aboutsummaryrefslogtreecommitdiffstats
path: root/src/block/serpent_sse2/serp_sse2_sbox.h
diff options
context:
space:
mode:
Diffstat (limited to 'src/block/serpent_sse2/serp_sse2_sbox.h')
-rw-r--r--src/block/serpent_sse2/serp_sse2_sbox.h434
1 files changed, 434 insertions, 0 deletions
diff --git a/src/block/serpent_sse2/serp_sse2_sbox.h b/src/block/serpent_sse2/serp_sse2_sbox.h
new file mode 100644
index 000000000..40c552e87
--- /dev/null
+++ b/src/block/serpent_sse2/serp_sse2_sbox.h
@@ -0,0 +1,434 @@
+/*
+* Serpent Sboxes in SSE2 form
+* (C) 2009 Jack Lloyd
+*
+* Distributed under the terms of the Botan license
+*/
+
+#ifndef SERPENT_SSE2_SBOXES_H__
+#define SERPENT_SSE2_SBOXES_H__
+
+#define SBoxE1(B0, B1, B2, B3) \
+ do { \
+ B3 = _mm_xor_si128(B3, B0); \
+ __m128i B4 = B1; \
+ B1 = _mm_and_si128(B1, B3); \
+ B4 = _mm_xor_si128(B4, B2); \
+ B1 = _mm_xor_si128(B1, B0); \
+ B0 = _mm_or_si128(B0, B3); \
+ B0 = _mm_xor_si128(B0, B4); \
+ B4 = _mm_xor_si128(B4, B3); \
+ B3 = _mm_xor_si128(B3, B2); \
+ B2 = _mm_or_si128(B2, B1); \
+ B2 = _mm_xor_si128(B2, B4); \
+ B4 = _mm_xor_si128(B4, all_ones); \
+ B4 = _mm_or_si128(B4, B1); \
+ B1 = _mm_xor_si128(B1, B3); \
+ B1 = _mm_xor_si128(B1, B4); \
+ B3 = _mm_or_si128(B3, B0); \
+ B1 = _mm_xor_si128(B1, B3); \
+ B4 = _mm_xor_si128(B4, B3); \
+ B3 = B0; \
+ B0 = B1; \
+ B1 = B4; \
+ } while(0);
+
+#define SBoxE2(B0, B1, B2, B3) \
+ do { \
+ B0 = _mm_xor_si128(B0, all_ones); \
+ B2 = _mm_xor_si128(B2, all_ones); \
+ __m128i B4 = B0; \
+ B0 = _mm_and_si128(B0, B1); \
+ B2 = _mm_xor_si128(B2, B0); \
+ B0 = _mm_or_si128(B0, B3); \
+ B3 = _mm_xor_si128(B3, B2); \
+ B1 = _mm_xor_si128(B1, B0); \
+ B0 = _mm_xor_si128(B0, B4); \
+ B4 = _mm_or_si128(B4, B1); \
+ B1 = _mm_xor_si128(B1, B3); \
+ B2 = _mm_or_si128(B2, B0); \
+ B2 = _mm_and_si128(B2, B4); \
+ B0 = _mm_xor_si128(B0, B1); \
+ B1 = _mm_and_si128(B1, B2); \
+ B1 = _mm_xor_si128(B1, B0); \
+ B0 = _mm_and_si128(B0, B2); \
+ B4 = _mm_xor_si128(B4, B0); \
+ B0 = B2; \
+ B2 = B3; \
+ B3 = B1; \
+ B1 = B4; \
+ } while(0);
+
+#define SBoxE3(B0, B1, B2, B3) \
+ do { \
+ __m128i B4 = B0; \
+ B0 = _mm_and_si128(B0, B2); \
+ B0 = _mm_xor_si128(B0, B3); \
+ B2 = _mm_xor_si128(B2, B1); \
+ B2 = _mm_xor_si128(B2, B0); \
+ B3 = _mm_or_si128(B3, B4); \
+ B3 = _mm_xor_si128(B3, B1); \
+ B4 = _mm_xor_si128(B4, B2); \
+ B1 = B3; \
+ B3 = _mm_or_si128(B3, B4); \
