--- /dev/null
+/**
+ * \file bn_mul.h
+ *
+ * \brief Multi-precision integer library
+ */
+/*
+ * Copyright (C) 2006-2015, ARM Limited, All Rights Reserved
+ * SPDX-License-Identifier: GPL-2.0
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with this program; if not, write to the Free Software Foundation, Inc.,
+ * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * This file is part of mbed TLS (https://tls.mbed.org)
+ */
+/*
+ * Multiply source vector [s] with b, add result
+ * to destination vector [d] and set carry c.
+ *
+ * Currently supports:
+ *
+ * . IA-32 (386+) . AMD64 / EM64T
+ * . IA-32 (SSE2) . Motorola 68000
+ * . PowerPC, 32-bit . MicroBlaze
+ * . PowerPC, 64-bit . TriCore
+ * . SPARC v8 . ARM v3+
+ * . Alpha . MIPS32
+ * . C, longlong . C, generic
+ */
+#ifndef MBEDTLS_BN_MUL_H
+#define MBEDTLS_BN_MUL_H
+
+#include "bignum.h"
+
+#if defined(MBEDTLS_HAVE_ASM)
+
+#ifndef asm
+#define asm __asm
+#endif
+
+/* armcc5 --gnu defines __GNUC__ but doesn't support GNU's extended asm */
+#if defined(__GNUC__) && \
+ ( !defined(__ARMCC_VERSION) || __ARMCC_VERSION >= 6000000 )
+
+/*
+ * Disable use of the i386 assembly code below if option -O0, to disable all
+ * compiler optimisations, is passed, detected with __OPTIMIZE__
+ * This is done as the number of registers used in the assembly code doesn't
+ * work with the -O0 option.
+ */
+#if defined(__i386__) && defined(__OPTIMIZE__)
+
+#define MULADDC_INIT \
+ asm( \
+ "movl %%ebx, %0 \n\t" \
+ "movl %5, %%esi \n\t" \
+ "movl %6, %%edi \n\t" \
+ "movl %7, %%ecx \n\t" \
+ "movl %8, %%ebx \n\t"
+
+#define MULADDC_CORE \
+ "lodsl \n\t" \
+ "mull %%ebx \n\t" \
+ "addl %%ecx, %%eax \n\t" \
+ "adcl $0, %%edx \n\t" \
+ "addl (%%edi), %%eax \n\t" \
+ "adcl $0, %%edx \n\t" \
+ "movl %%edx, %%ecx \n\t" \
+ "stosl \n\t"
+
+#if defined(MBEDTLS_HAVE_SSE2)
+
+#define MULADDC_HUIT \
+ "movd %%ecx, %%mm1 \n\t" \
+ "movd %%ebx, %%mm0 \n\t" \
+ "movd (%%edi), %%mm3 \n\t" \
+ "paddq %%mm3, %%mm1 \n\t" \
+ "movd (%%esi), %%mm2 \n\t" \
+ "pmuludq %%mm0, %%mm2 \n\t" \
+ "movd 4(%%esi), %%mm4 \n\t" \
+ "pmuludq %%mm0, %%mm4 \n\t" \
+ "movd 8(%%esi), %%mm6 \n\t" \
+ "pmuludq %%mm0, %%mm6 \n\t" \
+ "movd 12(%%esi), %%mm7 \n\t" \
+ "pmuludq %%mm0, %%mm7 \n\t" \
+ "paddq %%mm2, %%mm1 \n\t" \
+ "movd 4(%%edi), %%mm3 \n\t" \
+ "paddq %%mm4, %%mm3 \n\t" \
+ "movd 8(%%edi), %%mm5 \n\t" \
+ "paddq %%mm6, %%mm5 \n\t" \
+ "movd 12(%%edi), %%mm4 \n\t" \
+ "paddq %%mm4, %%mm7 \n\t" \
+ "movd %%mm1, (%%edi) \n\t" \
+ "movd 16(%%esi), %%mm2 \n\t" \
+ "pmuludq %%mm0, %%mm2 \n\t" \
+ "psrlq $32, %%mm1 \n\t" \
+ "movd 20(%%esi), %%mm4 \n\t" \
+ "pmuludq %%mm0, %%mm4 \n\t" \
+ "paddq %%mm3, %%mm1 \n\t" \
+ "movd 24(%%esi), %%mm6 \n\t" \
+ "pmuludq %%mm0, %%mm6 \n\t" \
+ "movd %%mm1, 4(%%edi) \n\t" \
+ "psrlq $32, %%mm1 \n\t" \
+ "movd 28(%%esi), %%mm3 \n\t" \
+ "pmuludq %%mm0, %%mm3 \n\t" \
