arno/hashcat
1
0
Fork 0
mirror of https://github.com/hashcat/hashcat.git synced 2025-01-06 22:01:15 +00:00
Code Issues Releases Wiki Activity

Fix -a 3 for -m 31400 in vector datatype mode

Browse Source
This commit is contained in:
jsteube 2023-03-18 23:16:01 +00:00
parent d820cfa48e
commit 074411418f
5 changed files with 643 additions and 801 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

194
OpenCL/m31400_a0-pure.cl
View File

@ -22,7 +22,7 @@
DECLSPEC void shift_buffer_by_offset (PRIVATE_AS u32 *w0, const u32 offset) DECLSPEC void shift_buffer_by_offset (PRIVATE_AS u32 *w0, const u32 offset)
{ {
const int offset_switch = offset / 4; const int offset_switch = offset / 4;
#if ((defined IS_AMD || defined IS_HIP) && HAS_VPERM == 0) || defined IS_GENERIC #if ((defined IS_AMD || defined IS_HIP) && HAS_VPERM == 0) || defined IS_GENERIC
switch (offset_switch) switch (offset_switch)
@ -71,7 +71,7 @@ const int offset_switch = offset / 4;
#endif #endif
#if (defined IS_AMD || defined IS_HIP) #if (defined IS_AMD || defined IS_HIP)
const int selector = l32_from_64_S (0x0706050403020100UL >> ((offset & 3) * 8)); const int selector = l32_from_64_S(0x0706050403020100UL >> ((offset & 3) * 8));
#endif #endif
switch (offset_switch) switch (offset_switch)
@ -114,17 +114,16 @@ const int offset_switch = offset / 4;
#endif #endif
} }
DECLSPEC void aes256_scrt_format (PRIVATE_AS u32 *aes_ks, PRIVATE_AS u32 *pw, const u32 pw_len, PRIVATE_AS u32 *hash, PRIVATE_AS u32 *out, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4)
DECLSPEC void aes256_scrt_format (PRIVATE_AS u32 *aes_ks, PRIVATE_AS u32 *pw, const int pw_len, PRIVATE_AS u32 *hash, PRIVATE_AS u32 *out, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4)
{ {
AES256_set_encrypt_key(aes_ks, hash, s_te0, s_te1, s_te2, s_te3); AES256_set_encrypt_key (aes_ks, hash, s_te0, s_te1, s_te2, s_te3);
shift_buffer_by_offset(hash,pw_len+4); shift_buffer_by_offset (hash, pw_len + 4);
hash[0] = hc_swap32_S(pw_len); hash[0] = hc_swap32_S (pw_len);
hash[1] |= hc_swap32_S(pw[0]); hash[1] |= hc_swap32_S (pw[0]);
hash[2] |= hc_swap32_S(pw[1]); hash[2] |= hc_swap32_S (pw[1]);
hash[3] |= hc_swap32_S(pw[2]); hash[3] |= hc_swap32_S (pw[2]);
AES256_encrypt (aes_ks, hash, out, s_te0, s_te1, s_te2, s_te3, s_te4); AES256_encrypt (aes_ks, hash, out, s_te0, s_te1, s_te2, s_te3, s_te4);
} }
@ -143,91 +142,6 @@ KERNEL_FQ void m31400_mxx (KERN_ATTR_RULES ())
* aes shared * aes shared
*/ */
#ifdef REAL_SHM
LOCAL_VK u32 s_te0[256];
LOCAL_VK u32 s_te1[256];
LOCAL_VK u32 s_te2[256];
LOCAL_VK u32 s_te3[256];
LOCAL_VK u32 s_te4[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_te0[i] = te0[i];
s_te1[i] = te1[i];
s_te2[i] = te2[i];
s_te3[i] = te3[i];
s_te4[i] = te4[i];
}
SYNC_THREADS();
#else
CONSTANT_AS u32a *s_te0 = te0;
CONSTANT_AS u32a *s_te1 = te1;
CONSTANT_AS u32a *s_te2 = te2;
CONSTANT_AS u32a *s_te3 = te3;
CONSTANT_AS u32a *s_te4 = te4;
#endif
if (gid >= GID_CNT) return;
/**
* base
*/
COPY_PW (pws[gid]);
/**
* loop
*/
u32 ks[60];
for (u32 il_pos = 0; il_pos < IL_CNT; il_pos++)
{
pw_t tmp = PASTE_PW;
tmp.pw_len = apply_rules (rules_buf[il_pos].cmds, tmp.i, tmp.pw_len);
sha256_ctx_t ctx;
sha256_init (&ctx);
sha256_update_swap (&ctx, tmp.i, tmp.pw_len);
sha256_final (&ctx);
u32 out[4]={0};
aes256_scrt_format(ks,tmp.i,tmp.pw_len,ctx.h,out,s_te0, s_te1, s_te2, s_te3, s_te4);
const u32 r0 = out[DGST_R0];
const u32 r1 = out[DGST_R1];
const u32 r2 = out[DGST_R2];
const u32 r3 = out[DGST_R3];
COMPARE_M_SCALAR (r0, r1, r2, r3);
}
}
KERNEL_FQ void m31400_sxx (KERN_ATTR_RULES ())
{
/**
* modifier
*/
const u64 lid = get_local_id (0);
const u64 gid = get_global_id (0);
const u64 lsz = get_local_size (0);
/**
* aes shared
*/
#ifdef REAL_SHM #ifdef REAL_SHM
LOCAL_VK u32 s_te0[256]; LOCAL_VK u32 s_te0[256];
@ -245,7 +159,91 @@ KERNEL_FQ void m31400_sxx (KERN_ATTR_RULES ())
s_te4[i] = te4[i]; s_te4[i] = te4[i];
} }
SYNC_THREADS (); SYNC_THREADS();
#else
CONSTANT_AS u32a *s_te0 = te0;
CONSTANT_AS u32a *s_te1 = te1;
CONSTANT_AS u32a *s_te2 = te2;
CONSTANT_AS u32a *s_te3 = te3;
CONSTANT_AS u32a *s_te4 = te4;
#endif
if (gid >= GID_CNT) return;
/**
* base
*/
COPY_PW (pws[gid]);
u32 ks[60];
