summaryrefslogtreecommitdiff
path: root/package/px5g/src/polarssl/bn_mul.h
diff options
context:
space:
mode:
authorjow <jow@3c298f89-4303-0410-b956-a3cf2f4a3e73>2010-03-25 15:46:39 +0000
committerjow <jow@3c298f89-4303-0410-b956-a3cf2f4a3e73>2010-03-25 15:46:39 +0000
commitafc9f776d270b46fffb6c09f05e338f22894c3d1 (patch)
tree040e278b00ff66581e0db568d2ed3f8da2aaf777 /package/px5g/src/polarssl/bn_mul.h
parent56d71638dbf48b8e4885e1e5ed5d0f1859674541 (diff)
[package] add px5g (moved from LuCI trunk)
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@20429 3c298f89-4303-0410-b956-a3cf2f4a3e73
Diffstat (limited to 'package/px5g/src/polarssl/bn_mul.h')
-rw-r--r--package/px5g/src/polarssl/bn_mul.h731
1 files changed, 731 insertions, 0 deletions
diff --git a/package/px5g/src/polarssl/bn_mul.h b/package/px5g/src/polarssl/bn_mul.h
new file mode 100644
index 0000000000..f6d34da58a
--- /dev/null
+++ b/package/px5g/src/polarssl/bn_mul.h
@@ -0,0 +1,731 @@
+/**
+ * \file bn_mul.h
+ *
+ * Based on XySSL: Copyright (C) 2006-2008 Christophe Devine
+ *
+ * Copyright (C) 2009 Paul Bakker <polarssl_maintainer at polarssl dot org>
+ *
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the names of PolarSSL or XySSL nor the names of its contributors
+ * may be used to endorse or promote products derived from this software
+ * without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
+ * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED
+ * TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
+ * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
+ * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
+ * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+ * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+/*
+ * 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 POLARSSL_BN_MUL_H
+#define POLARSSL_BN_MUL_H
+
+#include "polarssl/config.h"
+
+#if defined(POLARSSL_HAVE_ASM)
+
+#if defined(__GNUC__)
+#if defined(__i386__)
+
+#define MULADDC_INIT \
+ asm( "movl %%ebx, %0 " : "=m" (t)); \
+ asm( "movl %0, %%esi " :: "m" (s)); \
+ asm( "movl %0, %%edi " :: "m" (d)); \
+ asm( "movl %0, %%ecx " :: "m" (c)); \
+ asm( "movl %0, %%ebx " :: "m" (b));
+
+#define MULADDC_CORE \
+ asm( "lodsl " ); \
+ asm( "mull %ebx " ); \
+ asm( "addl %ecx, %eax " ); \
+ asm( "adcl $0, %edx " ); \
+ asm( "addl (%edi), %eax " ); \
+ asm( "adcl $0, %edx " ); \
+ asm( "movl %edx, %ecx " ); \
+ asm( "stosl " );
+
+#if defined(POLARSSL_HAVE_SSE2)