+ "paddq %%mm5, %%mm1 \n\t" \
+ "movd 16(%%edi), %%mm5 \n\t" \
+ "paddq %%mm5, %%mm2 \n\t" \
+ "movd %%mm1, 8(%%edi) \n\t" \
+ "psrlq $32, %%mm1 \n\t" \
+ "paddq %%mm7, %%mm1 \n\t" \
+ "movd 20(%%edi), %%mm5 \n\t" \
+ "paddq %%mm5, %%mm4 \n\t" \
+ "movd %%mm1, 12(%%edi) \n\t" \
+ "psrlq $32, %%mm1 \n\t" \
+ "paddq %%mm2, %%mm1 \n\t" \
+ "movd 24(%%edi), %%mm5 \n\t" \
+ "paddq %%mm5, %%mm6 \n\t" \
+ "movd %%mm1, 16(%%edi) \n\t" \
+ "psrlq $32, %%mm1 \n\t" \
+ "paddq %%mm4, %%mm1 \n\t" \
+ "movd 28(%%edi), %%mm5 \n\t" \
+ "paddq %%mm5, %%mm3 \n\t" \
+ "movd %%mm1, 20(%%edi) \n\t" \
+ "psrlq $32, %%mm1 \n\t" \
+ "paddq %%mm6, %%mm1 \n\t" \
+ "movd %%mm1, 24(%%edi) \n\t" \
+ "psrlq $32, %%mm1 \n\t" \
+ "paddq %%mm3, %%mm1 \n\t" \
+ "movd %%mm1, 28(%%edi) \n\t" \
+ "addl $32, %%edi \n\t" \
+ "addl $32, %%esi \n\t" \
+ "psrlq $32, %%mm1 \n\t" \
+ "movd %%mm1, %%ecx \n\t"
+
+#define MULADDC_STOP \
+ "emms \n\t" \
+ "movl %4, %%ebx \n\t" \
+ "movl %%ecx, %1 \n\t" \
+ "movl %%edi, %2 \n\t" \
+ "movl %%esi, %3 \n\t" \
+ : "=m" (t), "=m" (c), "=m" (d), "=m" (s) \
+ : "m" (t), "m" (s), "m" (d), "m" (c), "m" (b) \
+ : "eax", "ebx", "ecx", "edx", "esi", "edi" \
+ );
+
+#else
+
+#define MULADDC_STOP \
+ "movl %4, %%ebx \n\t" \
+ "movl %%ecx, %1 \n\t" \
+ "movl %%edi, %2 \n\t" \
+ "movl %%esi, %3 \n\t" \
+ : "=m" (t), "=m" (c), "=m" (d), "=m" (s) \
+ : "m" (t), "m" (s), "m" (d), "m" (c), "m" (b) \
+ : "eax", "ebx", "ecx", "edx", "esi", "edi" \
+ );
+#endif /* SSE2 */
+#endif /* i386 */
+
+#if defined(__amd64__) || defined (__x86_64__)
+
+#define MULADDC_INIT \
+ asm( \
+ "xorq %%r8, %%r8 \n\t"
+
+#define MULADDC_CORE \
+ "movq (%%rsi), %%rax \n\t" \
+ "mulq %%rbx \n\t" \
+ "addq $8, %%rsi \n\t" \
+ "addq %%rcx, %%rax \n\t" \
+ "movq %%r8, %%rcx \n\t" \
+ "adcq $0, %%rdx \n\t" \
+ "nop \n\t" \
+ "addq %%rax, (%%rdi) \n\t" \
+ "adcq %%rdx, %%rcx \n\t" \
+ "addq $8, %%rdi \n\t"
+
+#define MULADDC_STOP \
+ : "+c" (c), "+D" (d), "+S" (s) \
+ : "b" (b) \
+ : "rax", "rdx", "r8" \
+ );
+
+#endif /* AMD64 */
+
+#if defined(__mc68020__) || defined(__mcpu32__)
+
+#define MULADDC_INIT \
+ asm( \
+ "movl %3, %%a2 \n\t" \
+ "movl %4, %%a3 \n\t" \
+ "movl %5, %%d3 \n\t" \
+ "movl %6, %%d2 \n\t" \
+ "moveq #0, %%d0 \n\t"
+
+#define MULADDC_CORE \
+ "movel %%a2@+, %%d1 \n\t" \
+ "mulul %%d2, %%d4:%%d1 \n\t" \
+ "addl %%d3, %%d1 \n\t" \
+ "addxl %%d0, %%d4 \n\t" \
+ "moveq #0, %%d3 \n\t" \
+ "addl %%d1, %%a3@+ \n\t" \
+ "addxl %%d4, %%d3 \n\t"
+
+#define MULADDC_STOP \
+ "movl %%d3, %0 \n\t" \
+ "movl %%a3, %1 \n\t" \
+ "movl %%a2, %2 \n\t" \
+ : "=m" (c), "=m" (d), "=m" (s) \
+ : "m" (s), "m" (d), "m" (c), "m" (b) \
+ : "d0", "d1", "d2", "d3", "d4", "a2", "a3" \
+ );
+
+#define MULADDC_HUIT \
+ "movel %%a2@+, %%d1 \n\t" \
+ "mulul %%d2, %%d4:%%d1 \n\t" \
+ "addxl %%d3, %%d1 \n\t" \
+ "addxl %%d0, %%d4 \n\t" \