/**
* loop
*/
for (u32 il_pos = 0; il_pos < IL_CNT; il_pos++)
{
pw_t tmp = PASTE_PW;
tmp.pw_len = apply_rules (rules_buf[il_pos].cmds, tmp.i, tmp.pw_len);
sha256_ctx_t ctx;
sha256_init (&ctx);
sha256_update_swap (&ctx, tmp.i, tmp.pw_len);
sha256_final (&ctx);
u32 out[4] = { 0 };
aes256_scrt_format (ks, tmp.i, tmp.pw_len, ctx.h, out,s_te0, s_te1, s_te2, s_te3, s_te4);
const u32 r0 = out[DGST_R0];
const u32 r1 = out[DGST_R1];
const u32 r2 = out[DGST_R2];
const u32 r3 = out[DGST_R3];
COMPARE_M_SCALAR (r0, r1, r2, r3);
}
}
KERNEL_FQ void m31400_sxx (KERN_ATTR_RULES ())
{
/**
* modifier
*/
const u64 lid = get_local_id (0);
const u64 gid = get_global_id (0);
const u64 lsz = get_local_size (0);
/**
* aes shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_te0[256];
LOCAL_VK u32 s_te1[256];
LOCAL_VK u32 s_te2[256];
LOCAL_VK u32 s_te3[256];
LOCAL_VK u32 s_te4[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_te0[i] = te0[i];
s_te1[i] = te1[i];
s_te2[i] = te2[i];
s_te3[i] = te3[i];
s_te4[i] = te4[i];
}
SYNC_THREADS();
#else #else
@ -297,9 +295,9 @@ KERNEL_FQ void m31400_sxx (KERN_ATTR_RULES ())
sha256_final (&ctx); sha256_final (&ctx);
u32 out[4]={0}; u32 out[4] = { 0 };
aes256_scrt_format(ks,tmp.i,tmp.pw_len,ctx.h,out,s_te0, s_te1, s_te2, s_te3, s_te4); aes256_scrt_format (ks, tmp.i, tmp.pw_len, ctx.h, out,s_te0, s_te1, s_te2, s_te3, s_te4);
const u32 r0 = out[DGST_R0]; const u32 r0 = out[DGST_R0];
const u32 r1 = out[DGST_R1]; const u32 r1 = out[DGST_R1];

55
OpenCL/m31400_a1-pure.cl
View File

@ -22,7 +22,7 @@
DECLSPEC void shift_buffer_by_offset (PRIVATE_AS u32 *w0, const u32 offset) DECLSPEC void shift_buffer_by_offset (PRIVATE_AS u32 *w0, const u32 offset)
{ {
const int offset_switch = offset / 4; const int offset_switch = offset / 4;
#if ((defined IS_AMD || defined IS_HIP) && HAS_VPERM == 0) || defined IS_GENERIC #if ((defined IS_AMD || defined IS_HIP) && HAS_VPERM == 0) || defined IS_GENERIC
switch (offset_switch) switch (offset_switch)
@ -71,7 +71,7 @@ const int offset_switch = offset / 4;
#endif #endif
#if (defined IS_AMD || defined IS_HIP) #if (defined IS_AMD || defined IS_HIP)
const int selector = l32_from_64_S (0x0706050403020100UL >> ((offset & 3) * 8)); const int selector = l32_from_64_S(0x0706050403020100UL >> ((offset & 3) * 8));
#endif #endif
switch (offset_switch) switch (offset_switch)
@ -114,17 +114,16 @@ const int offset_switch = offset / 4;
#endif #endif
} }
DECLSPEC void aes256_scrt_format (PRIVATE_AS u32 *aes_ks, PRIVATE_AS u32 *pw, const u32 pw_len, PRIVATE_AS u32 *hash, PRIVATE_AS u32 *out, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4)
DECLSPEC void aes256_scrt_format (PRIVATE_AS u32 *aes_ks, PRIVATE_AS u32 *pw, const int pw_len, PRIVATE_AS u32 *hash, PRIVATE_AS u32 *out, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4)
{ {
AES256_set_encrypt_key(aes_ks, hash, s_te0, s_te1, s_te2, s_te3); AES256_set_encrypt_key (aes_ks, hash, s_te0, s_te1, s_te2, s_te3);
shift_buffer_by_offset(hash,pw_len+4); shift_buffer_by_offset (hash, pw_len + 4);
hash[0] = hc_swap32_S(pw_len); hash[0] = hc_swap32_S (pw_len);
hash[1] |= hc_swap32_S(pw[0]); hash[1] |= hc_swap32_S (pw[0]);
hash[2] |= hc_swap32_S(pw[1]); hash[2] |= hc_swap32_S (pw[1]);
hash[3] |= hc_swap32_S(pw[2]); hash[3] |= hc_swap32_S (pw[2]);
AES256_encrypt (aes_ks, hash, out, s_te0, s_te1, s_te2, s_te3, s_te4); AES256_encrypt (aes_ks, hash, out, s_te0, s_te1, s_te2, s_te3, s_te4);
} }
@ -143,7 +142,7 @@ KERNEL_FQ void m31400_mxx (KERN_ATTR_BASIC ())
* aes shared * aes shared
*/ */
#ifdef REAL_SHM #ifdef REAL_SHM
LOCAL_VK u32 s_te0[256]; LOCAL_VK u32 s_te0[256];
LOCAL_VK u32 s_te1[256]; LOCAL_VK u32 s_te1[256];
@ -162,7 +161,7 @@ KERNEL_FQ void m31400_mxx (KERN_ATTR_BASIC ())
SYNC_THREADS(); SYNC_THREADS();
#else #else
CONSTANT_AS u32a *s_te0 = te0; CONSTANT_AS u32a *s_te0 = te0;
CONSTANT_AS u32a *s_te1 = te1; CONSTANT_AS u32a *s_te1 = te1;
@ -170,7 +169,7 @@ KERNEL_FQ void m31400_mxx (KERN_ATTR_BASIC ())
CONSTANT_AS u32a *s_te3 = te3; CONSTANT_AS u32a *s_te3 = te3;
CONSTANT_AS u32a *s_te4 = te4; CONSTANT_AS u32a *s_te4 = te4;
#endif #endif
if (gid >= GID_CNT) return; if (gid >= GID_CNT) return;
