本帖最后由 MindMotion 于 2024-8-9 16:25 编辑
1. Dhrystone简介
Dhrystone是由Reinhold P. Weicker在1984年提出来的一个基准测试程序，其主要目的是测试处理器的整数运算和逻辑运算的性能。Dhrystone首先用Ada语言发布，后来Rick Richardson为Unix开发了用C语言编写的Version 1.1，这个版本也成功的推动了Dhrystone的广泛应用。
Dhrystone标准的测试方法很简单，就是单位时间内跑了多少次Dhrystone程序，其指标单位为DMIPS/MHz。MIPS是Million Instructions Per Second的缩写，每秒处理的百万级的机器语言指令数。DMIPS中的D是Dhrystone的缩写，它表示了在Dhrystone标准的测试方法下的MIPS。
DMIPS表示的是一个相对值，由于历史原因将VAX-11/780机器上每秒运行1757次 Dhrystone程序定义为1 DMIPS，因此在其他平台测试到的每秒Dhrystone次数应除以1757，才是DMIPS数值。
DMIPS/MHz：表示CPU在每1MHz的运行速度下可以执行多少个DMIPS，由于DMIPS与CPU频率具有正相关性，所以这一分数更容易比较不同的CPU在不同的时钟频率下运行Dhrystone的结果。
作为一项基准程序Dhrystone具有以下缺陷：
1) 代码与具有代表性的实际程序代码并不相同。
2) 易受编译器影响。
3) 代码量过小，在现代CPU中，它能够被放进指令缓存中，所以它并不能严格的测量取指性能。
2. 源码获取
Dhrystone程序的最新版本是2.1，其实际上于1988年便已停更。Dhrystone并没有官网，所以想下载其源程序可能会有很多来源，有各种语言版本的实现，以及各种平台下的移植程序。
Roy Longbottom，是一个来自英国政府计算机采购部门Central Computer and Telecommunications Agency (CCTA)的职员，他制作了一个PC性能测试结果网站，搜集了很多性能测试程序以及结果，其中便有Dhrystone，我们可以从他的网站下载Dhrystone源码(C语言版)。
下载地址：http://www.roylongbottom.org.uk/classic_benchmarks.tar.gz
源代码在\classic_benchmarks\source_code\dhrystone2\ 目录下，文件如下：
文件介绍：
dhry.h —— 关于兼容性的原型定义
dhry_1.c —— 主程序入口
dhry_2.c —— 算法子程序
Dhrystone其源码本是用作在PC上运行的，移植到ARM Cortex-M平台下裸系统运行，一般需要修改dhry.h和dhry_1.c文件、删除I/O的相关代码、实现计时函数和打印函数。
3. 移植
Dhrystone作为一项基准测试程序，C编译器的编译效率对测试结果也有很大影响，实测在IAR代码优化下的跑分较高。本文以MM32F5270 MCU为例，开发板型号为Mini-F5277-OB|MB-100，使用IAR 9.30.1 IDE，介绍Dhrystone程序的移植和相关配置。
3.1 创建工程
工程可以参考灵动微电子官网的MM32F5270 LibSamples（https://www.mindmotion.com.cn/products/mm32mcu/mm32f/mm32f_performance/mm32f5270/），也可以直接选择一个简单的工程稍作修改。
3.2 加入Dhrystone
在工程中新创建一个文件夹，命名为“Dhrystone”。
Dhrystone源代码在\classic_benchmarks\source_code\dhrystone2\目录，将dhry.h、dhry_1.c、dhry_2.c复制到该文件夹。
在项目栏添加DHRYSTONE组，并加入dhry_1.c、dhry_2.c文件。
在Option->C/C++ Complier菜单加入dhry.h文件包含路径。
工程结构如下：
3.3 文件修改
参考MM32F5270 LibSamples，其工程具有统一的代码结构，以及做好了硬件的初始化，为了便于移植，将dhry_1.c自带的main函数名称改为dhrystone，在硬件初始化之后调用。
Dhrystone本是用作在PC上运行的，需要删除计时部分、文件I/O部分，以适合在嵌入式平台运行。
计时部分：删除dhry_1.c文件里的#include语句、删除dhry.h文件里跟TIME宏相关的代码；
文件I/O部分：删除dhry_1.c文件里的#include语句、删除涉及Dhry.txt的代码。
修改后的代码如下：
dhry1.c文件：
/*
*************************************************************************
*
* "DHRYSTONE" Benchmark Program
* -----------------------------
*
* Version: C, Version 2.1
*
* File: dhry_1.c (part 2 of 3)
*
* Date: May 25, 1988
*
* Author: Reinhold P. Weicker
*
*************************************************************************
*
* #define options not used
*/
#include <stdlib.h>
#include <stdio.h>
#include "dhry.h"
/* Global Variables: */
Rec_Pointer Ptr_Glob, Next_Ptr_Glob;
int Int_Glob;
Boolean Bool_Glob;
char Ch_1_Glob, Ch_2_Glob;
int Arr_1_Glob [50];
int Arr_2_Glob [50] [50];
Enumeration Func_1 (Capital_Letter Ch_1_Par_Val, Capital_Letter Ch_2_Par_Val);
/*
forward declaration necessary since Enumeration may not simply be int
*/
#ifndef REG
Boolean Reg = false;
