summaryrefslogtreecommitdiffstats
path: root/arch/x86/crypto/twofish-x86_64-asm_64-3way.S
diff options
context:
space:
mode:
Diffstat (limited to 'arch/x86/crypto/twofish-x86_64-asm_64-3way.S')
-rw-r--r--arch/x86/crypto/twofish-x86_64-asm_64-3way.S305
1 files changed, 305 insertions, 0 deletions
diff --git a/arch/x86/crypto/twofish-x86_64-asm_64-3way.S b/arch/x86/crypto/twofish-x86_64-asm_64-3way.S
new file mode 100644
index 000000000..d2288bf38
--- /dev/null
+++ b/arch/x86/crypto/twofish-x86_64-asm_64-3way.S
@@ -0,0 +1,305 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Twofish Cipher 3-way parallel algorithm (x86_64)
+ *
+ * Copyright (C) 2011 Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
+ */
+
+#include <linux/linkage.h>
+
+.file "twofish-x86_64-asm-3way.S"
+.text
+
+/* structure of crypto context */
+#define s0 0
+#define s1 1024
+#define s2 2048
+#define s3 3072
+#define w 4096
+#define k 4128
+
+/**********************************************************************
+ 3-way twofish
+ **********************************************************************/
+#define CTX %rdi
+#define RIO %rdx
+
+#define RAB0 %rax
+#define RAB1 %rbx
+#define RAB2 %rcx
+
+#define RAB0d %eax
+#define RAB1d %ebx
+#define RAB2d %ecx
+
+#define RAB0bh %ah
+#define RAB1bh %bh
+#define RAB2bh %ch
+
+#define RAB0bl %al
+#define RAB1bl %bl
+#define RAB2bl %cl
+
+#define CD0 0x0(%rsp)
+#define CD1 0x8(%rsp)
+#define CD2 0x10(%rsp)
+
+# used only before/after all rounds
+#define RCD0 %r8
+#define RCD1 %r9
+#define RCD2 %r10
+
+# used only during rounds
+#define RX0 %r8
+#define RX1 %r9
+#define RX2 %r10
+
+#define RX0d %r8d
+#define RX1d %r9d
+#define RX2d %r10d
+
+#define RY0 %r11
+#define RY1 %r12
+#define RY2 %r13
+
+#define RY0d %r11d
+#define RY1d %r12d
+#define RY2d %r13d
+
+#define RT0 %rdx
+#define RT1 %rsi
+
+#define RT0d %edx
+#define RT1d %esi
+
+#define RT1bl %sil
+
+#define do16bit_ror(rot, op1, op2, T0, T1, tmp1, tmp2, ab, dst) \
+ movzbl ab ## bl, tmp2 ## d; \
+ movzbl ab ## bh, tmp1 ## d; \
+ rorq $(rot), ab; \
+ op1##l T0(CTX, tmp2, 4), dst ## d; \
+ op2##l T1(CTX, tmp1, 4), dst ## d;
+
+#define swap_ab_with_cd(ab, cd, tmp) \
+ movq cd, tmp; \
+ movq ab, cd; \
+ movq tmp, ab;
+
+/*
+ * Combined G1 & G2 function. Reordered with help of rotates to have moves
+ * at beginning.