@ -184,12 +183,12 @@ KERNEL_FQ void m31400_mxx (KERN_ATTR_BASIC ())
sha256_update_global_swap (&ctx0, pws[gid].i, pws[gid].pw_len); sha256_update_global_swap (&ctx0, pws[gid].i, pws[gid].pw_len);
u32 ks[60];
/** /**
* loop * loop
*/ */
u32 ks[60];
for (u32 il_pos = 0; il_pos < IL_CNT; il_pos++) for (u32 il_pos = 0; il_pos < IL_CNT; il_pos++)
{ {
sha256_ctx_t ctx = ctx0; sha256_ctx_t ctx = ctx0;
@ -197,16 +196,18 @@ KERNEL_FQ void m31400_mxx (KERN_ATTR_BASIC ())
sha256_update_global_swap (&ctx, combs_buf[il_pos].i, combs_buf[il_pos].pw_len); sha256_update_global_swap (&ctx, combs_buf[il_pos].i, combs_buf[il_pos].pw_len);
u32 pw_candidate[3]; u32 pw_candidate[3];
pw_candidate[0]= hc_swap32_S(ctx.w0[0]);
pw_candidate[1]= hc_swap32_S(ctx.w0[1]); pw_candidate[0] = hc_swap32_S (ctx.w0[0]);
pw_candidate[2]= hc_swap32_S(ctx.w0[2]); pw_candidate[1] = hc_swap32_S (ctx.w0[1]);
pw_candidate[2] = hc_swap32_S (ctx.w0[2]);
u32 pw_len=ctx.len; u32 pw_len=ctx.len;
sha256_final (&ctx); sha256_final (&ctx);
u32 out[4]={0}; u32 out[4] = { 0 };
aes256_scrt_format(ks,pw_candidate,pw_len,ctx.h,out,s_te0, s_te1, s_te2, s_te3, s_te4); aes256_scrt_format (ks, pw_candidate, pw_len, ctx.h, out, s_te0, s_te1, s_te2, s_te3, s_te4);
const u32 r0 = out[DGST_R0]; const u32 r0 = out[DGST_R0];
const u32 r1 = out[DGST_R1]; const u32 r1 = out[DGST_R1];
@ -248,7 +249,7 @@ KERNEL_FQ void m31400_sxx (KERN_ATTR_BASIC ())
s_te4[i] = te4[i]; s_te4[i] = te4[i];
} }
SYNC_THREADS (); SYNC_THREADS();
#else #else
@ -297,16 +298,18 @@ KERNEL_FQ void m31400_sxx (KERN_ATTR_BASIC ())
sha256_update_global_swap (&ctx, combs_buf[il_pos].i, combs_buf[il_pos].pw_len); sha256_update_global_swap (&ctx, combs_buf[il_pos].i, combs_buf[il_pos].pw_len);
u32 pw_candidate[3]; u32 pw_candidate[3];
pw_candidate[0]=hc_swap32_S(ctx.w0[0]);
pw_candidate[1]=hc_swap32_S(ctx.w0[1]); pw_candidate[0] = hc_swap32_S (ctx.w0[0]);
pw_candidate[2]=hc_swap32_S(ctx.w0[2]); pw_candidate[1] = hc_swap32_S (ctx.w0[1]);
pw_candidate[2] = hc_swap32_S (ctx.w0[2]);
u32 pw_len=ctx.len; u32 pw_len=ctx.len;
sha256_final (&ctx); sha256_final (&ctx);
u32 out[4]={0}; u32 out[4] = { 0 };
aes256_scrt_format(ks,pw_candidate,pw_len,ctx.h,out,s_te0, s_te1, s_te2, s_te3, s_te4); aes256_scrt_format (ks, pw_candidate, pw_len, ctx.h, out, s_te0, s_te1, s_te2, s_te3, s_te4);
const u32 r0 = out[DGST_R0]; const u32 r0 = out[DGST_R0];
const u32 r1 = out[DGST_R1]; const u32 r1 = out[DGST_R1];

472
OpenCL/m31400_a3-pure.cl
View File

@ -1,7 +1,7 @@
/** /**
* Author......: See docs/credits.txt * Author......: See docs/credits.txt
* License.....: MIT * License.....: MIT
*/ */
#define NEW_SIMD_CODE #define NEW_SIMD_CODE
@ -20,36 +20,36 @@
#include M2S(INCLUDE_PATH/inc_cipher_aes.cl) #include M2S(INCLUDE_PATH/inc_cipher_aes.cl)
#endif #endif
DECLSPEC void shift_buffer_by_offset(PRIVATE_AS u32 *w0, const u32 offset) DECLSPEC void shift_buffer_by_offset (PRIVATE_AS u32 *w0, const u32 offset)
{ {
const int offset_switch = offset / 4; const int offset_switch = offset / 4;
#if ((defined IS_AMD || defined IS_HIP) && HAS_VPERM == 0) || defined IS_GENERIC #if ((defined IS_AMD || defined IS_HIP) && HAS_VPERM == 0) || defined IS_GENERIC
switch (offset_switch) switch (offset_switch)
{ {
case 0: case 0:
w0[3] = hc_bytealign_be_S(w0[2], w0[3], offset); w0[3] = hc_bytealign_be_S (w0[2], w0[3], offset);
w0[2] = hc_bytealign_be_S(w0[1], w0[2], offset); w0[2] = hc_bytealign_be_S (w0[1], w0[2], offset);
w0[1] = hc_bytealign_be_S(w0[0], w0[1], offset); w0[1] = hc_bytealign_be_S (w0[0], w0[1], offset);
w0[0] = hc_bytealign_be_S(0, w0[0], offset); w0[0] = hc_bytealign_be_S ( 0, w0[0], offset);
break; break;
case 1: case 1:
w0[3] = hc_bytealign_be_S(w0[1], w0[2], offset); w0[3] = hc_bytealign_be_S (w0[1], w0[2], offset);
w0[2] = hc_bytealign_be_S(w0[0], w0[1], offset); w0[2] = hc_bytealign_be_S (w0[0], w0[1], offset);
w0[1] = hc_bytealign_be_S(0, w0[0], offset); w0[1] = hc_bytealign_be_S ( 0, w0[0], offset);
w0[0] = 0; w0[0] = 0;
break; break;
case 2: case 2:
w0[3] = hc_bytealign_be_S(w0[0], w0[1], offset); w0[3] = hc_bytealign_be_S (w0[0], w0[1], offset);
w0[2] = hc_bytealign_be_S(0, w0[0], offset); w0[2] = hc_bytealign_be_S ( 0, w0[0], offset);