#define REG
/* REG becomes defined as empty */
/* i.e. no register variables */
#else
Boolean Reg = true;
#endif
void Proc_1 (REG Rec_Pointer Ptr_Val_Par);
void Proc_2 (One_Fifty *Int_Par_Ref);
void Proc_3 (Rec_Pointer *Ptr_Ref_Par);
void Proc_4 ();
void Proc_5 ();
void Proc_6 (Enumeration Enum_Val_Par, Enumeration *Enum_Ref_Par);
void Proc_7 (One_Fifty Int_1_Par_Val, One_Fifty Int_2_Par_Val, One_Fifty *Int_Par_Ref);
void Proc_8 (Arr_1_Dim Arr_1_Par_Ref, Arr_2_Dim Arr_2_Par_Ref, int Int_1_Par_Val, int Int_2_Par_Val);
Boolean Func_2 (Str_30 Str_1_Par_Ref, Str_30 Str_2_Par_Ref);
/* variables for time measurement: */
#define Too_Small_Time 2 /* Measurements should last at least 2 seconds */
double User_Time;
double Microseconds, Dhrystones_Per_Second, Vax_Mips;
/* end of variables for time measurement */
void dhrystone(void)
/*****/
/* main program, corresponds to procedures */
/* Main and Proc_0 in the Ada version */
{
One_Fifty Int_1_Loc;
REG One_Fifty Int_2_Loc;
One_Fifty Int_3_Loc;
REG char Ch_Index;
Enumeration Enum_Loc;
Str_30 Str_1_Loc;
Str_30 Str_2_Loc;
REG int Run_Index;
REG int Number_Of_Runs;
/* Initializations */
Next_Ptr_Glob = (Rec_Pointer) malloc (sizeof (Rec_Type));
Ptr_Glob = (Rec_Pointer) malloc (sizeof (Rec_Type));
Ptr_Glob->Ptr_Comp = Next_Ptr_Glob;
Ptr_Glob->Discr = Ident_1;
Ptr_Glob->variant.var_1.Enum_Comp = Ident_3;
Ptr_Glob->variant.var_1.Int_Comp = 40;
strcpy (Ptr_Glob->variant.var_1.Str_Comp, "DHRYSTONE PROGRAM, SOME STRING");
strcpy (Str_1_Loc, "DHRYSTONE PROGRAM, 1'ST STRING");
Arr_2_Glob [8][7] = 10;
/* Was missing in published program. Without this statement, */
/* Arr_2_Glob [8][7] would have an undefined value. */
/* Warning: With 16-Bit processors and Number_Of_Runs > 32000, */
/* overflow may occur for this array element. */
printf ("Dhrystone Benchmark, Version 2.1 (Language: C or C++)\n");
getDetails();
if (Reg)
{
printf ("Program compiled with 'register' attribute\n");
}
else
{
printf ("Program compiled without 'register' attribute\n");
}
Number_Of_Runs = NUMBER_Of_RUNS;
printf ("Execution starts, %d runs through Dhrystone......\n", Number_Of_Runs);
/***************/
/* Start timer */
/***************/
start_time();
for (Run_Index = 1; Run_Index <= Number_Of_Runs; ++Run_Index)
{
Proc_5();
Proc_4();
/* Ch_1_Glob == 'A', Ch_2_Glob == 'B', Bool_Glob == true */
Int_1_Loc = 2;
Int_2_Loc = 3;
strcpy (Str_2_Loc, "DHRYSTONE PROGRAM, 2'ND STRING");
Enum_Loc = Ident_2;
Bool_Glob = ! Func_2 (Str_1_Loc, Str_2_Loc);
/* Bool_Glob == 1 */
while (Int_1_Loc < Int_2_Loc) /* loop body executed once */
{
Int_3_Loc = 5 * Int_1_Loc - Int_2_Loc;
/* Int_3_Loc == 7 */
Proc_7 (Int_1_Loc, Int_2_Loc, &Int_3_Loc);
/* Int_3_Loc == 7 */
Int_1_Loc += 1;
} /* while */
/* Int_1_Loc == 3, Int_2_Loc == 3, Int_3_Loc == 7 */
Proc_8 (Arr_1_Glob, Arr_2_Glob, Int_1_Loc, Int_3_Loc);