+ B3 = _mm_xor_si128(B3, B0); \
+ B0 = _mm_and_si128(B0, B1); \
+ B4 = _mm_xor_si128(B4, B0); \
+ B1 = _mm_xor_si128(B1, B3); \
+ B1 = _mm_xor_si128(B1, B4); \
+ B4 = _mm_xor_si128(B4, all_ones); \
+ B0 = B2; \
+ B2 = B1; \
+ B1 = B3; \
+ B3 = B4; \
+ } while(0);
+
+#define SBoxE4(B0, B1, B2, B3) \
+ do { \
+ __m128i B4 = B0; \
+ B0 = _mm_or_si128(B0, B3); \
+ B3 = _mm_xor_si128(B3, B1); \
+ B1 = _mm_and_si128(B1, B4); \
+ B4 = _mm_xor_si128(B4, B2); \
+ B2 = _mm_xor_si128(B2, B3); \
+ B3 = _mm_and_si128(B3, B0); \
+ B4 = _mm_or_si128(B4, B1); \
+ B3 = _mm_xor_si128(B3, B4); \
+ B0 = _mm_xor_si128(B0, B1); \
+ B4 = _mm_and_si128(B4, B0); \
+ B1 = _mm_xor_si128(B1, B3); \
+ B4 = _mm_xor_si128(B4, B2); \
+ B1 = _mm_or_si128(B1, B0); \
+ B1 = _mm_xor_si128(B1, B2); \
+ B0 = _mm_xor_si128(B0, B3); \
+ B2 = B1; \
+ B1 = _mm_or_si128(B1, B3); \
+ B0 = _mm_xor_si128(B0, B1); \
+ B1 = B2; \
+ B2 = B3; \
+ B3 = B4; \
+ } while(0);
+
+#define SBoxE5(B0, B1, B2, B3) \
+ do { \
+ B1 = _mm_xor_si128(B1, B3); \
+ B3 = _mm_xor_si128(B3, all_ones); \
+ B2 = _mm_xor_si128(B2, B3); \
+ B3 = _mm_xor_si128(B3, B0); \
+ __m128i B4 = B1; \
+ B1 = _mm_and_si128(B1, B3); \
+ B1 = _mm_xor_si128(B1, B2); \
+ B4 = _mm_xor_si128(B4, B3); \
+ B0 = _mm_xor_si128(B0, B4); \
+ B2 = _mm_and_si128(B2, B4); \
+ B2 = _mm_xor_si128(B2, B0); \
+ B0 = _mm_and_si128(B0, B1); \
+ B3 = _mm_xor_si128(B3, B0); \
+ B4 = _mm_or_si128(B4, B1); \
+ B4 = _mm_xor_si128(B4, B0); \
+ B0 = _mm_or_si128(B0, B3); \
+ B0 = _mm_xor_si128(B0, B2); \
+ B2 = _mm_and_si128(B2, B3); \
+ B0 = _mm_xor_si128(B0, all_ones); \
+ B4 = _mm_xor_si128(B4, B2); \
+ B2 = B0; \
+ B0 = B1; \
+ B1 = B4; \
+ } while(0);
+
+#define SBoxE6(B0, B1, B2, B3) \
+ do { \
+ B0 = _mm_xor_si128(B0, B1); \
+ B1 = _mm_xor_si128(B1, B3); \
+ B3 = _mm_xor_si128(B3, all_ones); \
+ __m128i B4 = B1; \
+ B1 = _mm_and_si128(B1, B0); \
+ B2 = _mm_xor_si128(B2, B3); \
+ B1 = _mm_xor_si128(B1, B2); \
+ B2 = _mm_or_si128(B2, B4); \
+ B4 = _mm_xor_si128(B4, B3); \
+ B3 = _mm_and_si128(B3, B1); \
+ B3 = _mm_xor_si128(B3, B0); \
+ B4 = _mm_xor_si128(B4, B1); \
+ B4 = _mm_xor_si128(B4, B2); \
+ B2 = _mm_xor_si128(B2, B0); \
+ B0 = _mm_and_si128(B0, B3); \
+ B2 = _mm_xor_si128(B2, all_ones); \
+ B0 = _mm_xor_si128(B0, B4); \
+ B4 = _mm_or_si128(B4, B3); \
+ B4 = _mm_xor_si128(B4, B2); \
+ B2 = B0; \
+ B0 = B1; \