+
+#define MULADDC_HUIT \
+ asm( "movd %ecx, %mm1 " ); \
+ asm( "movd %ebx, %mm0 " ); \
+ asm( "movd (%edi), %mm3 " ); \
+ asm( "paddq %mm3, %mm1 " ); \
+ asm( "movd (%esi), %mm2 " ); \
+ asm( "pmuludq %mm0, %mm2 " ); \
+ asm( "movd 4(%esi), %mm4 " ); \
+ asm( "pmuludq %mm0, %mm4 " ); \
+ asm( "movd 8(%esi), %mm6 " ); \
+ asm( "pmuludq %mm0, %mm6 " ); \
+ asm( "movd 12(%esi), %mm7 " ); \
+ asm( "pmuludq %mm0, %mm7 " ); \
+ asm( "paddq %mm2, %mm1 " ); \
+ asm( "movd 4(%edi), %mm3 " ); \
+ asm( "paddq %mm4, %mm3 " ); \
+ asm( "movd 8(%edi), %mm5 " ); \
+ asm( "paddq %mm6, %mm5 " ); \
+ asm( "movd 12(%edi), %mm4 " ); \
+ asm( "paddq %mm4, %mm7 " ); \
+ asm( "movd %mm1, (%edi) " ); \
+ asm( "movd 16(%esi), %mm2 " ); \
+ asm( "pmuludq %mm0, %mm2 " ); \
+ asm( "psrlq $32, %mm1 " ); \
+ asm( "movd 20(%esi), %mm4 " ); \
+ asm( "pmuludq %mm0, %mm4 " ); \
+ asm( "paddq %mm3, %mm1 " ); \
+ asm( "movd 24(%esi), %mm6 " ); \
+ asm( "pmuludq %mm0, %mm6 " ); \
+ asm( "movd %mm1, 4(%edi) " ); \
+ asm( "psrlq $32, %mm1 " ); \
+ asm( "movd 28(%esi), %mm3 " ); \
+ asm( "pmuludq %mm0, %mm3 " ); \
+ asm( "paddq %mm5, %mm1 " ); \
+ asm( "movd 16(%edi), %mm5 " ); \
+ asm( "paddq %mm5, %mm2 " ); \
+ asm( "movd %mm1, 8(%edi) " ); \
+ asm( "psrlq $32, %mm1 " ); \
+ asm( "paddq %mm7, %mm1 " ); \
+ asm( "movd 20(%edi), %mm5 " ); \
+ asm( "paddq %mm5, %mm4 " ); \
+ asm( "movd %mm1, 12(%edi) " ); \
+ asm( "psrlq $32, %mm1 " ); \
+ asm( "paddq %mm2, %mm1 " ); \
+ asm( "movd 24(%edi), %mm5 " ); \
+ asm( "paddq %mm5, %mm6 " ); \
+ asm( "movd %mm1, 16(%edi) " ); \
+ asm( "psrlq $32, %mm1 " ); \
+ asm( "paddq %mm4, %mm1 " ); \
+ asm( "movd 28(%edi), %mm5 " ); \
+ asm( "paddq %mm5, %mm3 " ); \
+ asm( "movd %mm1, 20(%edi) " ); \
+ asm( "psrlq $32, %mm1 " ); \
+ asm( "paddq %mm6, %mm1 " ); \
+ asm( "movd %mm1, 24(%edi) " ); \
+ asm( "psrlq $32, %mm1 " ); \
+ asm( "paddq %mm3, %mm1 " ); \
+ asm( "movd %mm1, 28(%edi) " ); \
+ asm( "addl $32, %edi " ); \
+ asm( "addl $32, %esi " ); \
+ asm( "psrlq $32, %mm1 " ); \
+ asm( "movd %mm1, %ecx " );
+
+#define MULADDC_STOP \
+ asm( "emms " ); \
+ asm( "movl %0, %%ebx " :: "m" (t)); \
+ asm( "movl %%ecx, %0 " : "=m" (c)); \
+ asm( "movl %%edi, %0 " : "=m" (d)); \
+ asm( "movl %%esi, %0 " : "=m" (s) :: \
+ "eax", "ecx", "edx", "esi", "edi" );
+
+#else
+
+#define MULADDC_STOP \
+ asm( "movl %0, %%ebx " :: "m" (t)); \
+ asm( "movl %%ecx, %0 " : "=m" (c)); \
+ asm( "movl %%edi, %0 " : "=m" (d)); \
