写在前面的话
这些算法，是我自己近几天在网上搜索下来，并通过自己调试。下列的各个算法，仅是每个算法最核心的部分。
如果需要源代码，可以私信我。 （内存分配、EDF还未写上）
高响应比调度算法
voidhighestResponseRatioNext(struct JCB* jcb){
createJCB(jcb); // 其实就是对所列数据结构进行初始化赋值
compare(jcb); //按到达时间排序
int i = 0, j = 0;
for(; finishNumber != JOBNUMBER; currentTime++){
float maxPriority = 0.0;
int indexPriority = 0;
if(currentTime < jcb[0].arriveTime)//当前时间小于第一个节点到来时间时，直接打印
printJob(jcb);
else{
for(int i = 0; i < JOBNUMBER; i++){
if(strcmp(jcb[i].processStatus, FINISH) != 0){
int waitTime = currentTime - jcb[i].arriveTime;
priority[i] = (waitTime + jcb[i].runTime) * 1.0 / jcb[i].runTime;
if(priority[i] > maxPriority){
maxPriority = priority[i];
indexPriority = i;
}
}
}
strcpy(JobArray[j++], jcb[indexPriority].jobName);
loop(jcb, indexPriority); //见下面一个函数
}
}
printInfo(jcb);//打印进程调度顺序，平均周转时间及平均带权周转时间
currentTime = 0;//当前时间置位
finishNumber = 0;//完成进程数量置位
}
void loop(struct JCB* jcb, int i){
jcb[i].startTime = currentTime;
jcb[i].endTime = jcb[i].startTime + jcb[i].runTime;
jcb[i].turnoverTime = jcb[i].endTime - jcb[i].arriveTime;
jcb[i].useWeightTurnoverTime = jcb[i].turnoverTime * 1.0 / jcb[i].runTime;
strcpy(jcb[i].processStatus, RUN);
while(true){
if(currentTime == jcb[i].endTime){
strcpy(jcb[i].processStatus, FINISH);
finishNumber++;
if(finishNumber == JOBNUMBER)
printJob(jcb);
currentTime--;
break;
}
else{
printJob(jcb);
currentTime++;
}
}
}
先来先服务
voidfirstComeFirstServed(struct JCB* jcb){
createJCB(jcb);
compare(jcb);
int i = 0;
//进程调度currentTime每次加1，直到进程全部被调度完成为止
for(; finishNumber != JOBNUMBER; currentTime++){
if(currentTime < jcb[0].arriveTime)//当前时间小于第一个节点到来时间时，直接打印
printJob(jcb);
else{
strcpy(JobArray[i], jcb[i].jobName);
loop(jcb, i); //见上一个函数
i++;
}
}
printInfo(jcb);//打印进程调度顺序，平均周转时间及平均带权周转时间
currentTime = 0;//静态变量当前时间置位
finishNumber = 0;//静态变量完成进程数量置位
}
时间片轮转调度算法
typedef struct pcb{
char id[10];//进程标识数
int arrivetime;//到达时间
int runtime;//进程已经占用的cpu时间
int needtime;//进程还需要的时间
char state[12];//进程运行状态：wait or runing
struct pcb *next;
}pcb,*PCB;
voidRR_RunProcess(){
//运行进程,简单轮转法Round Robin
PCB p,q,temp;
p=head->next;
while(1){
if(head->next==NULL)
{
printf("此时就绪队列中已无进程!\n");
return ;
}
else
{
while(p){
if((p->needtime>0)&&!(strcmp(p->state,"wait")))
{
printf("进程%s开始,\n",p->id );
strcpy(p->state,"run");
p->runtime+=TIME;
p->needtime-=TIME;
if(p->needtime<0)
p->needtime=0;
}
temp=p;//把该时间片内运行完的进程存到临时temp中
//把temp接到链表尾部,销毁P；
if(temp->needtime>0){//把该时间片内运行完的进程接到就绪队列的尾部
if(count>1)
{
head->next=temp->next;
tail->next=temp;
tail=tail->next;
strcpy(tail->state,"wait");
tail->next=NULL;
}
else if(count==1)
{
//当只有一个进程等待时，分开讨论
head->next=temp;
tail=temp;
strcpy(tail->state,"wait");
tail->next=NULL;
}
}
if(temp->needtime==0){//销毁就绪队列中已经结束的进程
count--;//此时就绪队列中进程数减1
printf("进程%s结束.\n",p->id);
head->next=temp->next;
free(temp);//撤销就绪队列中已经结束的进程
}
p=head->next;
}
}
}
}
银行家算法
typedef struct {
int A;
int B;
int C;
}RESOURCE;
//最大需求矩阵
RESOURCE Max[PROCESSES_NUMBER] =
{
{7,5,3},
{3,2,2},
{9,0,2},
{2,2,2},
{4,3,3}
};
//已分配资源数矩阵
RESOURCE Allocation[PROCESSES_NUMBER] =
{
{0,1,0},
{2,0,0},
{3,0,2},
{2,1,1},
{0,0,2}
};
//需求矩阵
RESOURCE Need[PROCESSES_NUMBER] =
{
{7,4,3},
{1,2,2},
{6,0,0},
{0,1,1},
{4,3,1}
};
//可用资源向量
RESOURCE Available = {3,3,2};
//试探分配