+ "addl %%d1, %%a3@+ \n\t" \
+ "movel %%a2@+, %%d1 \n\t" \
+ "mulul %%d2, %%d3:%%d1 \n\t" \
+ "addxl %%d4, %%d1 \n\t" \
+ "addxl %%d0, %%d3 \n\t" \
+ "addl %%d1, %%a3@+ \n\t" \
+ "movel %%a2@+, %%d1 \n\t" \
+ "mulul %%d2, %%d4:%%d1 \n\t" \
+ "addxl %%d3, %%d1 \n\t" \
+ "addxl %%d0, %%d4 \n\t" \
+ "addl %%d1, %%a3@+ \n\t" \
+ "movel %%a2@+, %%d1 \n\t" \
+ "mulul %%d2, %%d3:%%d1 \n\t" \
+ "addxl %%d4, %%d1 \n\t" \
+ "addxl %%d0, %%d3 \n\t" \
+ "addl %%d1, %%a3@+ \n\t" \
+ "movel %%a2@+, %%d1 \n\t" \
+ "mulul %%d2, %%d4:%%d1 \n\t" \
+ "addxl %%d3, %%d1 \n\t" \
+ "addxl %%d0, %%d4 \n\t" \
+ "addl %%d1, %%a3@+ \n\t" \
+ "movel %%a2@+, %%d1 \n\t" \
+ "mulul %%d2, %%d3:%%d1 \n\t" \
+ "addxl %%d4, %%d1 \n\t" \
+ "addxl %%d0, %%d3 \n\t" \
+ "addl %%d1, %%a3@+ \n\t" \
+ "movel %%a2@+, %%d1 \n\t" \
+ "mulul %%d2, %%d4:%%d1 \n\t" \
+ "addxl %%d3, %%d1 \n\t" \
+ "addxl %%d0, %%d4 \n\t" \
+ "addl %%d1, %%a3@+ \n\t" \
+ "movel %%a2@+, %%d1 \n\t" \
+ "mulul %%d2, %%d3:%%d1 \n\t" \
+ "addxl %%d4, %%d1 \n\t" \
+ "addxl %%d0, %%d3 \n\t" \
+ "addl %%d1, %%a3@+ \n\t" \
+ "addxl %%d0, %%d3 \n\t"
+
+#endif /* MC68000 */
+
+#if defined(__powerpc64__) || defined(__ppc64__)
+
+#if defined(__MACH__) && defined(__APPLE__)
+
+#define MULADDC_INIT \
+ asm( \
+ "ld r3, %3 \n\t" \
+ "ld r4, %4 \n\t" \
+ "ld r5, %5 \n\t" \
+ "ld r6, %6 \n\t" \
+ "addi r3, r3, -8 \n\t" \
+ "addi r4, r4, -8 \n\t" \
+ "addic r5, r5, 0 \n\t"
+
+#define MULADDC_CORE \
+ "ldu r7, 8(r3) \n\t" \
+ "mulld r8, r7, r6 \n\t" \
+ "mulhdu r9, r7, r6 \n\t" \
+ "adde r8, r8, r5 \n\t" \
+ "ld r7, 8(r4) \n\t" \
+ "addze r5, r9 \n\t" \
+ "addc r8, r8, r7 \n\t" \
+ "stdu r8, 8(r4) \n\t"
+
+#define MULADDC_STOP \
+ "addze r5, r5 \n\t" \
+ "addi r4, r4, 8 \n\t" \
+ "addi r3, r3, 8 \n\t" \
+ "std r5, %0 \n\t" \
+ "std r4, %1 \n\t" \
+ "std r3, %2 \n\t" \
+ : "=m" (c), "=m" (d), "=m" (s) \
+ : "m" (s), "m" (d), "m" (c), "m" (b) \
+ : "r3", "r4", "r5", "r6", "r7", "r8", "r9" \
+ );
+
+
+#else /* __MACH__ && __APPLE__ */
+
+#define MULADDC_INIT \
+ asm( \
+ "ld %%r3, %3 \n\t" \
+ "ld %%r4, %4 \n\t" \
+ "ld %%r5, %5 \n\t" \
+ "ld %%r6, %6 \n\t" \
+ "addi %%r3, %%r3, -8 \n\t" \
+ "addi %%r4, %%r4, -8 \n\t" \
+ "addic %%r5, %%r5, 0 \n\t"
+
+#define MULADDC_CORE \
+ "ldu %%r7, 8(%%r3) \n\t" \
+ "mulld %%r8, %%r7, %%r6 \n\t" \
+ "mulhdu %%r9, %%r7, %%r6 \n\t" \
+ "adde %%r8, %%r8, %%r5 \n\t" \
+ "ld %%r7, 8(%%r4) \n\t" \
+ "addze %%r5, %%r9 \n\t" \
+ "addc %%r8, %%r8, %%r7 \n\t" \
+ "stdu %%r8, 8(%%r4) \n\t"
+
+#define MULADDC_STOP \
+ "addze %%r5, %%r5 \n\t" \
+ "addi %%r4, %%r4, 8 \n\t" \
+ "addi %%r3, %%r3, 8 \n\t" \
+ "std %%r5, %0 \n\t" \
+ "std %%r4, %1 \n\t" \
+ "std %%r3, %2 \n\t" \
+ : "=m" (c), "=m" (d), "=m" (s) \
+ : "m" (s), "m" (d), "m" (c), "m" (b) \
+ : "r3", "r4", "r5", "r6", "r7", "r8", "r9" \
+ );
+
+#endif /* __MACH__ && __APPLE__ */
+