+ asm( "movl %%esi, %0 " : "=m" (s) :: \
+ "eax", "ecx", "edx", "esi", "edi" );
+
+#endif /* SSE2 */
+#endif /* i386 */
+
+#if defined(__amd64__) || defined (__x86_64__)
+
+#define MULADDC_INIT \
+ asm( "movq %0, %%rsi " :: "m" (s)); \
+ asm( "movq %0, %%rdi " :: "m" (d)); \
+ asm( "movq %0, %%rcx " :: "m" (c)); \
+ asm( "movq %0, %%rbx " :: "m" (b)); \
+ asm( "xorq %r8, %r8 " );
+
+#define MULADDC_CORE \
+ asm( "movq (%rsi),%rax " ); \
+ asm( "mulq %rbx " ); \
+ asm( "addq $8, %rsi " ); \
+ asm( "addq %rcx, %rax " ); \
+ asm( "movq %r8, %rcx " ); \
+ asm( "adcq $0, %rdx " ); \
+ asm( "nop " ); \
+ asm( "addq %rax, (%rdi) " ); \
+ asm( "adcq %rdx, %rcx " ); \
+ asm( "addq $8, %rdi " );
+
+#define MULADDC_STOP \
+ asm( "movq %%rcx, %0 " : "=m" (c)); \
+ asm( "movq %%rdi, %0 " : "=m" (d)); \
+ asm( "movq %%rsi, %0 " : "=m" (s) :: \
+ "rax", "rcx", "rdx", "rbx", "rsi", "rdi", "r8" );
+
+#endif /* AMD64 */
+
+#if defined(__mc68020__) || defined(__mcpu32__)
+
+#define MULADDC_INIT \
+ asm( "movl %0, %%a2 " :: "m" (s)); \
+ asm( "movl %0, %%a3 " :: "m" (d)); \
+ asm( "movl %0, %%d3 " :: "m" (c)); \
+ asm( "movl %0, %%d2 " :: "m" (b)); \
+ asm( "moveq #0, %d0 " );
+
+#define MULADDC_CORE \
+ asm( "movel %a2@+, %d1 " ); \
+ asm( "mulul %d2, %d4:%d1 " ); \
+ asm( "addl %d3, %d1 " ); \
+ asm( "addxl %d0, %d4 " ); \
+ asm( "moveq #0, %d3 " ); \
+ asm( "addl %d1, %a3@+ " ); \
+ asm( "addxl %d4, %d3 " );
+
+#define MULADDC_STOP \
+ asm( "movl %%d3, %0 " : "=m" (c)); \
+ asm( "movl %%a3, %0 " : "=m" (d)); \
+ asm( "movl %%a2, %0 " : "=m" (s) :: \
+ "d0", "d1", "d2", "d3", "d4", "a2", "a3" );
+
+#define MULADDC_HUIT \
+ asm( "movel %a2@+, %d1 " ); \
+ asm( "mulul %d2, %d4:%d1 " ); \
+ asm( "addxl %d3, %d1 " ); \
+ asm( "addxl %d0, %d4 " ); \
+ asm( "addl %d1, %a3@+ " ); \
+ asm( "movel %a2@+, %d1 " ); \
+ asm( "mulul %d2, %d3:%d1 " ); \
+ asm( "addxl %d4, %d1 " ); \
+ asm( "addxl %d0, %d3 " ); \
+ asm( "addl %d1, %a3@+ " ); \
+ asm( "movel %a2@+, %d1 " ); \
+ asm( "mulul %d2, %d4:%d1 " ); \
+ asm( "addxl %d3, %d1 " ); \
+ asm( "addxl %d0, %d4 " ); \
+ asm( "addl %d1, %a3@+ " ); \
+ asm( "movel %a2@+, %d1 " ); \
+ asm( "mulul %d2, %d3:%d1 " ); \
+ asm( "addxl %d4, %d1 " ); \
+ asm( "addxl %d0, %d3 " ); \
+ asm( "addl %d1, %a3@+ " ); \
+ asm( "movel %a2@+, %d1 " ); \
+ asm( "mulul %d2, %d4:%d1 " ); \
+ asm( "addxl %d3, %d1 " ); \
+ asm( "addxl %d0, %d4 " ); \
+ asm( "addl %d1, %a3@+ " ); \
+ asm( "movel %a2@+, %d1 " ); \
