NEW_SIMD_CODE
Move files from include/ to OpenCL/ if they are used within kernels Rename includes in OpenCL so that it's easier to recognize them as such
2016-05-25 21:04:26 +00:00
#include "inc_vendor.cl"
Move macros DGST_R0 - DGST_R3 to host, define dgst_size for opencl kernel from host; both at runtime
2016-06-26 21:39:42 +00:00
#include "inc_hash_constants.h"
Move files from include/ to OpenCL/ if they are used within kernels Rename includes in OpenCL so that it's easier to recognize them as such
2016-05-25 21:04:26 +00:00
#include "inc_hash_functions.cl"
#include "inc_types.cl"
#include "inc_common.cl"
#include "inc_simd.cl"
Converted _a3 kernels, use SIMD for CPU and GPU
2016-01-23 14:32:31 +00:00
#if VECT_SIZE == 1
#define uint_to_hex_upper8(i) (u32x) (l_bin2asc[(i)])
#elif VECT_SIZE == 2
#define uint_to_hex_upper8(i) (u32x) (l_bin2asc[(i).s0], l_bin2asc[(i).s1])
#elif VECT_SIZE == 4
#define uint_to_hex_upper8(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_upper8(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])
Allow and support vector-width 16, which is current maximum for OpenCL. Closes #226.
2016-02-16 15:42:08 +00:00
#elif VECT_SIZE == 16
#define uint_to_hex_upper8(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])
Converted _a3 kernels, use SIMD for CPU and GPU
2016-01-23 14:32:31 +00:00
#endif
Initial commit
2015-12-04 14:47:52 +00:00
Workaround memory allocation limit from OpenCL by using multiple buffers for scrypt
2016-06-28 09:03:04 +00:00
void m12600m (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const u32 pw_len, __global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global 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, __local u32 *l_bin2asc)
Initial commit
2015-12-04 14:47:52 +00:00
{
/**
* modifier
*/
const u32 gid = get_global_id (0);
const u32 lid = get_local_id (0);
/**
* salt
*/
u32 pc256[8];
pc256[0] = salt_bufs[salt_pos].salt_buf_pc[0];
pc256[1] = salt_bufs[salt_pos].salt_buf_pc[1];
pc256[2] = salt_bufs[salt_pos].salt_buf_pc[2];
pc256[3] = salt_bufs[salt_pos].salt_buf_pc[3];
pc256[4] = salt_bufs[salt_pos].salt_buf_pc[4];
pc256[5] = salt_bufs[salt_pos].salt_buf_pc[5];
pc256[6] = salt_bufs[salt_pos].salt_buf_pc[6];
pc256[7] = salt_bufs[salt_pos].salt_buf_pc[7];
/**
* loop
*/
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w0l = w0[0];
Initial commit
2015-12-04 14:47:52 +00:00
Fixed two major problems 1) SIMD code for all attack-mode Macro vector_accessible() was not refactored and missing completely. Had to rename variables rules_cnt, combs_cnt and bfs_cnt into il_cnt which was a good thing anyway as with new SIMD code they all act in the same way. 2) SIMD code for attack-mode 0 With new SIMD code, apply_rules_vect() has to return u32 not u32x. This has massive impact on all *_a0 kernels. I've rewritten most of them. Deep testing using test.sh is still required. Some kernel need more fixes: - Some are kind of completely incompatible like m10400 but they still use old check_* includes, we should get rid of them as they are no longer neccessary as we have simd.c - Some have a chance but require additional effort like m11500. We can use commented out "#define NEW_SIMD_CODE" to find them This change can have negative impact on -a0 performance for device that require vectorization. That is mostly CPU devices. New GPU's are all scalar, so they wont get hurt by this. This change also proofes that there's no way to efficiently vectorize kernel rules with new SIMD code, but it enables the addition of the rule functions like @ that we were missing for some long time. This is a TODO.
