如何在汇编语言X86中检测溢出条件 [英] How to detect overflow conditions in Assembly Language X86

问题描述

我有一个作业，其中我们必须编写两个函数.还必须使用处理器的条件代码检测溢出条件，并返回0表示已遇到错误.我能够编写函数.

I have an assignment in which we have to write two functions. Also must detect overflow conditions using the processor's condition codes and return 0 to indicate that an error has been encountered. I was able to write the functions.

 .file  "formula.c"  
    .text
.globl _nCr  
    .def    _nCr;   .scl    2;  .type   32; .endef  
_nCr:  
        pushl   %ebp  
    movl    %esp, %ebp  
    subl    $56, %esp  
    movl    8(%ebp), %eax  
    movl    %eax, (%esp)  
    testl %eax, %eax  
    call    _factorial  
    movl    %eax, -12(%ebp)  
    movl    12(%ebp), %eax  
    addl    $1, %eax  
    movl    %eax, (%esp)  
    call    _factorial  
    movl    %eax, -16(%ebp)  
    movl    12(%ebp), %eax  
    notl    %eax  
    addl    8(%ebp), %eax  
    movl    %eax, (%esp)  
    call    _factorial  
    movl    %eax, -20(%ebp)  
    movl    -16(%ebp), %eax  
    movl    %eax, %edx  
    imull   -20(%ebp), %edx  
    movl    %edx, -28(%ebp)  
    movl    -12(%ebp), %eax  
    movl    %eax, %edx  
    sarl    $31, %edx  
    idivl   -28(%ebp)  
    leave  
    ret  
.globl _factorial   
    .def    _factorial;  .scl    2;     .type   32;     .endef   
_factorial:  
    pushl   %ebp  
    movl    %esp, %ebp  
    subl    $16, %esp  
    movl    $1, -8(%ebp)  
    movl    $1, -4(%ebp)  
    jmp L3  
L4:   
    movl    -8(%ebp), %eax   
    imull   -4(%ebp), %eax  
    movl    %eax, -8(%ebp)  
    addl    $1, -4(%ebp)   
L3:
    movl    -4(%ebp), %eax  
    cmpl    8(%ebp), %eax  
    jle L4  
    movl    -8(%ebp), %eax  
    leave  
    ret  
    .def    ___main;    .scl    2;  .type   32; .endef  
    .section .rdata,"dr"  
    .align 4  

此功能基本上执行nCr = n! / (r! (n-r)!).当数字变大时，阶乘会发生溢出.

This function basically does nCr = n! / (r! (n-r)!). The overflow occurs in factorial when the numbers get larger.

我只是不明白如何设置溢出条件.

I just do not understand how I would set the overflow conditions.

推荐答案

在x86架构上，当执行诸如addl 8(%ebp), %eax之类的算术指令时，条件代码将设置在CPU状态字中.有些指令的行为取决于条件代码.

On the x86 architecture, when an arithmetic instruction executes such as addl 8(%ebp), %eax the condition codes are set in the CPU status word. There are instructions whose behavior depends on condition codes.

您可以让代码在给定条件下采用替代路径(执行分支). x86在Jxx助记符下提供了一系列条件分支指令:JA, JAE, JB, JBE, JC, JCXZ, ..., JZ.例如，JZ表示如果为零则跳转:如果指令产生的结果为零，则转移一个分支，并设置零标志. JO在溢出时跳转.

You can have the code take an alternate path (execute a branch) on a given condition. The x86 has a family of conditional branching instructions under the Jxx mnemonics: JA, JAE, JB, JBE, JC, JCXZ, ..., JZ. For instance JZ means jump if zero: take a branch if the instruction produced a zero result, setting the zero flag. JO is jump on overflow.

条件也可以转换为字节数据并存储到寄存器或存储器中.这对于编译如下C表达式很有用:

A condition can also be converted to a byte datum and stored into a register or memory. This is useful for compiling C expressions like:

 x = (y != 3); /* if (y != 3) x = 1; else x = 0 */

这是由SETx组指令完成的，该组指令也很多，如条件分支:SETA, SETAE, SETB, ..., SETZ.例如，如果零条件为真，则SETZ将给定字节设置为1.例如

It is done by the SETx group of instructions which are also numerous, like the conditional branches: SETA, SETAE, SETB, ..., SETZ. For instance SETZ will set a given byte to 1 if the zero condition is true. E.g.

 seto %bl  /* set bottom byte of B register to 1 if overflow flag set */

这篇关于如何在汇编语言X86中检测溢出条件的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！