+ */
+#define g1g2_3(ab, cd, Tx0, Tx1, Tx2, Tx3, Ty0, Ty1, Ty2, Ty3, x, y) \
+ /* G1,1 && G2,1 */ \
+ do16bit_ror(32, mov, xor, Tx0, Tx1, RT0, x ## 0, ab ## 0, x ## 0); \
+ do16bit_ror(48, mov, xor, Ty1, Ty2, RT0, y ## 0, ab ## 0, y ## 0); \
+ \
+ do16bit_ror(32, mov, xor, Tx0, Tx1, RT0, x ## 1, ab ## 1, x ## 1); \
+ do16bit_ror(48, mov, xor, Ty1, Ty2, RT0, y ## 1, ab ## 1, y ## 1); \
+ \
+ do16bit_ror(32, mov, xor, Tx0, Tx1, RT0, x ## 2, ab ## 2, x ## 2); \
+ do16bit_ror(48, mov, xor, Ty1, Ty2, RT0, y ## 2, ab ## 2, y ## 2); \
+ \
+ /* G1,2 && G2,2 */ \
+ do16bit_ror(32, xor, xor, Tx2, Tx3, RT0, RT1, ab ## 0, x ## 0); \
+ do16bit_ror(16, xor, xor, Ty3, Ty0, RT0, RT1, ab ## 0, y ## 0); \
+ swap_ab_with_cd(ab ## 0, cd ## 0, RT0); \
+ \
+ do16bit_ror(32, xor, xor, Tx2, Tx3, RT0, RT1, ab ## 1, x ## 1); \
+ do16bit_ror(16, xor, xor, Ty3, Ty0, RT0, RT1, ab ## 1, y ## 1); \
+ swap_ab_with_cd(ab ## 1, cd ## 1, RT0); \
+ \
+ do16bit_ror(32, xor, xor, Tx2, Tx3, RT0, RT1, ab ## 2, x ## 2); \
+ do16bit_ror(16, xor, xor, Ty3, Ty0, RT0, RT1, ab ## 2, y ## 2); \
+ swap_ab_with_cd(ab ## 2, cd ## 2, RT0);
+
+#define enc_round_end(ab, x, y, n) \
+ addl y ## d, x ## d; \
+ addl x ## d, y ## d; \
+ addl k+4*(2*(n))(CTX), x ## d; \
+ xorl ab ## d, x ## d; \
+ addl k+4*(2*(n)+1)(CTX), y ## d; \
+ shrq $32, ab; \
+ roll $1, ab ## d; \
+ xorl y ## d, ab ## d; \
+ shlq $32, ab; \
+ rorl $1, x ## d; \
+ orq x, ab;
+
+#define dec_round_end(ba, x, y, n) \
+ addl y ## d, x ## d; \
+ addl x ## d, y ## d; \
+ addl k+4*(2*(n))(CTX), x ## d; \
+ addl k+4*(2*(n)+1)(CTX), y ## d; \
+ xorl ba ## d, y ## d; \
+ shrq $32, ba; \
+ roll $1, ba ## d; \
+ xorl x ## d, ba ## d; \
+ shlq $32, ba; \
+ rorl $1, y ## d; \
+ orq y, ba;
+
+#define encrypt_round3(ab, cd, n) \
+ g1g2_3(ab, cd, s0, s1, s2, s3, s0, s1, s2, s3, RX, RY); \
+ \
+ enc_round_end(ab ## 0, RX0, RY0, n); \
+ enc_round_end(ab ## 1, RX1, RY1, n); \
+ enc_round_end(ab ## 2, RX2, RY2, n);
+
+#define decrypt_round3(ba, dc, n) \
+ g1g2_3(ba, dc, s1, s2, s3, s0, s3, s0, s1, s2, RY, RX); \
+ \
+ dec_round_end(ba ## 0, RX0, RY0, n); \
+ dec_round_end(ba ## 1, RX1, RY1, n); \
+ dec_round_end(ba ## 2, RX2, RY2, n);
+
+#define encrypt_cycle3(ab, cd, n) \
+ encrypt_round3(ab, cd, n*2); \
+ encrypt_round3(ab, cd, (n*2)+1);
+
+#define decrypt_cycle3(ba, dc, n) \
+ decrypt_round3(ba, dc, (n*2)+1); \
+ decrypt_round3(ba, dc, (n*2));
+
+#define push_cd() \
+ pushq RCD2; \
+ pushq RCD1; \
+ pushq RCD0;
+
+#define pop_cd() \
+ popq RCD0; \
+ popq RCD1; \