2016-02-27 16:18:54 +00:00
for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
Initial commit
2015-12-04 14:47:52 +00:00
{
Prepare NEW_SIMD_MODE for -a 1 kernels
2016-02-20 15:13:06 +00:00
const u32x w0r = ix_create_bft (bfs_buf, il_pos);
Initial commit
2015-12-04 14:47:52 +00:00
Converted _a3 kernels, use SIMD for CPU and GPU
2016-01-23 14:32:31 +00:00
const u32x w0lr = w0l | w0r;
Initial commit
2015-12-04 14:47:52 +00:00
/**
* sha1
*/
Converted _a3 kernels, use SIMD for CPU and GPU
2016-01-23 14:32:31 +00:00
u32x w0_t = w0lr;
u32x w1_t = w0[1];
u32x w2_t = w0[2];
u32x w3_t = w0[3];
u32x w4_t = w1[0];
u32x w5_t = w1[1];
u32x w6_t = w1[2];
u32x w7_t = w1[3];
u32x w8_t = w2[0];
u32x w9_t = w2[1];
u32x wa_t = w2[2];
u32x wb_t = w2[3];
u32x wc_t = w3[0];
u32x wd_t = w3[1];
u32x we_t = 0;
u32x wf_t = pw_len * 8;
u32x a = SHA1M_A;
u32x b = SHA1M_B;
u32x c = SHA1M_C;
u32x d = SHA1M_D;
u32x e = SHA1M_E;
u32x f = 0;
u32x g = 0;
u32x h = 0;
Initial commit
2015-12-04 14:47:52 +00:00
#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 = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
w3_t = 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 = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
w7_t = 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 = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
wb_t = 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 = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
wf_t = 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;
/**
* sha256
*/
w0_t = uint_to_hex_upper8 ((a >> 24) & 255) << 0
| uint_to_hex_upper8 ((a >> 16) & 255) << 16;
w1_t = uint_to_hex_upper8 ((a >> 8) & 255) << 0
| uint_to_hex_upper8 ((a >> 0) & 255) << 16;
w2_t = uint_to_hex_upper8 ((b >> 24) & 255) << 0
| uint_to_hex_upper8 ((b >> 16) & 255) << 16;
w3_t = uint_to_hex_upper8 ((b >> 8) & 255) << 0
| uint_to_hex_upper8 ((b >> 0) & 255) << 16;
w4_t = uint_to_hex_upper8 ((c >> 24) & 255) << 0
| uint_to_hex_upper8 ((c >> 16) & 255) << 16;
w5_t = uint_to_hex_upper8 ((c >> 8) & 255) << 0
| uint_to_hex_upper8 ((c >> 0) & 255) << 16;
w6_t = uint_to_hex_upper8 ((d >> 24) & 255) << 0
| uint_to_hex_upper8 ((d >> 16) & 255) << 16;
w7_t = uint_to_hex_upper8 ((d >> 8) & 255) << 0
| uint_to_hex_upper8 ((d >> 0) & 255) << 16;
w8_t = uint_to_hex_upper8 ((e >> 24) & 255) << 0
| uint_to_hex_upper8 ((e >> 16) & 255) << 16;
w9_t = uint_to_hex_upper8 ((e >> 8) & 255) << 0
| uint_to_hex_upper8 ((e >> 0) & 255) << 16;
Fix more append_* functions in kernels
2015-12-15 15:50:21 +00:00
w0_t = swap32 (w0_t);
w1_t = swap32 (w1_t);
w2_t = swap32 (w2_t);
w3_t = swap32 (w3_t);
w4_t = swap32 (w4_t);
w5_t = swap32 (w5_t);
w6_t = swap32 (w6_t);
w7_t = swap32 (w7_t);
w8_t = swap32 (w8_t);
w9_t = swap32 (w9_t);
Initial commit
2015-12-04 14:47:52 +00:00
wa_t = 0x80000000;
wb_t = 0;
wc_t = 0;
wd_t = 0;
we_t = 0;
wf_t = (64 + 40) * 8;
a = pc256[0];
b = pc256[1];
c = pc256[2];
d = pc256[3];
e = pc256[4];
f = pc256[5];
g = pc256[6];
h = pc256[7];
SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w0_t, SHA256C00);
SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w1_t, SHA256C01);
SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w2_t, SHA256C02);
SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w3_t, SHA256C03);
SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w4_t, SHA256C04);
SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w5_t, SHA256C05);
SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w6_t, SHA256C06);
SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w7_t, SHA256C07);
SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w8_t, SHA256C08);
SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w9_t, SHA256C09);
SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, wa_t, SHA256C0a);
SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, wb_t, SHA256C0b);
SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, wc_t, SHA256C0c);
SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, wd_t, SHA256C0d);
SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, we_t, SHA256C0e);
SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, wf_t, SHA256C0f);
w0_t = SHA256_EXPAND (we_t, w9_t, w1_t, w0_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w0_t, SHA256C10);
w1_t = SHA256_EXPAND (wf_t, wa_t, w2_t, w1_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w1_t, SHA256C11);
w2_t = SHA256_EXPAND (w0_t, wb_t, w3_t, w2_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w2_t, SHA256C12);
w3_t = SHA256_EXPAND (w1_t, wc_t, w4_t, w3_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w3_t, SHA256C13);
w4_t = SHA256_EXPAND (w2_t, wd_t, w5_t, w4_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w4_t, SHA256C14);
w5_t = SHA256_EXPAND (w3_t, we_t, w6_t, w5_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w5_t, SHA256C15);
w6_t = SHA256_EXPAND (w4_t, wf_t, w7_t, w6_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w6_t, SHA256C16);