void ProbeAlloc(int process,RESOURCE *res)
{
Available.A -= res->A;
Available.B -= res->B;
Available.C -= res->C;
Allocation[process].A += res->A;
Allocation[process].B += res->B;
Allocation[process].C += res->C;
Need[process].A -= res->A;
Need[process].B -= res->B;
Need[process].C -= res->C;
}
//若试探分配后进入不安全状态，将分配回滚
void RollBack(int process,RESOURCE *res)
{
Available.A += res->A;
Available.B += res->B;
Available.C += res->C;
Allocation[process].A -= res->A;
Allocation[process].B -= res->B;
Allocation[process].C -= res->C;
Need[process].A += res->A;
Need[process].B += res->B;
Need[process].C += res->C;
}
bool SafeCheck()
{
RESOURCE Work = Available;
bool Finish[PROCESSES_NUMBER] = {false,false,false,false,false};
int i;
int j = 0;
for (i = 0; i < PROCESSES_NUMBER; i++)
{
//该进程是否已执行完毕
if(Finish[i] == false)
{
//是否有足够的资源分配给该进程
if(Need[i].A <= Work.A && Need[i].B <= Work.B && Need[i].C <= Work.C)
{
//有则使其执行完成，并将已分配给该进程的资源全部回收
Work.A += Allocation[i].A;
Work.B += Allocation[i].B;
Work.C += Allocation[i].C;
Finish[i] = true;
safeSeq[j++] = i; //顺便记录下来安全序列
i = -1; //需要从开始重新进行遍历
}
}
}
//如果所有进程的Finish向量都为true则处于安全状态，否则为不安全状态
for (i = 0; i < PROCESSES_NUMBER; i++)
{
if (Finish[i] == false)
{
return false;
}
}
return true;
}
//资源分配请求
bool request(int process,RESOURCE *res)
{
//request向量需小于Need矩阵中对应的向量
if(res->A <= Need[process].A && res->B <= Need[process].B && res->C <= Need[process].C)
{
//request向量需小于Available向量
if(res->A <= Available.A && res->B <= Available.B && res->C <= Available.C)
{
//试探分配
ProbeAlloc(process,res);
//如果安全检查成立，则请求成功，否则将分配回滚并返回失败
if(SafeCheck())
{
return true;
}
else
{
printf("安全性检查失败。原因：系统将进入不安全状态，有可能引起死锁。\n");
printf("正在回滚...\n");
RollBack(process,res);
}
}
else
{
printf("安全性检查失败。原因：请求向量大于可利用资源向量。\n");
}
}
else
{
printf("安全性检查失败。原因：请求向量大于需求向量。\n");
}
return false;
}
页面调度算法
int fifo(MemBlock *pb,int m)
{
int lps=0;
double lpp;
int p = 0;
int index =0;
while(index
if(!in_mem(page[index],pb,M)){ //如果该页面不在物理块中
pb[p].pagenum = page[index];
p = (p+1)%M;
lps++;
for(int i=0;i
result[i][index] = pb[i].pagenum;
}
}
index++;
}
printf("FIFO算法所得缺页次数为 %d\n",lps);
lpp = (double)lps/PAGES;
printf("FIFO算法缺页率为 %0.4lf \n",lpp);
printf("页面号序列为:\n");
printArr1(page,PAGES);
printf("结果数列为:\n");
printRelArr(result);
return 0;
}
LRU算法
int getP(MemBlock *pb,int p)
{
int i;
bool out = true; //
for(i=0;i
if(pb[i].recenttime == -1){
p = i;
out = false;
break;
}
}
if(out){
for(i=0;i
if(pb[i].recenttime>pb[p].recenttime)
p = i;
}
}
return p;
}
int getEQnum(int num,MemBlock *pb)
{
int i;
int in = -1;
for(i=0;i
if(pb[i].pagenum == num){
in = i;
break;
}
}
return in;
}
void lru(MemBlock *pb,int m)
{
int lps=0;
double lpp;
int p = 0;
int index =0;
while(index
if(!in_mem(page[index],pb,m)){
p = getP(pb,p);
pb[p].pagenum = page[index];
pb[p].recenttime = 0;
lps++;
for(int i=0;i
result[i][index] = pb[i].pagenum;
}
}else{
int in = getEQnum(page[index],pb);
pb[in].recenttime = 0;
}
int i;
for(i=0;i
if(i!=p&&pb[i].recenttime!=-1){
pb[i].recenttime++;
}
}
index++;
}
printf("LRU算法所得缺页次数为 %d \n",lps);
lpp = 1.0*lps/PAGES;
printf("LRU算法缺页率为: %0.4lf\n",lpp);
printf("页面号序列为:\n");
printArr1(page,PAGES);
printf("LRU结果数组为:\n");
printRelArr(result);
}
0
0
已为社区贡献1条内容
所有评论(0)