+#elif defined(__powerpc__) || defined(__ppc__) /* end PPC64/begin PPC32 */
+
+#if defined(__MACH__) && defined(__APPLE__)
+
+#define MULADDC_INIT \
+ asm( \
+ "lwz r3, %3 \n\t" \
+ "lwz r4, %4 \n\t" \
+ "lwz r5, %5 \n\t" \
+ "lwz r6, %6 \n\t" \
+ "addi r3, r3, -4 \n\t" \
+ "addi r4, r4, -4 \n\t" \
+ "addic r5, r5, 0 \n\t"
+
+#define MULADDC_CORE \
+ "lwzu r7, 4(r3) \n\t" \
+ "mullw r8, r7, r6 \n\t" \
+ "mulhwu r9, r7, r6 \n\t" \
+ "adde r8, r8, r5 \n\t" \
+ "lwz r7, 4(r4) \n\t" \
+ "addze r5, r9 \n\t" \
+ "addc r8, r8, r7 \n\t" \
+ "stwu r8, 4(r4) \n\t"
+
+#define MULADDC_STOP \
+ "addze r5, r5 \n\t" \
+ "addi r4, r4, 4 \n\t" \
+ "addi r3, r3, 4 \n\t" \
+ "stw r5, %0 \n\t" \
+ "stw r4, %1 \n\t" \
+ "stw r3, %2 \n\t" \
+ : "=m" (c), "=m" (d), "=m" (s) \
+ : "m" (s), "m" (d), "m" (c), "m" (b) \
+ : "r3", "r4", "r5", "r6", "r7", "r8", "r9" \
+ );
+
+#else /* __MACH__ && __APPLE__ */
+
+#define MULADDC_INIT \
+ asm( \
+ "lwz %%r3, %3 \n\t" \
+ "lwz %%r4, %4 \n\t" \
+ "lwz %%r5, %5 \n\t" \
+ "lwz %%r6, %6 \n\t" \
+ "addi %%r3, %%r3, -4 \n\t" \
+ "addi %%r4, %%r4, -4 \n\t" \
+ "addic %%r5, %%r5, 0 \n\t"
+
+#define MULADDC_CORE \
+ "lwzu %%r7, 4(%%r3) \n\t" \
+ "mullw %%r8, %%r7, %%r6 \n\t" \
+ "mulhwu %%r9, %%r7, %%r6 \n\t" \
+ "adde %%r8, %%r8, %%r5 \n\t" \
+ "lwz %%r7, 4(%%r4) \n\t" \
+ "addze %%r5, %%r9 \n\t" \
+ "addc %%r8, %%r8, %%r7 \n\t" \
+ "stwu %%r8, 4(%%r4) \n\t"
+
+#define MULADDC_STOP \
+ "addze %%r5, %%r5 \n\t" \
+ "addi %%r4, %%r4, 4 \n\t" \
+ "addi %%r3, %%r3, 4 \n\t" \
+ "stw %%r5, %0 \n\t" \
+ "stw %%r4, %1 \n\t" \
+ "stw %%r3, %2 \n\t" \
+ : "=m" (c), "=m" (d), "=m" (s) \
+ : "m" (s), "m" (d), "m" (c), "m" (b) \
+ : "r3", "r4", "r5", "r6", "r7", "r8", "r9" \
+ );
+
+#endif /* __MACH__ && __APPLE__ */
+
+#endif /* PPC32 */
+
+/*
+ * The Sparc(64) assembly is reported to be broken.
+ * Disable it for now, until we're able to fix it.
+ */
+#if 0 && defined(__sparc__)
+#if defined(__sparc64__)
+
+#define MULADDC_INIT \
+ asm( \
+ "ldx %3, %%o0 \n\t" \
+ "ldx %4, %%o1 \n\t" \
+ "ld %5, %%o2 \n\t" \
+ "ld %6, %%o3 \n\t"
+
+#define MULADDC_CORE \
+ "ld [%%o0], %%o4 \n\t" \
+ "inc 4, %%o0 \n\t" \
+ "ld [%%o1], %%o5 \n\t" \
+ "umul %%o3, %%o4, %%o4 \n\t" \
+ "addcc %%o4, %%o2, %%o4 \n\t" \
+ "rd %%y, %%g1 \n\t" \
+ "addx %%g1, 0, %%g1 \n\t" \
+ "addcc %%o4, %%o5, %%o4 \n\t" \
+ "st %%o4, [%%o1] \n\t" \
+ "addx %%g1, 0, %%o2 \n\t" \
+ "inc 4, %%o1 \n\t"
+
+ #define MULADDC_STOP \
+ "st %%o2, %0 \n\t" \
+ "stx %%o1, %1 \n\t" \
+ "stx %%o0, %2 \n\t" \
+ : "=m" (c), "=m" (d), "=m" (s) \
+ : "m" (s), "m" (d), "m" (c), "m" (b) \
+ : "g1", "o0", "o1", "o2", "o3", "o4", \
+ "o5" \
+ );
+
+#else /* __sparc64__ */
+
+#define MULADDC_INIT \
+ asm( \
+ "ld %3, %%o0 \n\t" \
+ "ld %4, %%o1 \n\t" \
+ "ld %5, %%o2 \n\t" \
+ "ld %6, %%o3 \n\t"
+
+#define MULADDC_CORE \
+ "ld [%%o0], %%o4 \n\t" \
+ "inc 4, %%o0 \n\t" \
+ "ld [%%o1], %%o5 \n\t" \