+ B1 = B3; \
+ B3 = B4; \
+ } while(0);
+
+#define SBoxE7(B0, B1, B2, B3) \
+ do { \
+ B2 = _mm_xor_si128(B2, all_ones); \
+ __m128i B4 = B3; \
+ B3 = _mm_and_si128(B3, B0); \
+ B0 = _mm_xor_si128(B0, B4); \
+ B3 = _mm_xor_si128(B3, B2); \
+ B2 = _mm_or_si128(B2, B4); \
+ B1 = _mm_xor_si128(B1, B3); \
+ B2 = _mm_xor_si128(B2, B0); \
+ B0 = _mm_or_si128(B0, B1); \
+ B2 = _mm_xor_si128(B2, B1); \
+ B4 = _mm_xor_si128(B4, B0); \
+ B0 = _mm_or_si128(B0, B3); \
+ B0 = _mm_xor_si128(B0, B2); \
+ B4 = _mm_xor_si128(B4, B3); \
+ B4 = _mm_xor_si128(B4, B0); \
+ B3 = _mm_xor_si128(B3, all_ones); \
+ B2 = _mm_and_si128(B2, B4); \
+ B3 = _mm_xor_si128(B3, B2); \
+ B2 = B4; \
+ } while(0);
+
+#define SBoxE8(B0, B1, B2, B3) \
+ do { \
+ __m128i B4 = B1; \
+ B1 = _mm_or_si128(B1, B2); \
+ B1 = _mm_xor_si128(B1, B3); \
+ B4 = _mm_xor_si128(B4, B2); \
+ B2 = _mm_xor_si128(B2, B1); \
+ B3 = _mm_or_si128(B3, B4); \
+ B3 = _mm_and_si128(B3, B0); \
+ B4 = _mm_xor_si128(B4, B2); \
+ B3 = _mm_xor_si128(B3, B1); \
+ B1 = _mm_or_si128(B1, B4); \
+ B1 = _mm_xor_si128(B1, B0); \
+ B0 = _mm_or_si128(B0, B4); \
+ B0 = _mm_xor_si128(B0, B2); \
+ B1 = _mm_xor_si128(B1, B4); \
+ B2 = _mm_xor_si128(B2, B1); \
+ B1 = _mm_and_si128(B1, B0); \
+ B1 = _mm_xor_si128(B1, B4); \
+ B2 = _mm_xor_si128(B2, all_ones); \
+ B2 = _mm_or_si128(B2, B0); \
+ B4 = _mm_xor_si128(B4, B2); \
+ B2 = B1; \
+ B1 = B3; \
+ B3 = B0; \
+ B0 = B4; \
+ } while(0);
+
+#define SBoxD1(B0, B1, B2, B3) \
+ do \
+ { \
+ B2 = _mm_xor_si128(B2, all_ones); \
+ __m128i B4 = B1; \
+ B1 = _mm_or_si128(B1, B0); \
+ B4 = _mm_xor_si128(B4, all_ones); \
+ B1 = _mm_xor_si128(B1, B2); \
+ B2 = _mm_or_si128(B2, B4); \
+ B1 = _mm_xor_si128(B1, B3); \
+ B0 = _mm_xor_si128(B0, B4); \
+ B2 = _mm_xor_si128(B2, B0); \
+ B0 = _mm_and_si128(B0, B3); \
+ B4 = _mm_xor_si128(B4, B0); \
+ B0 = _mm_or_si128(B0, B1); \
+ B0 = _mm_xor_si128(B0, B2); \
+ B3 = _mm_xor_si128(B3, B4); \
+ B2 = _mm_xor_si128(B2, B1); \
+ B3 = _mm_xor_si128(B3, B0); \
+ B3 = _mm_xor_si128(B3, B1); \
+ B2 = _mm_and_si128(B2, B3); \
+ B4 = _mm_xor_si128(B4, B2); \
+ B2 = B1; \
+ B1 = B4; \
+ } while(0);
+
+#define SBoxD2(B0, B1, B2, B3) \
+ do \
+ { \
+ __m128i B4 = B1; \
+ B1 = _mm_xor_si128(B1, B3); \
+ B3 = _mm_and_si128(B3, B1); \
+ B4 = _mm_xor_si128(B4, B2); \
+ B3 = _mm_xor_si128(B3, B0); \
+ B0 = _mm_or_si128(B0, B1); \
+ B2 = _mm_xor_si128(B2, B3); \