w0[1] = 0; w0[1] = 0;
w0[0] = 0; w0[0] = 0;
break; break;
case 3: case 3:
w0[3] = hc_bytealign_be_S(0, w0[0], offset); w0[3] = hc_bytealign_be_S ( 0, w0[0], offset);
w0[2] = 0; w0[2] = 0;
w0[1] = 0; w0[1] = 0;
w0[0] = 0; w0[0] = 0;
@ -62,43 +62,43 @@ DECLSPEC void shift_buffer_by_offset(PRIVATE_AS u32 *w0, const u32 offset)
w0[0] = 0; w0[0] = 0;
break; break;
} }
#endif #endif
#if ((defined IS_AMD || defined IS_HIP) && HAS_VPERM == 1) || defined IS_NV #if ((defined IS_AMD || defined IS_HIP) && HAS_VPERM == 1) || defined IS_NV
#if defined IS_NV #if defined IS_NV
const int selector = (0x76543210 >> ((offset & 3) * 4)) & 0xffff; const int selector = (0x76543210 >> ((offset & 3) * 4)) & 0xffff;
#endif #endif
#if (defined IS_AMD || defined IS_HIP) #if (defined IS_AMD || defined IS_HIP)
const int selector = l32_from_64_S(0x0706050403020100UL >> ((offset & 3) * 8)); const int selector = l32_from_64_S(0x0706050403020100UL >> ((offset & 3) * 8));
#endif #endif
switch (offset_switch) switch (offset_switch)
{ {
case 0: case 0:
w0[3] = hc_byte_perm_S(w0[3], w0[2], selector); w0[3] = hc_byte_perm_S (w0[3], w0[2], selector);
w0[2] = hc_byte_perm_S(w0[2], w0[1], selector); w0[2] = hc_byte_perm_S (w0[2], w0[1], selector);
w0[1] = hc_byte_perm_S(w0[1], w0[0], selector); w0[1] = hc_byte_perm_S (w0[1], w0[0], selector);
w0[0] = hc_byte_perm_S(w0[0], 0, selector); w0[0] = hc_byte_perm_S (w0[0], 0, selector);
break; break;
case 1: case 1:
w0[3] = hc_byte_perm_S(w0[2], w0[1], selector); w0[3] = hc_byte_perm_S (w0[2], w0[1], selector);
w0[2] = hc_byte_perm_S(w0[1], w0[0], selector); w0[2] = hc_byte_perm_S (w0[1], w0[0], selector);
w0[1] = hc_byte_perm_S(w0[0], 0, selector); w0[1] = hc_byte_perm_S (w0[0], 0, selector);
w0[0] = 0; w0[0] = 0;
break; break;
case 2: case 2:
w0[3] = hc_byte_perm_S(w0[1], w0[0], selector); w0[3] = hc_byte_perm_S (w0[1], w0[0], selector);
w0[2] = hc_byte_perm_S(w0[0], 0, selector); w0[2] = hc_byte_perm_S (w0[0], 0, selector);
w0[1] = 0; w0[1] = 0;
w0[0] = 0; w0[0] = 0;
break; break;
case 3: case 3:
w0[3] = hc_byte_perm_S(w0[0], 0, selector); w0[3] = hc_byte_perm_S (w0[0], 0, selector);
w0[2] = 0; w0[2] = 0;
w0[1] = 0; w0[1] = 0;
w0[0] = 0; w0[0] = 0;
@ -111,369 +111,208 @@ DECLSPEC void shift_buffer_by_offset(PRIVATE_AS u32 *w0, const u32 offset)
w0[0] = 0; w0[0] = 0;
break; break;
} }
#endif #endif
} }
DECLSPEC void aes256_scrt_format (PRIVATE_AS u32 *aes_ks, PRIVATE_AS u32 *pw, const u32 pw_len, PRIVATE_AS u32 *hash, PRIVATE_AS u32 *out, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4) DECLSPEC void aes256_scrt_format (PRIVATE_AS u32 *aes_ks, PRIVATE_AS u32 *pw, const u32 pw_len, PRIVATE_AS u32 *hash, PRIVATE_AS u32 *out, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4)
{ {
AES256_set_encrypt_key(aes_ks, hash, s_te0, s_te1, s_te2, s_te3); AES256_set_encrypt_key (aes_ks, hash, s_te0, s_te1, s_te2, s_te3);
shift_buffer_by_offset(hash, pw_len + 4); shift_buffer_by_offset (hash, pw_len + 4);
hash[0] = hc_swap32_S(pw_len); hash[0] = hc_swap32_S (pw_len);
hash[1] |= hc_swap32_S(pw[0]); hash[1] |= hc_swap32_S (pw[0]);
hash[2] |= hc_swap32_S(pw[1]); hash[2] |= hc_swap32_S (pw[1]);
hash[3] |= hc_swap32_S(pw[2]); hash[3] |= hc_swap32_S (pw[2]);
AES256_encrypt(aes_ks, hash, out, s_te0, s_te1, s_te2, s_te3, s_te4); AES256_encrypt (aes_ks, hash, out, s_te0, s_te1, s_te2, s_te3, s_te4);
} }
DECLSPEC void aes256_scrt_format_VV(PRIVATE_AS u32 *aes_ks, PRIVATE_AS u32x *w, const u32 pw_len, PRIVATE_AS u32x *hash, PRIVATE_AS u32x *out, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4) DECLSPEC void aes256_scrt_format_VV (PRIVATE_AS u32 *aes_ks, PRIVATE_AS u32x *w, const u32 pw_len, PRIVATE_AS u32x *h, PRIVATE_AS u32x *out, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4)
{ {
#if VECT_SIZE == 1 #if VECT_SIZE == 1
aes256_scrt_format(aes_ks, w, pw_len, hash, out, s_te0, s_te1, s_te2, s_te3, s_te4); aes256_scrt_format (aes_ks, w, pw_len, h, out, s_te0, s_te1, s_te2, s_te3, s_te4);
#endif #endif
#if VECT_SIZE >= 2 #if VECT_SIZE >= 2
u32 tmp_w[16];
u32 tmp_w[4]; u32 tmp_h[8];
u32 tmp_hash[4];
u32 tmp_out[4]; u32 tmp_out[4];
//s0 //s0
tmp_w[0] = w[0].s0;
tmp_w[1] = w[1].s0;