+ "umul %%o3, %%o4, %%o4 \n\t" \
+ "addcc %%o4, %%o2, %%o4 \n\t" \
+ "rd %%y, %%g1 \n\t" \
+ "addx %%g1, 0, %%g1 \n\t" \
+ "addcc %%o4, %%o5, %%o4 \n\t" \
+ "st %%o4, [%%o1] \n\t" \
+ "addx %%g1, 0, %%o2 \n\t" \
+ "inc 4, %%o1 \n\t"
+
+#define MULADDC_STOP \
+ "st %%o2, %0 \n\t" \
+ "st %%o1, %1 \n\t" \
+ "st %%o0, %2 \n\t" \
+ : "=m" (c), "=m" (d), "=m" (s) \
+ : "m" (s), "m" (d), "m" (c), "m" (b) \
+ : "g1", "o0", "o1", "o2", "o3", "o4", \
+ "o5" \
+ );
+
+#endif /* __sparc64__ */
+#endif /* __sparc__ */
+
+#if defined(__microblaze__) || defined(microblaze)
+
+#define MULADDC_INIT \
+ asm( \
+ "lwi r3, %3 \n\t" \
+ "lwi r4, %4 \n\t" \
+ "lwi r5, %5 \n\t" \
+ "lwi r6, %6 \n\t" \
+ "andi r7, r6, 0xffff \n\t" \
+ "bsrli r6, r6, 16 \n\t"
+
+#define MULADDC_CORE \
+ "lhui r8, r3, 0 \n\t" \
+ "addi r3, r3, 2 \n\t" \
+ "lhui r9, r3, 0 \n\t" \
+ "addi r3, r3, 2 \n\t" \
+ "mul r10, r9, r6 \n\t" \
+ "mul r11, r8, r7 \n\t" \
+ "mul r12, r9, r7 \n\t" \
+ "mul r13, r8, r6 \n\t" \
+ "bsrli r8, r10, 16 \n\t" \
+ "bsrli r9, r11, 16 \n\t" \
+ "add r13, r13, r8 \n\t" \
+ "add r13, r13, r9 \n\t" \
+ "bslli r10, r10, 16 \n\t" \
+ "bslli r11, r11, 16 \n\t" \
+ "add r12, r12, r10 \n\t" \
+ "addc r13, r13, r0 \n\t" \
+ "add r12, r12, r11 \n\t" \
+ "addc r13, r13, r0 \n\t" \
+ "lwi r10, r4, 0 \n\t" \
+ "add r12, r12, r10 \n\t" \
+ "addc r13, r13, r0 \n\t" \
+ "add r12, r12, r5 \n\t" \
+ "addc r5, r13, r0 \n\t" \
+ "swi r12, r4, 0 \n\t" \
+ "addi r4, r4, 4 \n\t"
+
+#define MULADDC_STOP \
+ "swi r5, %0 \n\t" \
+ "swi r4, %1 \n\t" \
+ "swi r3, %2 \n\t" \
+ : "=m" (c), "=m" (d), "=m" (s) \
+ : "m" (s), "m" (d), "m" (c), "m" (b) \
+ : "r3", "r4", "r5", "r6", "r7", "r8", \
+ "r9", "r10", "r11", "r12", "r13" \
+ );
+
+#endif /* MicroBlaze */
+
+#if defined(__tricore__)
+
+#define MULADDC_INIT \
+ asm( \
+ "ld.a %%a2, %3 \n\t" \
+ "ld.a %%a3, %4 \n\t" \
+ "ld.w %%d4, %5 \n\t" \
+ "ld.w %%d1, %6 \n\t" \
+ "xor %%d5, %%d5 \n\t"
+
+#define MULADDC_CORE \
+ "ld.w %%d0, [%%a2+] \n\t" \
+ "madd.u %%e2, %%e4, %%d0, %%d1 \n\t" \
+ "ld.w %%d0, [%%a3] \n\t" \
+ "addx %%d2, %%d2, %%d0 \n\t" \
+ "addc %%d3, %%d3, 0 \n\t" \
+ "mov %%d4, %%d3 \n\t" \
+ "st.w [%%a3+], %%d2 \n\t"
+
+#define MULADDC_STOP \
+ "st.w %0, %%d4 \n\t" \
+ "st.a %1, %%a3 \n\t" \
+ "st.a %2, %%a2 \n\t" \
+ : "=m" (c), "=m" (d), "=m" (s) \
+ : "m" (s), "m" (d), "m" (c), "m" (b) \
+ : "d0", "d1", "e2", "d4", "a2", "a3" \
+ );
+
+#endif /* TriCore */
+
+/*
+ * gcc -O0 by default uses r7 for the frame pointer, so it complains about our
+ * use of r7 below, unless -fomit-frame-pointer is passed. Unfortunately,
+ * passing that option is not easy when building with yotta.
+ *
+ * On the other hand, -fomit-frame-pointer is implied by any -Ox options with
+ * x !=0, which we can detect using __OPTIMIZE__ (which is also defined by
+ * clang and armcc5 under the same conditions).