w7_t = SHA256_EXPAND (w5_t, w0_t, w8_t, w7_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w7_t, SHA256C17);
w8_t = SHA256_EXPAND (w6_t, w1_t, w9_t, w8_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w8_t, SHA256C18);
w9_t = SHA256_EXPAND (w7_t, w2_t, wa_t, w9_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w9_t, SHA256C19);
wa_t = SHA256_EXPAND (w8_t, w3_t, wb_t, wa_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, wa_t, SHA256C1a);
wb_t = SHA256_EXPAND (w9_t, w4_t, wc_t, wb_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, wb_t, SHA256C1b);
wc_t = SHA256_EXPAND (wa_t, w5_t, wd_t, wc_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, wc_t, SHA256C1c);
wd_t = SHA256_EXPAND (wb_t, w6_t, we_t, wd_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, wd_t, SHA256C1d);
we_t = SHA256_EXPAND (wc_t, w7_t, wf_t, we_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, we_t, SHA256C1e);
wf_t = SHA256_EXPAND (wd_t, w8_t, w0_t, wf_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, wf_t, SHA256C1f);
w0_t = SHA256_EXPAND (we_t, w9_t, w1_t, w0_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w0_t, SHA256C20);
w1_t = SHA256_EXPAND (wf_t, wa_t, w2_t, w1_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w1_t, SHA256C21);
w2_t = SHA256_EXPAND (w0_t, wb_t, w3_t, w2_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w2_t, SHA256C22);
w3_t = SHA256_EXPAND (w1_t, wc_t, w4_t, w3_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w3_t, SHA256C23);
w4_t = SHA256_EXPAND (w2_t, wd_t, w5_t, w4_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w4_t, SHA256C24);
w5_t = SHA256_EXPAND (w3_t, we_t, w6_t, w5_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w5_t, SHA256C25);
w6_t = SHA256_EXPAND (w4_t, wf_t, w7_t, w6_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w6_t, SHA256C26);
w7_t = SHA256_EXPAND (w5_t, w0_t, w8_t, w7_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w7_t, SHA256C27);
w8_t = SHA256_EXPAND (w6_t, w1_t, w9_t, w8_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w8_t, SHA256C28);
w9_t = SHA256_EXPAND (w7_t, w2_t, wa_t, w9_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w9_t, SHA256C29);
wa_t = SHA256_EXPAND (w8_t, w3_t, wb_t, wa_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, wa_t, SHA256C2a);
wb_t = SHA256_EXPAND (w9_t, w4_t, wc_t, wb_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, wb_t, SHA256C2b);
wc_t = SHA256_EXPAND (wa_t, w5_t, wd_t, wc_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, wc_t, SHA256C2c);
wd_t = SHA256_EXPAND (wb_t, w6_t, we_t, wd_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, wd_t, SHA256C2d);
we_t = SHA256_EXPAND (wc_t, w7_t, wf_t, we_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, we_t, SHA256C2e);
wf_t = SHA256_EXPAND (wd_t, w8_t, w0_t, wf_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, wf_t, SHA256C2f);
w0_t = SHA256_EXPAND (we_t, w9_t, w1_t, w0_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w0_t, SHA256C30);
w1_t = SHA256_EXPAND (wf_t, wa_t, w2_t, w1_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w1_t, SHA256C31);
w2_t = SHA256_EXPAND (w0_t, wb_t, w3_t, w2_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w2_t, SHA256C32);
w3_t = SHA256_EXPAND (w1_t, wc_t, w4_t, w3_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w3_t, SHA256C33);
w4_t = SHA256_EXPAND (w2_t, wd_t, w5_t, w4_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w4_t, SHA256C34);
w5_t = SHA256_EXPAND (w3_t, we_t, w6_t, w5_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w5_t, SHA256C35);
w6_t = SHA256_EXPAND (w4_t, wf_t, w7_t, w6_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w6_t, SHA256C36);
w7_t = SHA256_EXPAND (w5_t, w0_t, w8_t, w7_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w7_t, SHA256C37);
w8_t = SHA256_EXPAND (w6_t, w1_t, w9_t, w8_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w8_t, SHA256C38);
w9_t = SHA256_EXPAND (w7_t, w2_t, wa_t, w9_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w9_t, SHA256C39);
wa_t = SHA256_EXPAND (w8_t, w3_t, wb_t, wa_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, wa_t, SHA256C3a);
wb_t = SHA256_EXPAND (w9_t, w4_t, wc_t, wb_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, wb_t, SHA256C3b);
wc_t = SHA256_EXPAND (wa_t, w5_t, wd_t, wc_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, wc_t, SHA256C3c);
wd_t = SHA256_EXPAND (wb_t, w6_t, we_t, wd_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, wd_t, SHA256C3d);
we_t = SHA256_EXPAND (wc_t, w7_t, wf_t, we_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, we_t, SHA256C3e);
wf_t = SHA256_EXPAND (wd_t, w8_t, w0_t, wf_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, wf_t, SHA256C3f);
Converted _a3 kernels, use SIMD for CPU and GPU
2016-01-23 14:32:31 +00:00
COMPARE_M_SIMD (d, h, c, g);
Initial commit
2015-12-04 14:47:52 +00:00
}
}
Workaround memory allocation limit from OpenCL by using multiple buffers for scrypt
2016-06-28 09:03:04 +00:00