+ B0 = _mm_xor_si128(B0, B4); \
+ B0 = _mm_or_si128(B0, B2); \
+ B1 = _mm_xor_si128(B1, B3); \
+ B0 = _mm_xor_si128(B0, B1); \
+ B1 = _mm_or_si128(B1, B3); \
+ B1 = _mm_xor_si128(B1, B0); \
+ B4 = _mm_xor_si128(B4, all_ones); \
+ B4 = _mm_xor_si128(B4, B1); \
+ B1 = _mm_or_si128(B1, B0); \
+ B1 = _mm_xor_si128(B1, B0); \
+ B1 = _mm_or_si128(B1, B4); \
+ B3 = _mm_xor_si128(B3, B1); \
+ B1 = B0; \
+ B0 = B4; \
+ B4 = B2; \
+ B2 = B3; \
+ B3 = B4; \
+ } while(0);
+
+#define SBoxD3(B0, B1, B2, B3) \
+ do \
+ { \
+ B2 = _mm_xor_si128(B2, B3); \
+ B3 = _mm_xor_si128(B3, B0); \
+ __m128i B4 = B3; \
+ B3 = _mm_and_si128(B3, B2); \
+ B3 = _mm_xor_si128(B3, B1); \
+ B1 = _mm_or_si128(B1, B2); \
+ B1 = _mm_xor_si128(B1, B4); \
+ B4 = _mm_and_si128(B4, B3); \
+ B2 = _mm_xor_si128(B2, B3); \
+ B4 = _mm_and_si128(B4, B0); \
+ B4 = _mm_xor_si128(B4, B2); \
+ B2 = _mm_and_si128(B2, B1); \
+ B2 = _mm_or_si128(B2, B0); \
+ B3 = _mm_xor_si128(B3, all_ones); \
+ B2 = _mm_xor_si128(B2, B3); \
+ B0 = _mm_xor_si128(B0, B3); \
+ B0 = _mm_and_si128(B0, B1); \
+ B3 = _mm_xor_si128(B3, B4); \
+ B3 = _mm_xor_si128(B3, B0); \
+ B0 = B1; \
+ B1 = B4; \
+ } while(0);
+
+#define SBoxD4(B0, B1, B2, B3) \
+ do \
+ { \
+ __m128i B4 = B2; \
+ B2 = _mm_xor_si128(B2, B1); \
+ B0 = _mm_xor_si128(B0, B2); \
+ B4 = _mm_and_si128(B4, B2); \
+ B4 = _mm_xor_si128(B4, B0); \
+ B0 = _mm_and_si128(B0, B1); \
+ B1 = _mm_xor_si128(B1, B3); \
+ B3 = _mm_or_si128(B3, B4); \
+ B2 = _mm_xor_si128(B2, B3); \
+ B0 = _mm_xor_si128(B0, B3); \
+ B1 = _mm_xor_si128(B1, B4); \
+ B3 = _mm_and_si128(B3, B2); \
+ B3 = _mm_xor_si128(B3, B1); \
+ B1 = _mm_xor_si128(B1, B0); \
+ B1 = _mm_or_si128(B1, B2); \
+ B0 = _mm_xor_si128(B0, B3); \
+ B1 = _mm_xor_si128(B1, B4); \
+ B0 = _mm_xor_si128(B0, B1); \
+ B4 = B0; \
+ B0 = B2; \
+ B2 = B3; \
+ B3 = B4; \
+ } while(0);
+
+#define SBoxD5(B0, B1, B2, B3) \
+ do \
+ { \
+ __m128i B4 = B2; \
+ B2 = _mm_and_si128(B2, B3); \
+ B2 = _mm_xor_si128(B2, B1); \
+ B1 = _mm_or_si128(B1, B3); \
+ B1 = _mm_and_si128(B1, B0); \
+ B4 = _mm_xor_si128(B4, B2); \
+ B4 = _mm_xor_si128(B4, B1); \
+ B1 = _mm_and_si128(B1, B2); \
+ B0 = _mm_xor_si128(B0, all_ones); \
+ B3 = _mm_xor_si128(B3, B4); \
+ B1 = _mm_xor_si128(B1, B3); \
+ B3 = _mm_and_si128(B3, B0); \
+ B3 = _mm_xor_si128(B3, B2); \
+ B0 = _mm_xor_si128(B0, B1); \
+ B2 = _mm_and_si128(B2, B0); \