+ *
+ * So, only use the optimized assembly below for optimized build, which avoids
+ * the build error and is pretty reasonable anyway.
+ */
+#if defined(__GNUC__) && !defined(__OPTIMIZE__)
+#define MULADDC_CANNOT_USE_R7
+#endif
+
+#if defined(__arm__) && !defined(MULADDC_CANNOT_USE_R7)
+
+#if defined(__thumb__) && !defined(__thumb2__)
+
+#define MULADDC_INIT \
+ asm( \
+ "ldr r0, %3 \n\t" \
+ "ldr r1, %4 \n\t" \
+ "ldr r2, %5 \n\t" \
+ "ldr r3, %6 \n\t" \
+ "lsr r7, r3, #16 \n\t" \
+ "mov r9, r7 \n\t" \
+ "lsl r7, r3, #16 \n\t" \
+ "lsr r7, r7, #16 \n\t" \
+ "mov r8, r7 \n\t"
+
+#define MULADDC_CORE \
+ "ldmia r0!, {r6} \n\t" \
+ "lsr r7, r6, #16 \n\t" \
+ "lsl r6, r6, #16 \n\t" \
+ "lsr r6, r6, #16 \n\t" \
+ "mov r4, r8 \n\t" \
+ "mul r4, r6 \n\t" \
+ "mov r3, r9 \n\t" \
+ "mul r6, r3 \n\t" \
+ "mov r5, r9 \n\t" \
+ "mul r5, r7 \n\t" \
+ "mov r3, r8 \n\t" \
+ "mul r7, r3 \n\t" \
+ "lsr r3, r6, #16 \n\t" \
+ "add r5, r5, r3 \n\t" \
+ "lsr r3, r7, #16 \n\t" \
+ "add r5, r5, r3 \n\t" \
+ "add r4, r4, r2 \n\t" \
+ "mov r2, #0 \n\t" \
+ "adc r5, r2 \n\t" \
+ "lsl r3, r6, #16 \n\t" \
+ "add r4, r4, r3 \n\t" \
+ "adc r5, r2 \n\t" \
+ "lsl r3, r7, #16 \n\t" \
+ "add r4, r4, r3 \n\t" \
+ "adc r5, r2 \n\t" \
+ "ldr r3, [r1] \n\t" \
+ "add r4, r4, r3 \n\t" \
+ "adc r2, r5 \n\t" \
+ "stmia r1!, {r4} \n\t"
+
+#define MULADDC_STOP \
+ "str r2, %0 \n\t" \
+ "str r1, %1 \n\t" \
+ "str r0, %2 \n\t" \
+ : "=m" (c), "=m" (d), "=m" (s) \
+ : "m" (s), "m" (d), "m" (c), "m" (b) \
+ : "r0", "r1", "r2", "r3", "r4", "r5", \
+ "r6", "r7", "r8", "r9", "cc" \
+ );
+
+#else
+
+#define MULADDC_INIT \
+ asm( \
+ "ldr r0, %3 \n\t" \
+ "ldr r1, %4 \n\t" \
+ "ldr r2, %5 \n\t" \
+ "ldr r3, %6 \n\t"
+
+#define MULADDC_CORE \
+ "ldr r4, [r0], #4 \n\t" \
+ "mov r5, #0 \n\t" \
+ "ldr r6, [r1] \n\t" \
+ "umlal r2, r5, r3, r4 \n\t" \
+ "adds r7, r6, r2 \n\t" \
+ "adc r2, r5, #0 \n\t" \
+ "str r7, [r1], #4 \n\t"
+
+#define MULADDC_STOP \
+ "str r2, %0 \n\t" \
+ "str r1, %1 \n\t" \
+ "str r0, %2 \n\t" \
+ : "=m" (c), "=m" (d), "=m" (s) \
+ : "m" (s), "m" (d), "m" (c), "m" (b) \
+ : "r0", "r1", "r2", "r3", "r4", "r5", \
+ "r6", "r7", "cc" \
+ );
+
+#endif /* Thumb */
+
+#endif /* ARMv3 */
+
+#if defined(__alpha__)
+
+#define MULADDC_INIT \
+ asm( \
+ "ldq $1, %3 \n\t" \
+ "ldq $2, %4 \n\t" \
+ "ldq $3, %5 \n\t" \
+ "ldq $4, %6 \n\t"
+
+#define MULADDC_CORE \
+ "ldq $6, 0($1) \n\t" \
+ "addq $1, 8, $1 \n\t" \
+ "mulq $6, $4, $7 \n\t" \
+ "umulh $6, $4, $6 \n\t" \
+ "addq $7, $3, $7 \n\t" \
+ "cmpult $7, $3, $3 \n\t" \
+ "ldq $5, 0($2) \n\t" \
+ "addq $7, $5, $7 \n\t" \
+ "cmpult $7, $5, $5 \n\t" \
+ "stq $7, 0($2) \n\t" \
+ "addq $2, 8, $2 \n\t" \