tmp_w[2] = w[2].s0;
tmp_w[3] = w[3].s0;
tmp_hash[0] = hash[0].s0; for (u32 i = 0; i < 64; i++) tmp_w[i] = w[i].s0;
tmp_hash[1] = hash[1].s0; for (u32 i = 0; i < 8; i++) tmp_h[i] = h[i].s0;
tmp_hash[2] = hash[2].s0;
tmp_hash[3] = hash[3].s0;
aes256_scrt_format (aes_ks, tmp_w, pw_len, tmp_h, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4);
aes256_scrt_format(aes_ks, tmp_w, pw_len, tmp_hash, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4); for (u32 i = 0; i < 4; i++) out[i].s0 = tmp_out[i];
out[0].s0 = tmp_out[0];
out[1].s0 = tmp_out[1];
out[2].s0 = tmp_out[2];
out[3].s0 = tmp_out[3];
//s1 //s1
tmp_w[0] = w[0].s1;
tmp_w[1] = w[1].s1;
tmp_w[2] = w[2].s1;
tmp_w[3] = w[3].s1;
tmp_hash[0] = hash[0].s1; for (u32 i = 0; i < 64; i++) tmp_w[i] = w[i].s1;
tmp_hash[1] = hash[1].s1; for (u32 i = 0; i < 8; i++) tmp_h[i] = h[i].s1;
tmp_hash[2] = hash[2].s1;
tmp_hash[3] = hash[3].s1;
aes256_scrt_format (aes_ks, tmp_w, pw_len, tmp_h, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4);
aes256_scrt_format(aes_ks, tmp_w, pw_len, tmp_hash, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4); for (u32 i = 0; i < 4; i++) out[i].s1 = tmp_out[i];
out[0].s1 = tmp_out[0]; #endif
out[1].s1 = tmp_out[1];
out[2].s1 = tmp_out[2];
out[3].s1 = tmp_out[3];
#if VECT_SIZE >= 4
#endif
#if VECT_SIZE >= 4
//s2 //s2
tmp_w[0] = w[0].s2;
tmp_w[1] = w[1].s2;
tmp_w[2] = w[2].s2;
tmp_w[3] = w[3].s2;
tmp_hash[0] = hash[0].s2; for (u32 i = 0; i < 64; i++) tmp_w[i] = w[i].s2;
tmp_hash[1] = hash[1].s2; for (u32 i = 0; i < 8; i++) tmp_h[i] = h[i].s2;
tmp_hash[2] = hash[2].s2;
tmp_hash[3] = hash[3].s2;
aes256_scrt_format(aes_ks, tmp_w, pw_len, tmp_hash, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4); aes256_scrt_format (aes_ks, tmp_w, pw_len, tmp_h, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4);
out[0].s2 = tmp_out[0];
out[1].s2 = tmp_out[1];
out[2].s2 = tmp_out[2];
out[3].s2 = tmp_out[3];
for (u32 i = 0; i < 4; i++) out[i].s2 = tmp_out[i];
//s3 //s3
tmp_w[0] = w[0].s3;
tmp_w[1] = w[1].s3;
tmp_w[2] = w[2].s3;
tmp_w[3] = w[3].s3;
tmp_hash[0] = hash[0].s3; for (u32 i = 0; i < 64; i++) tmp_w[i] = w[i].s3;
tmp_hash[1] = hash[1].s3; for (u32 i = 0; i < 8; i++) tmp_h[i] = h[i].s3;
tmp_hash[2] = hash[2].s3;
tmp_hash[3] = hash[3].s3;
aes256_scrt_format(aes_ks, tmp_w, pw_len, tmp_hash, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4); aes256_scrt_format (aes_ks, tmp_w, pw_len, tmp_h, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4);
out[0].s3 = tmp_out[0]; for (u32 i = 0; i < 4; i++) out[i].s3 = tmp_out[i];
out[1].s3 = tmp_out[1];
out[2].s3 = tmp_out[2];
out[3].s3 = tmp_out[3];
#endif
#endif #if VECT_SIZE >= 8
#if VECT_SIZE >= 8
//s4 //s4
tmp_w[0] = w[0].s4;
tmp_w[1] = w[1].s4;
tmp_w[2] = w[2].s4;
tmp_w[3] = w[3].s4;
tmp_hash[0] = hash[0].s4; for (u32 i = 0; i < 64; i++) tmp_w[i] = w[i].s4;
tmp_hash[1] = hash[1].s4; for (u32 i = 0; i < 8; i++) tmp_h[i] = h[i].s4;
tmp_hash[2] = hash[2].s4;
tmp_hash[3] = hash[3].s4;
aes256_scrt_format (aes_ks, tmp_w, pw_len, tmp_h, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4);
aes256_scrt_format(aes_ks, tmp_w, pw_len, tmp_hash, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4); for (u32 i = 0; i < 4; i++) out[i].s4 = tmp_out[i];
out[0].s4 = tmp_out[0];
out[1].s4 = tmp_out[1];
out[2].s4 = tmp_out[2];
out[3].s4 = tmp_out[3];
//s5 //s5
tmp_w[0] = w[0].s5;
tmp_w[1] = w[1].s5;
tmp_w[2] = w[2].s5;
tmp_w[3] = w[3].s5;
tmp_hash[0] = hash[0].s5; for (u32 i = 0; i < 64; i++) tmp_w[i] = w[i].s5;
tmp_hash[1] = hash[1].s5; for (u32 i = 0; i < 8; i++) tmp_h[i] = h[i].s5;
tmp_hash[2] = hash[2].s5;
tmp_hash[3] = hash[3].s5;
aes256_scrt_format(aes_ks, tmp_w, pw_len, tmp_hash, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4); aes256_scrt_format (aes_ks, tmp_w, pw_len, tmp_h, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4);
out[0].s5 = tmp_out[0];
out[1].s5 = tmp_out[1];
out[2].s5 = tmp_out[2];
out[3].s5 = tmp_out[3];
for (u32 i = 0; i < 4; i++) out[i].s5 = tmp_out[i];
//s6 //s6
tmp_w[0] = w[0].s6;
tmp_w[1] = w[1].s6;
tmp_w[2] = w[2].s6;
tmp_w[3] = w[3].s6;
tmp_hash[0] = hash[0].s6; for (u32 i = 0; i < 64; i++) tmp_w[i] = w[i].s6;