void m12600s (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const u32 pw_len, __global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global 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, __local u32 *l_bin2asc)
Initial commit
2015-12-04 14:47:52 +00:00
{
/**
* modifier
*/
const u32 gid = get_global_id (0);
const u32 lid = get_local_id (0);
/**
* salt
*/
u32 pc256[8];
pc256[0] = salt_bufs[salt_pos].salt_buf_pc[0];
pc256[1] = salt_bufs[salt_pos].salt_buf_pc[1];
pc256[2] = salt_bufs[salt_pos].salt_buf_pc[2];
pc256[3] = salt_bufs[salt_pos].salt_buf_pc[3];
pc256[4] = salt_bufs[salt_pos].salt_buf_pc[4];
pc256[5] = salt_bufs[salt_pos].salt_buf_pc[5];
pc256[6] = salt_bufs[salt_pos].salt_buf_pc[6];
pc256[7] = salt_bufs[salt_pos].salt_buf_pc[7];
Cleanup -m 126xx kernels to latest standard
2016-04-18 14:44:54 +00:00
/**
* 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]
};
Initial commit
2015-12-04 14:47:52 +00:00
/**
* loop
*/
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w0l = w0[0];
Initial commit
2015-12-04 14:47:52 +00:00
Fixed two major problems 1) SIMD code for all attack-mode Macro vector_accessible() was not refactored and missing completely. Had to rename variables rules_cnt, combs_cnt and bfs_cnt into il_cnt which was a good thing anyway as with new SIMD code they all act in the same way. 2) SIMD code for attack-mode 0 With new SIMD code, apply_rules_vect() has to return u32 not u32x. This has massive impact on all *_a0 kernels. I've rewritten most of them. Deep testing using test.sh is still required. Some kernel need more fixes: - Some are kind of completely incompatible like m10400 but they still use old check_* includes, we should get rid of them as they are no longer neccessary as we have simd.c - Some have a chance but require additional effort like m11500. We can use commented out "#define NEW_SIMD_CODE" to find them This change can have negative impact on -a0 performance for device that require vectorization. That is mostly CPU devices. New GPU's are all scalar, so they wont get hurt by this. This change also proofes that there's no way to efficiently vectorize kernel rules with new SIMD code, but it enables the addition of the rule functions like @ that we were missing for some long time. This is a TODO.
2016-02-27 16:18:54 +00:00
for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
Initial commit
2015-12-04 14:47:52 +00:00
{
Prepare NEW_SIMD_MODE for -a 1 kernels
2016-02-20 15:13:06 +00:00
const u32x w0r = ix_create_bft (bfs_buf, il_pos);
Initial commit
2015-12-04 14:47:52 +00:00
Converted _a3 kernels, use SIMD for CPU and GPU
2016-01-23 14:32:31 +00:00
const u32x w0lr = w0l | w0r;
Initial commit
2015-12-04 14:47:52 +00:00
/**
* sha1
*/
Converted _a3 kernels, use SIMD for CPU and GPU
2016-01-23 14:32:31 +00:00
u32x w0_t = w0lr;
u32x w1_t = w0[1];
u32x w2_t = w0[2];
u32x w3_t = w0[3];
u32x w4_t = w1[0];
u32x w5_t = w1[1];
u32x w6_t = w1[2];
u32x w7_t = w1[3];
u32x w8_t = w2[0];
u32x w9_t = w2[1];
u32x wa_t = w2[2];
u32x wb_t = w2[3];
u32x wc_t = w3[0];
u32x wd_t = w3[1];
u32x we_t = 0;
u32x wf_t = pw_len * 8;
u32x a = SHA1M_A;
u32x b = SHA1M_B;
u32x c = SHA1M_C;
u32x d = SHA1M_D;
u32x e = SHA1M_E;
u32x f = 0;
u32x g = 0;
u32x h = 0;
Initial commit
2015-12-04 14:47:52 +00:00
#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 = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
w3_t = 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 = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
w7_t = 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 = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
wb_t = 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 = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
wf_t = 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;
/**
* sha256
*/
w0_t = uint_to_hex_upper8 ((a >> 24) & 255) << 0
| uint_to_hex_upper8 ((a >> 16) & 255) << 16;
w1_t = uint_to_hex_upper8 ((a >> 8) & 255) << 0
| uint_to_hex_upper8 ((a >> 0) & 255) << 16;
w2_t = uint_to_hex_upper8 ((b >> 24) & 255) << 0
| uint_to_hex_upper8 ((b >> 16) & 255) << 16;
w3_t = uint_to_hex_upper8 ((b >> 8) & 255) << 0
| uint_to_hex_upper8 ((b >> 0) & 255) << 16;
w4_t = uint_to_hex_upper8 ((c >> 24) & 255) << 0
| uint_to_hex_upper8 ((c >> 16) & 255) << 16;
w5_t = uint_to_hex_upper8 ((c >> 8) & 255) << 0
| uint_to_hex_upper8 ((c >> 0) & 255) << 16;
w6_t = uint_to_hex_upper8 ((d >> 24) & 255) << 0
| uint_to_hex_upper8 ((d >> 16) & 255) << 16;
w7_t = uint_to_hex_upper8 ((d >> 8) & 255) << 0
| uint_to_hex_upper8 ((d >> 0) & 255) << 16;
w8_t = uint_to_hex_upper8 ((e >> 24) & 255) << 0
| uint_to_hex_upper8 ((e >> 16) & 255) << 16;
w9_t = uint_to_hex_upper8 ((e >> 8) & 255) << 0
| uint_to_hex_upper8 ((e >> 0) & 255) << 16;
Fix more append_* functions in kernels
2015-12-15 15:50:21 +00:00
w0_t = swap32 (w0_t);
w1_t = swap32 (w1_t);
w2_t = swap32 (w2_t);
w3_t = swap32 (w3_t);
w4_t = swap32 (w4_t);
w5_t = swap32 (w5_t);
w6_t = swap32 (w6_t);