+ "addq $6, $3, $3 \n\t" \
+ "addq $5, $3, $3 \n\t"
+
+#define MULADDC_STOP \
+ "stq $3, %0 \n\t" \
+ "stq $2, %1 \n\t" \
+ "stq $1, %2 \n\t" \
+ : "=m" (c), "=m" (d), "=m" (s) \
+ : "m" (s), "m" (d), "m" (c), "m" (b) \
+ : "$1", "$2", "$3", "$4", "$5", "$6", "$7" \
+ );
+#endif /* Alpha */
+
+#if defined(__mips__) && !defined(__mips64)
+
+#define MULADDC_INIT \
+ asm( \
+ "lw $10, %3 \n\t" \
+ "lw $11, %4 \n\t" \
+ "lw $12, %5 \n\t" \
+ "lw $13, %6 \n\t"
+
+#define MULADDC_CORE \
+ "lw $14, 0($10) \n\t" \
+ "multu $13, $14 \n\t" \
+ "addi $10, $10, 4 \n\t" \
+ "mflo $14 \n\t" \
+ "mfhi $9 \n\t" \
+ "addu $14, $12, $14 \n\t" \
+ "lw $15, 0($11) \n\t" \
+ "sltu $12, $14, $12 \n\t" \
+ "addu $15, $14, $15 \n\t" \
+ "sltu $14, $15, $14 \n\t" \
+ "addu $12, $12, $9 \n\t" \
+ "sw $15, 0($11) \n\t" \
+ "addu $12, $12, $14 \n\t" \
+ "addi $11, $11, 4 \n\t"
+
+#define MULADDC_STOP \
+ "sw $12, %0 \n\t" \
+ "sw $11, %1 \n\t" \
+ "sw $10, %2 \n\t" \
+ : "=m" (c), "=m" (d), "=m" (s) \
+ : "m" (s), "m" (d), "m" (c), "m" (b) \
+ : "$9", "$10", "$11", "$12", "$13", "$14", "$15" \
+ );
+
+#endif /* MIPS */
+#endif /* GNUC */
+
+#if (defined(_MSC_VER) && defined(_M_IX86)) || defined(__WATCOMC__)
+
+#define MULADDC_INIT \
+ __asm mov esi, s \
+ __asm mov edi, d \
+ __asm mov ecx, c \
+ __asm mov ebx, b
+
+#define MULADDC_CORE \
+ __asm lodsd \
+ __asm mul ebx \
+ __asm add eax, ecx \
+ __asm adc edx, 0 \
+ __asm add eax, [edi] \
+ __asm adc edx, 0 \
+ __asm mov ecx, edx \
+ __asm stosd
+
+#if defined(MBEDTLS_HAVE_SSE2)
+
+#define EMIT __asm _emit
+
+#define MULADDC_HUIT \
+ EMIT 0x0F EMIT 0x6E EMIT 0xC9 \
+ EMIT 0x0F EMIT 0x6E EMIT 0xC3 \
+ EMIT 0x0F EMIT 0x6E EMIT 0x1F \
+ EMIT 0x0F EMIT 0xD4 EMIT 0xCB \
+ EMIT 0x0F EMIT 0x6E EMIT 0x16 \
+ EMIT 0x0F EMIT 0xF4 EMIT 0xD0 \
+ EMIT 0x0F EMIT 0x6E EMIT 0x66 EMIT 0x04 \
+ EMIT 0x0F EMIT 0xF4 EMIT 0xE0 \
+ EMIT 0x0F EMIT 0x6E EMIT 0x76 EMIT 0x08 \
+ EMIT 0x0F EMIT 0xF4 EMIT 0xF0 \
+ EMIT 0x0F EMIT 0x6E EMIT 0x7E EMIT 0x0C \
+ EMIT 0x0F EMIT 0xF4 EMIT 0xF8 \
+ EMIT 0x0F EMIT 0xD4 EMIT 0xCA \
+ EMIT 0x0F EMIT 0x6E EMIT 0x5F EMIT 0x04 \
+ EMIT 0x0F EMIT 0xD4 EMIT 0xDC \
+ EMIT 0x0F EMIT 0x6E EMIT 0x6F EMIT 0x08 \
+ EMIT 0x0F EMIT 0xD4 EMIT 0xEE \
+ EMIT 0x0F EMIT 0x6E EMIT 0x67 EMIT 0x0C \
+ EMIT 0x0F EMIT 0xD4 EMIT 0xFC \
+ EMIT 0x0F EMIT 0x7E EMIT 0x0F \
+ EMIT 0x0F EMIT 0x6E EMIT 0x56 EMIT 0x10 \
+ EMIT 0x0F EMIT 0xF4 EMIT 0xD0 \
+ EMIT 0x0F EMIT 0x73 EMIT 0xD1 EMIT 0x20 \
+ EMIT 0x0F EMIT 0x6E EMIT 0x66 EMIT 0x14 \
+ EMIT 0x0F EMIT 0xF4 EMIT 0xE0 \
+ EMIT 0x0F EMIT 0xD4 EMIT 0xCB \
+ EMIT 0x0F EMIT 0x6E EMIT 0x76 EMIT 0x18 \
+ EMIT 0x0F EMIT 0xF4 EMIT 0xF0 \