+ popq RCD2;
+
+#define inpack3(in, n, xy, m) \
+ movq 4*(n)(in), xy ## 0; \
+ xorq w+4*m(CTX), xy ## 0; \
+ \
+ movq 4*(4+(n))(in), xy ## 1; \
+ xorq w+4*m(CTX), xy ## 1; \
+ \
+ movq 4*(8+(n))(in), xy ## 2; \
+ xorq w+4*m(CTX), xy ## 2;
+
+#define outunpack3(op, out, n, xy, m) \
+ xorq w+4*m(CTX), xy ## 0; \
+ op ## q xy ## 0, 4*(n)(out); \
+ \
+ xorq w+4*m(CTX), xy ## 1; \
+ op ## q xy ## 1, 4*(4+(n))(out); \
+ \
+ xorq w+4*m(CTX), xy ## 2; \
+ op ## q xy ## 2, 4*(8+(n))(out);
+
+#define inpack_enc3() \
+ inpack3(RIO, 0, RAB, 0); \
+ inpack3(RIO, 2, RCD, 2);
+
+#define outunpack_enc3(op) \
+ outunpack3(op, RIO, 2, RAB, 6); \
+ outunpack3(op, RIO, 0, RCD, 4);
+
+#define inpack_dec3() \
+ inpack3(RIO, 0, RAB, 4); \
+ rorq $32, RAB0; \
+ rorq $32, RAB1; \
+ rorq $32, RAB2; \
+ inpack3(RIO, 2, RCD, 6); \
+ rorq $32, RCD0; \
+ rorq $32, RCD1; \
+ rorq $32, RCD2;
+
+#define outunpack_dec3() \
+ rorq $32, RCD0; \
+ rorq $32, RCD1; \
+ rorq $32, RCD2; \
+ outunpack3(mov, RIO, 0, RCD, 0); \
+ rorq $32, RAB0; \
+ rorq $32, RAB1; \
+ rorq $32, RAB2; \
+ outunpack3(mov, RIO, 2, RAB, 2);
+
+SYM_FUNC_START(__twofish_enc_blk_3way)
+ /* input:
+ * %rdi: ctx, CTX
+ * %rsi: dst
+ * %rdx: src, RIO
+ * %rcx: bool, if true: xor output
+ */
+ pushq %r13;
+ pushq %r12;
+ pushq %rbx;
+
+ pushq %rcx; /* bool xor */
+ pushq %rsi; /* dst */
+
+ inpack_enc3();
+
+ push_cd();
+ encrypt_cycle3(RAB, CD, 0);
+ encrypt_cycle3(RAB, CD, 1);
+ encrypt_cycle3(RAB, CD, 2);
+ encrypt_cycle3(RAB, CD, 3);
+ encrypt_cycle3(RAB, CD, 4);
+ encrypt_cycle3(RAB, CD, 5);
+ encrypt_cycle3(RAB, CD, 6);
+ encrypt_cycle3(RAB, CD, 7);
+ pop_cd();
+
+ popq RIO; /* dst */
+ popq RT1; /* bool xor */
+
+ testb RT1bl, RT1bl;
+ jnz .L__enc_xor3;
+
+ outunpack_enc3(mov);
+
+ popq %rbx;
+ popq %r12;
+ popq %r13;
+ RET;
+
+.L__enc_xor3:
+ outunpack_enc3(xor);
+
+ popq %rbx;
+ popq %r12;
+ popq %r13;
+ RET;
+SYM_FUNC_END(__twofish_enc_blk_3way)
+
+SYM_FUNC_START(twofish_dec_blk_3way)
+ /* input:
+ * %rdi: ctx, CTX
+ * %rsi: dst
+ * %rdx: src, RIO
+ */
+ pushq %r13;
+ pushq %r12;
+ pushq %rbx;
+
+ pushq %rsi; /* dst */
+
+ inpack_dec3();
+
+ push_cd();
+ decrypt_cycle3(RAB, CD, 7);
+ decrypt_cycle3(RAB, CD, 6);
+ decrypt_cycle3(RAB, CD, 5);
+ decrypt_cycle3(RAB, CD, 4);
+ decrypt_cycle3(RAB, CD, 3);
+ decrypt_cycle3(RAB, CD, 2);
+ decrypt_cycle3(RAB, CD, 1);
+ decrypt_cycle3(RAB, CD, 0);
+ pop_cd();
+
+ popq RIO; /* dst */
+
+ outunpack_dec3();
+
+ popq %rbx;
+ popq %r12;
+ popq %r13;
+ RET;
+SYM_FUNC_END(twofish_dec_blk_3way)