w7_t = swap32 (w7_t);
w8_t = swap32 (w8_t);
w9_t = swap32 (w9_t);
Initial commit
2015-12-04 14:47:52 +00:00
wa_t = 0x80000000;
wb_t = 0;
wc_t = 0;
wd_t = 0;
we_t = 0;
wf_t = (64 + 40) * 8;
a = pc256[0];
b = pc256[1];
c = pc256[2];
d = pc256[3];
e = pc256[4];
f = pc256[5];
g = pc256[6];
h = pc256[7];
SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w0_t, SHA256C00);
SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w1_t, SHA256C01);
SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w2_t, SHA256C02);
SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w3_t, SHA256C03);
SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w4_t, SHA256C04);
SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w5_t, SHA256C05);
SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w6_t, SHA256C06);
SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w7_t, SHA256C07);
SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w8_t, SHA256C08);
SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w9_t, SHA256C09);
SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, wa_t, SHA256C0a);
SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, wb_t, SHA256C0b);
SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, wc_t, SHA256C0c);
SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, wd_t, SHA256C0d);
SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, we_t, SHA256C0e);
SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, wf_t, SHA256C0f);
w0_t = SHA256_EXPAND (we_t, w9_t, w1_t, w0_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w0_t, SHA256C10);
w1_t = SHA256_EXPAND (wf_t, wa_t, w2_t, w1_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w1_t, SHA256C11);
w2_t = SHA256_EXPAND (w0_t, wb_t, w3_t, w2_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w2_t, SHA256C12);
w3_t = SHA256_EXPAND (w1_t, wc_t, w4_t, w3_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w3_t, SHA256C13);
w4_t = SHA256_EXPAND (w2_t, wd_t, w5_t, w4_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w4_t, SHA256C14);
w5_t = SHA256_EXPAND (w3_t, we_t, w6_t, w5_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w5_t, SHA256C15);
w6_t = SHA256_EXPAND (w4_t, wf_t, w7_t, w6_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w6_t, SHA256C16);
w7_t = SHA256_EXPAND (w5_t, w0_t, w8_t, w7_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w7_t, SHA256C17);
w8_t = SHA256_EXPAND (w6_t, w1_t, w9_t, w8_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w8_t, SHA256C18);
w9_t = SHA256_EXPAND (w7_t, w2_t, wa_t, w9_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w9_t, SHA256C19);
wa_t = SHA256_EXPAND (w8_t, w3_t, wb_t, wa_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, wa_t, SHA256C1a);
wb_t = SHA256_EXPAND (w9_t, w4_t, wc_t, wb_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, wb_t, SHA256C1b);
wc_t = SHA256_EXPAND (wa_t, w5_t, wd_t, wc_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, wc_t, SHA256C1c);
wd_t = SHA256_EXPAND (wb_t, w6_t, we_t, wd_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, wd_t, SHA256C1d);
we_t = SHA256_EXPAND (wc_t, w7_t, wf_t, we_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, we_t, SHA256C1e);
wf_t = SHA256_EXPAND (wd_t, w8_t, w0_t, wf_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, wf_t, SHA256C1f);
w0_t = SHA256_EXPAND (we_t, w9_t, w1_t, w0_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w0_t, SHA256C20);
w1_t = SHA256_EXPAND (wf_t, wa_t, w2_t, w1_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w1_t, SHA256C21);
w2_t = SHA256_EXPAND (w0_t, wb_t, w3_t, w2_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w2_t, SHA256C22);
w3_t = SHA256_EXPAND (w1_t, wc_t, w4_t, w3_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w3_t, SHA256C23);
w4_t = SHA256_EXPAND (w2_t, wd_t, w5_t, w4_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w4_t, SHA256C24);
w5_t = SHA256_EXPAND (w3_t, we_t, w6_t, w5_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w5_t, SHA256C25);
w6_t = SHA256_EXPAND (w4_t, wf_t, w7_t, w6_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w6_t, SHA256C26);
w7_t = SHA256_EXPAND (w5_t, w0_t, w8_t, w7_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w7_t, SHA256C27);
w8_t = SHA256_EXPAND (w6_t, w1_t, w9_t, w8_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w8_t, SHA256C28);
w9_t = SHA256_EXPAND (w7_t, w2_t, wa_t, w9_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w9_t, SHA256C29);
wa_t = SHA256_EXPAND (w8_t, w3_t, wb_t, wa_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, wa_t, SHA256C2a);
wb_t = SHA256_EXPAND (w9_t, w4_t, wc_t, wb_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, wb_t, SHA256C2b);
wc_t = SHA256_EXPAND (wa_t, w5_t, wd_t, wc_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, wc_t, SHA256C2c);
wd_t = SHA256_EXPAND (wb_t, w6_t, we_t, wd_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, wd_t, SHA256C2d);