/* Int_Glob == 5 */
Proc_1 (Ptr_Glob);
for (Ch_Index = 'A'; Ch_Index <= Ch_2_Glob; ++Ch_Index)
/* loop body executed twice */
{
if (Enum_Loc == Func_1 (Ch_Index, 'C'))
/* then, not executed */
{
Proc_6 (Ident_1, &Enum_Loc);
strcpy (Str_2_Loc, "DHRYSTONE PROGRAM, 3'RD STRING");
Int_2_Loc = Run_Index;
Int_Glob = Run_Index;
}
}
/* Int_1_Loc == 3, Int_2_Loc == 3, Int_3_Loc == 7 */
Int_2_Loc = Int_2_Loc * Int_1_Loc;
Int_1_Loc = Int_2_Loc / Int_3_Loc;
Int_2_Loc = 7 * (Int_2_Loc - Int_3_Loc) - Int_1_Loc;
/* Int_1_Loc == 1, Int_2_Loc == 13, Int_3_Loc == 7 */
Proc_2 (&Int_1_Loc);
/* Int_1_Loc == 5 */
} /* loop "for Run_Index" */
/**************/
/* Stop timer */
/**************/
User_Time = end_time();
printf ("Execution ends!\n");
printf ("%12.0f runs %6.2f seconds \n",(double)Number_Of_Runs, (double)User_Time);
printf ("\n");
printf ("Final values:\n");
printf ("Int_Glob: ");
if (Int_Glob == 5) printf ("O.K. ");
else printf ("WRONG ");
printf ("%d\n", Int_Glob);
printf ("Bool_Glob: ");
if (Bool_Glob == 1) printf ("O.K. ");
else printf ("WRONG ");
printf ("%d\n", Bool_Glob);
printf ("Ch_1_Glob: ");
if (Ch_1_Glob == 'A') printf ("O.K. ");
else printf ("WRONG ");
printf ("%c\n", Ch_1_Glob);
printf ("Ch_2_Glob: ");
if (Ch_2_Glob == 'B') printf ("O.K. ");
else printf ("WRONG ");
printf ("%c\n", Ch_2_Glob);
printf ("Arr_1_Glob[8]: ");
if (Arr_1_Glob[8] == 7) printf ("O.K. ");
else printf ("WRONG ");
printf ("%d\n", Arr_1_Glob[8]);
printf ("Arr_2_Glob8/7: ");
if (Arr_2_Glob[8][7] == Number_Of_Runs + 10)
printf ("O.K. ");
else printf ("WRONG ");
printf ("%10d\n", Arr_2_Glob[8][7]);
printf ("Ptr_Glob-> ");
printf (" Ptr_Comp: * %d\n", (int) Ptr_Glob->Ptr_Comp);
printf (" Discr: ");
if (Ptr_Glob->Discr == 0) printf ("O.K. ");
else printf ("WRONG ");
printf ("%d\n", Ptr_Glob->Discr);
printf ("Enum_Comp: ");
if (Ptr_Glob->variant.var_1.Enum_Comp == 2)
printf ("O.K. ");
else printf ("WRONG ");
printf ("%d\n", Ptr_Glob->variant.var_1.Enum_Comp);
printf (" Int_Comp: ");
if (Ptr_Glob->variant.var_1.Int_Comp == 17) printf ("O.K. ");
else printf ("WRONG ");
printf ("%d\n", Ptr_Glob->variant.var_1.Int_Comp);
printf ("Str_Comp: ");
if (strcmp(Ptr_Glob->variant.var_1.Str_Comp,
"DHRYSTONE PROGRAM, SOME STRING") == 0)
printf ("O.K. ");
else printf ("WRONG ");
printf ("%s\n", Ptr_Glob->variant.var_1.Str_Comp);
printf ("Next_Ptr_Glob-> ");
printf (" Ptr_Comp: * %d", (int) Next_Ptr_Glob->Ptr_Comp);
printf (" same as above\n");
printf (" Discr: ");
if (Next_Ptr_Glob->Discr == 0)
printf ("O.K. ");
else printf ("WRONG ");
printf ("%d\n", Next_Ptr_Glob->Discr);
printf ("Enum_Comp: ");
if (Next_Ptr_Glob->variant.var_1.Enum_Comp == 1)
printf ("O.K. ");
else printf ("WRONG ");
printf ("%d\n", Next_Ptr_Glob->variant.var_1.Enum_Comp);
printf (" Int_Comp: ");
if (Next_Ptr_Glob->variant.var_1.Int_Comp == 18)
printf ("O.K. ");