+ B3 = _mm_xor_si128(B3, B0); \
+ B2 = _mm_xor_si128(B2, B4); \
+ B2 = _mm_or_si128(B2, B3); \
+ B3 = _mm_xor_si128(B3, B0); \
+ B2 = _mm_xor_si128(B2, B1); \
+ B1 = B3; \
+ B3 = B4; \
+ } while(0);
+
+#define SBoxD6(B0, B1, B2, B3) \
+ do \
+ { \
+ B1 = _mm_xor_si128(B1, all_ones); \
+ __m128i B4 = B3; \
+ B2 = _mm_xor_si128(B2, B1); \
+ B3 = _mm_or_si128(B3, B0); \
+ B3 = _mm_xor_si128(B3, B2); \
+ B2 = _mm_or_si128(B2, B1); \
+ B2 = _mm_and_si128(B2, B0); \
+ B4 = _mm_xor_si128(B4, B3); \
+ B2 = _mm_xor_si128(B2, B4); \
+ B4 = _mm_or_si128(B4, B0); \
+ B4 = _mm_xor_si128(B4, B1); \
+ B1 = _mm_and_si128(B1, B2); \
+ B1 = _mm_xor_si128(B1, B3); \
+ B4 = _mm_xor_si128(B4, B2); \
+ B3 = _mm_and_si128(B3, B4); \
+ B4 = _mm_xor_si128(B4, B1); \
+ B3 = _mm_xor_si128(B3, B4); \
+ B4 = _mm_xor_si128(B4, all_ones); \
+ B3 = _mm_xor_si128(B3, B0); \
+ B0 = B1; \
+ B1 = B4; \
+ B4 = B3; \
+ B3 = B2; \
+ B2 = B4; \
+ } while(0);
+
+#define SBoxD7(B0, B1, B2, B3) \
+ do \
+ { \
+ B0 = _mm_xor_si128(B0, B2); \
+ __m128i B4 = B2; \
+ B2 = _mm_and_si128(B2, B0); \
+ B4 = _mm_xor_si128(B4, B3); \
+ B2 = _mm_xor_si128(B2, all_ones); \
+ B3 = _mm_xor_si128(B3, B1); \
+ B2 = _mm_xor_si128(B2, B3); \
+ B4 = _mm_or_si128(B4, B0); \
+ B0 = _mm_xor_si128(B0, B2); \
+ B3 = _mm_xor_si128(B3, B4); \
+ B4 = _mm_xor_si128(B4, B1); \
+ B1 = _mm_and_si128(B1, B3); \
+ B1 = _mm_xor_si128(B1, B0); \
+ B0 = _mm_xor_si128(B0, B3); \
+ B0 = _mm_or_si128(B0, B2); \
+ B3 = _mm_xor_si128(B3, B1); \
+ B4 = _mm_xor_si128(B4, B0); \
+ B0 = B1; \
+ B1 = B2; \
+ B2 = B4; \
+ } while(0);
+
+#define SBoxD8(B0, B1, B2, B3) \
+ do \
+ { \
+ __m128i B4 = B2; \
+ B2 = _mm_xor_si128(B2, B0); \
+ B0 = _mm_and_si128(B0, B3); \
+ B4 = _mm_or_si128(B4, B3); \
+ B2 = _mm_xor_si128(B2, all_ones); \
+ B3 = _mm_xor_si128(B3, B1); \
+ B1 = _mm_or_si128(B1, B0); \
+ B0 = _mm_xor_si128(B0, B2); \
+ B2 = _mm_and_si128(B2, B4); \
+ B3 = _mm_and_si128(B3, B4); \
+ B1 = _mm_xor_si128(B1, B2); \
+ B2 = _mm_xor_si128(B2, B0); \
+ B0 = _mm_or_si128(B0, B2); \
+ B4 = _mm_xor_si128(B4, B1); \
+ B0 = _mm_xor_si128(B0, B3); \
+ B3 = _mm_xor_si128(B3, B4); \
+ B4 = _mm_or_si128(B4, B0); \
+ B3 = _mm_xor_si128(B3, B2); \
+ B4 = _mm_xor_si128(B4, B2); \
+ B2 = B1; \
+ B1 = B0; \
+ B0 = B3; \
+ B3 = B4; \
+ } while(0);
+
+#endif
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-18 18:16:06 +0000