we_t = SHA256_EXPAND (wc_t, w7_t, wf_t, we_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, we_t, SHA256C2e);
wf_t = SHA256_EXPAND (wd_t, w8_t, w0_t, wf_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, wf_t, SHA256C2f);
w0_t = SHA256_EXPAND (we_t, w9_t, w1_t, w0_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w0_t, SHA256C30);
w1_t = SHA256_EXPAND (wf_t, wa_t, w2_t, w1_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w1_t, SHA256C31);
w2_t = SHA256_EXPAND (w0_t, wb_t, w3_t, w2_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w2_t, SHA256C32);
w3_t = SHA256_EXPAND (w1_t, wc_t, w4_t, w3_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w3_t, SHA256C33);
w4_t = SHA256_EXPAND (w2_t, wd_t, w5_t, w4_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w4_t, SHA256C34);
w5_t = SHA256_EXPAND (w3_t, we_t, w6_t, w5_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w5_t, SHA256C35);
w6_t = SHA256_EXPAND (w4_t, wf_t, w7_t, w6_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w6_t, SHA256C36);
w7_t = SHA256_EXPAND (w5_t, w0_t, w8_t, w7_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w7_t, SHA256C37);
w8_t = SHA256_EXPAND (w6_t, w1_t, w9_t, w8_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w8_t, SHA256C38);
w9_t = SHA256_EXPAND (w7_t, w2_t, wa_t, w9_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w9_t, SHA256C39);
wa_t = SHA256_EXPAND (w8_t, w3_t, wb_t, wa_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, wa_t, SHA256C3a);
wb_t = SHA256_EXPAND (w9_t, w4_t, wc_t, wb_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, wb_t, SHA256C3b);
wc_t = SHA256_EXPAND (wa_t, w5_t, wd_t, wc_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, wc_t, SHA256C3c);
Cleanup -m 126xx kernels to latest standard
2016-04-18 14:44:54 +00:00
if (MATCHES_NONE_VS (d, search[0])) continue;
Initial commit
2015-12-04 14:47:52 +00:00
wd_t = SHA256_EXPAND (wb_t, w6_t, we_t, wd_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, wd_t, SHA256C3d);
we_t = SHA256_EXPAND (wc_t, w7_t, wf_t, we_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, we_t, SHA256C3e);
wf_t = SHA256_EXPAND (wd_t, w8_t, w0_t, wf_t); SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, wf_t, SHA256C3f);
Converted _a3 kernels, use SIMD for CPU and GPU
2016-01-23 14:32:31 +00:00
COMPARE_S_SIMD (d, h, c, g);
Initial commit
2015-12-04 14:47:52 +00:00
}
}
Workaround memory allocation limit from OpenCL by using multiple buffers for scrypt
2016-06-28 09:03:04 +00:00
__kernel void m12600_m04 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global 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)
Initial commit
2015-12-04 14:47:52 +00:00
{
/**
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
* modifier
Initial commit
2015-12-04 14:47:52 +00:00
*/
const u32 gid = get_global_id (0);
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
const u32 lid = get_local_id (0);
const u32 lsz = get_local_size (0);
Initial commit
2015-12-04 14:47:52 +00:00
/**
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
* bin2asc table
Initial commit
2015-12-04 14:47:52 +00:00
*/
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
__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) << 8
| ((i1 < 10) ? '0' + i1 : 'A' - 10 + i1) << 0;
}
barrier (CLK_LOCAL_MEM_FENCE);
if (gid >= gid_max) return;
Initial commit
2015-12-04 14:47:52 +00:00
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
/**
* modifier
*/
Initial commit
2015-12-04 14:47:52 +00:00
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w0[4];
Initial commit
2015-12-04 14:47:52 +00:00
w0[0] = pws[gid].i[ 0];
w0[1] = pws[gid].i[ 1];
w0[2] = pws[gid].i[ 2];
w0[3] = pws[gid].i[ 3];
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w1[4];
Initial commit
2015-12-04 14:47:52 +00:00
w1[0] = 0;
w1[1] = 0;
w1[2] = 0;
w1[3] = 0;
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w2[4];
Initial commit
2015-12-04 14:47:52 +00:00
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w3[4];
Initial commit
2015-12-04 14:47:52 +00:00
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = pws[gid].i[15];
const u32 pw_len = pws[gid].pw_len;
/**
* main
*/
Workaround memory allocation limit from OpenCL by using multiple buffers for scrypt
2016-06-28 09:03:04 +00:00
m12600m (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_scryptV0_buf, d_scryptV1_buf, d_scryptV2_buf, d_scryptV3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, l_bin2asc);
Initial commit
2015-12-04 14:47:52 +00:00
}
Workaround memory allocation limit from OpenCL by using multiple buffers for scrypt
2016-06-28 09:03:04 +00:00
__kernel void m12600_m08 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global 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)
Initial commit
2015-12-04 14:47:52 +00:00
{
/**
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
* modifier
Initial commit
2015-12-04 14:47:52 +00:00
*/
const u32 gid = get_global_id (0);
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
const u32 lid = get_local_id (0);