+ asm( "mulul %d2, %d3:%d1 " ); \
+ asm( "addxl %d4, %d1 " ); \
+ asm( "addxl %d0, %d3 " ); \
+ asm( "addl %d1, %a3@+ " ); \
+ asm( "movel %a2@+, %d1 " ); \
+ asm( "mulul %d2, %d4:%d1 " ); \
+ asm( "addxl %d3, %d1 " ); \
+ asm( "addxl %d0, %d4 " ); \
+ asm( "addl %d1, %a3@+ " ); \
+ asm( "movel %a2@+, %d1 " ); \
+ asm( "mulul %d2, %d3:%d1 " ); \
+ asm( "addxl %d4, %d1 " ); \
+ asm( "addxl %d0, %d3 " ); \
+ asm( "addl %d1, %a3@+ " ); \
+ asm( "addxl %d0, %d3 " );
+
+#endif /* MC68000 */
+
+#if defined(__powerpc__) || defined(__ppc__)
+#if defined(__powerpc64__) || defined(__ppc64__)
+
+#if defined(__MACH__) && defined(__APPLE__)
+
+#define MULADDC_INIT \
+ asm( "ld r3, %0 " :: "m" (s)); \
+ asm( "ld r4, %0 " :: "m" (d)); \
+ asm( "ld r5, %0 " :: "m" (c)); \
+ asm( "ld r6, %0 " :: "m" (b)); \
+ asm( "addi r3, r3, -8 " ); \
+ asm( "addi r4, r4, -8 " ); \
+ asm( "addic r5, r5, 0 " );
+
+#define MULADDC_CORE \
+ asm( "ldu r7, 8(r3) " ); \
+ asm( "mulld r8, r7, r6 " ); \
+ asm( "mulhdu r9, r7, r6 " ); \
+ asm( "adde r8, r8, r5 " ); \
+ asm( "ld r7, 8(r4) " ); \
+ asm( "addze r5, r9 " ); \
+ asm( "addc r8, r8, r7 " ); \
+ asm( "stdu r8, 8(r4) " );
+
+#define MULADDC_STOP \
+ asm( "addze r5, r5 " ); \
+ asm( "addi r4, r4, 8 " ); \
+ asm( "addi r3, r3, 8 " ); \
+ asm( "std r5, %0 " : "=m" (c)); \
+ asm( "std r4, %0 " : "=m" (d)); \
+ asm( "std r3, %0 " : "=m" (s) :: \
+ "r3", "r4", "r5", "r6", "r7", "r8", "r9" );
+
+#else
+
+#define MULADDC_INIT \
+ asm( "ld %%r3, %0 " :: "m" (s)); \
+ asm( "ld %%r4, %0 " :: "m" (d)); \
+ asm( "ld %%r5, %0 " :: "m" (c)); \
+ asm( "ld %%r6, %0 " :: "m" (b)); \
+ asm( "addi %r3, %r3, -8 " ); \
+ asm( "addi %r4, %r4, -8 " ); \
+ asm( "addic %r5, %r5, 0 " );
+
+#define MULADDC_CORE \
+ asm( "ldu %r7, 8(%r3) " ); \
+ asm( "mulld %r8, %r7, %r6 " ); \
+ asm( "mulhdu %r9, %r7, %r6 " ); \
+ asm( "adde %r8, %r8, %r5 " ); \
+ asm( "ld %r7, 8(%r4) " ); \
+ asm( "addze %r5, %r9 " ); \
+ asm( "addc %r8, %r8, %r7 " ); \
+ asm( "stdu %r8, 8(%r4) " );
+
+#define MULADDC_STOP \
+ asm( "addze %r5, %r5 " ); \
+ asm( "addi %r4, %r4, 8 " ); \
+ asm( "addi %r3, %r3, 8 " ); \
+ asm( "std %%r5, %0 " : "=m" (c)); \
+ asm( "std %%r4, %0 " : "=m" (d)); \
+ asm( "std %%r3, %0 " : "=m" (s) :: \
+ "r3", "r4", "r5", "r6", "r7", "r8", "r9" );
+
+#endif
+
+#else /* PPC32 */
+
+#if defined(__MACH__) && defined(__APPLE__)
+
+#define MULADDC_INIT \
+ asm( "lwz r3, %0 " :: "m" (s)); \
+ asm( "lwz r4, %0 " :: "m" (d)); \