tmp_hash[1] = hash[1].s6; for (u32 i = 0; i < 8; i++) tmp_h[i] = h[i].s6;
tmp_hash[2] = hash[2].s6;
tmp_hash[3] = hash[3].s6;
aes256_scrt_format(aes_ks, tmp_w, pw_len, tmp_hash, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4); aes256_scrt_format (aes_ks, tmp_w, pw_len, tmp_h, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4);
out[0].s6 = tmp_out[0];
out[1].s6 = tmp_out[1];
out[2].s6 = tmp_out[2];
out[3].s6 = tmp_out[3];
for (u32 i = 0; i < 4; i++) out[i].s6 = tmp_out[i];
//s7 //s7
tmp_w[0] = w[0].s7;
tmp_w[1] = w[1].s7;
tmp_w[2] = w[2].s7;
tmp_w[3] = w[3].s7;
tmp_hash[0] = hash[0].s7; for (u32 i = 0; i < 64; i++) tmp_w[i] = w[i].s7;
tmp_hash[1] = hash[1].s7; for (u32 i = 0; i < 8; i++) tmp_h[i] = h[i].s7;
tmp_hash[2] = hash[2].s7;
tmp_hash[3] = hash[3].s7;
aes256_scrt_format(aes_ks, tmp_w, pw_len, tmp_hash, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4); aes256_scrt_format (aes_ks, tmp_w, pw_len, tmp_h, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4);
out[0].s7 = tmp_out[0]; for (u32 i = 0; i < 4; i++) out[i].s7 = tmp_out[i];
out[1].s7 = tmp_out[1];
out[2].s7 = tmp_out[2];
out[3].s7 = tmp_out[3];
#endif #endif
#if VECT_SIZE >= 16 #if VECT_SIZE >= 16
//s8 //s8
tmp_w[0] = w[0].s8;
tmp_w[1] = w[1].s8;
tmp_w[2] = w[2].s8;
tmp_w[3] = w[3].s8;
tmp_hash[0] = hash[0].s8; for (u32 i = 0; i < 64; i++) tmp_w[i] = w[i].s8;
tmp_hash[1] = hash[1].s8; for (u32 i = 0; i < 8; i++) tmp_h[i] = h[i].s8;
tmp_hash[2] = hash[2].s8;
tmp_hash[3] = hash[3].s8;
aes256_scrt_format(aes_ks, tmp_w, pw_len, tmp_hash, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4); aes256_scrt_format (aes_ks, tmp_w, pw_len, tmp_h, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4);
out[0].s8 = tmp_out[0];
out[1].s8 = tmp_out[1];
out[2].s8 = tmp_out[2];
out[3].s8 = tmp_out[3];
for (u32 i = 0; i < 4; i++) out[i].s8 = tmp_out[i];
//s9 //s9
tmp_w[0] = w[0].s9;
tmp_w[1] = w[1].s9;
tmp_w[2] = w[2].s9;
tmp_w[3] = w[3].s9;
tmp_hash[0] = hash[0].s9; for (u32 i = 0; i < 64; i++) tmp_w[i] = w[i].s9;
tmp_hash[1] = hash[1].s9; for (u32 i = 0; i < 8; i++) tmp_h[i] = h[i].s9;
tmp_hash[2] = hash[2].s9;
tmp_hash[3] = hash[3].s9;
aes256_scrt_format(aes_ks, tmp_w, pw_len, tmp_hash, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4); aes256_scrt_format (aes_ks, tmp_w, pw_len, tmp_h, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4);
out[0].s9 = tmp_out[0]; for (u32 i = 0; i < 4; i++) out[i].s9 = tmp_out[i];
out[1].s9 = tmp_out[1];
out[2].s9 = tmp_out[2];
out[3].s9 = tmp_out[3];
//sa
//s10 for (u32 i = 0; i < 64; i++) tmp_w[i] = w[i].sa;
tmp_w[0] = w[0].sa; for (u32 i = 0; i < 8; i++) tmp_h[i] = h[i].sa;
tmp_w[1] = w[1].sa;
tmp_w[2] = w[2].sa;
tmp_w[3] = w[3].sa;
tmp_hash[0] = hash[0].sa; aes256_scrt_format (aes_ks, tmp_w, pw_len, tmp_h, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4);
tmp_hash[1] = hash[1].sa;
tmp_hash[2] = hash[2].sa;
tmp_hash[3] = hash[3].sa;
for (u32 i = 0; i < 4; i++) out[i].sa = tmp_out[i];
aes256_scrt_format(aes_ks, tmp_w, pw_len, tmp_hash, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4); //sb
out[0].sa = tmp_out[0]; for (u32 i = 0; i < 64; i++) tmp_w[i] = w[i].sb;
out[1].sa = tmp_out[1]; for (u32 i = 0; i < 8; i++) tmp_h[i] = h[i].sb;
out[2].sa = tmp_out[2];
out[3].sa = tmp_out[3];
aes256_scrt_format (aes_ks, tmp_w, pw_len, tmp_h, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4);
//s11 for (u32 i = 0; i < 4; i++) out[i].sb = tmp_out[i];
tmp_w[0] = w[0].sb;
tmp_w[1] = w[1].sb;
tmp_w[2] = w[2].sb;
tmp_w[3] = w[3].sb;
tmp_hash[0] = hash[0].sb; //sc
tmp_hash[1] = hash[1].sb;
tmp_hash[2] = hash[2].sb;
tmp_hash[3] = hash[3].sb;
aes256_scrt_format(aes_ks, tmp_w, pw_len, tmp_hash, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4); for (u32 i = 0; i < 64; i++) tmp_w[i] = w[i].sc;
for (u32 i = 0; i < 8; i++) tmp_h[i] = h[i].sc;
out[0].sb = tmp_out[0]; aes256_scrt_format (aes_ks, tmp_w, pw_len, tmp_h, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4);
out[1].sb = tmp_out[1];
out[2].sb = tmp_out[2];
out[3].sb = tmp_out[3];
for (u32 i = 0; i < 4; i++) out[i].sc = tmp_out[i];
//s12 //sd
tmp_w[0] = w[0].sc;
tmp_w[1] = w[1].sc;
tmp_w[2] = w[2].sc;
tmp_w[3] = w[3].sc;
tmp_hash[0] = hash[0].sc; for (u32 i = 0; i < 64; i++) tmp_w[i] = w[i].sd;