const u32 lsz = get_local_size (0);
Initial commit
2015-12-04 14:47:52 +00:00
/**
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
* bin2asc table
Initial commit
2015-12-04 14:47:52 +00:00
*/
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
__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) << 8
| ((i1 < 10) ? '0' + i1 : 'A' - 10 + i1) << 0;
}
barrier (CLK_LOCAL_MEM_FENCE);
if (gid >= gid_max) return;
/**
* modifier
*/
Initial commit
2015-12-04 14:47:52 +00:00
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w0[4];
Initial commit
2015-12-04 14:47:52 +00:00
w0[0] = pws[gid].i[ 0];
w0[1] = pws[gid].i[ 1];
w0[2] = pws[gid].i[ 2];
w0[3] = pws[gid].i[ 3];
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w1[4];
Initial commit
2015-12-04 14:47:52 +00:00
w1[0] = pws[gid].i[ 4];
w1[1] = pws[gid].i[ 5];
w1[2] = pws[gid].i[ 6];
w1[3] = pws[gid].i[ 7];
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w2[4];
Initial commit
2015-12-04 14:47:52 +00:00
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w3[4];
Initial commit
2015-12-04 14:47:52 +00:00
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = pws[gid].i[15];
const u32 pw_len = pws[gid].pw_len;
/**
* main
*/
Workaround memory allocation limit from OpenCL by using multiple buffers for scrypt
2016-06-28 09:03:04 +00:00
m12600m (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_scryptV0_buf, d_scryptV1_buf, d_scryptV2_buf, d_scryptV3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, l_bin2asc);
Initial commit
2015-12-04 14:47:52 +00:00
}
Workaround memory allocation limit from OpenCL by using multiple buffers for scrypt
2016-06-28 09:03:04 +00:00
__kernel void m12600_m16 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global 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)
Initial commit
2015-12-04 14:47:52 +00:00
{
/**
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
* modifier
Initial commit
2015-12-04 14:47:52 +00:00
*/
const u32 gid = get_global_id (0);
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
const u32 lid = get_local_id (0);
const u32 lsz = get_local_size (0);
Initial commit
2015-12-04 14:47:52 +00:00
/**
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
* bin2asc table
Initial commit
2015-12-04 14:47:52 +00:00
*/
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
__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) << 8
| ((i1 < 10) ? '0' + i1 : 'A' - 10 + i1) << 0;
}
barrier (CLK_LOCAL_MEM_FENCE);
if (gid >= gid_max) return;
/**
* modifier
*/
Initial commit
2015-12-04 14:47:52 +00:00
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w0[4];
Initial commit
2015-12-04 14:47:52 +00:00
w0[0] = pws[gid].i[ 0];
w0[1] = pws[gid].i[ 1];
w0[2] = pws[gid].i[ 2];
w0[3] = pws[gid].i[ 3];
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w1[4];
Initial commit
2015-12-04 14:47:52 +00:00
w1[0] = pws[gid].i[ 4];
w1[1] = pws[gid].i[ 5];
w1[2] = pws[gid].i[ 6];
w1[3] = pws[gid].i[ 7];
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w2[4];
Initial commit
2015-12-04 14:47:52 +00:00
w2[0] = pws[gid].i[ 8];
w2[1] = pws[gid].i[ 9];
w2[2] = pws[gid].i[10];
w2[3] = pws[gid].i[11];
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w3[4];
Initial commit
2015-12-04 14:47:52 +00:00
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;
/**
* main
*/
Workaround memory allocation limit from OpenCL by using multiple buffers for scrypt
2016-06-28 09:03:04 +00:00
m12600m (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_scryptV0_buf, d_scryptV1_buf, d_scryptV2_buf, d_scryptV3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, l_bin2asc);
Initial commit
2015-12-04 14:47:52 +00:00
}
Workaround memory allocation limit from OpenCL by using multiple buffers for scrypt
2016-06-28 09:03:04 +00:00
__kernel void m12600_s04 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global 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)
Initial commit
2015-12-04 14:47:52 +00:00
{
/**
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
* modifier
Initial commit
2015-12-04 14:47:52 +00:00
*/
const u32 gid = get_global_id (0);
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
const u32 lid = get_local_id (0);
const u32 lsz = get_local_size (0);
Initial commit
2015-12-04 14:47:52 +00:00
/**
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
* bin2asc table
Initial commit
2015-12-04 14:47:52 +00:00
*/
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
__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) << 8
| ((i1 < 10) ? '0' + i1 : 'A' - 10 + i1) << 0;
}
barrier (CLK_LOCAL_MEM_FENCE);
if (gid >= gid_max) return;
/**
* modifier
*/
Initial commit
2015-12-04 14:47:52 +00:00
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w0[4];
Initial commit
2015-12-04 14:47:52 +00:00
w0[0] = pws[gid].i[ 0];
w0[1] = pws[gid].i[ 1];
w0[2] = pws[gid].i[ 2];
w0[3] = pws[gid].i[ 3];
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w1[4];
Initial commit