+ asm( "lwz r5, %0 " :: "m" (c)); \
+ asm( "lwz r6, %0 " :: "m" (b)); \
+ asm( "addi r3, r3, -4 " ); \
+ asm( "addi r4, r4, -4 " ); \
+ asm( "addic r5, r5, 0 " );
+
+#define MULADDC_CORE \
+ asm( "lwzu r7, 4(r3) " ); \
+ asm( "mullw r8, r7, r6 " ); \
+ asm( "mulhwu r9, r7, r6 " ); \
+ asm( "adde r8, r8, r5 " ); \
+ asm( "lwz r7, 4(r4) " ); \
+ asm( "addze r5, r9 " ); \
+ asm( "addc r8, r8, r7 " ); \
+ asm( "stwu r8, 4(r4) " );
+
+#define MULADDC_STOP \
+ asm( "addze r5, r5 " ); \
+ asm( "addi r4, r4, 4 " ); \
+ asm( "addi r3, r3, 4 " ); \
+ asm( "stw r5, %0 " : "=m" (c)); \
+ asm( "stw r4, %0 " : "=m" (d)); \
+ asm( "stw r3, %0 " : "=m" (s) :: \
+ "r3", "r4", "r5", "r6", "r7", "r8", "r9" );
+
+#else
+
+#define MULADDC_INIT \
+ asm( "lwz %%r3, %0 " :: "m" (s)); \
+ asm( "lwz %%r4, %0 " :: "m" (d)); \
+ asm( "lwz %%r5, %0 " :: "m" (c)); \
+ asm( "lwz %%r6, %0 " :: "m" (b)); \
+ asm( "addi %r3, %r3, -4 " ); \
+ asm( "addi %r4, %r4, -4 " ); \
+ asm( "addic %r5, %r5, 0 " );
+
+#define MULADDC_CORE \
+ asm( "lwzu %r7, 4(%r3) " ); \
+ asm( "mullw %r8, %r7, %r6 " ); \
+ asm( "mulhwu %r9, %r7, %r6 " ); \
+ asm( "adde %r8, %r8, %r5 " ); \
+ asm( "lwz %r7, 4(%r4) " ); \
+ asm( "addze %r5, %r9 " ); \
+ asm( "addc %r8, %r8, %r7 " ); \
+ asm( "stwu %r8, 4(%r4) " );
+
+#define MULADDC_STOP \
+ asm( "addze %r5, %r5 " ); \
+ asm( "addi %r4, %r4, 4 " ); \
+ asm( "addi %r3, %r3, 4 " ); \
+ asm( "stw %%r5, %0 " : "=m" (c)); \
+ asm( "stw %%r4, %0 " : "=m" (d)); \
+ asm( "stw %%r3, %0 " : "=m" (s) :: \
+ "r3", "r4", "r5", "r6", "r7", "r8", "r9" );
+
+#endif
+
+#endif /* PPC32 */
+#endif /* PPC64 */
+
+#if defined(__sparc__)
+
+#define MULADDC_INIT \
+ asm( "ld %0, %%o0 " :: "m" (s)); \
+ asm( "ld %0, %%o1 " :: "m" (d)); \
+ asm( "ld %0, %%o2 " :: "m" (c)); \
+ asm( "ld %0, %%o3 " :: "m" (b));
+
+#define MULADDC_CORE \
+ asm( "ld [%o0], %o4 " ); \
+ asm( "inc 4, %o0 " ); \
+ asm( "ld [%o1], %o5 " ); \
+ asm( "umul %o3, %o4, %o4 " ); \
+ asm( "addcc %o4, %o2, %o4 " ); \
+ asm( "rd %y, %g1 " ); \
+ asm( "addx %g1, 0, %g1 " ); \
+ asm( "addcc %o4, %o5, %o4 " ); \
+ asm( "st %o4, [%o1] " ); \
+ asm( "addx %g1, 0, %o2 " ); \
+ asm( "inc 4, %o1 " );
+
+#define MULADDC_STOP \
+ asm( "st %%o2, %0 " : "=m" (c)); \
+ asm( "st %%o1, %0 " : "=m" (d)); \
+ asm( "st %%o0, %0 " : "=m" (s) :: \
+ "g1", "o0", "o1", "o2", "o3", "o4", "o5" );
+
+#endif /* SPARCv8 */
+
+#if defined(__microblaze__) || defined(microblaze)
+