else printf ("WRONG ");
printf ("%d\n", Next_Ptr_Glob->variant.var_1.Int_Comp);
printf ("Str_Comp: ");
if (strcmp(Next_Ptr_Glob->variant.var_1.Str_Comp,
"DHRYSTONE PROGRAM, SOME STRING") == 0)
printf ("O.K. ");
else printf ("WRONG ");
printf ("%s\n", Next_Ptr_Glob->variant.var_1.Str_Comp);
printf ("Int_1_Loc: ");
if (Int_1_Loc == 5)
printf ("O.K. ");
else printf ("WRONG ");
printf ("%d\n", Int_1_Loc);
printf ("Int_2_Loc: ");
if (Int_2_Loc == 13)
printf ("O.K. ");
else printf ("WRONG ");
printf ("%d\n", Int_2_Loc);
printf ("Int_3_Loc: ");
if (Int_3_Loc == 7)
printf ("O.K. ");
else printf ("WRONG ");
printf ("%d\n", Int_3_Loc);
printf ("Enum_Loc: ");
if (Enum_Loc == 1)
printf ("O.K. ");
else printf ("WRONG ");
printf ("%d\n", Enum_Loc);
printf ("Str_1_Loc: ");
if (strcmp(Str_1_Loc, "DHRYSTONE PROGRAM, 1'ST STRING") == 0)
printf ("O.K. ");
else printf ("WRONG ");
printf ("%s\n", Str_1_Loc);
printf ("Str_2_Loc: ");
if (strcmp(Str_2_Loc, "DHRYSTONE PROGRAM, 2'ND STRING") == 0)
printf ("O.K. ");
else printf ("WRONG ");
printf ("%s\n", Str_2_Loc);
printf ("\n");
if (User_Time < Too_Small_Time)
{
printf ("Measured time too small to obtain meaningful results\n");
printf ("Please increase number of runs\n");
printf ("\n");
}
else
{
Microseconds = User_Time * Mic_secs_Per_Second / (double) Number_Of_Runs;
Dhrystones_Per_Second = (double) Number_Of_Runs / User_Time;
Vax_Mips = Dhrystones_Per_Second / 1757.0;
printf ("Microseconds for one run through Dhrystone: ");
printf ("%12.2lf \n", Microseconds);
printf ("Dhrystones per Second: ");
printf ("%10.0lf \n", Dhrystones_Per_Second);
printf ("VAX MIPS rating = ");
printf ("%12.2lf \n",Vax_Mips);
printf ("\n");
printf ("\n");
printf ("\n");
}
}
void Proc_1 (REG Rec_Pointer Ptr_Val_Par)
/******************/
/* executed once */
{
REG Rec_Pointer Next_Record = Ptr_Val_Par->Ptr_Comp;
/* == Ptr_Glob_Next */
/* Local variable, initialized with Ptr_Val_Par->Ptr_Comp, */
/* corresponds to "rename" in Ada, "with" in Pascal */
structassign (*Ptr_Val_Par->Ptr_Comp, *Ptr_Glob);
Ptr_Val_Par->variant.var_1.Int_Comp = 5;
Next_Record->variant.var_1.Int_Comp = Ptr_Val_Par->variant.var_1.Int_Comp;
Next_Record->Ptr_Comp = Ptr_Val_Par->Ptr_Comp;
Proc_3 (&Next_Record->Ptr_Comp);
/* Ptr_Val_Par->Ptr_Comp->Ptr_Comp = Ptr_Glob->Ptr_Comp */
if (Next_Record->Discr == Ident_1)
/* then, executed */
{
Next_Record->variant.var_1.Int_Comp = 6;
Proc_6 (Ptr_Val_Par->variant.var_1.Enum_Comp, &Next_Record->variant.var_1.Enum_Comp);
Next_Record->Ptr_Comp = Ptr_Glob->Ptr_Comp;
Proc_7 (Next_Record->variant.var_1.Int_Comp, 10, &Next_Record->variant.var_1.Int_Comp);
}
else /* not executed */
structassign (*Ptr_Val_Par, *Ptr_Val_Par->Ptr_Comp);
} /* Proc_1 */
void Proc_2 (One_Fifty *Int_Par_Ref)
/******************/
/* executed once */
/* *Int_Par_Ref == 1, becomes 4 */
{
One_Fifty Int_Loc;