tmp_hash[1] = hash[1].sc; for (u32 i = 0; i < 8; i++) tmp_h[i] = h[i].sd;
tmp_hash[2] = hash[2].sc;
tmp_hash[3] = hash[3].sc;
aes256_scrt_format(aes_ks, tmp_w, pw_len, tmp_hash, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4); aes256_scrt_format (aes_ks, tmp_w, pw_len, tmp_h, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4);
out[0].sc = tmp_out[0]; for (u32 i = 0; i < 4; i++) out[i].sd = tmp_out[i];
out[1].sc = tmp_out[1];
out[2].sc = tmp_out[2];
out[3].sc = tmp_out[3];
//se
//s13 for (u32 i = 0; i < 64; i++) tmp_w[i] = w[i].se;
tmp_w[0] = w[0].sd; for (u32 i = 0; i < 8; i++) tmp_h[i] = h[i].se;
tmp_w[1] = w[1].sd;
tmp_w[2] = w[2].sd;
tmp_w[3] = w[3].sd;
tmp_hash[0] = hash[0].sd; aes256_scrt_format (aes_ks, tmp_w, pw_len, tmp_h, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4);
tmp_hash[1] = hash[1].sd;
tmp_hash[2] = hash[2].sd;
tmp_hash[3] = hash[3].sd;
aes256_scrt_format(aes_ks, tmp_w, pw_len, tmp_hash, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4); for (u32 i = 0; i < 4; i++) out[i].se = tmp_out[i];
out[0].sd = tmp_out[0]; //sf
out[1].sd = tmp_out[1];
out[2].sd = tmp_out[2];
out[3].sd = tmp_out[3];
//s14 for (u32 i = 0; i < 64; i++) tmp_w[i] = w[i].sf;
tmp_w[0] = w[0].se; for (u32 i = 0; i < 8; i++) tmp_h[i] = h[i].sf;
tmp_w[1] = w[1].se;
tmp_w[2] = w[2].se;
tmp_w[3] = w[3].se;
tmp_hash[0] = hash[0].se; aes256_scrt_format (aes_ks, tmp_w, pw_len, tmp_h, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4);
tmp_hash[1] = hash[1].se;
tmp_hash[2] = hash[2].se;
tmp_hash[3] = hash[3].se;
aes256_scrt_format(aes_ks, tmp_w, pw_len, tmp_hash, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4); for (u32 i = 0; i < 4; i++) out[i].sf = tmp_out[i];
out[0].se = tmp_out[0];
out[1].se = tmp_out[1];
out[2].se = tmp_out[2];
out[3].se = tmp_out[3];
//s15
tmp_w[0] = w[0].sf;
tmp_w[1] = w[1].sf;
tmp_w[2] = w[2].sf;
tmp_w[3] = w[3].sf;
tmp_hash[0] = hash[0].sf;
tmp_hash[1] = hash[1].sf;
tmp_hash[2] = hash[2].sf;
tmp_hash[3] = hash[3].sf;
aes256_scrt_format(aes_ks, tmp_w, pw_len, tmp_hash, tmp_out, s_te0, s_te1, s_te2, s_te3, s_te4);
out[0].sf = tmp_out[0];
out[1].sf = tmp_out[1];
out[2].sf = tmp_out[2];
out[3].sf = tmp_out[3];
#endif
#endif
} }
KERNEL_FQ void m31400_mxx(KERN_ATTR_VECTOR()) KERNEL_FQ void m31400_mxx (KERN_ATTR_VECTOR())
{ {
/** /**
* modifier * modifier
*/ */
const u64 lid = get_local_id(0); const u64 lid = get_local_id (0);
const u64 gid = get_global_id(0); const u64 gid = get_global_id (0);
const u64 lsz = get_local_size(0); const u64 lsz = get_local_size (0);
/** /**
* aes shared * aes shared
*/ */
#ifdef REAL_SHM #ifdef REAL_SHM
LOCAL_VK u32 s_te0[256]; LOCAL_VK u32 s_te0[256];
LOCAL_VK u32 s_te1[256]; LOCAL_VK u32 s_te1[256];
@ -492,7 +331,7 @@ KERNEL_FQ void m31400_mxx(KERN_ATTR_VECTOR())
SYNC_THREADS(); SYNC_THREADS();
#else #else
CONSTANT_AS u32a *s_te0 = te0; CONSTANT_AS u32a *s_te0 = te0;
CONSTANT_AS u32a *s_te1 = te1; CONSTANT_AS u32a *s_te1 = te1;
@ -500,10 +339,9 @@ KERNEL_FQ void m31400_mxx(KERN_ATTR_VECTOR())
CONSTANT_AS u32a *s_te3 = te3; CONSTANT_AS u32a *s_te3 = te3;
CONSTANT_AS u32a *s_te4 = te4; CONSTANT_AS u32a *s_te4 = te4;
#endif #endif
if (gid >= GID_CNT) if (gid >= GID_CNT) return;
return;
/** /**
* base * base
@ -518,8 +356,6 @@ KERNEL_FQ void m31400_mxx(KERN_ATTR_VECTOR())
w[idx] = pws[gid].i[idx]; w[idx] = pws[gid].i[idx];
} }
u32 aes_ks[60];
/** /**
* loop * loop
*/ */
@ -536,40 +372,42 @@ KERNEL_FQ void m31400_mxx(KERN_ATTR_VECTOR())
sha256_ctx_vector_t ctx; sha256_ctx_vector_t ctx;
sha256_init_vector(&ctx); sha256_init_vector (&ctx);
sha256_update_vector_swap(&ctx, w, pw_len); sha256_update_vector_swap (&ctx, w, pw_len);
sha256_final_vector(&ctx); sha256_final_vector (&ctx);
u32x out[4] = {0}; u32x out[4] = {0};
aes256_scrt_format_VV(aes_ks, w, pw_len, ctx.h, out, s_te0, s_te1, s_te2, s_te3, s_te4); u32 aes_ks[60];
aes256_scrt_format_VV (aes_ks, w, pw_len, ctx.h, out, s_te0, s_te1, s_te2, s_te3, s_te4);
const u32x r0 = out[DGST_R0]; const u32x r0 = out[DGST_R0];
const u32x r1 = out[DGST_R1]; const u32x r1 = out[DGST_R1];
const u32x r2 = out[DGST_R2]; const u32x r2 = out[DGST_R2];
