• p5_190635
    了解作者
  • 2.7MB
    文件大小
  • rar
    文件格式
  • 0
    收藏次数
  • VIP专享
    资源类型
  • 0
    下载次数
  • 2022-06-15 02:49
    上传日期
UCOS-II在DOS下的壳,具体说明见《嵌入式实时操作系统uCOS-II》(UC作者写的UC使用手册的中文版)
UCOS2.rar
内容介绍
/*------------------------------------------------------------------------ * filename - matherr.c * * function(s) * matherr - user-modifiable math error handler *-----------------------------------------------------------------------*/ /* * C/C++ Run Time Library - Version 5.0 * * Copyright (c) 1987, 1992 by Borland International * All Rights Reserved. * */ #include <math.h> #ifdef _Windows #include <_win.h> #endif #ifdef UNIX_matherr #include <stdio.h> #include <process.h> char *whyS [] = { "argument domain error", "argument singularity ", "overflow range error ", "underflow range error", "total loss of significance", "partial loss of significance" }; /*------------------------------------------------------------------------* Name matherr - user-modifiable math error handler Usage #include <math.h> int matherr(struct exception *e); Prototype in math.h Description When exceptions are detected in the math library then a call is made to _matherr() with all the available information. That function does very little, except to map the exception "why" into either ERANGE or EDOMAIN in errno. Its main purpose is to act as a focal point for changes in error handling. For example, if you were writing a spreadsheet you might replace this function with one which pops up an error window explaining something like: "log (-2.0) caused domain error, in cell J7" and then longjmp() to a reset state in the spreadsheet and await the next command from the user. The default version of Turbo C's matherr routine masks underflow and precision errors; others errors are considered fatal. It serves as a hook that you can replace when writing your own math error handling routine. The rationale for masking underflow and precision errors is that these are not errors according to the ANSI C spec. Consequently, you will get exp(-1000) = 0 sin(1e100) = NAN without any error or warning, even though there is a total loss of precision in both cases. You can trap these errors by modifying matherr. The possible errors are DOMAIN, SING, OVERFLOW, UNDERFLOW, TLOSS, PLOSS and listed in <math.h>. As explained above, UNDERFLOW and TLOSS are masked by the default matherr. PLOSS is not supported by TC and is not generated by any library functions. The remaining errors, DOMAIN, SING, and OVERFLOW, are fatal with the default matherr. You can modify matherr to be a custom error handling routine (such as one that catches and resolves certain type of errors); the modified matherr should return 0 if it failed to resolve the error, or non-zero if the error was resolved. When matherr returns non-zero, no error message is printed, and errno is not changed. The important thing is that we don't know what error handling you want, but you are assured that all errors will arrive at matherr() with all the information you need to design a custom format. We do not ship as standard the function named matherr() which may be familiar to UNIX users, since the ANSI x3j11 draft specifies an incompatible style. This version is as close as we could get without breaking the ANSI rules. You can, however, convert this version to the UNIX style if you prefer. The necessary code is included but switched off. Return value The default return value for matherr is simply 0. matherr can also modify e->retval, which propagates through _matherr back to the original caller. When matherr returns 0, (indicating that it was not able to resolve the error) _matherr sets errno and prints an error message. When matherr returns non-zero, (indicating that it was able to resolve the error) errno is not set and no messages are printed. *-------------------------------------------------------------------------*/ int _FARFUNC matherr (struct exception *e) { #ifdef _Windows char errMsg[ 80 ]; sprintf (errMsg, "%s (%8g,%8g): %s\n", e->name, e->arg1, e->arg2, whyS [e->type - 1]); _errorExitBox( errMsg, 1 ); #else fprintf (stderr, "%s (%8g,%8g): %s\n", e->name, e->arg1, e->arg2, whyS [e->type - 1]); exit (1); #endif } #else int _FARFUNC matherr(struct exception *e) { if (e->type == UNDERFLOW) { /* flush underflow to 0 */ e->retval = 0; return 1; } if (e->type == TLOSS) { /* total loss of precision, but ignore the problem */ return 1; } /* all other errors are fatal */ return 0; } #endif
评论
    相关推荐
    • PROTEUS模拟DOS
      一个proteus下模拟DOS系统的程序和仿真原理图,特别好
    • mbedostutorial:用于嵌入式编程的Mbed OS教程
      mbedostutorial:用于嵌入式编程的Mbed OS教程
    • RDOS operating system-开源
      适用于PC和基于386+的嵌入式系统的快速,多线程OS。 运行许多DOS,DPMI和Win32控制台程序。 100%组装,带分段保护。
    • 嵌入式必备字模软件
      字模软件,嵌入式开发必备工具,界面开发,字库建设。可以根据需要的字建立字库,也可以绘制任何你需要的图形来生成一个数组,非常好用的界面设置软件。
    • DOS图形记事本(基于嵌入式GUI)源代码及设计文档
      DOS图形记事本(基于嵌入式GUI)源代码及设计文档 http://www.ecgui.com 原创软件!
    • EasyARM软件 嵌入式
      easyARM软件 嵌入式系统 arm easyARM软件 嵌入式系统 arm
    • DOS/32 Advanced DOS Extender-开源
      DOS / 32 Advanced DOS Extender是DOS / 4GW DOS Extender及其兼容版本的替代产品。 快速,小型和灵活的DOS / 32A可以在许多环境中使用,从嵌入式系统到DOS仿真器,无论是开发人员还是最终用户都可以使用。
    • 嵌入式ARMlpc2214
      基于周立功公司的开发板,内核是ARM7的lpc2214芯片
    • 嵌入式系统外文翻译
      嵌入式毕业设计有关嵌入式操作系统的外文翻译,一般的嵌入式项目都可以用。 嵌入式毕业设计有关嵌入式操作系统的外文翻译,一般的嵌入式项目都可以用。
    • dos启动vxworks
      dos嵌入式行业的应用还是比较方便的,vxworks开发初期,对于一些硬件设置和调试可能经常需要,我就经常vxworks和dos交替使用.