+ EMIT 0x0F EMIT 0x7E EMIT 0x4F EMIT 0x04 \
+ EMIT 0x0F EMIT 0x73 EMIT 0xD1 EMIT 0x20 \
+ EMIT 0x0F EMIT 0x6E EMIT 0x5E EMIT 0x1C \
+ EMIT 0x0F EMIT 0xF4 EMIT 0xD8 \
+ EMIT 0x0F EMIT 0xD4 EMIT 0xCD \
+ EMIT 0x0F EMIT 0x6E EMIT 0x6F EMIT 0x10 \
+ EMIT 0x0F EMIT 0xD4 EMIT 0xD5 \
+ EMIT 0x0F EMIT 0x7E EMIT 0x4F EMIT 0x08 \
+ EMIT 0x0F EMIT 0x73 EMIT 0xD1 EMIT 0x20 \
+ EMIT 0x0F EMIT 0xD4 EMIT 0xCF \
+ EMIT 0x0F EMIT 0x6E EMIT 0x6F EMIT 0x14 \
+ EMIT 0x0F EMIT 0xD4 EMIT 0xE5 \
+ EMIT 0x0F EMIT 0x7E EMIT 0x4F EMIT 0x0C \
+ EMIT 0x0F EMIT 0x73 EMIT 0xD1 EMIT 0x20 \
+ EMIT 0x0F EMIT 0xD4 EMIT 0xCA \
+ EMIT 0x0F EMIT 0x6E EMIT 0x6F EMIT 0x18 \
+ EMIT 0x0F EMIT 0xD4 EMIT 0xF5 \
+ EMIT 0x0F EMIT 0x7E EMIT 0x4F EMIT 0x10 \
+ EMIT 0x0F EMIT 0x73 EMIT 0xD1 EMIT 0x20 \
+ EMIT 0x0F EMIT 0xD4 EMIT 0xCC \
+ EMIT 0x0F EMIT 0x6E EMIT 0x6F EMIT 0x1C \
+ EMIT 0x0F EMIT 0xD4 EMIT 0xDD \
+ EMIT 0x0F EMIT 0x7E EMIT 0x4F EMIT 0x14 \
+ EMIT 0x0F EMIT 0x73 EMIT 0xD1 EMIT 0x20 \
+ EMIT 0x0F EMIT 0xD4 EMIT 0xCE \
+ EMIT 0x0F EMIT 0x7E EMIT 0x4F EMIT 0x18 \
+ EMIT 0x0F EMIT 0x73 EMIT 0xD1 EMIT 0x20 \
+ EMIT 0x0F EMIT 0xD4 EMIT 0xCB \
+ EMIT 0x0F EMIT 0x7E EMIT 0x4F EMIT 0x1C \
+ EMIT 0x83 EMIT 0xC7 EMIT 0x20 \
+ EMIT 0x83 EMIT 0xC6 EMIT 0x20 \
+ EMIT 0x0F EMIT 0x73 EMIT 0xD1 EMIT 0x20 \
+ EMIT 0x0F EMIT 0x7E EMIT 0xC9
+
+#define MULADDC_STOP \
+ EMIT 0x0F EMIT 0x77 \
+ __asm mov c, ecx \
+ __asm mov d, edi \
+ __asm mov s, esi \
+
+#else
+
+#define MULADDC_STOP \
+ __asm mov c, ecx \
+ __asm mov d, edi \
+ __asm mov s, esi \
+
+#endif /* SSE2 */
+#endif /* MSVC */
+
+#endif /* MBEDTLS_HAVE_ASM */
+
+#if !defined(MULADDC_CORE)
+#if defined(MBEDTLS_HAVE_UDBL)
+
+#define MULADDC_INIT \
+{ \
+ mbedtls_t_udbl r; \
+ mbedtls_mpi_uint r0, r1;
+
+#define MULADDC_CORE \
+ r = *(s++) * (mbedtls_t_udbl) b; \
+ r0 = (mbedtls_mpi_uint) r; \
+ r1 = (mbedtls_mpi_uint)( r >> biL ); \
+ r0 += c; r1 += (r0 < c); \
+ r0 += *d; r1 += (r0 < *d); \
+ c = r1; *(d++) = r0;
+
+#define MULADDC_STOP \
+}
+
+#else
+#define MULADDC_INIT \
+{ \
+ mbedtls_mpi_uint s0, s1, b0, b1; \
+ mbedtls_mpi_uint r0, r1, rx, ry; \
+ b0 = ( b << biH ) >> biH; \
+ b1 = ( b >> biH );
+
+#define MULADDC_CORE \
+ s0 = ( *s << biH ) >> biH; \
+ s1 = ( *s >> biH ); s++; \
+ rx = s0 * b1; r0 = s0 * b0; \
+ ry = s1 * b0; r1 = s1 * b1; \
+ r1 += ( rx >> biH ); \
+ r1 += ( ry >> biH ); \
+ rx <<= biH; ry <<= biH; \
+ r0 += rx; r1 += (r0 < rx); \
+ r0 += ry; r1 += (r0 < ry); \
+ r0 += c; r1 += (r0 < c); \
+ r0 += *d; r1 += (r0 < *d); \
+ c = r1; *(d++) = r0;
+
+#define MULADDC_STOP \
+}
+
+#endif /* C (generic) */
+#endif /* C (longlong) */
+
+#endif /* bn_mul.h */
projects / proxmark3-svn / blobdiff