+#define MULADDC_INIT \
+ asm( "lwi r3, %0 " :: "m" (s)); \
+ asm( "lwi r4, %0 " :: "m" (d)); \
+ asm( "lwi r5, %0 " :: "m" (c)); \
+ asm( "lwi r6, %0 " :: "m" (b)); \
+ asm( "andi r7, r6, 0xffff" ); \
+ asm( "bsrli r6, r6, 16 " );
+
+#define MULADDC_CORE \
+ asm( "lhui r8, r3, 0 " ); \
+ asm( "addi r3, r3, 2 " ); \
+ asm( "lhui r9, r3, 0 " ); \
+ asm( "addi r3, r3, 2 " ); \
+ asm( "mul r10, r9, r6 " ); \
+ asm( "mul r11, r8, r7 " ); \
+ asm( "mul r12, r9, r7 " ); \
+ asm( "mul r13, r8, r6 " ); \
+ asm( "bsrli r8, r10, 16 " ); \
+ asm( "bsrli r9, r11, 16 " ); \
+ asm( "add r13, r13, r8 " ); \
+ asm( "add r13, r13, r9 " ); \
+ asm( "bslli r10, r10, 16 " ); \
+ asm( "bslli r11, r11, 16 " ); \
+ asm( "add r12, r12, r10 " ); \
+ asm( "addc r13, r13, r0 " ); \
+ asm( "add r12, r12, r11 " ); \
+ asm( "addc r13, r13, r0 " ); \
+ asm( "lwi r10, r4, 0 " ); \
+ asm( "add r12, r12, r10 " ); \
+ asm( "addc r13, r13, r0 " ); \
+ asm( "add r12, r12, r5 " ); \
+ asm( "addc r5, r13, r0 " ); \
+ asm( "swi r12, r4, 0 " ); \
+ asm( "addi r4, r4, 4 " );
+
+#define MULADDC_STOP \
+ asm( "swi r5, %0 " : "=m" (c)); \
+ asm( "swi r4, %0 " : "=m" (d)); \
+ asm( "swi r3, %0 " : "=m" (s) :: \
+ "r3", "r4" , "r5" , "r6" , "r7" , "r8" , \
+ "r9", "r10", "r11", "r12", "r13" );
+
+#endif /* MicroBlaze */
+
+#if defined(__tricore__)
+
+#define MULADDC_INIT \
+ asm( "ld.a %%a2, %0 " :: "m" (s)); \
+ asm( "ld.a %%a3, %0 " :: "m" (d)); \
+ asm( "ld.w %%d4, %0 " :: "m" (c)); \
+ asm( "ld.w %%d1, %0 " :: "m" (b)); \
+ asm( "xor %d5, %d5 " );
+
+#define MULADDC_CORE \
+ asm( "ld.w %d0, [%a2+] " ); \
+ asm( "madd.u %e2, %e4, %d0, %d1 " ); \
+ asm( "ld.w %d0, [%a3] " ); \
+ asm( "addx %d2, %d2, %d0 " ); \
+ asm( "addc %d3, %d3, 0 " ); \
+ asm( "mov %d4, %d3 " ); \
+ asm( "st.w [%a3+], %d2 " );
+
+#define MULADDC_STOP \
+ asm( "st.w %0, %%d4 " : "=m" (c)); \
+ asm( "st.a %0, %%a3 " : "=m" (d)); \
+ asm( "st.a %0, %%a2 " : "=m" (s) :: \
+ "d0", "d1", "e2", "d4", "a2", "a3" );
+
+#endif /* TriCore */
+
+#if defined(__arm__)
+
+#define MULADDC_INIT \
+ asm( "ldr r0, %0 " :: "m" (s)); \
+ asm( "ldr r1, %0 " :: "m" (d)); \
+ asm( "ldr r2, %0 " :: "m" (c)); \
+ asm( "ldr r3, %0 " :: "m" (b));
+
+#define MULADDC_CORE \
+ asm( "ldr r4, [r0], #4 " ); \
+ asm( "mov r5, #0 " ); \
+ asm( "ldr r6, [r1] " ); \
+ asm( "umlal r2, r5, r3, r4 " ); \
+ asm( "adds r7, r6, r2 " ); \
+ asm( "adc r2, r5, #0 " ); \
+ asm( "str r7, [r1], #4 " );
+
+#define MULADDC_STOP \