2015-12-04 14:47:52 +00:00
w1[0] = 0;
w1[1] = 0;
w1[2] = 0;
w1[3] = 0;
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w2[4];
Initial commit
2015-12-04 14:47:52 +00:00
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w3[4];
Initial commit
2015-12-04 14:47:52 +00:00
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = pws[gid].i[15];
const u32 pw_len = pws[gid].pw_len;
/**
* main
*/
Workaround memory allocation limit from OpenCL by using multiple buffers for scrypt
2016-06-28 09:03:04 +00:00
m12600s (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_scryptV0_buf, d_scryptV1_buf, d_scryptV2_buf, d_scryptV3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, l_bin2asc);
Initial commit
2015-12-04 14:47:52 +00:00
}
Workaround memory allocation limit from OpenCL by using multiple buffers for scrypt
2016-06-28 09:03:04 +00:00
__kernel void m12600_s08 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global 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)
Initial commit
2015-12-04 14:47:52 +00:00
{
/**
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
* modifier
Initial commit
2015-12-04 14:47:52 +00:00
*/
const u32 gid = get_global_id (0);
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
const u32 lid = get_local_id (0);
const u32 lsz = get_local_size (0);
Initial commit
2015-12-04 14:47:52 +00:00
/**
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
* bin2asc table
Initial commit
2015-12-04 14:47:52 +00:00
*/
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
__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) << 8
| ((i1 < 10) ? '0' + i1 : 'A' - 10 + i1) << 0;
}
barrier (CLK_LOCAL_MEM_FENCE);
if (gid >= gid_max) return;
/**
* modifier
*/
Initial commit
2015-12-04 14:47:52 +00:00
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w0[4];
Initial commit
2015-12-04 14:47:52 +00:00
w0[0] = pws[gid].i[ 0];
w0[1] = pws[gid].i[ 1];
w0[2] = pws[gid].i[ 2];
w0[3] = pws[gid].i[ 3];
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w1[4];
Initial commit
2015-12-04 14:47:52 +00:00
w1[0] = pws[gid].i[ 4];
w1[1] = pws[gid].i[ 5];
w1[2] = pws[gid].i[ 6];
w1[3] = pws[gid].i[ 7];
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w2[4];
Initial commit
2015-12-04 14:47:52 +00:00
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w3[4];
Initial commit
2015-12-04 14:47:52 +00:00
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = pws[gid].i[15];
const u32 pw_len = pws[gid].pw_len;
/**
* main
*/
Workaround memory allocation limit from OpenCL by using multiple buffers for scrypt
2016-06-28 09:03:04 +00:00
m12600s (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_scryptV0_buf, d_scryptV1_buf, d_scryptV2_buf, d_scryptV3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, l_bin2asc);
Initial commit
2015-12-04 14:47:52 +00:00
}
Workaround memory allocation limit from OpenCL by using multiple buffers for scrypt
2016-06-28 09:03:04 +00:00
__kernel void m12600_s16 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global 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)
Initial commit
2015-12-04 14:47:52 +00:00
{
/**
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
* modifier
Initial commit
2015-12-04 14:47:52 +00:00
*/
const u32 gid = get_global_id (0);
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
const u32 lid = get_local_id (0);
const u32 lsz = get_local_size (0);
Initial commit
2015-12-04 14:47:52 +00:00
/**
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
* bin2asc table
Initial commit
2015-12-04 14:47:52 +00:00
*/
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
__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) << 8
| ((i1 < 10) ? '0' + i1 : 'A' - 10 + i1) << 0;
}
barrier (CLK_LOCAL_MEM_FENCE);
if (gid >= gid_max) return;
/**
* modifier
*/
Initial commit
2015-12-04 14:47:52 +00:00
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w0[4];
Initial commit
2015-12-04 14:47:52 +00:00
w0[0] = pws[gid].i[ 0];
w0[1] = pws[gid].i[ 1];
w0[2] = pws[gid].i[ 2];
w0[3] = pws[gid].i[ 3];
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w1[4];
Initial commit
2015-12-04 14:47:52 +00:00
w1[0] = pws[gid].i[ 4];
w1[1] = pws[gid].i[ 5];
w1[2] = pws[gid].i[ 6];
w1[3] = pws[gid].i[ 7];
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w2[4];
Initial commit
2015-12-04 14:47:52 +00:00
w2[0] = pws[gid].i[ 8];
w2[1] = pws[gid].i[ 9];
w2[2] = pws[gid].i[10];
w2[3] = pws[gid].i[11];
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 w3[4];
Initial commit
2015-12-04 14:47:52 +00:00
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;
/**
* main
*/
Workaround memory allocation limit from OpenCL by using multiple buffers for scrypt
2016-06-28 09:03:04 +00:00
m12600s (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_scryptV0_buf, d_scryptV1_buf, d_scryptV2_buf, d_scryptV3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, l_bin2asc);
Initial commit
2015-12-04 14:47:52 +00:00
}
Reference in New Issue Copy Permalink