# 2016q3 Homework2 (phonebook-concurrent)
contributed by <`HahaSula`>
>未曾進行課程表單填寫，可能造成課程成績登記錯誤，請立即補填寫個人基本資料表單，以免權益受損[name=課程助教]
###### tags: `HahaSula` `concurrent`

---
## 預期研究目標
* concurrent vs
* POSIX THREAD
* 學習效能分析工具
* code refactoring 練習
* 探索 clz 的應用

## 硬體規格

## concurrent? vs Parallelism?
看完部分指定連結
>> 動作也太慢了吧？之後幾周都是這種等級的自修教材 [name=jserv]

問題癥結點?
concurrent 問題?
* concurrent
## 閱讀吳彥寬的實驗 (觀看 phonebook--concurrent)

尚未實作，先提出想法

append 改善

header部分 Head entryStart 重複了(使用上重複且其中一個僅使用一次)
eptr 是一個mmap對印file的結尾 如果能用一個int 計數器不知效果會不會比較好?

phonebook-opt.h
typedef struct _append_a {
    char *ptr;
    char *eptr;
    int tid;
    int nthread;
    entry *entryStart;
    entry *pHead;
    entry *pLast;
} append_a;

每一個thread 在file裡面的shift 是thread * Length而且為一量每一輪都要進行累加，想法承接上面，傳入entry計算量 這樣shift也不用累加(不知道哪種對cache比較好)
##### 想法示意圖

phonebook-opt.c
void append(void *arg)
{
    struct timespec start, end;
    double cpu_time;

    clock_gettime(CLOCK_REALTIME, &start);

    append_a *app = (append_a *) arg;

    int count = 0;
    entry *j = app->entryStart;
    for (char *i = app->ptr; i < app->eptr;
            i += MAX_LAST_NAME_SIZE * app->nthread,
            j += app->nthread,count++) {
        app->pLast->pNext = j;
        app->pLast = app->pLast->pNext;

        app->pLast->lastName = i;
        dprintf("thread %d append string = %s\n",
                app->tid, app->pLast->lastName);
        app->pLast->pNext = NULL;
    }
    clock_gettime(CLOCK_REALTIME, &end);
    cpu_time = diff_in_second(start, end);

    dprintf("thread take %lf sec, count %d\n", cpu_time, count);

    pthread_exit(NULL);
}

phead 在上面有做一個malloc 而且有一個e的aline(沒有用到)
且在malloc之後又做一個 phead = phead->next的動作(NULL)

main thread 本身應該也可以去分配append 這樣join 的waiting time 應該可以減少

最後在append各thread 的entry list 的時候可以把[1]給提出來 for迴圈裡面就不用再做branch(for 迴圈不大 應該成效不彰)

main.c

entry *pHead, *e;
    pHead = (entry *) malloc(sizeof(entry));
    printf("size of entry : %lu bytes\n", sizeof(entry));
    e = pHead;
    e->pNext = NULL;
----
for (int i = 0; i < THREAD_NUM; i++)
        pthread_create( &tid[i], NULL, (void *) &append, (void *) app[i]);

    for (int i = 0; i < THREAD_NUM; i++)
        pthread_join(tid[i], NULL);

    entry *etmp;
    pHead = pHead->pNext;
    for (int i = 0; i < THREAD_NUM; i++) {
        if (i == 0) {
            pHead = app[i]->pHead->pNext;
            dprintf("Connect %d head string %s %p\n", i,
                    app[i]->pHead->pNext->lastName, app[i]->ptr);
        } else {
            etmp->pNext = app[i]->pHead->pNext;
            dprintf("Connect %d head string %s %p\n", i,
                    app[i]->pHead->pNext->lastName, app[i]->ptr);
        }

        etmp = app[i]->pLast;
        dprintf("Connect %d tail string %s %p\n", i,
                app[i]->pLast->lastName, app[i]->ptr);
        dprintf("round %d\n", i);
    }

thread pool沒做
concurrent-ll 沒分析

###### tags: `HahaSula` `concurrent`