Enumeration Enum_Loc;
Int_Loc = *Int_Par_Ref + 10;
do /* executed once */
if (Ch_1_Glob == 'A')
/* then, executed */
{
Int_Loc -= 1;
*Int_Par_Ref = Int_Loc - Int_Glob;
Enum_Loc = Ident_1;
} /* if */
while (Enum_Loc != Ident_1); /* true */
} /* Proc_2 */
void Proc_3 (Rec_Pointer *Ptr_Ref_Par)
/******************/
/* executed once */
/* Ptr_Ref_Par becomes Ptr_Glob */
{
if (Ptr_Glob != Null)
/* then, executed */
*Ptr_Ref_Par = Ptr_Glob->Ptr_Comp;
Proc_7 (10, Int_Glob, &Ptr_Glob->variant.var_1.Int_Comp);
} /* Proc_3 */
void Proc_4 (void) /* without parameters */
/*******/
/* executed once */
{
Boolean Bool_Loc;
Bool_Loc = Ch_1_Glob == 'A';
Bool_Glob = Bool_Loc | Bool_Glob;
Ch_2_Glob = 'B';
} /* Proc_4 */
void Proc_5 (void) /* without parameters */
/*******/
/* executed once */
{
Ch_1_Glob = 'A';
Bool_Glob = false;
} /* Proc_5 */
/* Procedure for the assignment of structures, */
/* if the C compiler doesn't support this feature */
#ifdef NOSTRUCTASSIGN
memcpy (d, s, l)
register char *d;
register char *s;
register int l;
{
while (l--)
*d++ = *s++;
}
#endif
dhry.h文件：
/* Compiler and system dependent definitions: */
#define Mic_secs_Per_Second 1000000.0
/* Berkeley UNIX C returns process times in seconds/HZ */
#ifdef NOSTRUCTASSIGN
#define structassign(d, s) memcpy(&(d), &(s), sizeof(d))
#else
#define structassign(d, s) d = s
#endif
#ifdef NOENUM
#define Ident_1 0
#define Ident_2 1
#define Ident_3 2
#define Ident_4 3
#define Ident_5 4
typedef int Enumeration;
#else
typedef enum {Ident_1, Ident_2, Ident_3, Ident_4, Ident_5}
Enumeration;
#endif
/* for boolean and enumeration types in Ada, Pascal */
/* General definitions: */
#include <stdio.h>
#include <string.h>
/* for strcpy, strcmp */
#define Null 0
/* Value of a Null pointer */
#define true 1
#define false 0
typedef int One_Thirty;
typedef int One_Fifty;
typedef char Capital_Letter;
typedef int Boolean;
typedef char Str_30 [31];
typedef int Arr_1_Dim [50];
typedef int Arr_2_Dim [50] [50];
typedef struct record
{
struct record *Ptr_Comp;
Enumeration Discr;
union {
struct {
Enumeration Enum_Comp;
int Int_Comp;
char Str_Comp [31];
} var_1;
struct {
Enumeration E_Comp_2;
char Str_2_Comp [31];
} var_2;
struct {
char Ch_1_Comp;
char Ch_2_Comp;
} var_3;
} variant;
} Rec_Type, *Rec_Pointer;
打印函数：
工程的platform.c文件已经做好了UART初始化以及重定向相关代码，无需额外添加，源码中需要打印的地方直接用printf即可。
计时函数：
Dhrystone测试需要计算跑dhrystone程序的时间，以推算出单位时间内运行dhrystone程序的次数。这里选择使用TIM1来记录时间，定义start_time函数和end_time函数获取运行程序的时间，这两个函数已经在dhrystone主函数中调用。
为了便于计算，TIM1配置为向上计数模式。需要注意系统时钟频率，配置范围，防止溢出。
3.4 编译器优化
在Option->C/C++ Compiler界面Optimizaition选项设置优化等级如下：
4. 测试
Dhrystone测试需要配置的参数：
NUMBER_Of_RUNS，这个值是自己定义的，但要保证Dhrystone程序运行时间大于2s。
定时器的PSC和ARR，一般配置时钟频率为1KHZ，即1ms计数1次，预装载值可配置为最大值。
连接串口，运行程序，观察上位机调试助手打印情况：
其中VAX MIPS rating就是所谓的DMPIS，上面截图数据是在8MHZ跑的，一般DMIPS/MHZ信息还需要除以频率（单位MHZ），即12.10/8 = 1.5125。
|
楼主
标题置顶
标题高亮