+ asm( "str r2, %0 " : "=m" (c)); \
+ asm( "str r1, %0 " : "=m" (d)); \
+ asm( "str r0, %0 " : "=m" (s) :: \
+ "r0", "r1", "r2", "r3", "r4", "r5", "r6", "r7" );
+
+#endif /* ARMv3 */
+
+#if defined(__alpha__)
+
+#define MULADDC_INIT \
+ asm( "ldq $1, %0 " :: "m" (s)); \
+ asm( "ldq $2, %0 " :: "m" (d)); \
+ asm( "ldq $3, %0 " :: "m" (c)); \
+ asm( "ldq $4, %0 " :: "m" (b));
+
+#define MULADDC_CORE \
+ asm( "ldq $6, 0($1) " ); \
+ asm( "addq $1, 8, $1 " ); \
+ asm( "mulq $6, $4, $7 " ); \
+ asm( "umulh $6, $4, $6 " ); \
+ asm( "addq $7, $3, $7 " ); \
+ asm( "cmpult $7, $3, $3 " ); \
+ asm( "ldq $5, 0($2) " ); \
+ asm( "addq $7, $5, $7 " ); \
+ asm( "cmpult $7, $5, $5 " ); \
+ asm( "stq $7, 0($2) " ); \
+ asm( "addq $2, 8, $2 " ); \
+ asm( "addq $6, $3, $3 " ); \
+ asm( "addq $5, $3, $3 " );
+
+#define MULADDC_STOP \
+ asm( "stq $3, %0 " : "=m" (c)); \
+ asm( "stq $2, %0 " : "=m" (d)); \
+ asm( "stq $1, %0 " : "=m" (s) :: \
+ "$1", "$2", "$3", "$4", "$5", "$6", "$7" );
+
+#endif /* Alpha */
+
+#if defined(__mips__)
+
+#define MULADDC_INIT \
+ asm( "lw $10, %0 " :: "m" (s)); \
+ asm( "lw $11, %0 " :: "m" (d)); \
+ asm( "lw $12, %0 " :: "m" (c)); \
+ asm( "lw $13, %0 " :: "m" (b));
+
+#define MULADDC_CORE \
+ asm( "lw $14, 0($10) " ); \
+ asm( "multu $13, $14 " ); \
+ asm( "addi $10, $10, 4 " ); \
+ asm( "mflo $14 " ); \
+ asm( "mfhi $9 " ); \
+ asm( "addu $14, $12, $14 " ); \
+ asm( "lw $15, 0($11) " ); \
+ asm( "sltu $12, $14, $12 " ); \
+ asm( "addu $15, $14, $15 " ); \
+ asm( "sltu $14, $15, $14 " ); \
+ asm( "addu $12, $12, $9 " ); \
+ asm( "sw $15, 0($11) " ); \
+ asm( "addu $12, $12, $14 " ); \
+ asm( "addi $11, $11, 4 " );
+
+#define MULADDC_STOP \
+ asm( "sw $12, %0 " : "=m" (c)); \
+ asm( "sw $11, %0 " : "=m" (d)); \
+ asm( "sw $10, %0 " : "=m" (s) :: \
+ "$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(POLARSSL_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 /* POLARSSL_HAVE_ASM */
+
+#if !defined(MULADDC_CORE)
+#if defined(POLARSSL_HAVE_LONGLONG)
+
+#define MULADDC_INIT \
+{ \
+ t_dbl r; \
+ t_int r0, r1;
+
+#define MULADDC_CORE \
+ r = *(s++) * (t_dbl) b; \
+ r0 = r; \
+ r1 = r >> biL; \
+ r0 += c; r1 += (r0 < c); \
+ r0 += *d; r1 += (r0 < *d); \
+ c = r1; *(d++) = r0;
+
+#define MULADDC_STOP \
+}
+
+#else
+#define MULADDC_INIT \
+{ \
+ t_int s0, s1, b0, b1; \
+ t_int 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 */
generated by cgit v1.2.3 (git 2.39.1) at 2024-11-08 16:48:50 +0000