From b3d3b31c3e6af6bbeb2212c7fb63f79e7593bc3e Mon Sep 17 00:00:00 2001 From: Gabriele Gristina Date: Thu, 10 Feb 2022 21:53:08 +0100 Subject: [PATCH] Metal: added support for vectors up to 4 --- OpenCL/inc_hash_blake2b.cl | 4 +- OpenCL/inc_hash_md4.h | 2 +- OpenCL/inc_hash_md5.h | 2 +- OpenCL/inc_hash_ripemd160.h | 4 +- OpenCL/inc_hash_sha1.h | 2 +- OpenCL/inc_hash_sha224.h | 2 +- OpenCL/inc_hash_sha256.h | 2 +- OpenCL/inc_hash_sha384.h | 2 +- OpenCL/inc_hash_sha512.h | 2 +- OpenCL/inc_platform.cl | 40 ---- OpenCL/inc_simd.h | 338 +++++++++++++++++----------------- OpenCL/inc_types.h | 8 + OpenCL/inc_vendor.h | 10 + OpenCL/m00300_a0-optimized.cl | 20 +- OpenCL/m00300_a1-optimized.cl | 20 +- OpenCL/m00300_a3-optimized.cl | 20 +- OpenCL/m01100_a0-optimized.cl | 16 +- OpenCL/m01100_a1-optimized.cl | 16 +- OpenCL/m01100_a3-optimized.cl | 16 +- OpenCL/m02610_a0-optimized.cl | 16 +- OpenCL/m02610_a1-optimized.cl | 16 +- OpenCL/m02610_a3-optimized.cl | 16 +- OpenCL/m02710_a0-optimized.cl | 32 ++-- OpenCL/m02710_a1-optimized.cl | 32 ++-- OpenCL/m02710_a3-optimized.cl | 32 ++-- OpenCL/m02810_a0-optimized.cl | 32 ++-- OpenCL/m02810_a1-optimized.cl | 32 ++-- OpenCL/m02810_a3-optimized.cl | 32 ++-- OpenCL/m03500_a0-optimized.cl | 32 ++-- OpenCL/m03500_a1-optimized.cl | 32 ++-- OpenCL/m03500_a3-optimized.cl | 32 ++-- OpenCL/m03710_a0-optimized.cl | 16 +- OpenCL/m03710_a1-optimized.cl | 16 +- OpenCL/m03710_a3-optimized.cl | 16 +- OpenCL/m03910_a0-optimized.cl | 32 ++-- OpenCL/m03910_a1-optimized.cl | 32 ++-- OpenCL/m03910_a3-optimized.cl | 32 ++-- OpenCL/m04010_a0-optimized.cl | 16 +- OpenCL/m04010_a1-optimized.cl | 16 +- OpenCL/m04010_a3-optimized.cl | 16 +- OpenCL/m04110_a0-optimized.cl | 16 +- OpenCL/m04110_a1-optimized.cl | 16 +- OpenCL/m04110_a3-optimized.cl | 16 +- OpenCL/m04310_a0-optimized.cl | 16 +- OpenCL/m04310_a1-optimized.cl | 16 +- OpenCL/m04310_a3-optimized.cl | 16 +- OpenCL/m04400_a0-optimized.cl | 20 +- OpenCL/m04400_a1-optimized.cl | 20 +- OpenCL/m04400_a3-optimized.cl | 20 +- OpenCL/m04500_a0-optimized.cl | 20 +- OpenCL/m04500_a1-optimized.cl | 20 +- OpenCL/m04500_a3-optimized.cl | 20 +- OpenCL/m04510_a0-optimized.cl | 20 +- OpenCL/m04510_a1-optimized.cl | 20 +- OpenCL/m04510_a3-optimized.cl | 20 +- OpenCL/m04520_a0-optimized.cl | 40 ++-- OpenCL/m04520_a1-optimized.cl | 40 ++-- OpenCL/m04520_a3-optimized.cl | 40 ++-- OpenCL/m04700_a0-optimized.cl | 16 +- OpenCL/m04700_a1-optimized.cl | 16 +- OpenCL/m04700_a3-optimized.cl | 16 +- OpenCL/m04710_a0-optimized.cl | 16 +- OpenCL/m04710_a1-optimized.cl | 16 +- OpenCL/m04710_a3-optimized.cl | 16 +- OpenCL/m05000_a0-optimized.cl | 20 +- OpenCL/m05000_a1-optimized.cl | 20 +- OpenCL/m05000_a3-optimized.cl | 20 +- OpenCL/m05100_a0-optimized.cl | 16 +- OpenCL/m05100_a1-optimized.cl | 16 +- OpenCL/m05100_a3-optimized.cl | 16 +- OpenCL/m05500_a0-optimized.cl | 20 +- OpenCL/m05500_a1-optimized.cl | 20 +- OpenCL/m05500_a3-optimized.cl | 20 +- OpenCL/m09900_a0-optimized.cl | 16 +- OpenCL/m09900_a1-optimized.cl | 16 +- OpenCL/m09900_a3-optimized.cl | 16 +- OpenCL/m11000_a0-optimized.cl | 16 +- OpenCL/m11000_a1-optimized.cl | 16 +- OpenCL/m11000_a3-optimized.cl | 16 +- OpenCL/m11100_a0-optimized.cl | 16 +- OpenCL/m11100_a1-optimized.cl | 16 +- OpenCL/m11100_a3-optimized.cl | 16 +- OpenCL/m11200_a0-optimized.cl | 60 +++--- OpenCL/m11200_a1-optimized.cl | 60 +++--- OpenCL/m11200_a3-optimized.cl | 60 +++--- OpenCL/m12600_a0-optimized.cl | 20 +- OpenCL/m12600_a1-optimized.cl | 20 +- OpenCL/m12600_a3-optimized.cl | 20 +- OpenCL/m13300_a0-optimized.cl | 16 +- OpenCL/m13300_a1-optimized.cl | 16 +- OpenCL/m13300_a3-optimized.cl | 16 +- OpenCL/m15500_a0-optimized.cl | 20 +- OpenCL/m15500_a1-optimized.cl | 20 +- OpenCL/m15500_a3-optimized.cl | 20 +- OpenCL/m20710_a0-optimized.cl | 48 ++--- OpenCL/m20710_a1-optimized.cl | 48 ++--- OpenCL/m20710_a3-optimized.cl | 50 ++--- OpenCL/m20800_a0-optimized.cl | 16 +- OpenCL/m20800_a1-optimized.cl | 16 +- OpenCL/m20800_a3-optimized.cl | 16 +- OpenCL/m20900_a0-optimized.cl | 52 +++--- OpenCL/m20900_a1-optimized.cl | 52 +++--- OpenCL/m20900_a3-optimized.cl | 54 +++--- OpenCL/m21100_a0-optimized.cl | 16 +- OpenCL/m21100_a1-optimized.cl | 16 +- OpenCL/m21100_a3-optimized.cl | 16 +- OpenCL/m21200_a0-optimized.cl | 32 ++-- OpenCL/m21200_a1-optimized.cl | 32 ++-- OpenCL/m21200_a3-optimized.cl | 34 ++-- OpenCL/m21400_a0-optimized.cl | 32 ++-- OpenCL/m21400_a1-optimized.cl | 32 ++-- OpenCL/m21400_a3-optimized.cl | 32 ++-- OpenCL/m24300_a0-optimized.cl | 40 ++-- OpenCL/m24300_a1-optimized.cl | 40 ++-- OpenCL/m24300_a3-optimized.cl | 40 ++-- OpenCL/m24700_a0-optimized.cl | 16 +- OpenCL/m24700_a1-optimized.cl | 16 +- OpenCL/m24700_a3-optimized.cl | 16 +- OpenCL/m24900_a0-optimized.cl | 16 +- OpenCL/m24900_a1-optimized.cl | 16 +- OpenCL/m24900_a3-optimized.cl | 16 +- src/backend.c | 7 + src/ext_metal.m | 9 +- 123 files changed, 1512 insertions(+), 1520 deletions(-) diff --git a/OpenCL/inc_hash_blake2b.cl b/OpenCL/inc_hash_blake2b.cl index 6f6146e77..9df986d1c 100644 --- a/OpenCL/inc_hash_blake2b.cl +++ b/OpenCL/inc_hash_blake2b.cl @@ -601,9 +601,9 @@ DECLSPEC void blake2b_transform_vector (PRIVATE_AS u64x *h, PRIVATE_AS const u64 v[ 9] = BLAKE2B_IV_01; v[10] = BLAKE2B_IV_02; v[11] = BLAKE2B_IV_03; - v[12] = BLAKE2B_IV_04 ^ t0; + v[12] = make_u64x (BLAKE2B_IV_04) ^ t0; v[13] = BLAKE2B_IV_05; // ^ t1; - v[14] = BLAKE2B_IV_06 ^ f0; + v[14] = make_u64x (BLAKE2B_IV_06) ^ f0; v[15] = BLAKE2B_IV_07; // ^ f1; BLAKE2B_ROUND_VECTOR ( 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15); diff --git a/OpenCL/inc_hash_md4.h b/OpenCL/inc_hash_md4.h index a4fa88eb9..31f607e8f 100644 --- a/OpenCL/inc_hash_md4.h +++ b/OpenCL/inc_hash_md4.h @@ -31,7 +31,7 @@ #define MD4_STEP(f,a,b,c,d,x,K,s) \ { \ - a += K; \ + a += make_u32x (K); \ a = hc_add3 (a, x, f (b, c, d)); \ a = hc_rotl32 (a, s); \ } diff --git a/OpenCL/inc_hash_md5.h b/OpenCL/inc_hash_md5.h index ae2e40793..bb9fff32f 100644 --- a/OpenCL/inc_hash_md5.h +++ b/OpenCL/inc_hash_md5.h @@ -36,7 +36,7 @@ #define MD5_STEP(f,a,b,c,d,x,K,s) \ { \ - a += K; \ + a += make_u32x (K); \ a = hc_add3 (a, x, f (b, c, d)); \ a = hc_rotl32 (a, s); \ a += b; \ diff --git a/OpenCL/inc_hash_ripemd160.h b/OpenCL/inc_hash_ripemd160.h index 15e8d1cb7..7450c8cc5 100644 --- a/OpenCL/inc_hash_ripemd160.h +++ b/OpenCL/inc_hash_ripemd160.h @@ -32,7 +32,7 @@ #define RIPEMD160_STEP(f,a,b,c,d,e,x,K,s) \ { \ - a += K; \ + a += make_u32x (K); \ a += x; \ a += f (b, c, d); \ a = hc_rotl32 (a, s); \ @@ -54,7 +54,7 @@ #define RIPEMD160_STEP_WORKAROUND_BUG(f,a,b,c,d,e,x,K,s) \ { \ - a += K; \ + a += make_u32x (K); \ a += x; \ a += f (b, c, d); \ a = ROTATE_LEFT_WORKAROUND_BUG (a, s); \ diff --git a/OpenCL/inc_hash_sha1.h b/OpenCL/inc_hash_sha1.h index 947ff8e68..2d537975e 100644 --- a/OpenCL/inc_hash_sha1.h +++ b/OpenCL/inc_hash_sha1.h @@ -28,7 +28,7 @@ #define SHA1_STEP(f,a,b,c,d,e,x) \ { \ - e += K; \ + e += make_u32x (K); \ e = hc_add3 (e, x, f (b, c, d)); \ e += hc_rotl32 (a, 5u); \ b = hc_rotl32 (b, 30u); \ diff --git a/OpenCL/inc_hash_sha224.h b/OpenCL/inc_hash_sha224.h index e497a0536..71aa8d443 100644 --- a/OpenCL/inc_hash_sha224.h +++ b/OpenCL/inc_hash_sha224.h @@ -41,7 +41,7 @@ #define SHA224_STEP(F0,F1,a,b,c,d,e,f,g,h,x,K) \ { \ - h = hc_add3 (h, K, x); \ + h = hc_add3 (h, make_u32x (K), x); \ h = hc_add3 (h, SHA224_S3 (e), F1 (e,f,g)); \ d += h; \ h = hc_add3 (h, SHA224_S2 (a), F0 (a,b,c)); \ diff --git a/OpenCL/inc_hash_sha256.h b/OpenCL/inc_hash_sha256.h index 018c1024e..e1104121d 100644 --- a/OpenCL/inc_hash_sha256.h +++ b/OpenCL/inc_hash_sha256.h @@ -41,7 +41,7 @@ #define SHA256_STEP(F0,F1,a,b,c,d,e,f,g,h,x,K) \ { \ - h = hc_add3 (h, K, x); \ + h = hc_add3 (h, make_u32x (K), x); \ h = hc_add3 (h, SHA256_S3 (e), F1 (e,f,g)); \ d += h; \ h = hc_add3 (h, SHA256_S2 (a), F0 (a,b,c)); \ diff --git a/OpenCL/inc_hash_sha384.h b/OpenCL/inc_hash_sha384.h index 802168fb5..882421756 100644 --- a/OpenCL/inc_hash_sha384.h +++ b/OpenCL/inc_hash_sha384.h @@ -44,7 +44,7 @@ #define SHA384_STEP(F0,F1,a,b,c,d,e,f,g,h,x,K) \ { \ - h += K; \ + h += make_u64x (K); \ h += x; \ h += SHA384_S1 (e); \ h += F0 (e, f, g); \ diff --git a/OpenCL/inc_hash_sha512.h b/OpenCL/inc_hash_sha512.h index 94001474a..af2b1bf8b 100644 --- a/OpenCL/inc_hash_sha512.h +++ b/OpenCL/inc_hash_sha512.h @@ -44,7 +44,7 @@ #define SHA512_STEP(F0,F1,a,b,c,d,e,f,g,h,x,K) \ { \ - h += K; \ + h += make_u64x (K); \ h += x; \ h += SHA512_S1 (e); \ h += F0 (e, f, g); \ diff --git a/OpenCL/inc_platform.cl b/OpenCL/inc_platform.cl index a3fdd392b..06ba76553 100644 --- a/OpenCL/inc_platform.cl +++ b/OpenCL/inc_platform.cl @@ -326,46 +326,6 @@ DECLSPEC u32 hc_atomic_or (volatile GLOBAL_AS u32 *p, volatile const u32 val) return atomic_fetch_or_explicit (pd, val, memory_order_relaxed); } -DECLSPEC u32x rotl32 (const u32x a, const int n) -{ - return ((a << n) | ((a >> (32 - n)))); -} - -DECLSPEC u32x rotr32 (const u32x a, const int n) -{ - return ((a >> n) | ((a << (32 - n)))); -} - -DECLSPEC u32 rotl32_S (const u32 a, const int n) -{ - return ((a << n) | ((a >> (32 - n)))); -} - -DECLSPEC u32 rotr32_S (const u32 a, const int n) -{ - return ((a >> n) | ((a << (32 - n)))); -} - -DECLSPEC u64x rotl64 (const u64x a, const int n) -{ - return ((a << n) | ((a >> (64 - n)))); -} - -DECLSPEC u64x rotr64 (const u64x a, const int n) -{ - return ((a >> n) | ((a << (64 - n)))); -} - -DECLSPEC u64 rotl64_S (const u64 a, const int n) -{ - return ((a << n) | ((a >> (64 - n)))); -} - -DECLSPEC u64 rotr64_S (const u64 a, const int n) -{ - return ((a >> n) | ((a << (64 - n)))); -} - #define FIXED_THREAD_COUNT(n) #define SYNC_THREADS() threadgroup_barrier (mem_flags::mem_threadgroup) #endif // IS_METAL diff --git a/OpenCL/inc_simd.h b/OpenCL/inc_simd.h index 836f6e197..91d4084bd 100644 --- a/OpenCL/inc_simd.h +++ b/OpenCL/inc_simd.h @@ -17,11 +17,11 @@ { \ if (((h0) == search[0]) && ((h1) == search[1]) && ((h2) == search[2]) && ((h3) == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0) \ + if (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0) \ { \ - mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos, 0, 0); \ + mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos, 0, 0); \ } \ } \ } @@ -37,13 +37,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp0, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp0, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0) \ + if (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos, 0, 0); \ } \ @@ -66,9 +66,9 @@ { \ if (((h0).s0 == search[0]) && ((h1).s0 == search[1]) && ((h2).s0 == search[2]) && ((h3).s0 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 0) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 0) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 0, 0, 0); \ } \ @@ -76,9 +76,9 @@ \ if (((h0).s1 == search[0]) && ((h1).s1 == search[1]) && ((h2).s1 == search[2]) && ((h3).s1 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 1) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 1) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 1, 0, 0); \ } \ @@ -97,13 +97,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp0, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp0, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 0) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 0) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 0, 0, 0); \ } \ @@ -117,13 +117,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp1, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp1, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 1) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 1) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 1, 0, 0); \ } \ @@ -144,9 +144,9 @@ { \ if (((h0).s0 == search[0]) && ((h1).s0 == search[1]) && ((h2).s0 == search[2]) && ((h3).s0 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 0) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 0) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 0, 0, 0); \ } \ @@ -154,9 +154,9 @@ \ if (((h0).s1 == search[0]) && ((h1).s1 == search[1]) && ((h2).s1 == search[2]) && ((h3).s1 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 1) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 1) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 1, 0, 0); \ } \ @@ -164,9 +164,9 @@ \ if (((h0).s2 == search[0]) && ((h1).s2 == search[1]) && ((h2).s2 == search[2]) && ((h3).s2 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 2) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 2) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 2, 0, 0); \ } \ @@ -174,9 +174,9 @@ \ if (((h0).s3 == search[0]) && ((h1).s3 == search[1]) && ((h2).s3 == search[2]) && ((h3).s3 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 3) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 3) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 3, 0, 0); \ } \ @@ -197,13 +197,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp0, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp0, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 0) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 0) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 0, 0, 0); \ } \ @@ -217,13 +217,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp1, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp1, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 1) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 1) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 1, 0, 0); \ } \ @@ -237,13 +237,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp2, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp2, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 2) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 2) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 2, 0, 0); \ } \ @@ -257,13 +257,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp3, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp3, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 3) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 3) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 3, 0, 0); \ } \ @@ -284,9 +284,9 @@ { \ if (((h0).s0 == search[0]) && ((h1).s0 == search[1]) && ((h2).s0 == search[2]) && ((h3).s0 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 0) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 0) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 0, 0, 0); \ } \ @@ -294,9 +294,9 @@ \ if (((h0).s1 == search[0]) && ((h1).s1 == search[1]) && ((h2).s1 == search[2]) && ((h3).s1 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 1) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 1) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 1, 0, 0); \ } \ @@ -304,9 +304,9 @@ \ if (((h0).s2 == search[0]) && ((h1).s2 == search[1]) && ((h2).s2 == search[2]) && ((h3).s2 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 2) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 2) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 2, 0, 0); \ } \ @@ -314,18 +314,18 @@ \ if (((h0).s3 == search[0]) && ((h1).s3 == search[1]) && ((h2).s3 == search[2]) && ((h3).s3 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 3) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 3) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 3, 0, 0); \ } \ } \ if (((h0).s4 == search[0]) && ((h1).s4 == search[1]) && ((h2).s4 == search[2]) && ((h3).s4 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 4) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 4) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 4, 0, 0); \ } \ @@ -333,9 +333,9 @@ \ if (((h0).s5 == search[0]) && ((h1).s5 == search[1]) && ((h2).s5 == search[2]) && ((h3).s5 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 5) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 5) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 5, 0, 0); \ } \ @@ -343,9 +343,9 @@ \ if (((h0).s6 == search[0]) && ((h1).s6 == search[1]) && ((h2).s6 == search[2]) && ((h3).s6 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 6) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 6) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 6, 0, 0); \ } \ @@ -353,9 +353,9 @@ \ if (((h0).s7 == search[0]) && ((h1).s7 == search[1]) && ((h2).s7 == search[2]) && ((h3).s7 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 7) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 7) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 7, 0, 0); \ } \ @@ -380,13 +380,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp0, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp0, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 0) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 0) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 0, 0, 0); \ } \ @@ -400,13 +400,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp1, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp1, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 1) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 1) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 1, 0, 0); \ } \ @@ -420,13 +420,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp2, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp2, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 2) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 2) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 2, 0, 0); \ } \ @@ -440,13 +440,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp3, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp3, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 3) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 3) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 3, 0, 0); \ } \ @@ -459,13 +459,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp4, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp4, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 4) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 4) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 4, 0, 0); \ } \ @@ -479,13 +479,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp5, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp5, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 5) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 5) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 5, 0, 0); \ } \ @@ -499,13 +499,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp6, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp6, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 6) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 6) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 6, 0, 0); \ } \ @@ -519,13 +519,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp7, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp7, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 7) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 7) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 7, 0, 0); \ } \ @@ -546,9 +546,9 @@ { \ if (((h0).s0 == search[0]) && ((h1).s0 == search[1]) && ((h2).s0 == search[2]) && ((h3).s0 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 0) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 0) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 0, 0, 0); \ } \ @@ -556,9 +556,9 @@ \ if (((h0).s1 == search[0]) && ((h1).s1 == search[1]) && ((h2).s1 == search[2]) && ((h3).s1 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 1) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 1) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 1, 0, 0); \ } \ @@ -566,9 +566,9 @@ \ if (((h0).s2 == search[0]) && ((h1).s2 == search[1]) && ((h2).s2 == search[2]) && ((h3).s2 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 2) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 2) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 2, 0, 0); \ } \ @@ -576,18 +576,18 @@ \ if (((h0).s3 == search[0]) && ((h1).s3 == search[1]) && ((h2).s3 == search[2]) && ((h3).s3 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 3) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 3) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 3, 0, 0); \ } \ } \ if (((h0).s4 == search[0]) && ((h1).s4 == search[1]) && ((h2).s4 == search[2]) && ((h3).s4 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 4) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 4) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 4, 0, 0); \ } \ @@ -595,9 +595,9 @@ \ if (((h0).s5 == search[0]) && ((h1).s5 == search[1]) && ((h2).s5 == search[2]) && ((h3).s5 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 5) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 5) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 5, 0, 0); \ } \ @@ -605,9 +605,9 @@ \ if (((h0).s6 == search[0]) && ((h1).s6 == search[1]) && ((h2).s6 == search[2]) && ((h3).s6 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 6) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 6) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 6, 0, 0); \ } \ @@ -615,9 +615,9 @@ \ if (((h0).s7 == search[0]) && ((h1).s7 == search[1]) && ((h2).s7 == search[2]) && ((h3).s7 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 7) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 7) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 7, 0, 0); \ } \ @@ -625,9 +625,9 @@ \ if (((h0).s8 == search[0]) && ((h1).s8 == search[1]) && ((h2).s8 == search[2]) && ((h3).s8 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 8) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 8) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 8, 0, 0); \ } \ @@ -635,9 +635,9 @@ \ if (((h0).s9 == search[0]) && ((h1).s9 == search[1]) && ((h2).s9 == search[2]) && ((h3).s9 == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 9) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 9) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 9, 0, 0); \ } \ @@ -645,9 +645,9 @@ \ if (((h0).sa == search[0]) && ((h1).sa == search[1]) && ((h2).sa == search[2]) && ((h3).sa == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 10) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 10) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 10, 0, 0); \ } \ @@ -655,9 +655,9 @@ \ if (((h0).sb == search[0]) && ((h1).sb == search[1]) && ((h2).sb == search[2]) && ((h3).sb == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 11) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 11) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 11, 0, 0); \ } \ @@ -665,9 +665,9 @@ \ if (((h0).sc == search[0]) && ((h1).sc == search[1]) && ((h2).sc == search[2]) && ((h3).sc == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 12) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 12) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 12, 0, 0); \ } \ @@ -675,9 +675,9 @@ \ if (((h0).sd == search[0]) && ((h1).sd == search[1]) && ((h2).sd == search[2]) && ((h3).sd == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 13) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 13) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 13, 0, 0); \ } \ @@ -685,9 +685,9 @@ \ if (((h0).se == search[0]) && ((h1).se == search[1]) && ((h2).se == search[2]) && ((h3).se == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 14) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 14) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 14, 0, 0); \ } \ @@ -695,9 +695,9 @@ \ if (((h0).sf == search[0]) && ((h1).sf == search[1]) && ((h2).sf == search[2]) && ((h3).sf == search[3])) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + 0; \ \ - if (vector_accessible (il_pos, IL_CNT, 15) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 15) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, 0, final_hash_pos, gid, il_pos + 15, 0, 0); \ } \ @@ -730,13 +730,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp00, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp00, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 0) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 0) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 0, 0, 0); \ } \ @@ -750,13 +750,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp01, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp01, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 1) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 1) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 1, 0, 0); \ } \ @@ -770,13 +770,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp02, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp02, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 2) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 2) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 2, 0, 0); \ } \ @@ -790,13 +790,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp03, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp03, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 3) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 3) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 3, 0, 0); \ } \ @@ -810,13 +810,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp04, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp04, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 4) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 4) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 4, 0, 0); \ } \ @@ -830,13 +830,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp05, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp05, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 5) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 5) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 5, 0, 0); \ } \ @@ -850,13 +850,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp06, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp06, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 6) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 6) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 6, 0, 0); \ } \ @@ -870,13 +870,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp07, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp07, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 7) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 7) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 7, 0, 0); \ } \ @@ -890,13 +890,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp08, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp08, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 8) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 8) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 8, 0, 0); \ } \ @@ -910,13 +910,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp09, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp09, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 9) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 9) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 9, 0, 0); \ } \ @@ -930,13 +930,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp10, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp10, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 10) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 10) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 10, 0, 0); \ } \ @@ -950,13 +950,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp11, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp11, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 11) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 11) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 11, 0, 0); \ } \ @@ -970,13 +970,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp12, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp12, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 12) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 12) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 12, 0, 0); \ } \ @@ -990,13 +990,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp13, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp13, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 13) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 13) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 13, 0, 0); \ } \ @@ -1010,13 +1010,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp14, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp14, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 14) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 14) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 14, 0, 0); \ } \ @@ -1030,13 +1030,13 @@ BITMAP_SHIFT1, \ BITMAP_SHIFT2)) \ { \ - int digest_pos = find_hash (digest_tp15, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ + int digest_pos = find_hash (digest_tp15, DIGESTS_CNT, &digests_buf[DIGESTS_OFFSET_HOST]); \ \ if (digest_pos != -1) \ { \ - const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ + const u32 final_hash_pos = DIGESTS_OFFSET_HOST + digest_pos; \ \ - if (vector_accessible (il_pos, IL_CNT, 15) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ + if (vector_accessible (il_pos, IL_CNT, 15) && (hc_atomic_inc (&hashes_shown[final_hash_pos]) == 0)) \ { \ mark_hash (plains_buf, d_return_buf, SALT_POS_HOST, DIGESTS_CNT, digest_pos, final_hash_pos, gid, il_pos + 15, 0, 0); \ } \ @@ -1162,21 +1162,21 @@ #elif VECT_SIZE == 2 #define VECTOR_ELEMENT(v, n) (n == 0 ? (v).s0 : (v).s1) #elif VECT_SIZE == 4 -#define VECTOR_ELEMENT(v, n) \ - (n < 2 ? \ - (n == 0 ? (v).s0 : (v).s1) : \ - (n == 2 ? (v).s2 : (v).s3) \ +#define VECTOR_ELEMENT(v, n) \ + (n < 2 ? \ + (n == 0 ? (v).s0 : (v).s1) : \ + (n == 2 ? (v).s2 : (v).s3) \ ) #elif VECT_SIZE == 8 -#define VECTOR_ELEMENT(v, n) \ - (n < 4 ? \ - (n < 2 ? \ - (n == 0 ? (v).s0 : (v).s1) : \ - (n == 2 ? (v).s2 : (v).s3) \ - ) : (n < 6 ? \ - (n == 4 ? (v).s4 : (v).s5) : \ - (n == 6 ? (v).s6 : (v).s7) \ - ) \ +#define VECTOR_ELEMENT(v, n) \ + (n < 4 ? \ + (n < 2 ? \ + (n == 0 ? (v).s0 : (v).s1) : \ + (n == 2 ? (v).s2 : (v).s3) \ + ) : (n < 6 ? \ + (n == 4 ? (v).s4 : (v).s5) : \ + (n == 6 ? (v).s6 : (v).s7) \ + ) \ ) #elif VECT_SIZE == 16 #define VECTOR_ELEMENT(v, n) \ diff --git a/OpenCL/inc_types.h b/OpenCL/inc_types.h index 918e3a7f9..2c7ae1f9d 100644 --- a/OpenCL/inc_types.h +++ b/OpenCL/inc_types.h @@ -109,6 +109,7 @@ typedef u64 u64x; #define make_u64x (u64) #else + #if defined IS_CUDA || defined IS_HIP #if VECT_SIZE == 2 @@ -871,10 +872,17 @@ typedef VTYPE(ushort, VECT_SIZE) u16x; typedef VTYPE(uint, VECT_SIZE) u32x; typedef VTYPE(ullong, VECT_SIZE) u64x; +#ifndef IS_METAL #define make_u8x (u8x) #define make_u16x (u16x) #define make_u32x (u32x) #define make_u64x (u64x) +#else +#define make_u8x u8x +#define make_u16x u16x +#define make_u32x u32x +#define make_u64x u64x +#endif #endif #endif diff --git a/OpenCL/inc_vendor.h b/OpenCL/inc_vendor.h index ac13a7c23..ec7e27c47 100644 --- a/OpenCL/inc_vendor.h +++ b/OpenCL/inc_vendor.h @@ -197,4 +197,14 @@ using namespace metal; //#define USE_SWIZZLE #endif +#ifdef IS_METAL +#define USE_ROTATE + +// Metal support max VECT_SIZE = 4 +#define s0 x +#define s1 y +#define s2 z +#define s3 w #endif + +#endif // _INC_VENDOR_H diff --git a/OpenCL/m00300_a0-optimized.cl b/OpenCL/m00300_a0-optimized.cl index 4f29cda35..594d190dc 100644 --- a/OpenCL/m00300_a0-optimized.cl +++ b/OpenCL/m00300_a0-optimized.cl @@ -184,11 +184,11 @@ KERNEL_FQ void m00300_m04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); w0_t = a; w1_t = b; @@ -507,11 +507,11 @@ KERNEL_FQ void m00300_s04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); w0_t = a; w1_t = b; diff --git a/OpenCL/m00300_a1-optimized.cl b/OpenCL/m00300_a1-optimized.cl index ef836bda8..77b0ffb0c 100644 --- a/OpenCL/m00300_a1-optimized.cl +++ b/OpenCL/m00300_a1-optimized.cl @@ -240,11 +240,11 @@ KERNEL_FQ void m00300_m04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); w0_t = a; w1_t = b; @@ -621,11 +621,11 @@ KERNEL_FQ void m00300_s04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); w0_t = a; w1_t = b; diff --git a/OpenCL/m00300_a3-optimized.cl b/OpenCL/m00300_a3-optimized.cl index 9e5e2b064..185cd9938 100644 --- a/OpenCL/m00300_a3-optimized.cl +++ b/OpenCL/m00300_a3-optimized.cl @@ -251,11 +251,11 @@ DECLSPEC void m00300m (PRIVATE_AS u32 *w, const u32 pw_len, KERN_ATTR_FUNC_VECTO SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_78s ^ w0s07 ^ w0s08 ^ w0s15 ^ w0s18 ^ w0s20)); SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_79s ^ w0s08 ^ w0s22)); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); u32x w0_t = a; u32x w1_t = b; @@ -635,11 +635,11 @@ DECLSPEC void m00300s (PRIVATE_AS u32 *w, const u32 pw_len, KERN_ATTR_FUNC_VECTO SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_78s ^ w0s07 ^ w0s08 ^ w0s15 ^ w0s18 ^ w0s20)); SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_79s ^ w0s08 ^ w0s22)); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); u32x w0_t = a; u32x w1_t = b; diff --git a/OpenCL/m01100_a0-optimized.cl b/OpenCL/m01100_a0-optimized.cl index 6d0c99eae..9c8b4194f 100644 --- a/OpenCL/m01100_a0-optimized.cl +++ b/OpenCL/m01100_a0-optimized.cl @@ -145,10 +145,10 @@ KERNEL_FQ void m01100_m04 (KERN_ATTR_RULES ()) MD4_STEP (MD4_H , c, d, a, b, w1[3], MD4C02, MD4S22); MD4_STEP (MD4_H , b, c, d, a, w3[3], MD4C02, MD4S23); - a += MD4M_A; - b += MD4M_B; - c += MD4M_C; - d += MD4M_D; + a += make_u32x (MD4M_A); + b += make_u32x (MD4M_B); + c += make_u32x (MD4M_C); + d += make_u32x (MD4M_D); w0[0] = a; w0[1] = b; @@ -376,10 +376,10 @@ KERNEL_FQ void m01100_s04 (KERN_ATTR_RULES ()) MD4_STEP (MD4_H , c, d, a, b, w1[3], MD4C02, MD4S22); MD4_STEP (MD4_H , b, c, d, a, w3[3], MD4C02, MD4S23); - a += MD4M_A; - b += MD4M_B; - c += MD4M_C; - d += MD4M_D; + a += make_u32x (MD4M_A); + b += make_u32x (MD4M_B); + c += make_u32x (MD4M_C); + d += make_u32x (MD4M_D); w0[0] = a; w0[1] = b; diff --git a/OpenCL/m01100_a1-optimized.cl b/OpenCL/m01100_a1-optimized.cl index bd9bbb3a7..72e6d71fa 100644 --- a/OpenCL/m01100_a1-optimized.cl +++ b/OpenCL/m01100_a1-optimized.cl @@ -205,10 +205,10 @@ KERNEL_FQ void m01100_m04 (KERN_ATTR_BASIC ()) MD4_STEP (MD4_H , c, d, a, b, w1[3], MD4C02, MD4S22); MD4_STEP (MD4_H , b, c, d, a, w3[3], MD4C02, MD4S23); - a += MD4M_A; - b += MD4M_B; - c += MD4M_C; - d += MD4M_D; + a += make_u32x (MD4M_A); + b += make_u32x (MD4M_B); + c += make_u32x (MD4M_C); + d += make_u32x (MD4M_D); w0[0] = a; w0[1] = b; @@ -498,10 +498,10 @@ KERNEL_FQ void m01100_s04 (KERN_ATTR_BASIC ()) MD4_STEP (MD4_H , c, d, a, b, w1[3], MD4C02, MD4S22); MD4_STEP (MD4_H , b, c, d, a, w3[3], MD4C02, MD4S23); - a += MD4M_A; - b += MD4M_B; - c += MD4M_C; - d += MD4M_D; + a += make_u32x (MD4M_A); + b += make_u32x (MD4M_B); + c += make_u32x (MD4M_C); + d += make_u32x (MD4M_D); w0[0] = a; w0[1] = b; diff --git a/OpenCL/m01100_a3-optimized.cl b/OpenCL/m01100_a3-optimized.cl index 38548d52c..d26ac1708 100644 --- a/OpenCL/m01100_a3-optimized.cl +++ b/OpenCL/m01100_a3-optimized.cl @@ -159,10 +159,10 @@ DECLSPEC void m01100m (LOCAL_AS salt_t *s_salt_buf, PRIVATE_AS u32 *w, const u32 MD4_STEP0(MD4_H , c, d, a, b, H_w7c02, MD4S22); MD4_STEP0(MD4_H , b, c, d, a, H_wfc02, MD4S23); - a += MD4M_A; - b += MD4M_B; - c += MD4M_C; - d += MD4M_D; + a += make_u32x (MD4M_A); + b += make_u32x (MD4M_B); + c += make_u32x (MD4M_C); + d += make_u32x (MD4M_D); u32x w0_t[4]; u32x w1_t[4]; @@ -403,10 +403,10 @@ DECLSPEC void m01100s (LOCAL_AS salt_t *s_salt_buf, PRIVATE_AS u32 *w, const u32 MD4_STEP0(MD4_H , c, d, a, b, H_w7c02, MD4S22); MD4_STEP0(MD4_H , b, c, d, a, H_wfc02, MD4S23); - a += MD4M_A; - b += MD4M_B; - c += MD4M_C; - d += MD4M_D; + a += make_u32x (MD4M_A); + b += make_u32x (MD4M_B); + c += make_u32x (MD4M_C); + d += make_u32x (MD4M_D); u32x w0_t[4]; u32x w1_t[4]; diff --git a/OpenCL/m02610_a0-optimized.cl b/OpenCL/m02610_a0-optimized.cl index 2c056e460..47ccd84a5 100644 --- a/OpenCL/m02610_a0-optimized.cl +++ b/OpenCL/m02610_a0-optimized.cl @@ -196,10 +196,10 @@ KERNEL_FQ void m02610_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -492,10 +492,10 @@ KERNEL_FQ void m02610_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; diff --git a/OpenCL/m02610_a1-optimized.cl b/OpenCL/m02610_a1-optimized.cl index 77cc1c038..9f05b58d8 100644 --- a/OpenCL/m02610_a1-optimized.cl +++ b/OpenCL/m02610_a1-optimized.cl @@ -253,10 +253,10 @@ KERNEL_FQ void m02610_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -608,10 +608,10 @@ KERNEL_FQ void m02610_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; diff --git a/OpenCL/m02610_a3-optimized.cl b/OpenCL/m02610_a3-optimized.cl index 6acbcf832..bdbf35318 100644 --- a/OpenCL/m02610_a3-optimized.cl +++ b/OpenCL/m02610_a3-optimized.cl @@ -173,10 +173,10 @@ DECLSPEC void m02610m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -440,10 +440,10 @@ DECLSPEC void m02610s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; diff --git a/OpenCL/m02710_a0-optimized.cl b/OpenCL/m02710_a0-optimized.cl index 730f87dc9..99fca1af1 100644 --- a/OpenCL/m02710_a0-optimized.cl +++ b/OpenCL/m02710_a0-optimized.cl @@ -196,10 +196,10 @@ KERNEL_FQ void m02710_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -300,10 +300,10 @@ KERNEL_FQ void m02710_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - const u32x r_a = a + MD5M_A; - const u32x r_b = b + MD5M_B; - const u32x r_c = c + MD5M_C; - const u32x r_d = d + MD5M_D; + const u32x r_a = a + make_u32x (MD5M_A); + const u32x r_b = b + make_u32x (MD5M_B); + const u32x r_c = c + make_u32x (MD5M_C); + const u32x r_d = d + make_u32x (MD5M_D); const u32x r_14 = (32 + salt_len) * 8; @@ -577,10 +577,10 @@ KERNEL_FQ void m02710_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -681,10 +681,10 @@ KERNEL_FQ void m02710_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - const u32x r_a = a + MD5M_A; - const u32x r_b = b + MD5M_B; - const u32x r_c = c + MD5M_C; - const u32x r_d = d + MD5M_D; + const u32x r_a = a + make_u32x (MD5M_A); + const u32x r_b = b + make_u32x (MD5M_B); + const u32x r_c = c + make_u32x (MD5M_C); + const u32x r_d = d + make_u32x (MD5M_D); const u32x r_14 = (32 + salt_len) * 8; diff --git a/OpenCL/m02710_a1-optimized.cl b/OpenCL/m02710_a1-optimized.cl index f2bdf0873..5e8b2d4fe 100644 --- a/OpenCL/m02710_a1-optimized.cl +++ b/OpenCL/m02710_a1-optimized.cl @@ -253,10 +253,10 @@ KERNEL_FQ void m02710_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -357,10 +357,10 @@ KERNEL_FQ void m02710_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - const u32x r_a = a + MD5M_A; - const u32x r_b = b + MD5M_B; - const u32x r_c = c + MD5M_C; - const u32x r_d = d + MD5M_D; + const u32x r_a = a + make_u32x (MD5M_A); + const u32x r_b = b + make_u32x (MD5M_B); + const u32x r_c = c + make_u32x (MD5M_C); + const u32x r_d = d + make_u32x (MD5M_D); const u32x r_14 = (32 + salt_len) * 8; @@ -693,10 +693,10 @@ KERNEL_FQ void m02710_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -797,10 +797,10 @@ KERNEL_FQ void m02710_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - const u32x r_a = a + MD5M_A; - const u32x r_b = b + MD5M_B; - const u32x r_c = c + MD5M_C; - const u32x r_d = d + MD5M_D; + const u32x r_a = a + make_u32x (MD5M_A); + const u32x r_b = b + make_u32x (MD5M_B); + const u32x r_c = c + make_u32x (MD5M_C); + const u32x r_d = d + make_u32x (MD5M_D); const u32x r_14 = (32 + salt_len) * 8; diff --git a/OpenCL/m02710_a3-optimized.cl b/OpenCL/m02710_a3-optimized.cl index 674c2c9ac..3c8404772 100644 --- a/OpenCL/m02710_a3-optimized.cl +++ b/OpenCL/m02710_a3-optimized.cl @@ -173,10 +173,10 @@ DECLSPEC void m02710m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -277,10 +277,10 @@ DECLSPEC void m02710m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - const u32x r_a = a + MD5M_A; - const u32x r_b = b + MD5M_B; - const u32x r_c = c + MD5M_C; - const u32x r_d = d + MD5M_D; + const u32x r_a = a + make_u32x (MD5M_A); + const u32x r_b = b + make_u32x (MD5M_B); + const u32x r_c = c + make_u32x (MD5M_C); + const u32x r_d = d + make_u32x (MD5M_D); const u32x r_14 = (32 + salt_len) * 8; @@ -525,10 +525,10 @@ DECLSPEC void m02710s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -629,10 +629,10 @@ DECLSPEC void m02710s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - const u32x r_a = a + MD5M_A; - const u32x r_b = b + MD5M_B; - const u32x r_c = c + MD5M_C; - const u32x r_d = d + MD5M_D; + const u32x r_a = a + make_u32x (MD5M_A); + const u32x r_b = b + make_u32x (MD5M_B); + const u32x r_c = c + make_u32x (MD5M_C); + const u32x r_d = d + make_u32x (MD5M_D); const u32x r_14 = (32 + salt_len) * 8; diff --git a/OpenCL/m02810_a0-optimized.cl b/OpenCL/m02810_a0-optimized.cl index d8e783d83..38853d0e2 100644 --- a/OpenCL/m02810_a0-optimized.cl +++ b/OpenCL/m02810_a0-optimized.cl @@ -194,10 +194,10 @@ KERNEL_FQ void m02810_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = salt_buf0[0]; w0[1] = salt_buf0[1]; @@ -298,10 +298,10 @@ KERNEL_FQ void m02810_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - const u32x r_a = a + MD5M_A; - const u32x r_b = b + MD5M_B; - const u32x r_c = c + MD5M_C; - const u32x r_d = d + MD5M_D; + const u32x r_a = a + make_u32x (MD5M_A); + const u32x r_b = b + make_u32x (MD5M_B); + const u32x r_c = c + make_u32x (MD5M_C); + const u32x r_d = d + make_u32x (MD5M_D); const u32x r_00 = 0x80; const u32x r_14 = 64 * 8; @@ -574,10 +574,10 @@ KERNEL_FQ void m02810_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = salt_buf0[0]; w0[1] = salt_buf0[1]; @@ -678,10 +678,10 @@ KERNEL_FQ void m02810_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - const u32x r_a = a + MD5M_A; - const u32x r_b = b + MD5M_B; - const u32x r_c = c + MD5M_C; - const u32x r_d = d + MD5M_D; + const u32x r_a = a + make_u32x (MD5M_A); + const u32x r_b = b + make_u32x (MD5M_B); + const u32x r_c = c + make_u32x (MD5M_C); + const u32x r_d = d + make_u32x (MD5M_D); const u32x r_00 = 0x80; const u32x r_14 = 64 * 8; diff --git a/OpenCL/m02810_a1-optimized.cl b/OpenCL/m02810_a1-optimized.cl index 11af80b72..5b61d0e25 100644 --- a/OpenCL/m02810_a1-optimized.cl +++ b/OpenCL/m02810_a1-optimized.cl @@ -251,10 +251,10 @@ KERNEL_FQ void m02810_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = salt_buf0[0]; w0[1] = salt_buf0[1]; @@ -355,10 +355,10 @@ KERNEL_FQ void m02810_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - const u32x r_a = a + MD5M_A; - const u32x r_b = b + MD5M_B; - const u32x r_c = c + MD5M_C; - const u32x r_d = d + MD5M_D; + const u32x r_a = a + make_u32x (MD5M_A); + const u32x r_b = b + make_u32x (MD5M_B); + const u32x r_c = c + make_u32x (MD5M_C); + const u32x r_d = d + make_u32x (MD5M_D); const u32x r_00 = 0x80; const u32x r_14 = 64 * 8; @@ -690,10 +690,10 @@ KERNEL_FQ void m02810_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = salt_buf0[0]; w0[1] = salt_buf0[1]; @@ -794,10 +794,10 @@ KERNEL_FQ void m02810_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - const u32x r_a = a + MD5M_A; - const u32x r_b = b + MD5M_B; - const u32x r_c = c + MD5M_C; - const u32x r_d = d + MD5M_D; + const u32x r_a = a + make_u32x (MD5M_A); + const u32x r_b = b + make_u32x (MD5M_B); + const u32x r_c = c + make_u32x (MD5M_C); + const u32x r_d = d + make_u32x (MD5M_D); const u32x r_00 = 0x80; const u32x r_14 = 64 * 8; diff --git a/OpenCL/m02810_a3-optimized.cl b/OpenCL/m02810_a3-optimized.cl index 0933ed04a..5f80b2be3 100644 --- a/OpenCL/m02810_a3-optimized.cl +++ b/OpenCL/m02810_a3-optimized.cl @@ -171,10 +171,10 @@ DECLSPEC void m02810m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = salt_buf0[0]; w0_t[1] = salt_buf0[1]; @@ -275,10 +275,10 @@ DECLSPEC void m02810m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - const u32x r_a = a + MD5M_A; - const u32x r_b = b + MD5M_B; - const u32x r_c = c + MD5M_C; - const u32x r_d = d + MD5M_D; + const u32x r_a = a + make_u32x (MD5M_A); + const u32x r_b = b + make_u32x (MD5M_B); + const u32x r_c = c + make_u32x (MD5M_C); + const u32x r_d = d + make_u32x (MD5M_D); const u32x r_00 = 0x80; const u32x r_14 = 64 * 8; @@ -522,10 +522,10 @@ DECLSPEC void m02810s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = salt_buf0[0]; w0_t[1] = salt_buf0[1]; @@ -626,10 +626,10 @@ DECLSPEC void m02810s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - const u32x r_a = a + MD5M_A; - const u32x r_b = b + MD5M_B; - const u32x r_c = c + MD5M_C; - const u32x r_d = d + MD5M_D; + const u32x r_a = a + make_u32x (MD5M_A); + const u32x r_b = b + make_u32x (MD5M_B); + const u32x r_c = c + make_u32x (MD5M_C); + const u32x r_d = d + make_u32x (MD5M_D); const u32x r_00 = 0x80; const u32x r_14 = 64 * 8; diff --git a/OpenCL/m03500_a0-optimized.cl b/OpenCL/m03500_a0-optimized.cl index e01f881e9..b24cdecb0 100644 --- a/OpenCL/m03500_a0-optimized.cl +++ b/OpenCL/m03500_a0-optimized.cl @@ -196,10 +196,10 @@ KERNEL_FQ void m03500_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -302,10 +302,10 @@ KERNEL_FQ void m03500_m04 (KERN_ATTR_RULES ()) //STEP 3 - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -598,10 +598,10 @@ KERNEL_FQ void m03500_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -704,10 +704,10 @@ KERNEL_FQ void m03500_s04 (KERN_ATTR_RULES ()) //STEP3 - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; diff --git a/OpenCL/m03500_a1-optimized.cl b/OpenCL/m03500_a1-optimized.cl index febe29ada..707718fce 100644 --- a/OpenCL/m03500_a1-optimized.cl +++ b/OpenCL/m03500_a1-optimized.cl @@ -253,10 +253,10 @@ KERNEL_FQ void m03500_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -359,10 +359,10 @@ KERNEL_FQ void m03500_m04 (KERN_ATTR_BASIC ()) //STEP3 - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -714,10 +714,10 @@ KERNEL_FQ void m03500_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -820,10 +820,10 @@ KERNEL_FQ void m03500_s04 (KERN_ATTR_BASIC ()) //STEP3 - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; diff --git a/OpenCL/m03500_a3-optimized.cl b/OpenCL/m03500_a3-optimized.cl index 2cd096bca..1c7972e41 100644 --- a/OpenCL/m03500_a3-optimized.cl +++ b/OpenCL/m03500_a3-optimized.cl @@ -173,10 +173,10 @@ DECLSPEC void m03500m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -279,10 +279,10 @@ DECLSPEC void m03500m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w //STEP 3 - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -546,10 +546,10 @@ DECLSPEC void m03500s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -652,10 +652,10 @@ DECLSPEC void m03500s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w //STEP 3 - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; diff --git a/OpenCL/m03710_a0-optimized.cl b/OpenCL/m03710_a0-optimized.cl index 3294edd01..54498ca59 100644 --- a/OpenCL/m03710_a0-optimized.cl +++ b/OpenCL/m03710_a0-optimized.cl @@ -198,10 +198,10 @@ KERNEL_FQ void m03710_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -525,10 +525,10 @@ KERNEL_FQ void m03710_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; diff --git a/OpenCL/m03710_a1-optimized.cl b/OpenCL/m03710_a1-optimized.cl index 39bc4fe62..c4254ae8d 100644 --- a/OpenCL/m03710_a1-optimized.cl +++ b/OpenCL/m03710_a1-optimized.cl @@ -255,10 +255,10 @@ KERNEL_FQ void m03710_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -641,10 +641,10 @@ KERNEL_FQ void m03710_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; diff --git a/OpenCL/m03710_a3-optimized.cl b/OpenCL/m03710_a3-optimized.cl index ac9b92139..c2d71c934 100644 --- a/OpenCL/m03710_a3-optimized.cl +++ b/OpenCL/m03710_a3-optimized.cl @@ -175,10 +175,10 @@ DECLSPEC void m03710m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -473,10 +473,10 @@ DECLSPEC void m03710s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; diff --git a/OpenCL/m03910_a0-optimized.cl b/OpenCL/m03910_a0-optimized.cl index f3c090135..e8a305f72 100644 --- a/OpenCL/m03910_a0-optimized.cl +++ b/OpenCL/m03910_a0-optimized.cl @@ -194,10 +194,10 @@ KERNEL_FQ void m03910_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -298,10 +298,10 @@ KERNEL_FQ void m03910_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - const u32x r_a = a + MD5M_A; - const u32x r_b = b + MD5M_B; - const u32x r_c = c + MD5M_C; - const u32x r_d = d + MD5M_D; + const u32x r_a = a + make_u32x (MD5M_A); + const u32x r_b = b + make_u32x (MD5M_B); + const u32x r_c = c + make_u32x (MD5M_C); + const u32x r_d = d + make_u32x (MD5M_D); const u32x r_00 = 0x80; const u32x r_14 = 64 * 8; @@ -574,10 +574,10 @@ KERNEL_FQ void m03910_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -678,10 +678,10 @@ KERNEL_FQ void m03910_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - const u32x r_a = a + MD5M_A; - const u32x r_b = b + MD5M_B; - const u32x r_c = c + MD5M_C; - const u32x r_d = d + MD5M_D; + const u32x r_a = a + make_u32x (MD5M_A); + const u32x r_b = b + make_u32x (MD5M_B); + const u32x r_c = c + make_u32x (MD5M_C); + const u32x r_d = d + make_u32x (MD5M_D); const u32x r_00 = 0x80; const u32x r_14 = 64 * 8; diff --git a/OpenCL/m03910_a1-optimized.cl b/OpenCL/m03910_a1-optimized.cl index 12edbb197..c759bde95 100644 --- a/OpenCL/m03910_a1-optimized.cl +++ b/OpenCL/m03910_a1-optimized.cl @@ -251,10 +251,10 @@ KERNEL_FQ void m03910_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -355,10 +355,10 @@ KERNEL_FQ void m03910_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - const u32x r_a = a + MD5M_A; - const u32x r_b = b + MD5M_B; - const u32x r_c = c + MD5M_C; - const u32x r_d = d + MD5M_D; + const u32x r_a = a + make_u32x (MD5M_A); + const u32x r_b = b + make_u32x (MD5M_B); + const u32x r_c = c + make_u32x (MD5M_C); + const u32x r_d = d + make_u32x (MD5M_D); const u32x r_00 = 0x80; const u32x r_14 = 64 * 8; @@ -690,10 +690,10 @@ KERNEL_FQ void m03910_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -794,10 +794,10 @@ KERNEL_FQ void m03910_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - const u32x r_a = a + MD5M_A; - const u32x r_b = b + MD5M_B; - const u32x r_c = c + MD5M_C; - const u32x r_d = d + MD5M_D; + const u32x r_a = a + make_u32x (MD5M_A); + const u32x r_b = b + make_u32x (MD5M_B); + const u32x r_c = c + make_u32x (MD5M_C); + const u32x r_d = d + make_u32x (MD5M_D); const u32x r_00 = 0x80; const u32x r_14 = 64 * 8; diff --git a/OpenCL/m03910_a3-optimized.cl b/OpenCL/m03910_a3-optimized.cl index eb966866f..9137008a4 100644 --- a/OpenCL/m03910_a3-optimized.cl +++ b/OpenCL/m03910_a3-optimized.cl @@ -171,10 +171,10 @@ DECLSPEC void m03910m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -275,10 +275,10 @@ DECLSPEC void m03910m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - const u32x r_a = a + MD5M_A; - const u32x r_b = b + MD5M_B; - const u32x r_c = c + MD5M_C; - const u32x r_d = d + MD5M_D; + const u32x r_a = a + make_u32x (MD5M_A); + const u32x r_b = b + make_u32x (MD5M_B); + const u32x r_c = c + make_u32x (MD5M_C); + const u32x r_d = d + make_u32x (MD5M_D); const u32x r_00 = 0x80; const u32x r_14 = 64 * 8; @@ -522,10 +522,10 @@ DECLSPEC void m03910s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -626,10 +626,10 @@ DECLSPEC void m03910s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - const u32x r_a = a + MD5M_A; - const u32x r_b = b + MD5M_B; - const u32x r_c = c + MD5M_C; - const u32x r_d = d + MD5M_D; + const u32x r_a = a + make_u32x (MD5M_A); + const u32x r_b = b + make_u32x (MD5M_B); + const u32x r_c = c + make_u32x (MD5M_C); + const u32x r_d = d + make_u32x (MD5M_D); const u32x r_00 = 0x80; const u32x r_14 = 64 * 8; diff --git a/OpenCL/m04010_a0-optimized.cl b/OpenCL/m04010_a0-optimized.cl index 0cd84e0b3..596fa9a0c 100644 --- a/OpenCL/m04010_a0-optimized.cl +++ b/OpenCL/m04010_a0-optimized.cl @@ -224,10 +224,10 @@ KERNEL_FQ void m04010_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -577,10 +577,10 @@ KERNEL_FQ void m04010_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; diff --git a/OpenCL/m04010_a1-optimized.cl b/OpenCL/m04010_a1-optimized.cl index 6028d3caa..d50efb929 100644 --- a/OpenCL/m04010_a1-optimized.cl +++ b/OpenCL/m04010_a1-optimized.cl @@ -280,10 +280,10 @@ KERNEL_FQ void m04010_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -691,10 +691,10 @@ KERNEL_FQ void m04010_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; diff --git a/OpenCL/m04010_a3-optimized.cl b/OpenCL/m04010_a3-optimized.cl index b40375523..6a52abb00 100644 --- a/OpenCL/m04010_a3-optimized.cl +++ b/OpenCL/m04010_a3-optimized.cl @@ -195,10 +195,10 @@ DECLSPEC void m04010m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -513,10 +513,10 @@ DECLSPEC void m04010s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; diff --git a/OpenCL/m04110_a0-optimized.cl b/OpenCL/m04110_a0-optimized.cl index b81017557..c2e04ac36 100644 --- a/OpenCL/m04110_a0-optimized.cl +++ b/OpenCL/m04110_a0-optimized.cl @@ -269,10 +269,10 @@ KERNEL_FQ void m04110_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -667,10 +667,10 @@ KERNEL_FQ void m04110_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; diff --git a/OpenCL/m04110_a1-optimized.cl b/OpenCL/m04110_a1-optimized.cl index 7e1222c44..cba0aac04 100644 --- a/OpenCL/m04110_a1-optimized.cl +++ b/OpenCL/m04110_a1-optimized.cl @@ -327,10 +327,10 @@ KERNEL_FQ void m04110_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -785,10 +785,10 @@ KERNEL_FQ void m04110_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; diff --git a/OpenCL/m04110_a3-optimized.cl b/OpenCL/m04110_a3-optimized.cl index ff185f8f7..c01dfa5ae 100644 --- a/OpenCL/m04110_a3-optimized.cl +++ b/OpenCL/m04110_a3-optimized.cl @@ -223,10 +223,10 @@ DECLSPEC void m04110m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -569,10 +569,10 @@ DECLSPEC void m04110s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; diff --git a/OpenCL/m04310_a0-optimized.cl b/OpenCL/m04310_a0-optimized.cl index cf8c6ebc3..3dca3c0a4 100644 --- a/OpenCL/m04310_a0-optimized.cl +++ b/OpenCL/m04310_a0-optimized.cl @@ -196,10 +196,10 @@ KERNEL_FQ void m04310_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_upper8 ((a >> 0) & 255) << 0 | uint_to_hex_upper8 ((a >> 8) & 255) << 16; @@ -492,10 +492,10 @@ KERNEL_FQ void m04310_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_upper8 ((a >> 0) & 255) << 0 | uint_to_hex_upper8 ((a >> 8) & 255) << 16; diff --git a/OpenCL/m04310_a1-optimized.cl b/OpenCL/m04310_a1-optimized.cl index f37914982..b6128578c 100644 --- a/OpenCL/m04310_a1-optimized.cl +++ b/OpenCL/m04310_a1-optimized.cl @@ -253,10 +253,10 @@ KERNEL_FQ void m04310_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_upper8 ((a >> 0) & 255) << 0 | uint_to_hex_upper8 ((a >> 8) & 255) << 16; @@ -608,10 +608,10 @@ KERNEL_FQ void m04310_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_upper8 ((a >> 0) & 255) << 0 | uint_to_hex_upper8 ((a >> 8) & 255) << 16; diff --git a/OpenCL/m04310_a3-optimized.cl b/OpenCL/m04310_a3-optimized.cl index 8bfeda5e7..0e2cb3e45 100644 --- a/OpenCL/m04310_a3-optimized.cl +++ b/OpenCL/m04310_a3-optimized.cl @@ -173,10 +173,10 @@ DECLSPEC void m04310m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = uint_to_hex_upper8 ((a >> 0) & 255) << 0 | uint_to_hex_upper8 ((a >> 8) & 255) << 16; @@ -440,10 +440,10 @@ DECLSPEC void m04310s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = uint_to_hex_upper8 ((a >> 0) & 255) << 0 | uint_to_hex_upper8 ((a >> 8) & 255) << 16; diff --git a/OpenCL/m04400_a0-optimized.cl b/OpenCL/m04400_a0-optimized.cl index cbf088c08..22a622863 100644 --- a/OpenCL/m04400_a0-optimized.cl +++ b/OpenCL/m04400_a0-optimized.cl @@ -214,11 +214,11 @@ KERNEL_FQ void m04400_m04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * md5 @@ -536,11 +536,11 @@ KERNEL_FQ void m04400_s04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * md5 diff --git a/OpenCL/m04400_a1-optimized.cl b/OpenCL/m04400_a1-optimized.cl index 05bd0e5f7..152fd04e7 100644 --- a/OpenCL/m04400_a1-optimized.cl +++ b/OpenCL/m04400_a1-optimized.cl @@ -270,11 +270,11 @@ KERNEL_FQ void m04400_m04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * md5 @@ -650,11 +650,11 @@ KERNEL_FQ void m04400_s04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * md5 diff --git a/OpenCL/m04400_a3-optimized.cl b/OpenCL/m04400_a3-optimized.cl index 329fec83f..e53ae6b16 100644 --- a/OpenCL/m04400_a3-optimized.cl +++ b/OpenCL/m04400_a3-optimized.cl @@ -168,11 +168,11 @@ DECLSPEC void m04400m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * md5 @@ -438,11 +438,11 @@ DECLSPEC void m04400s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * md5 diff --git a/OpenCL/m04500_a0-optimized.cl b/OpenCL/m04500_a0-optimized.cl index c78e10db9..1adb1c5a1 100644 --- a/OpenCL/m04500_a0-optimized.cl +++ b/OpenCL/m04500_a0-optimized.cl @@ -213,11 +213,11 @@ KERNEL_FQ void m04500_m04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * 2nd SHA1 @@ -568,11 +568,11 @@ KERNEL_FQ void m04500_s04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * 2nd SHA1 diff --git a/OpenCL/m04500_a1-optimized.cl b/OpenCL/m04500_a1-optimized.cl index 8c999a4d5..cb4a9af2b 100644 --- a/OpenCL/m04500_a1-optimized.cl +++ b/OpenCL/m04500_a1-optimized.cl @@ -269,11 +269,11 @@ KERNEL_FQ void m04500_m04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * 2nd SHA1 @@ -682,11 +682,11 @@ KERNEL_FQ void m04500_s04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * 2nd SHA1 diff --git a/OpenCL/m04500_a3-optimized.cl b/OpenCL/m04500_a3-optimized.cl index d552860b9..d03a86232 100644 --- a/OpenCL/m04500_a3-optimized.cl +++ b/OpenCL/m04500_a3-optimized.cl @@ -167,11 +167,11 @@ DECLSPEC void m04500m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * 2nd SHA1 @@ -470,11 +470,11 @@ DECLSPEC void m04500s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * 2nd SHA1 diff --git a/OpenCL/m04510_a0-optimized.cl b/OpenCL/m04510_a0-optimized.cl index 280526e12..1d7469a9f 100644 --- a/OpenCL/m04510_a0-optimized.cl +++ b/OpenCL/m04510_a0-optimized.cl @@ -312,11 +312,11 @@ KERNEL_FQ void m04510_m04 (KERN_ATTR_RULES ()) SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); // Update sha1 state - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * 2nd SHA1 @@ -1327,11 +1327,11 @@ KERNEL_FQ void m04510_s04 (KERN_ATTR_RULES ()) SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); // Update sha1 state - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * 2nd SHA1 diff --git a/OpenCL/m04510_a1-optimized.cl b/OpenCL/m04510_a1-optimized.cl index 5b0bea450..3f11a56cf 100644 --- a/OpenCL/m04510_a1-optimized.cl +++ b/OpenCL/m04510_a1-optimized.cl @@ -367,11 +367,11 @@ KERNEL_FQ void m04510_m04 (KERN_ATTR_BASIC ()) SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); // Update sha1 state - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * 2nd SHA1 @@ -1439,11 +1439,11 @@ KERNEL_FQ void m04510_s04 (KERN_ATTR_BASIC ()) SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); // Update sha1 state - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * 2nd SHA1 diff --git a/OpenCL/m04510_a3-optimized.cl b/OpenCL/m04510_a3-optimized.cl index 306daf8a0..dda50f136 100644 --- a/OpenCL/m04510_a3-optimized.cl +++ b/OpenCL/m04510_a3-optimized.cl @@ -266,11 +266,11 @@ DECLSPEC void m04510m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); // Update sha1 state - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * 2nd SHA1 @@ -1229,11 +1229,11 @@ DECLSPEC void m04510s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); // Update sha1 state - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * 2nd SHA1 diff --git a/OpenCL/m04520_a0-optimized.cl b/OpenCL/m04520_a0-optimized.cl index 4d474d836..52aa26687 100644 --- a/OpenCL/m04520_a0-optimized.cl +++ b/OpenCL/m04520_a0-optimized.cl @@ -241,11 +241,11 @@ KERNEL_FQ void m04520_m04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); u32x t0[4]; u32x t1[4]; @@ -424,11 +424,11 @@ KERNEL_FQ void m04520_m04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); t0[0] = c0[0]; t0[1] = c0[1]; @@ -821,11 +821,11 @@ KERNEL_FQ void m04520_s04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); u32x t0[4]; u32x t1[4]; @@ -1004,11 +1004,11 @@ KERNEL_FQ void m04520_s04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); t0[0] = c0[0]; t0[1] = c0[1]; diff --git a/OpenCL/m04520_a1-optimized.cl b/OpenCL/m04520_a1-optimized.cl index 899c3aa25..58a4c9fbd 100644 --- a/OpenCL/m04520_a1-optimized.cl +++ b/OpenCL/m04520_a1-optimized.cl @@ -297,11 +297,11 @@ KERNEL_FQ void m04520_m04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); u32x t0[4]; u32x t1[4]; @@ -480,11 +480,11 @@ KERNEL_FQ void m04520_m04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); t0[0] = c0[0]; t0[1] = c0[1]; @@ -935,11 +935,11 @@ KERNEL_FQ void m04520_s04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); u32x t0[4]; u32x t1[4]; @@ -1118,11 +1118,11 @@ KERNEL_FQ void m04520_s04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); t0[0] = c0[0]; t0[1] = c0[1]; diff --git a/OpenCL/m04520_a3-optimized.cl b/OpenCL/m04520_a3-optimized.cl index 3ddaa9bef..563da5b9d 100644 --- a/OpenCL/m04520_a3-optimized.cl +++ b/OpenCL/m04520_a3-optimized.cl @@ -195,11 +195,11 @@ DECLSPEC void m04520m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); u32x t0[4]; u32x t1[4]; @@ -378,11 +378,11 @@ DECLSPEC void m04520m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); t0[0] = c0[0]; t0[1] = c0[1]; @@ -723,11 +723,11 @@ DECLSPEC void m04520s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); u32x t0[4]; u32x t1[4]; @@ -906,11 +906,11 @@ DECLSPEC void m04520s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); t0[0] = c0[0]; t0[1] = c0[1]; diff --git a/OpenCL/m04700_a0-optimized.cl b/OpenCL/m04700_a0-optimized.cl index 811009b89..d89787e1b 100644 --- a/OpenCL/m04700_a0-optimized.cl +++ b/OpenCL/m04700_a0-optimized.cl @@ -174,10 +174,10 @@ KERNEL_FQ void m04700_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); /* * sha1 @@ -485,10 +485,10 @@ KERNEL_FQ void m04700_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); /* * sha1 diff --git a/OpenCL/m04700_a1-optimized.cl b/OpenCL/m04700_a1-optimized.cl index 20c8b2398..54b141681 100644 --- a/OpenCL/m04700_a1-optimized.cl +++ b/OpenCL/m04700_a1-optimized.cl @@ -227,10 +227,10 @@ KERNEL_FQ void m04700_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); /* * sha1 @@ -593,10 +593,10 @@ KERNEL_FQ void m04700_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); /* * sha1 diff --git a/OpenCL/m04700_a3-optimized.cl b/OpenCL/m04700_a3-optimized.cl index acaa860e5..91d329c4b 100644 --- a/OpenCL/m04700_a3-optimized.cl +++ b/OpenCL/m04700_a3-optimized.cl @@ -142,10 +142,10 @@ DECLSPEC void m04700m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); /* * sha1 @@ -415,10 +415,10 @@ DECLSPEC void m04700s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); /* * sha1 diff --git a/OpenCL/m04710_a0-optimized.cl b/OpenCL/m04710_a0-optimized.cl index d7098e09a..4cd2bd83e 100644 --- a/OpenCL/m04710_a0-optimized.cl +++ b/OpenCL/m04710_a0-optimized.cl @@ -202,10 +202,10 @@ KERNEL_FQ void m04710_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); // sha1 @@ -918,10 +918,10 @@ KERNEL_FQ void m04710_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); // sha1 diff --git a/OpenCL/m04710_a1-optimized.cl b/OpenCL/m04710_a1-optimized.cl index 0c7c10390..b8a104e69 100644 --- a/OpenCL/m04710_a1-optimized.cl +++ b/OpenCL/m04710_a1-optimized.cl @@ -255,10 +255,10 @@ KERNEL_FQ void m04710_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); // sha1 @@ -1026,10 +1026,10 @@ KERNEL_FQ void m04710_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); // sha1 diff --git a/OpenCL/m04710_a3-optimized.cl b/OpenCL/m04710_a3-optimized.cl index 1c2d6e95f..6058d7057 100644 --- a/OpenCL/m04710_a3-optimized.cl +++ b/OpenCL/m04710_a3-optimized.cl @@ -170,10 +170,10 @@ DECLSPEC void m04710m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); // sha1 @@ -848,10 +848,10 @@ DECLSPEC void m04710s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); // sha1 diff --git a/OpenCL/m05000_a0-optimized.cl b/OpenCL/m05000_a0-optimized.cl index 7073c7e70..af98d5922 100644 --- a/OpenCL/m05000_a0-optimized.cl +++ b/OpenCL/m05000_a0-optimized.cl @@ -314,11 +314,11 @@ KERNEL_FQ void m05000_m04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * 2nd SHA1 @@ -770,11 +770,11 @@ KERNEL_FQ void m05000_s04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * 2nd SHA1 diff --git a/OpenCL/m05000_a1-optimized.cl b/OpenCL/m05000_a1-optimized.cl index dc9b8c801..84d227e1f 100644 --- a/OpenCL/m05000_a1-optimized.cl +++ b/OpenCL/m05000_a1-optimized.cl @@ -372,11 +372,11 @@ KERNEL_FQ void m05000_m04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * 2nd SHA1 @@ -888,11 +888,11 @@ KERNEL_FQ void m05000_s04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * 2nd SHA1 diff --git a/OpenCL/m05000_a3-optimized.cl b/OpenCL/m05000_a3-optimized.cl index 329eda996..2b7de0385 100644 --- a/OpenCL/m05000_a3-optimized.cl +++ b/OpenCL/m05000_a3-optimized.cl @@ -286,11 +286,11 @@ DECLSPEC void m05000m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * 2nd SHA1 @@ -708,11 +708,11 @@ DECLSPEC void m05000s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * 2nd SHA1 diff --git a/OpenCL/m05100_a0-optimized.cl b/OpenCL/m05100_a0-optimized.cl index 78387e438..17ea18179 100644 --- a/OpenCL/m05100_a0-optimized.cl +++ b/OpenCL/m05100_a0-optimized.cl @@ -143,10 +143,10 @@ KERNEL_FQ void m05100_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); u32x z = 0; @@ -303,10 +303,10 @@ KERNEL_FQ void m05100_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); u32x z = 0; diff --git a/OpenCL/m05100_a1-optimized.cl b/OpenCL/m05100_a1-optimized.cl index 50fc822c4..33805c09b 100644 --- a/OpenCL/m05100_a1-optimized.cl +++ b/OpenCL/m05100_a1-optimized.cl @@ -196,10 +196,10 @@ KERNEL_FQ void m05100_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); u32x z = 0; @@ -411,10 +411,10 @@ KERNEL_FQ void m05100_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); u32x z = 0; diff --git a/OpenCL/m05100_a3-optimized.cl b/OpenCL/m05100_a3-optimized.cl index c47ea57c5..7bb0ab257 100644 --- a/OpenCL/m05100_a3-optimized.cl +++ b/OpenCL/m05100_a3-optimized.cl @@ -133,10 +133,10 @@ DECLSPEC void m05100m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, t0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, t2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); u32x z = 0; @@ -277,10 +277,10 @@ DECLSPEC void m05100s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, t0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, t2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); u32x z = 0; diff --git a/OpenCL/m05500_a0-optimized.cl b/OpenCL/m05500_a0-optimized.cl index 9a597096f..f3eee1661 100644 --- a/OpenCL/m05500_a0-optimized.cl +++ b/OpenCL/m05500_a0-optimized.cl @@ -664,15 +664,15 @@ KERNEL_FQ void m05500_m04 (KERN_ATTR_RULES ()) MD4_STEP (MD4_H , a, b, c, d, w0_t[3], MD4C02, MD4S20); MD4_STEP (MD4_H , d, a, b, c, w2_t[3], MD4C02, MD4S21); - if (MATCHES_NONE_VS (((d + MD4M_D) >> 16), s2)) continue; + if (MATCHES_NONE_VS (((d + make_u32x (MD4M_D)) >> 16), s2)) continue; MD4_STEP (MD4_H , c, d, a, b, w1_t[3], MD4C02, MD4S22); MD4_STEP (MD4_H , b, c, d, a, w3_t[3], MD4C02, MD4S23); - a += MD4M_A; - b += MD4M_B; - c += MD4M_C; - d += MD4M_D; + a += make_u32x (MD4M_A); + b += make_u32x (MD4M_B); + c += make_u32x (MD4M_C); + d += make_u32x (MD4M_D); /** * DES1 @@ -893,15 +893,15 @@ KERNEL_FQ void m05500_s04 (KERN_ATTR_RULES ()) MD4_STEP (MD4_H , a, b, c, d, w0_t[3], MD4C02, MD4S20); MD4_STEP (MD4_H , d, a, b, c, w2_t[3], MD4C02, MD4S21); - if (MATCHES_NONE_VS (((d + MD4M_D) >> 16), s2)) continue; + if (MATCHES_NONE_VS (((d + make_u32x (MD4M_D)) >> 16), s2)) continue; MD4_STEP (MD4_H , c, d, a, b, w1_t[3], MD4C02, MD4S22); MD4_STEP (MD4_H , b, c, d, a, w3_t[3], MD4C02, MD4S23); - a += MD4M_A; - b += MD4M_B; - c += MD4M_C; - d += MD4M_D; + a += make_u32x (MD4M_A); + b += make_u32x (MD4M_B); + c += make_u32x (MD4M_C); + d += make_u32x (MD4M_D); /** * DES1 diff --git a/OpenCL/m05500_a1-optimized.cl b/OpenCL/m05500_a1-optimized.cl index 62bb0a13a..a102db121 100644 --- a/OpenCL/m05500_a1-optimized.cl +++ b/OpenCL/m05500_a1-optimized.cl @@ -715,15 +715,15 @@ KERNEL_FQ void m05500_m04 (KERN_ATTR_BASIC ()) MD4_STEP (MD4_H , a, b, c, d, w0_t[3], MD4C02, MD4S20); MD4_STEP (MD4_H , d, a, b, c, w2_t[3], MD4C02, MD4S21); - if (MATCHES_NONE_VS (((d + MD4M_D) >> 16), s2)) continue; + if (MATCHES_NONE_VS (((d + make_u32x (MD4M_D)) >> 16), s2)) continue; MD4_STEP (MD4_H , c, d, a, b, w1_t[3], MD4C02, MD4S22); MD4_STEP (MD4_H , b, c, d, a, w3_t[3], MD4C02, MD4S23); - a += MD4M_A; - b += MD4M_B; - c += MD4M_C; - d += MD4M_D; + a += make_u32x (MD4M_A); + b += make_u32x (MD4M_B); + c += make_u32x (MD4M_C); + d += make_u32x (MD4M_D); /** * DES1 @@ -1002,15 +1002,15 @@ KERNEL_FQ void m05500_s04 (KERN_ATTR_BASIC ()) MD4_STEP (MD4_H , a, b, c, d, w0_t[3], MD4C02, MD4S20); MD4_STEP (MD4_H , d, a, b, c, w2_t[3], MD4C02, MD4S21); - if (MATCHES_NONE_VS (((d + MD4M_D) >> 16), s2)) continue; + if (MATCHES_NONE_VS (((d + make_u32x (MD4M_D)) >> 16), s2)) continue; MD4_STEP (MD4_H , c, d, a, b, w1_t[3], MD4C02, MD4S22); MD4_STEP (MD4_H , b, c, d, a, w3_t[3], MD4C02, MD4S23); - a += MD4M_A; - b += MD4M_B; - c += MD4M_C; - d += MD4M_D; + a += make_u32x (MD4M_A); + b += make_u32x (MD4M_B); + c += make_u32x (MD4M_C); + d += make_u32x (MD4M_D); /** * DES1 diff --git a/OpenCL/m05500_a3-optimized.cl b/OpenCL/m05500_a3-optimized.cl index 2ad1d9a62..421e683ca 100644 --- a/OpenCL/m05500_a3-optimized.cl +++ b/OpenCL/m05500_a3-optimized.cl @@ -602,15 +602,15 @@ DECLSPEC void m05500m (SHM_TYPE u32 (*s_SPtrans)[64], SHM_TYPE u32 (*s_skb)[64], MD4_STEP (MD4_H , a, b, c, d, w0_t[3], MD4C02, MD4S20); MD4_STEP (MD4_H , d, a, b, c, w2_t[3], MD4C02, MD4S21); - if (MATCHES_NONE_VS (((d + MD4M_D) >> 16), s2)) continue; + if (MATCHES_NONE_VS (((d + make_u32x (MD4M_D)) >> 16), s2)) continue; MD4_STEP (MD4_H , c, d, a, b, w1_t[3], MD4C02, MD4S22); MD4_STEP (MD4_H , b, c, d, a, w3_t[3], MD4C02, MD4S23); - a += MD4M_A; - b += MD4M_B; - c += MD4M_C; - d += MD4M_D; + a += make_u32x (MD4M_A); + b += make_u32x (MD4M_B); + c += make_u32x (MD4M_C); + d += make_u32x (MD4M_D); /** * DES1 @@ -768,15 +768,15 @@ DECLSPEC void m05500s (SHM_TYPE u32 (*s_SPtrans)[64], SHM_TYPE u32 (*s_skb)[64], MD4_STEP (MD4_H , a, b, c, d, w0_t[3], MD4C02, MD4S20); MD4_STEP (MD4_H , d, a, b, c, w2_t[3], MD4C02, MD4S21); - if (MATCHES_NONE_VS (((d + MD4M_D) >> 16), s2)) continue; + if (MATCHES_NONE_VS (((d + make_u32x (MD4M_D)) >> 16), s2)) continue; MD4_STEP (MD4_H , c, d, a, b, w1_t[3], MD4C02, MD4S22); MD4_STEP (MD4_H , b, c, d, a, w3_t[3], MD4C02, MD4S23); - a += MD4M_A; - b += MD4M_B; - c += MD4M_C; - d += MD4M_D; + a += make_u32x (MD4M_A); + b += make_u32x (MD4M_B); + c += make_u32x (MD4M_C); + d += make_u32x (MD4M_D); /** * DES1 diff --git a/OpenCL/m09900_a0-optimized.cl b/OpenCL/m09900_a0-optimized.cl index de3555042..2aa8ae62f 100644 --- a/OpenCL/m09900_a0-optimized.cl +++ b/OpenCL/m09900_a0-optimized.cl @@ -138,10 +138,10 @@ KERNEL_FQ void m09900_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); u32x r_a = a; u32x r_b = b; @@ -384,10 +384,10 @@ KERNEL_FQ void m09900_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); u32x r_a = a; u32x r_b = b; diff --git a/OpenCL/m09900_a1-optimized.cl b/OpenCL/m09900_a1-optimized.cl index ab4927163..fb52d6aa1 100644 --- a/OpenCL/m09900_a1-optimized.cl +++ b/OpenCL/m09900_a1-optimized.cl @@ -196,10 +196,10 @@ KERNEL_FQ void m09900_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); u32x r_a = a; u32x r_b = b; @@ -502,10 +502,10 @@ KERNEL_FQ void m09900_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); u32x r_a = a; u32x r_b = b; diff --git a/OpenCL/m09900_a3-optimized.cl b/OpenCL/m09900_a3-optimized.cl index 60b009a0d..c2d7e76b8 100644 --- a/OpenCL/m09900_a3-optimized.cl +++ b/OpenCL/m09900_a3-optimized.cl @@ -205,10 +205,10 @@ DECLSPEC void m09900m (PRIVATE_AS u32 *w, const u32 pw_len, KERN_ATTR_FUNC_VECTO MD5_STEP0(MD5_I , c, d, a, b, I_w2c3e, MD5S32); MD5_STEP0(MD5_I , b, c, d, a, I_w9c3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); u32x r_a = a; u32x r_b = b; @@ -495,10 +495,10 @@ DECLSPEC void m09900s (PRIVATE_AS u32 *w, const u32 pw_len, KERN_ATTR_FUNC_VECTO MD5_STEP0(MD5_I , c, d, a, b, I_w2c3e, MD5S32); MD5_STEP0(MD5_I , b, c, d, a, I_w9c3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); u32x r_a = a; u32x r_b = b; diff --git a/OpenCL/m11000_a0-optimized.cl b/OpenCL/m11000_a0-optimized.cl index 1f2a59a1b..05bf90ac3 100644 --- a/OpenCL/m11000_a0-optimized.cl +++ b/OpenCL/m11000_a0-optimized.cl @@ -199,10 +199,10 @@ KERNEL_FQ void m11000_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); u32x r_a = a; u32x r_b = b; @@ -508,10 +508,10 @@ KERNEL_FQ void m11000_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); u32x r_a = a; u32x r_b = b; diff --git a/OpenCL/m11000_a1-optimized.cl b/OpenCL/m11000_a1-optimized.cl index 01aed20a6..2500deba6 100644 --- a/OpenCL/m11000_a1-optimized.cl +++ b/OpenCL/m11000_a1-optimized.cl @@ -255,10 +255,10 @@ KERNEL_FQ void m11000_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); u32x r_a = a; u32x r_b = b; @@ -622,10 +622,10 @@ KERNEL_FQ void m11000_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); u32x r_a = a; u32x r_b = b; diff --git a/OpenCL/m11000_a3-optimized.cl b/OpenCL/m11000_a3-optimized.cl index 44e571e5f..a7cb4055a 100644 --- a/OpenCL/m11000_a3-optimized.cl +++ b/OpenCL/m11000_a3-optimized.cl @@ -170,10 +170,10 @@ DECLSPEC void m11000m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); u32x r_a = a; u32x r_b = b; @@ -444,10 +444,10 @@ DECLSPEC void m11000s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); u32x r_a = a; u32x r_b = b; diff --git a/OpenCL/m11100_a0-optimized.cl b/OpenCL/m11100_a0-optimized.cl index 1c38aaea5..23ba22a5a 100644 --- a/OpenCL/m11100_a0-optimized.cl +++ b/OpenCL/m11100_a0-optimized.cl @@ -240,10 +240,10 @@ KERNEL_FQ void m11100_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -586,10 +586,10 @@ KERNEL_FQ void m11100_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; diff --git a/OpenCL/m11100_a1-optimized.cl b/OpenCL/m11100_a1-optimized.cl index 296e483b7..015a2fdf4 100644 --- a/OpenCL/m11100_a1-optimized.cl +++ b/OpenCL/m11100_a1-optimized.cl @@ -298,10 +298,10 @@ KERNEL_FQ void m11100_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -704,10 +704,10 @@ KERNEL_FQ void m11100_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; diff --git a/OpenCL/m11100_a3-optimized.cl b/OpenCL/m11100_a3-optimized.cl index 23ed4d97a..af87664e3 100644 --- a/OpenCL/m11100_a3-optimized.cl +++ b/OpenCL/m11100_a3-optimized.cl @@ -214,10 +214,10 @@ DECLSPEC void m11100m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; @@ -528,10 +528,10 @@ DECLSPEC void m11100s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0 | uint_to_hex_lower8 ((a >> 8) & 255) << 16; diff --git a/OpenCL/m11200_a0-optimized.cl b/OpenCL/m11200_a0-optimized.cl index aa7a96f4a..9cf4eb094 100644 --- a/OpenCL/m11200_a0-optimized.cl +++ b/OpenCL/m11200_a0-optimized.cl @@ -196,11 +196,11 @@ KERNEL_FQ void m11200_m04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - const u32x plain_sha1_a = a + SHA1M_A; - const u32x plain_sha1_b = b + SHA1M_B; - const u32x plain_sha1_c = c + SHA1M_C; - const u32x plain_sha1_d = d + SHA1M_D; - const u32x plain_sha1_e = e + SHA1M_E; + const u32x plain_sha1_a = a + make_u32x (SHA1M_A); + const u32x plain_sha1_b = b + make_u32x (SHA1M_B); + const u32x plain_sha1_c = c + make_u32x (SHA1M_C); + const u32x plain_sha1_d = d + make_u32x (SHA1M_D); + const u32x plain_sha1_e = e + make_u32x (SHA1M_E); /** * sha1 (sha1 ($pass)) @@ -325,11 +325,11 @@ KERNEL_FQ void m11200_m04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * sha1 ($salt . sha1 (sha1 ($pass))) @@ -454,11 +454,11 @@ KERNEL_FQ void m11200_m04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); a ^= plain_sha1_a; b ^= plain_sha1_b; @@ -670,11 +670,11 @@ KERNEL_FQ void m11200_s04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - const u32x plain_sha1_a = a + SHA1M_A; - const u32x plain_sha1_b = b + SHA1M_B; - const u32x plain_sha1_c = c + SHA1M_C; - const u32x plain_sha1_d = d + SHA1M_D; - const u32x plain_sha1_e = e + SHA1M_E; + const u32x plain_sha1_a = a + make_u32x (SHA1M_A); + const u32x plain_sha1_b = b + make_u32x (SHA1M_B); + const u32x plain_sha1_c = c + make_u32x (SHA1M_C); + const u32x plain_sha1_d = d + make_u32x (SHA1M_D); + const u32x plain_sha1_e = e + make_u32x (SHA1M_E); /** * sha1 (sha1 ($pass)) @@ -799,11 +799,11 @@ KERNEL_FQ void m11200_s04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * sha1 ($salt . sha1 (sha1 ($pass))) @@ -928,11 +928,11 @@ KERNEL_FQ void m11200_s04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); a ^= plain_sha1_a; b ^= plain_sha1_b; diff --git a/OpenCL/m11200_a1-optimized.cl b/OpenCL/m11200_a1-optimized.cl index 40928999f..67bd63aef 100644 --- a/OpenCL/m11200_a1-optimized.cl +++ b/OpenCL/m11200_a1-optimized.cl @@ -252,11 +252,11 @@ KERNEL_FQ void m11200_m04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - const u32x plain_sha1_a = a + SHA1M_A; - const u32x plain_sha1_b = b + SHA1M_B; - const u32x plain_sha1_c = c + SHA1M_C; - const u32x plain_sha1_d = d + SHA1M_D; - const u32x plain_sha1_e = e + SHA1M_E; + const u32x plain_sha1_a = a + make_u32x (SHA1M_A); + const u32x plain_sha1_b = b + make_u32x (SHA1M_B); + const u32x plain_sha1_c = c + make_u32x (SHA1M_C); + const u32x plain_sha1_d = d + make_u32x (SHA1M_D); + const u32x plain_sha1_e = e + make_u32x (SHA1M_E); /** * sha1 (sha1 ($pass)) @@ -381,11 +381,11 @@ KERNEL_FQ void m11200_m04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * sha1 ($salt . sha1 (sha1 ($pass))) @@ -510,11 +510,11 @@ KERNEL_FQ void m11200_m04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); a ^= plain_sha1_a; b ^= plain_sha1_b; @@ -784,11 +784,11 @@ KERNEL_FQ void m11200_s04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - const u32x plain_sha1_a = a + SHA1M_A; - const u32x plain_sha1_b = b + SHA1M_B; - const u32x plain_sha1_c = c + SHA1M_C; - const u32x plain_sha1_d = d + SHA1M_D; - const u32x plain_sha1_e = e + SHA1M_E; + const u32x plain_sha1_a = a + make_u32x (SHA1M_A); + const u32x plain_sha1_b = b + make_u32x (SHA1M_B); + const u32x plain_sha1_c = c + make_u32x (SHA1M_C); + const u32x plain_sha1_d = d + make_u32x (SHA1M_D); + const u32x plain_sha1_e = e + make_u32x (SHA1M_E); /** * sha1 (sha1 ($pass)) @@ -913,11 +913,11 @@ KERNEL_FQ void m11200_s04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * sha1 ($salt . sha1 (sha1 ($pass))) @@ -1042,11 +1042,11 @@ KERNEL_FQ void m11200_s04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); a ^= plain_sha1_a; b ^= plain_sha1_b; diff --git a/OpenCL/m11200_a3-optimized.cl b/OpenCL/m11200_a3-optimized.cl index e85af1c49..e74a76733 100644 --- a/OpenCL/m11200_a3-optimized.cl +++ b/OpenCL/m11200_a3-optimized.cl @@ -167,11 +167,11 @@ DECLSPEC void m11200m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - const u32x plain_sha1_a = a + SHA1M_A; - const u32x plain_sha1_b = b + SHA1M_B; - const u32x plain_sha1_c = c + SHA1M_C; - const u32x plain_sha1_d = d + SHA1M_D; - const u32x plain_sha1_e = e + SHA1M_E; + const u32x plain_sha1_a = a + make_u32x (SHA1M_A); + const u32x plain_sha1_b = b + make_u32x (SHA1M_B); + const u32x plain_sha1_c = c + make_u32x (SHA1M_C); + const u32x plain_sha1_d = d + make_u32x (SHA1M_D); + const u32x plain_sha1_e = e + make_u32x (SHA1M_E); /** * sha1 (sha1 ($pass)) @@ -296,11 +296,11 @@ DECLSPEC void m11200m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * sha1 ($salt . sha1 (sha1 ($pass))) @@ -425,11 +425,11 @@ DECLSPEC void m11200m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); a ^= plain_sha1_a; b ^= plain_sha1_b; @@ -606,11 +606,11 @@ DECLSPEC void m11200s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - const u32x plain_sha1_a = a + SHA1M_A; - const u32x plain_sha1_b = b + SHA1M_B; - const u32x plain_sha1_c = c + SHA1M_C; - const u32x plain_sha1_d = d + SHA1M_D; - const u32x plain_sha1_e = e + SHA1M_E; + const u32x plain_sha1_a = a + make_u32x (SHA1M_A); + const u32x plain_sha1_b = b + make_u32x (SHA1M_B); + const u32x plain_sha1_c = c + make_u32x (SHA1M_C); + const u32x plain_sha1_d = d + make_u32x (SHA1M_D); + const u32x plain_sha1_e = e + make_u32x (SHA1M_E); /** * sha1 (sha1 ($pass)) @@ -735,11 +735,11 @@ DECLSPEC void m11200s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * sha1 ($salt . sha1 (sha1 ($pass))) @@ -864,11 +864,11 @@ DECLSPEC void m11200s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); a ^= plain_sha1_a; b ^= plain_sha1_b; diff --git a/OpenCL/m12600_a0-optimized.cl b/OpenCL/m12600_a0-optimized.cl index 275dcd445..323d8c0e6 100644 --- a/OpenCL/m12600_a0-optimized.cl +++ b/OpenCL/m12600_a0-optimized.cl @@ -232,11 +232,11 @@ KERNEL_FQ void m12600_m04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * sha256 @@ -584,11 +584,11 @@ KERNEL_FQ void m12600_s04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * sha256 diff --git a/OpenCL/m12600_a1-optimized.cl b/OpenCL/m12600_a1-optimized.cl index 0d074286d..3c5b709a8 100644 --- a/OpenCL/m12600_a1-optimized.cl +++ b/OpenCL/m12600_a1-optimized.cl @@ -288,11 +288,11 @@ KERNEL_FQ void m12600_m04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * sha256 @@ -698,11 +698,11 @@ KERNEL_FQ void m12600_s04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * sha256 diff --git a/OpenCL/m12600_a3-optimized.cl b/OpenCL/m12600_a3-optimized.cl index 8c3d44e96..e73899b05 100644 --- a/OpenCL/m12600_a3-optimized.cl +++ b/OpenCL/m12600_a3-optimized.cl @@ -186,11 +186,11 @@ DECLSPEC void m12600m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * sha256 @@ -486,11 +486,11 @@ DECLSPEC void m12600s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); /** * sha256 diff --git a/OpenCL/m13300_a0-optimized.cl b/OpenCL/m13300_a0-optimized.cl index c866c6692..5110045c3 100644 --- a/OpenCL/m13300_a0-optimized.cl +++ b/OpenCL/m13300_a0-optimized.cl @@ -184,10 +184,10 @@ KERNEL_FQ void m13300_m04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - e += SHA1M_E; - d += SHA1M_D; - c += SHA1M_C; + a += make_u32x (SHA1M_A); + e += make_u32x (SHA1M_E); + d += make_u32x (SHA1M_D); + c += make_u32x (SHA1M_C); e &= 0x00000000; @@ -383,10 +383,10 @@ KERNEL_FQ void m13300_s04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - e += SHA1M_E; - d += SHA1M_D; - c += SHA1M_C; + a += make_u32x (SHA1M_A); + e += make_u32x (SHA1M_E); + d += make_u32x (SHA1M_D); + c += make_u32x (SHA1M_C); e &= 0x00000000; diff --git a/OpenCL/m13300_a1-optimized.cl b/OpenCL/m13300_a1-optimized.cl index e4acc0d54..051eba2db 100644 --- a/OpenCL/m13300_a1-optimized.cl +++ b/OpenCL/m13300_a1-optimized.cl @@ -240,10 +240,10 @@ KERNEL_FQ void m13300_m04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - e += SHA1M_E; - d += SHA1M_D; - c += SHA1M_C; + a += make_u32x (SHA1M_A); + e += make_u32x (SHA1M_E); + d += make_u32x (SHA1M_D); + c += make_u32x (SHA1M_C); e &= 0x00000000; @@ -497,10 +497,10 @@ KERNEL_FQ void m13300_s04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - e += SHA1M_E; - d += SHA1M_D; - c += SHA1M_C; + a += make_u32x (SHA1M_A); + e += make_u32x (SHA1M_E); + d += make_u32x (SHA1M_D); + c += make_u32x (SHA1M_C); e &= 0x00000000; diff --git a/OpenCL/m13300_a3-optimized.cl b/OpenCL/m13300_a3-optimized.cl index 5f34542c4..dfeff4acd 100644 --- a/OpenCL/m13300_a3-optimized.cl +++ b/OpenCL/m13300_a3-optimized.cl @@ -251,10 +251,10 @@ DECLSPEC void m13300m (PRIVATE_AS u32 *w, const u32 pw_len, KERN_ATTR_FUNC_VECTO SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_78s ^ w0s07 ^ w0s08 ^ w0s15 ^ w0s18 ^ w0s20)); SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_79s ^ w0s08 ^ w0s22)); - a += SHA1M_A; - e += SHA1M_E; - d += SHA1M_D; - c += SHA1M_C; + a += make_u32x (SHA1M_A); + e += make_u32x (SHA1M_E); + d += make_u32x (SHA1M_D); + c += make_u32x (SHA1M_C); e &= 0x00000000; @@ -511,10 +511,10 @@ DECLSPEC void m13300s (PRIVATE_AS u32 *w, const u32 pw_len, KERN_ATTR_FUNC_VECTO SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_78s ^ w0s07 ^ w0s08 ^ w0s15 ^ w0s18 ^ w0s20)); SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_79s ^ w0s08 ^ w0s22)); - a += SHA1M_A; - e += SHA1M_E; - d += SHA1M_D; - c += SHA1M_C; + a += make_u32x (SHA1M_A); + e += make_u32x (SHA1M_E); + d += make_u32x (SHA1M_D); + c += make_u32x (SHA1M_C); e &= 0x00000000; diff --git a/OpenCL/m15500_a0-optimized.cl b/OpenCL/m15500_a0-optimized.cl index 8e40ce6b2..55d5cf935 100644 --- a/OpenCL/m15500_a0-optimized.cl +++ b/OpenCL/m15500_a0-optimized.cl @@ -248,11 +248,11 @@ KERNEL_FQ void m15500_m04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); a &= 0xff000000; b &= 0x0000ffff; @@ -516,11 +516,11 @@ KERNEL_FQ void m15500_s04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); a &= 0xff000000; b &= 0x0000ffff; diff --git a/OpenCL/m15500_a1-optimized.cl b/OpenCL/m15500_a1-optimized.cl index 7d8d2c506..6d253ba54 100644 --- a/OpenCL/m15500_a1-optimized.cl +++ b/OpenCL/m15500_a1-optimized.cl @@ -306,11 +306,11 @@ KERNEL_FQ void m15500_m04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); a &= 0xff000000; b &= 0x0000ffff; @@ -634,11 +634,11 @@ KERNEL_FQ void m15500_s04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); a &= 0xff000000; b &= 0x0000ffff; diff --git a/OpenCL/m15500_a3-optimized.cl b/OpenCL/m15500_a3-optimized.cl index 909c61f57..f6d37dd9f 100644 --- a/OpenCL/m15500_a3-optimized.cl +++ b/OpenCL/m15500_a3-optimized.cl @@ -299,11 +299,11 @@ DECLSPEC void m15500m (PRIVATE_AS u32 *w, const u32 pw_len, KERN_ATTR_FUNC_VECTO SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_78s ^ w0s07 ^ w0s08 ^ w0s15 ^ w0s18 ^ w0s20)); SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_79s ^ w0s08 ^ w0s22)); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); a &= 0xff000000; b &= 0x0000ffff; @@ -564,11 +564,11 @@ DECLSPEC void m15500s (PRIVATE_AS u32 *w, const u32 pw_len, KERN_ATTR_FUNC_VECTO SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_78s ^ w0s07 ^ w0s08 ^ w0s15 ^ w0s18 ^ w0s20)); SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_79s ^ w0s08 ^ w0s22)); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); a &= 0xff000000; b &= 0x0000ffff; diff --git a/OpenCL/m20710_a0-optimized.cl b/OpenCL/m20710_a0-optimized.cl index ee00961b3..b1ce011ea 100644 --- a/OpenCL/m20710_a0-optimized.cl +++ b/OpenCL/m20710_a0-optimized.cl @@ -230,14 +230,14 @@ KERNEL_FQ void m20710_m04 (KERN_ATTR_RULES ()) 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); - a += SHA256M_A; - b += SHA256M_B; - c += SHA256M_C; - d += SHA256M_D; - e += SHA256M_E; - f += SHA256M_F; - g += SHA256M_G; - h += SHA256M_H; + a += make_u32x (SHA256M_A); + b += make_u32x (SHA256M_B); + c += make_u32x (SHA256M_C); + d += make_u32x (SHA256M_D); + e += make_u32x (SHA256M_E); + f += make_u32x (SHA256M_F); + g += make_u32x (SHA256M_G); + h += make_u32x (SHA256M_H); // final sha256 @@ -609,12 +609,12 @@ KERNEL_FQ void m20710_m04 (KERN_ATTR_RULES ()) // a += digest[0] - SHA256M_A; // b += digest[1] - SHA256M_B; - c += digest[2] - SHA256M_C; - d += digest[3] - SHA256M_D; + c += digest[2] - make_u32x (SHA256M_C); + d += digest[3] - make_u32x (SHA256M_D); // e += digest[4] - SHA256M_E; // f += digest[5] - SHA256M_F; - g += digest[6] - SHA256M_G; - h += digest[7] - SHA256M_H; + g += digest[6] - make_u32x (SHA256M_G); + h += digest[7] - make_u32x (SHA256M_H); COMPARE_M_SIMD (d, h, c, g); } @@ -828,14 +828,14 @@ KERNEL_FQ void m20710_s04 (KERN_ATTR_RULES ()) 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); - a += SHA256M_A; - b += SHA256M_B; - c += SHA256M_C; - d += SHA256M_D; - e += SHA256M_E; - f += SHA256M_F; - g += SHA256M_G; - h += SHA256M_H; + a += make_u32x (SHA256M_A); + b += make_u32x (SHA256M_B); + c += make_u32x (SHA256M_C); + d += make_u32x (SHA256M_D); + e += make_u32x (SHA256M_E); + f += make_u32x (SHA256M_F); + g += make_u32x (SHA256M_G); + h += make_u32x (SHA256M_H); // final sha256 @@ -1213,12 +1213,12 @@ KERNEL_FQ void m20710_s04 (KERN_ATTR_RULES ()) // a += digest[0] - SHA256M_A; // b += digest[1] - SHA256M_B; - c += digest[2] - SHA256M_C; - d += digest[3] - SHA256M_D; + c += digest[2] - make_u32x (SHA256M_C); + d += digest[3] - make_u32x (SHA256M_D); // e += digest[4] - SHA256M_E; // f += digest[5] - SHA256M_F; - g += digest[6] - SHA256M_G; - h += digest[7] - SHA256M_H; + g += digest[6] - make_u32x (SHA256M_G); + h += digest[7] - make_u32x (SHA256M_H); COMPARE_S_SIMD (d, h, c, g); } diff --git a/OpenCL/m20710_a1-optimized.cl b/OpenCL/m20710_a1-optimized.cl index 2eea55a90..e84e13d35 100644 --- a/OpenCL/m20710_a1-optimized.cl +++ b/OpenCL/m20710_a1-optimized.cl @@ -286,14 +286,14 @@ KERNEL_FQ void m20710_m04 (KERN_ATTR_BASIC ()) 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); - a += SHA256M_A; - b += SHA256M_B; - c += SHA256M_C; - d += SHA256M_D; - e += SHA256M_E; - f += SHA256M_F; - g += SHA256M_G; - h += SHA256M_H; + a += make_u32x (SHA256M_A); + b += make_u32x (SHA256M_B); + c += make_u32x (SHA256M_C); + d += make_u32x (SHA256M_D); + e += make_u32x (SHA256M_E); + f += make_u32x (SHA256M_F); + g += make_u32x (SHA256M_G); + h += make_u32x (SHA256M_H); // final sha256 @@ -665,12 +665,12 @@ KERNEL_FQ void m20710_m04 (KERN_ATTR_BASIC ()) // a += digest[0] - SHA256M_A; // b += digest[1] - SHA256M_B; - c += digest[2] - SHA256M_C; - d += digest[3] - SHA256M_D; + c += digest[2] - make_u32x (SHA256M_C); + d += digest[3] - make_u32x (SHA256M_D); // e += digest[4] - SHA256M_E; // f += digest[5] - SHA256M_F; - g += digest[6] - SHA256M_G; - h += digest[7] - SHA256M_H; + g += digest[6] - make_u32x (SHA256M_G); + h += digest[7] - make_u32x (SHA256M_H); COMPARE_M_SIMD (d, h, c, g); } @@ -942,14 +942,14 @@ KERNEL_FQ void m20710_s04 (KERN_ATTR_BASIC ()) 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); - a += SHA256M_A; - b += SHA256M_B; - c += SHA256M_C; - d += SHA256M_D; - e += SHA256M_E; - f += SHA256M_F; - g += SHA256M_G; - h += SHA256M_H; + a += make_u32x (SHA256M_A); + b += make_u32x (SHA256M_B); + c += make_u32x (SHA256M_C); + d += make_u32x (SHA256M_D); + e += make_u32x (SHA256M_E); + f += make_u32x (SHA256M_F); + g += make_u32x (SHA256M_G); + h += make_u32x (SHA256M_H); // final sha256 @@ -1327,12 +1327,12 @@ KERNEL_FQ void m20710_s04 (KERN_ATTR_BASIC ()) // a += digest[0] - SHA256M_A; // b += digest[1] - SHA256M_B; - c += digest[2] - SHA256M_C; - d += digest[3] - SHA256M_D; + c += digest[2] - make_u32x (SHA256M_C); + d += digest[3] - make_u32x (SHA256M_D); // e += digest[4] - SHA256M_E; // f += digest[5] - SHA256M_F; - g += digest[6] - SHA256M_G; - h += digest[7] - SHA256M_H; + g += digest[6] - make_u32x (SHA256M_G); + h += digest[7] - make_u32x (SHA256M_H); COMPARE_S_SIMD (d, h, c, g); } diff --git a/OpenCL/m20710_a3-optimized.cl b/OpenCL/m20710_a3-optimized.cl index c7ac9c79c..ab959ef93 100644 --- a/OpenCL/m20710_a3-optimized.cl +++ b/OpenCL/m20710_a3-optimized.cl @@ -184,14 +184,14 @@ DECLSPEC void m20710m (PRIVATE_AS u32 *w, const u32 pw_len, KERN_ATTR_FUNC_VECTO 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); - a += SHA256M_A; - b += SHA256M_B; - c += SHA256M_C; - d += SHA256M_D; - e += SHA256M_E; - f += SHA256M_F; - g += SHA256M_G; - h += SHA256M_H; + a += make_u32x (SHA256M_A); + b += make_u32x (SHA256M_B); + c += make_u32x (SHA256M_C); + d += make_u32x (SHA256M_D); + e += make_u32x (SHA256M_E); + f += make_u32x (SHA256M_F); + g += make_u32x (SHA256M_G); + h += make_u32x (SHA256M_H); // final sha256 @@ -563,12 +563,12 @@ DECLSPEC void m20710m (PRIVATE_AS u32 *w, const u32 pw_len, KERN_ATTR_FUNC_VECTO // a += digest[0] - SHA256M_A; // b += digest[1] - SHA256M_B; - c += digest[2] - SHA256M_C; - d += digest[3] - SHA256M_D; + c += digest[2] - make_u32x (SHA256M_C); + d += digest[3] - make_u32x (SHA256M_D); // e += digest[4] - SHA256M_E; // f += digest[5] - SHA256M_F; - g += digest[6] - SHA256M_G; - h += digest[7] - SHA256M_H; + g += digest[6] - make_u32x (SHA256M_G); + h += digest[7] - make_u32x (SHA256M_H); COMPARE_M_SIMD (d, h, c, g); } @@ -730,14 +730,14 @@ DECLSPEC void m20710s (PRIVATE_AS u32 *w, const u32 pw_len, KERN_ATTR_FUNC_VECTO 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); - a += SHA256M_A; - b += SHA256M_B; - c += SHA256M_C; - d += SHA256M_D; - e += SHA256M_E; - f += SHA256M_F; - g += SHA256M_G; - h += SHA256M_H; + a += make_u32x (SHA256M_A); + b += make_u32x (SHA256M_B); + c += make_u32x (SHA256M_C); + d += make_u32x (SHA256M_D); + e += make_u32x (SHA256M_E); + f += make_u32x (SHA256M_F); + g += make_u32x (SHA256M_G); + h += make_u32x (SHA256M_H); // final sha256 @@ -1104,7 +1104,7 @@ DECLSPEC void m20710s (PRIVATE_AS u32 *w, const u32 pw_len, KERN_ATTR_FUNC_VECTO 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); - if (MATCHES_NONE_VS (h+digest[7]-SHA256M_H, search[1])) continue; + if (MATCHES_NONE_VS ((h + digest[7] - make_u32x (SHA256M_H)), search[1])) continue; 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); @@ -1112,12 +1112,12 @@ DECLSPEC void m20710s (PRIVATE_AS u32 *w, const u32 pw_len, KERN_ATTR_FUNC_VECTO // a += digest[0] - SHA256M_A; // b += digest[1] - SHA256M_B; - c += digest[2] - SHA256M_C; - d += digest[3] - SHA256M_D; + c += digest[2] - make_u32x (SHA256M_C); + d += digest[3] - make_u32x (SHA256M_D); // e += digest[4] - SHA256M_E; // f += digest[5] - SHA256M_F; - g += digest[6] - SHA256M_G; - h += digest[7] - SHA256M_H; + g += digest[6] - make_u32x (SHA256M_G); + h += digest[7] - make_u32x (SHA256M_H); COMPARE_S_SIMD (d, h, c, g); } diff --git a/OpenCL/m20800_a0-optimized.cl b/OpenCL/m20800_a0-optimized.cl index 3ef99bb96..2eca0f810 100644 --- a/OpenCL/m20800_a0-optimized.cl +++ b/OpenCL/m20800_a0-optimized.cl @@ -191,10 +191,10 @@ KERNEL_FQ void m20800_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); /* * sha256 @@ -492,10 +492,10 @@ KERNEL_FQ void m20800_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); /* * sha256 diff --git a/OpenCL/m20800_a1-optimized.cl b/OpenCL/m20800_a1-optimized.cl index b7cbe5d88..fa01e62d3 100644 --- a/OpenCL/m20800_a1-optimized.cl +++ b/OpenCL/m20800_a1-optimized.cl @@ -244,10 +244,10 @@ KERNEL_FQ void m20800_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); /* * sha256 @@ -600,10 +600,10 @@ KERNEL_FQ void m20800_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); /* * sha256 diff --git a/OpenCL/m20800_a3-optimized.cl b/OpenCL/m20800_a3-optimized.cl index 9524b13b0..3523584a9 100644 --- a/OpenCL/m20800_a3-optimized.cl +++ b/OpenCL/m20800_a3-optimized.cl @@ -159,10 +159,10 @@ DECLSPEC void m20800m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); /* * sha256 @@ -422,10 +422,10 @@ DECLSPEC void m20800s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); /* * sha256 diff --git a/OpenCL/m20900_a0-optimized.cl b/OpenCL/m20900_a0-optimized.cl index d4a3fa08e..d55cbb761 100644 --- a/OpenCL/m20900_a0-optimized.cl +++ b/OpenCL/m20900_a0-optimized.cl @@ -214,11 +214,11 @@ KERNEL_FQ void m20900_m04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); const u32x a0 = a; const u32x b0 = b; @@ -322,10 +322,10 @@ KERNEL_FQ void m20900_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); const u32x a1 = hc_swap32 (a); const u32x b1 = hc_swap32 (b); @@ -722,10 +722,10 @@ KERNEL_FQ void m20900_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += digest[0] - MD5M_A; - b += digest[1] - MD5M_B; - c += digest[2] - MD5M_C; - d += digest[3] - MD5M_D; + a += digest[0] - make_u32x (MD5M_A); + b += digest[1] - make_u32x (MD5M_B); + c += digest[2] - make_u32x (MD5M_C); + d += digest[3] - make_u32x (MD5M_D); COMPARE_M_SIMD (a, d, c, b); } @@ -936,11 +936,11 @@ KERNEL_FQ void m20900_s04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); const u32x a0 = a; const u32x b0 = b; @@ -1044,10 +1044,10 @@ KERNEL_FQ void m20900_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); const u32x a1 = hc_swap32 (a); const u32x b1 = hc_swap32 (b); @@ -1447,10 +1447,10 @@ KERNEL_FQ void m20900_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += digest[0] - MD5M_A; - b += digest[1] - MD5M_B; - c += digest[2] - MD5M_C; - d += digest[3] - MD5M_D; + a += digest[0] - make_u32x (MD5M_A); + b += digest[1] - make_u32x (MD5M_B); + c += digest[2] - make_u32x (MD5M_C); + d += digest[3] - make_u32x (MD5M_D); COMPARE_S_SIMD (a, d, c, b); } diff --git a/OpenCL/m20900_a1-optimized.cl b/OpenCL/m20900_a1-optimized.cl index 6e0bd6d20..07c3f210f 100644 --- a/OpenCL/m20900_a1-optimized.cl +++ b/OpenCL/m20900_a1-optimized.cl @@ -270,11 +270,11 @@ KERNEL_FQ void m20900_m04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); const u32x a0 = a; const u32x b0 = b; @@ -378,10 +378,10 @@ KERNEL_FQ void m20900_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); const u32x a1 = hc_swap32 (a); const u32x b1 = hc_swap32 (b); @@ -778,10 +778,10 @@ KERNEL_FQ void m20900_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += digest[0] - MD5M_A; - b += digest[1] - MD5M_B; - c += digest[2] - MD5M_C; - d += digest[3] - MD5M_D; + a += digest[0] - make_u32x (MD5M_A); + b += digest[1] - make_u32x (MD5M_B); + c += digest[2] - make_u32x (MD5M_C); + d += digest[3] - make_u32x (MD5M_D); COMPARE_M_SIMD (a, d, c, b); } @@ -1050,11 +1050,11 @@ KERNEL_FQ void m20900_s04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); const u32x a0 = a; const u32x b0 = b; @@ -1158,10 +1158,10 @@ KERNEL_FQ void m20900_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); const u32x a1 = hc_swap32 (a); const u32x b1 = hc_swap32 (b); @@ -1561,10 +1561,10 @@ KERNEL_FQ void m20900_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += digest[0] - MD5M_A; - b += digest[1] - MD5M_B; - c += digest[2] - MD5M_C; - d += digest[3] - MD5M_D; + a += digest[0] - make_u32x (MD5M_A); + b += digest[1] - make_u32x (MD5M_B); + c += digest[2] - make_u32x (MD5M_C); + d += digest[3] - make_u32x (MD5M_D); COMPARE_S_SIMD (a, d, c, b); } diff --git a/OpenCL/m20900_a3-optimized.cl b/OpenCL/m20900_a3-optimized.cl index 09141655b..adeb2b1e6 100644 --- a/OpenCL/m20900_a3-optimized.cl +++ b/OpenCL/m20900_a3-optimized.cl @@ -168,11 +168,11 @@ DECLSPEC void m20900m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); const u32x a0 = a; const u32x b0 = b; @@ -276,10 +276,10 @@ DECLSPEC void m20900m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); const u32x a1 = hc_swap32 (a); const u32x b1 = hc_swap32 (b); @@ -676,10 +676,10 @@ DECLSPEC void m20900m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += digest[0] - MD5M_A; - b += digest[1] - MD5M_B; - c += digest[2] - MD5M_C; - d += digest[3] - MD5M_D; + a += digest[0] - make_u32x (MD5M_A); + b += digest[1] - make_u32x (MD5M_B); + c += digest[2] - make_u32x (MD5M_C); + d += digest[3] - make_u32x (MD5M_D); COMPARE_M_SIMD (a, d, c, b); } @@ -838,11 +838,11 @@ DECLSPEC void m20900s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); const u32x a0 = a; const u32x b0 = b; @@ -946,10 +946,10 @@ DECLSPEC void m20900s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); const u32x a1 = hc_swap32 (a); const u32x b1 = hc_swap32 (b); @@ -1343,16 +1343,16 @@ DECLSPEC void m20900s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33); MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30); - if (MATCHES_NONE_VS ((a+digest[0]-MD5M_A), search[0])) continue; + if (MATCHES_NONE_VS ((a + digest[0] - make_u32x (MD5M_A)), search[0])) continue; MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31); MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += digest[0] - MD5M_A; - b += digest[1] - MD5M_B; - c += digest[2] - MD5M_C; - d += digest[3] - MD5M_D; + a += digest[0] - make_u32x (MD5M_A); + b += digest[1] - make_u32x (MD5M_B); + c += digest[2] - make_u32x (MD5M_C); + d += digest[3] - make_u32x (MD5M_D); COMPARE_S_SIMD (a, d, c, b); } diff --git a/OpenCL/m21100_a0-optimized.cl b/OpenCL/m21100_a0-optimized.cl index 2c88174aa..6c6654355 100644 --- a/OpenCL/m21100_a0-optimized.cl +++ b/OpenCL/m21100_a0-optimized.cl @@ -246,10 +246,10 @@ KERNEL_FQ void m21100_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); /* * sha1 @@ -629,10 +629,10 @@ KERNEL_FQ void m21100_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); /* * sha1 diff --git a/OpenCL/m21100_a1-optimized.cl b/OpenCL/m21100_a1-optimized.cl index 7035433c1..4a20328c3 100644 --- a/OpenCL/m21100_a1-optimized.cl +++ b/OpenCL/m21100_a1-optimized.cl @@ -302,10 +302,10 @@ KERNEL_FQ void m21100_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); /* * sha1 @@ -743,10 +743,10 @@ KERNEL_FQ void m21100_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); /* * sha1 diff --git a/OpenCL/m21100_a3-optimized.cl b/OpenCL/m21100_a3-optimized.cl index 989f295d7..e5b0e86ce 100644 --- a/OpenCL/m21100_a3-optimized.cl +++ b/OpenCL/m21100_a3-optimized.cl @@ -218,10 +218,10 @@ DECLSPEC void m21100m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); /* * sha1 @@ -491,10 +491,10 @@ DECLSPEC void m21100s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); /* * sha1 diff --git a/OpenCL/m21200_a0-optimized.cl b/OpenCL/m21200_a0-optimized.cl index cd3cf1d51..bb1c0ea8e 100644 --- a/OpenCL/m21200_a0-optimized.cl +++ b/OpenCL/m21200_a0-optimized.cl @@ -212,10 +212,10 @@ KERNEL_FQ void m21200_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); const u32x a1 = hc_swap32 (a); const u32x b1 = hc_swap32 (b); @@ -433,10 +433,10 @@ KERNEL_FQ void m21200_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += digest[0] - MD5M_A; - b += digest[1] - MD5M_B; - c += digest[2] - MD5M_C; - d += digest[3] - MD5M_D; + a += digest[0] - make_u32x (MD5M_A); + b += digest[1] - make_u32x (MD5M_B); + c += digest[2] - make_u32x (MD5M_C); + d += digest[3] - make_u32x (MD5M_D); COMPARE_M_SIMD (a, d, c, b); } @@ -646,10 +646,10 @@ KERNEL_FQ void m21200_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); const u32x a1 = hc_swap32 (a); const u32x b1 = hc_swap32 (b); @@ -870,10 +870,10 @@ KERNEL_FQ void m21200_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += digest[0] - MD5M_A; - b += digest[1] - MD5M_B; - c += digest[2] - MD5M_C; - d += digest[3] - MD5M_D; + a += digest[0] - make_u32x (MD5M_A); + b += digest[1] - make_u32x (MD5M_B); + c += digest[2] - make_u32x (MD5M_C); + d += digest[3] - make_u32x (MD5M_D); COMPARE_S_SIMD (a, d, c, b); } diff --git a/OpenCL/m21200_a1-optimized.cl b/OpenCL/m21200_a1-optimized.cl index d06154844..28262d244 100644 --- a/OpenCL/m21200_a1-optimized.cl +++ b/OpenCL/m21200_a1-optimized.cl @@ -268,10 +268,10 @@ KERNEL_FQ void m21200_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); const u32x a1 = hc_swap32 (a); const u32x b1 = hc_swap32 (b); @@ -489,10 +489,10 @@ KERNEL_FQ void m21200_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += digest[0] - MD5M_A; - b += digest[1] - MD5M_B; - c += digest[2] - MD5M_C; - d += digest[3] - MD5M_D; + a += digest[0] - make_u32x (MD5M_A); + b += digest[1] - make_u32x (MD5M_B); + c += digest[2] - make_u32x (MD5M_C); + d += digest[3] - make_u32x (MD5M_D); COMPARE_M_SIMD (a, d, c, b); } @@ -760,10 +760,10 @@ KERNEL_FQ void m21200_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); const u32x a1 = hc_swap32 (a); const u32x b1 = hc_swap32 (b); @@ -984,10 +984,10 @@ KERNEL_FQ void m21200_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += digest[0] - MD5M_A; - b += digest[1] - MD5M_B; - c += digest[2] - MD5M_C; - d += digest[3] - MD5M_D; + a += digest[0] - make_u32x (MD5M_A); + b += digest[1] - make_u32x (MD5M_B); + c += digest[2] - make_u32x (MD5M_C); + d += digest[3] - make_u32x (MD5M_D); COMPARE_S_SIMD (a, d, c, b); } diff --git a/OpenCL/m21200_a3-optimized.cl b/OpenCL/m21200_a3-optimized.cl index 772c7e60c..74292b4f1 100644 --- a/OpenCL/m21200_a3-optimized.cl +++ b/OpenCL/m21200_a3-optimized.cl @@ -166,10 +166,10 @@ DECLSPEC void m21200m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); const u32x a1 = hc_swap32 (a); const u32x b1 = hc_swap32 (b); @@ -387,10 +387,10 @@ DECLSPEC void m21200m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += digest[0] - MD5M_A; - b += digest[1] - MD5M_B; - c += digest[2] - MD5M_C; - d += digest[3] - MD5M_D; + a += digest[0] - make_u32x (MD5M_A); + b += digest[1] - make_u32x (MD5M_B); + c += digest[2] - make_u32x (MD5M_C); + d += digest[3] - make_u32x (MD5M_D); COMPARE_M_SIMD (a, d, c, b); } @@ -548,10 +548,10 @@ DECLSPEC void m21200s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); const u32x a1 = hc_swap32 (a); const u32x b1 = hc_swap32 (b); @@ -766,16 +766,16 @@ DECLSPEC void m21200s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33); MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30); - if (MATCHES_NONE_VS ((a+digest[0]-MD5M_A), search[0])) continue; + if (MATCHES_NONE_VS ((a + digest[0] - make_u32x (MD5M_A)), search[0])) continue; MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31); MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); - a += digest[0] - MD5M_A; - b += digest[1] - MD5M_B; - c += digest[2] - MD5M_C; - d += digest[3] - MD5M_D; + a += digest[0] - make_u32x (MD5M_A); + b += digest[1] - make_u32x (MD5M_B); + c += digest[2] - make_u32x (MD5M_C); + d += digest[3] - make_u32x (MD5M_D); COMPARE_S_SIMD (a, d, c, b); } diff --git a/OpenCL/m21400_a0-optimized.cl b/OpenCL/m21400_a0-optimized.cl index b7e5a037c..7148b8aa0 100644 --- a/OpenCL/m21400_a0-optimized.cl +++ b/OpenCL/m21400_a0-optimized.cl @@ -179,14 +179,14 @@ KERNEL_FQ void m21400_m04 (KERN_ATTR_RULES ()) * sha256 */ - w0_t = a + SHA256M_A; - w1_t = b + SHA256M_B; - w2_t = c + SHA256M_C; - w3_t = d + SHA256M_D; - w4_t = e + SHA256M_E; - w5_t = f + SHA256M_F; - w6_t = g + SHA256M_G; - w7_t = h + SHA256M_H; + w0_t = a + make_u32x (SHA256M_A); + w1_t = b + make_u32x (SHA256M_B); + w2_t = c + make_u32x (SHA256M_C); + w3_t = d + make_u32x (SHA256M_D); + w4_t = e + make_u32x (SHA256M_E); + w5_t = f + make_u32x (SHA256M_F); + w6_t = g + make_u32x (SHA256M_G); + w7_t = h + make_u32x (SHA256M_H); w8_t = 0x80000000; w9_t = 0; wa_t = 0; @@ -464,14 +464,14 @@ KERNEL_FQ void m21400_s04 (KERN_ATTR_RULES ()) * sha256 */ - w0_t = a + SHA256M_A; - w1_t = b + SHA256M_B; - w2_t = c + SHA256M_C; - w3_t = d + SHA256M_D; - w4_t = e + SHA256M_E; - w5_t = f + SHA256M_F; - w6_t = g + SHA256M_G; - w7_t = h + SHA256M_H; + w0_t = a + make_u32x (SHA256M_A); + w1_t = b + make_u32x (SHA256M_B); + w2_t = c + make_u32x (SHA256M_C); + w3_t = d + make_u32x (SHA256M_D); + w4_t = e + make_u32x (SHA256M_E); + w5_t = f + make_u32x (SHA256M_F); + w6_t = g + make_u32x (SHA256M_G); + w7_t = h + make_u32x (SHA256M_H); w8_t = 0x80000000; w9_t = 0; wa_t = 0; diff --git a/OpenCL/m21400_a1-optimized.cl b/OpenCL/m21400_a1-optimized.cl index cbf78d5df..c71550a34 100644 --- a/OpenCL/m21400_a1-optimized.cl +++ b/OpenCL/m21400_a1-optimized.cl @@ -235,14 +235,14 @@ KERNEL_FQ void m21400_m04 (KERN_ATTR_BASIC ()) * sha256 */ - w0_t = a + SHA256M_A; - w1_t = b + SHA256M_B; - w2_t = c + SHA256M_C; - w3_t = d + SHA256M_D; - w4_t = e + SHA256M_E; - w5_t = f + SHA256M_F; - w6_t = g + SHA256M_G; - w7_t = h + SHA256M_H; + w0_t = a + make_u32x (SHA256M_A); + w1_t = b + make_u32x (SHA256M_B); + w2_t = c + make_u32x (SHA256M_C); + w3_t = d + make_u32x (SHA256M_D); + w4_t = e + make_u32x (SHA256M_E); + w5_t = f + make_u32x (SHA256M_F); + w6_t = g + make_u32x (SHA256M_G); + w7_t = h + make_u32x (SHA256M_H); w8_t = 0x80000000; w9_t = 0; wa_t = 0; @@ -578,14 +578,14 @@ KERNEL_FQ void m21400_s04 (KERN_ATTR_BASIC ()) * sha256 */ - w0_t = a + SHA256M_A; - w1_t = b + SHA256M_B; - w2_t = c + SHA256M_C; - w3_t = d + SHA256M_D; - w4_t = e + SHA256M_E; - w5_t = f + SHA256M_F; - w6_t = g + SHA256M_G; - w7_t = h + SHA256M_H; + w0_t = a + make_u32x (SHA256M_A); + w1_t = b + make_u32x (SHA256M_B); + w2_t = c + make_u32x (SHA256M_C); + w3_t = d + make_u32x (SHA256M_D); + w4_t = e + make_u32x (SHA256M_E); + w5_t = f + make_u32x (SHA256M_F); + w6_t = g + make_u32x (SHA256M_G); + w7_t = h + make_u32x (SHA256M_H); w8_t = 0x80000000; w9_t = 0; wa_t = 0; diff --git a/OpenCL/m21400_a3-optimized.cl b/OpenCL/m21400_a3-optimized.cl index 6ffd6a1be..8fbbeb2b5 100644 --- a/OpenCL/m21400_a3-optimized.cl +++ b/OpenCL/m21400_a3-optimized.cl @@ -146,14 +146,14 @@ DECLSPEC void m21400m (PRIVATE_AS u32 *w, const u32 pw_len, KERN_ATTR_FUNC_VECTO * sha256 */ - w0_t = a + SHA256M_A; - w1_t = b + SHA256M_B; - w2_t = c + SHA256M_C; - w3_t = d + SHA256M_D; - w4_t = e + SHA256M_E; - w5_t = f + SHA256M_F; - w6_t = g + SHA256M_G; - w7_t = h + SHA256M_H; + w0_t = a + make_u32x (SHA256M_A); + w1_t = b + make_u32x (SHA256M_B); + w2_t = c + make_u32x (SHA256M_C); + w3_t = d + make_u32x (SHA256M_D); + w4_t = e + make_u32x (SHA256M_E); + w5_t = f + make_u32x (SHA256M_F); + w6_t = g + make_u32x (SHA256M_G); + w7_t = h + make_u32x (SHA256M_H); w8_t = 0x80000000; w9_t = 0; wa_t = 0; @@ -392,14 +392,14 @@ DECLSPEC void m21400s (PRIVATE_AS u32 *w, const u32 pw_len, KERN_ATTR_FUNC_VECTO * sha256 */ - w0_t = a + SHA256M_A; - w1_t = b + SHA256M_B; - w2_t = c + SHA256M_C; - w3_t = d + SHA256M_D; - w4_t = e + SHA256M_E; - w5_t = f + SHA256M_F; - w6_t = g + SHA256M_G; - w7_t = h + SHA256M_H; + w0_t = a + make_u32x (SHA256M_A); + w1_t = b + make_u32x (SHA256M_B); + w2_t = c + make_u32x (SHA256M_C); + w3_t = d + make_u32x (SHA256M_D); + w4_t = e + make_u32x (SHA256M_E); + w5_t = f + make_u32x (SHA256M_F); + w6_t = g + make_u32x (SHA256M_G); + w7_t = h + make_u32x (SHA256M_H); w8_t = 0x80000000; w9_t = 0; wa_t = 0; diff --git a/OpenCL/m24300_a0-optimized.cl b/OpenCL/m24300_a0-optimized.cl index 90e733fdd..5f595ac64 100644 --- a/OpenCL/m24300_a0-optimized.cl +++ b/OpenCL/m24300_a0-optimized.cl @@ -288,11 +288,11 @@ KERNEL_FQ void m24300_m04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); u32x t0[4]; u32x t1[4]; @@ -471,11 +471,11 @@ KERNEL_FQ void m24300_m04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); t0[0] = c0[0]; t0[1] = c0[1]; @@ -916,11 +916,11 @@ KERNEL_FQ void m24300_s04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); u32x t0[4]; u32x t1[4]; @@ -1099,11 +1099,11 @@ KERNEL_FQ void m24300_s04 (KERN_ATTR_RULES ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); t0[0] = c0[0]; t0[1] = c0[1]; diff --git a/OpenCL/m24300_a1-optimized.cl b/OpenCL/m24300_a1-optimized.cl index 1ac4636ba..153ba475a 100644 --- a/OpenCL/m24300_a1-optimized.cl +++ b/OpenCL/m24300_a1-optimized.cl @@ -342,11 +342,11 @@ KERNEL_FQ void m24300_m04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); u32x t0[4]; u32x t1[4]; @@ -525,11 +525,11 @@ KERNEL_FQ void m24300_m04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); t0[0] = c0[0]; t0[1] = c0[1]; @@ -1025,11 +1025,11 @@ KERNEL_FQ void m24300_s04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); u32x t0[4]; u32x t1[4]; @@ -1208,11 +1208,11 @@ KERNEL_FQ void m24300_s04 (KERN_ATTR_BASIC ()) we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); t0[0] = c0[0]; t0[1] = c0[1]; diff --git a/OpenCL/m24300_a3-optimized.cl b/OpenCL/m24300_a3-optimized.cl index df62cb1d1..c48755420 100644 --- a/OpenCL/m24300_a3-optimized.cl +++ b/OpenCL/m24300_a3-optimized.cl @@ -241,11 +241,11 @@ DECLSPEC void m24300m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); u32x t0[4]; u32x t1[4]; @@ -424,11 +424,11 @@ DECLSPEC void m24300m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); t0[0] = c0[0]; t0[1] = c0[1]; @@ -815,11 +815,11 @@ DECLSPEC void m24300s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); u32x t0[4]; u32x t1[4]; @@ -998,11 +998,11 @@ DECLSPEC void m24300s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); - a += SHA1M_A; - b += SHA1M_B; - c += SHA1M_C; - d += SHA1M_D; - e += SHA1M_E; + a += make_u32x (SHA1M_A); + b += make_u32x (SHA1M_B); + c += make_u32x (SHA1M_C); + d += make_u32x (SHA1M_D); + e += make_u32x (SHA1M_E); t0[0] = c0[0]; t0[1] = c0[1]; diff --git a/OpenCL/m24700_a0-optimized.cl b/OpenCL/m24700_a0-optimized.cl index 553589479..8a351fb6f 100644 --- a/OpenCL/m24700_a0-optimized.cl +++ b/OpenCL/m24700_a0-optimized.cl @@ -137,10 +137,10 @@ KERNEL_FQ void m24700_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = a; w0[1] = b & 0xff; w0[1] |= 0x8000; @@ -381,10 +381,10 @@ KERNEL_FQ void m24700_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = a; w0[1] = b & 0xff; w0[1] |= 0x8000; diff --git a/OpenCL/m24700_a1-optimized.cl b/OpenCL/m24700_a1-optimized.cl index 94d257636..65f3bd1ac 100644 --- a/OpenCL/m24700_a1-optimized.cl +++ b/OpenCL/m24700_a1-optimized.cl @@ -194,10 +194,10 @@ KERNEL_FQ void m24700_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = a; w0[1] = b & 0xff; w0[1] |= 0x8000; @@ -497,10 +497,10 @@ KERNEL_FQ void m24700_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0[0] = a; w0[1] = b & 0xff; w0[1] |= 0x8000; diff --git a/OpenCL/m24700_a3-optimized.cl b/OpenCL/m24700_a3-optimized.cl index 65863064b..231f129de 100644 --- a/OpenCL/m24700_a3-optimized.cl +++ b/OpenCL/m24700_a3-optimized.cl @@ -133,10 +133,10 @@ DECLSPEC void m24700m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = a; w0_t[1] = b & 0xff; w0_t[1] |= 0x8000; @@ -367,10 +367,10 @@ DECLSPEC void m24700s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); w0_t[0] = a; w0_t[1] = b & 0xff; w0_t[1] |= 0x8000; diff --git a/OpenCL/m24900_a0-optimized.cl b/OpenCL/m24900_a0-optimized.cl index ee1277268..f22a0c59b 100644 --- a/OpenCL/m24900_a0-optimized.cl +++ b/OpenCL/m24900_a0-optimized.cl @@ -139,10 +139,10 @@ KERNEL_FQ void m24900_m04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); const u32x a0 = (((a >> 0) & 0xff) + ((a >> 8) & 0xff)) % 62; const u32x a1 = (((a >> 16) & 0xff) + ((a >> 24) & 0xff)) % 62; @@ -305,10 +305,10 @@ KERNEL_FQ void m24900_s04 (KERN_ATTR_RULES ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); const u32x a0 = (((a >> 0) & 0xff) + ((a >> 8) & 0xff)) % 62; const u32x a1 = (((a >> 16) & 0xff) + ((a >> 24) & 0xff)) % 62; diff --git a/OpenCL/m24900_a1-optimized.cl b/OpenCL/m24900_a1-optimized.cl index ff2e37bcc..9bff24186 100644 --- a/OpenCL/m24900_a1-optimized.cl +++ b/OpenCL/m24900_a1-optimized.cl @@ -197,10 +197,10 @@ KERNEL_FQ void m24900_m04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); const u32x a0 = (((a >> 0) & 0xff) + ((a >> 8) & 0xff)) % 62; const u32x a1 = (((a >> 16) & 0xff) + ((a >> 24) & 0xff)) % 62; @@ -422,10 +422,10 @@ KERNEL_FQ void m24900_s04 (KERN_ATTR_BASIC ()) MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); const u32x a0 = (((a >> 0) & 0xff) + ((a >> 8) & 0xff)) % 62; const u32x a1 = (((a >> 16) & 0xff) + ((a >> 24) & 0xff)) % 62; diff --git a/OpenCL/m24900_a3-optimized.cl b/OpenCL/m24900_a3-optimized.cl index 71969b811..432dab27b 100644 --- a/OpenCL/m24900_a3-optimized.cl +++ b/OpenCL/m24900_a3-optimized.cl @@ -133,10 +133,10 @@ DECLSPEC void m24900m (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, t0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, t2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); const u32x a0 = (((a >> 0) & 0xff) + ((a >> 8) & 0xff)) % 62; const u32x a1 = (((a >> 16) & 0xff) + ((a >> 24) & 0xff)) % 62; @@ -287,10 +287,10 @@ DECLSPEC void m24900s (PRIVATE_AS u32 *w0, PRIVATE_AS u32 *w1, PRIVATE_AS u32 *w MD5_STEP (MD5_I , c, d, a, b, t0[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, t2[1], MD5C3f, MD5S33); - a += MD5M_A; - b += MD5M_B; - c += MD5M_C; - d += MD5M_D; + a += make_u32x (MD5M_A); + b += make_u32x (MD5M_B); + c += make_u32x (MD5M_C); + d += make_u32x (MD5M_D); const u32x a0 = (((a >> 0) & 0xff) + ((a >> 8) & 0xff)) % 62; const u32x a1 = (((a >> 16) & 0xff) + ((a >> 24) & 0xff)) % 62; diff --git a/src/backend.c b/src/backend.c index da6577e22..2bf7ddca1 100644 --- a/src/backend.c +++ b/src/backend.c @@ -8823,6 +8823,13 @@ int backend_session_begin (hashcat_ctx_t *hashcat_ctx) vector_width = user_options->backend_vector_width; } + // Metal supports vectors up to 4 + + if (device_param->is_metal == true && vector_width > 4) + { + vector_width = 4; + } + // We can't have SIMD in kernels where we have an unknown final password length // It also turns out that pure kernels (that have a higher register pressure) // actually run faster on scalar GPU (like 1080) without SIMD diff --git a/src/ext_metal.m b/src/ext_metal.m index c4e8a3a03..6bed7641b 100644 --- a/src/ext_metal.m +++ b/src/ext_metal.m @@ -166,7 +166,14 @@ static int hc_mtlBuildOptionsToDict (void *hashcat_ctx, const char *build_option } else { - //event_log_warning (hashcat_ctx, "%s(): skipping malformed build option: %s", __func__, [key UTF8String]); + #ifdef DEBUG + const char *tmp = [key UTF8String]; + + if (tmp != nil && strlen (tmp) > 0) + { + event_log_warning (hashcat_ctx, "%s(): skipping malformed build option: '%s'", __func__, tmp); + } + #endif continue; }