const u32x r3 = out[DGST_R3]; const u32x r3 = out[DGST_R3];
COMPARE_M_SIMD(r0, r1, r2, r3); COMPARE_M_SIMD (r0, r1, r2, r3);
} }
} }
KERNEL_FQ void m31400_sxx(KERN_ATTR_VECTOR()) KERNEL_FQ void m31400_sxx (KERN_ATTR_VECTOR())
{ {
/** /**
* modifier * modifier
*/ */
const u64 lid = get_local_id(0); const u64 lid = get_local_id (0);
const u64 gid = get_global_id(0); const u64 gid = get_global_id (0);
const u64 lsz = get_local_size(0); const u64 lsz = get_local_size (0);
/** /**
* aes shared * aes shared
*/ */
#ifdef REAL_SHM #ifdef REAL_SHM
LOCAL_VK u32 s_te0[256]; LOCAL_VK u32 s_te0[256];
LOCAL_VK u32 s_te1[256]; LOCAL_VK u32 s_te1[256];
@ -588,7 +426,7 @@ KERNEL_FQ void m31400_sxx(KERN_ATTR_VECTOR())
SYNC_THREADS(); SYNC_THREADS();
#else #else
CONSTANT_AS u32a *s_te0 = te0; CONSTANT_AS u32a *s_te0 = te0;
CONSTANT_AS u32a *s_te1 = te1; CONSTANT_AS u32a *s_te1 = te1;
@ -596,10 +434,10 @@ KERNEL_FQ void m31400_sxx(KERN_ATTR_VECTOR())
CONSTANT_AS u32a *s_te3 = te3; CONSTANT_AS u32a *s_te3 = te3;
CONSTANT_AS u32a *s_te4 = te4; CONSTANT_AS u32a *s_te4 = te4;
#endif #endif
if (gid >= GID_CNT) return;
if (gid >= GID_CNT)
return;
/** /**
* digest * digest
*/ */
@ -609,7 +447,8 @@ KERNEL_FQ void m31400_sxx(KERN_ATTR_VECTOR())
digests_buf[DIGESTS_OFFSET_HOST].digest_buf[DGST_R0], digests_buf[DIGESTS_OFFSET_HOST].digest_buf[DGST_R0],
digests_buf[DIGESTS_OFFSET_HOST].digest_buf[DGST_R1], digests_buf[DIGESTS_OFFSET_HOST].digest_buf[DGST_R1],
digests_buf[DIGESTS_OFFSET_HOST].digest_buf[DGST_R2], digests_buf[DIGESTS_OFFSET_HOST].digest_buf[DGST_R2],
digests_buf[DIGESTS_OFFSET_HOST].digest_buf[DGST_R3]}; digests_buf[DIGESTS_OFFSET_HOST].digest_buf[DGST_R3]
};
/** /**
* base * base
@ -628,7 +467,6 @@ KERNEL_FQ void m31400_sxx(KERN_ATTR_VECTOR())
* loop * loop
*/ */
u32 aes_ks[60];
u32x w0l = w[0]; u32x w0l = w[0];
for (u32 il_pos = 0; il_pos < IL_CNT; il_pos += VECT_SIZE) for (u32 il_pos = 0; il_pos < IL_CNT; il_pos += VECT_SIZE)
@ -641,21 +479,23 @@ KERNEL_FQ void m31400_sxx(KERN_ATTR_VECTOR())
sha256_ctx_vector_t ctx; sha256_ctx_vector_t ctx;
sha256_init_vector(&ctx); sha256_init_vector (&ctx);
sha256_update_vector_swap(&ctx, w, pw_len); sha256_update_vector_swap (&ctx, w, pw_len);
sha256_final_vector(&ctx); sha256_final_vector (&ctx);
u32x out[4] = {0}; u32x out[4] = {0};
aes256_scrt_format_VV(aes_ks, w, pw_len, ctx.h, out, s_te0, s_te1, s_te2, s_te3, s_te4); u32 aes_ks[60];
aes256_scrt_format_VV (aes_ks, w, pw_len, ctx.h, out, s_te0, s_te1, s_te2, s_te3, s_te4);
const u32x r0 = out[DGST_R0]; const u32x r0 = out[DGST_R0];
const u32x r1 = out[DGST_R1]; const u32x r1 = out[DGST_R1];
const u32x r2 = out[DGST_R2]; const u32x r2 = out[DGST_R2];
const u32x r3 = out[DGST_R3]; const u32x r3 = out[DGST_R3];
COMPARE_S_SIMD(r0, r1, r2, r3); COMPARE_S_SIMD (r0, r1, r2, r3);
} }
} }

1
docs/changes.txt
View File

@ -20,6 +20,7 @@
- Added hash-mode: GPG (AES-128/AES-256 (SHA-256($pass))) - Added hash-mode: GPG (AES-128/AES-256 (SHA-256($pass)))
- Added hash-mode: GPG (AES-128/AES-256 (SHA-512($pass))) - Added hash-mode: GPG (AES-128/AES-256 (SHA-512($pass)))
- Added hash-mode: MetaMask Wallet (short hash, plaintext check) - Added hash-mode: MetaMask Wallet (short hash, plaintext check)
- Added hash-mode: SecureCRT MasterPassphrase v2
- Added hash-mode: Veeam VB - Added hash-mode: Veeam VB
- Added hash-mode: bcrypt(sha256($pass)) - Added hash-mode: bcrypt(sha256($pass))
- Added hash-mode: HMAC-RIPEMD160 (key = $pass) - Added hash-mode: HMAC-RIPEMD160 (key = $pass)

2
src/modules/module_31400.c
View File

@ -18,7 +18,7 @@ static const u32 DGST_POS0 = 0;
static const u32 DGST_POS1 = 1; static const u32 DGST_POS1 = 1;
static const u32 DGST_POS2 = 2; static const u32 DGST_POS2 = 2;
static const u32 DGST_POS3 = 3; static const u32 DGST_POS3 = 3;
static const u32 DGST_SIZE = DGST_SIZE_4_8; static const u32 DGST_SIZE = DGST_SIZE_4_4;
static const u32 HASH_CATEGORY = HASH_CATEGORY_RAW_HASH; static const u32 HASH_CATEGORY = HASH_CATEGORY_RAW_HASH;
static const char *HASH_NAME = "SecureCRT MasterPassphrase v2"; static const char *HASH_NAME = "SecureCRT MasterPassphrase v2";
static const u64 KERN_TYPE = 31400; static const u64 KERN_TYPE = 31400;