Redis（四）：del/unlink 命令源码解析

    // 标识只有一个 w, 说明就是一个普通的写操作，没啥好说的
    {"del",delCommand,-2,"w",0,NULL,1,-1,1,0,0}
    // 标识为 wF, 说明它是一个快速写的操作，其实就是有一个异步优化的过程，稍后详解
    {"unlink",unlinkCommand,-2,"wF",0,NULL,1,-1,1,0,0}

// db.c, del 命令处理    
void delCommand(client *c) {
    // 同步删除
    delGenericCommand(c,0);
}

/* This command implements DEL and LAZYDEL. */
void delGenericCommand(client *c, int lazy) {
    int numdel = 0, j;

    for (j = 1; j < c->argc; j++) {
        // 自动过期数据清理
        expireIfNeeded(c->db,c->argv[j]);
        // 此处分同步删除和异步删除, 主要差别在于对于复杂数据类型的删除方面，如hash,list,set...
        // 针对 string 的删除是完全一样的
        int deleted  = lazy ? dbAsyncDelete(c->db,c->argv[j]) :
                              dbSyncDelete(c->db,c->argv[j]);
        // 写命令的传播问题
        if (deleted) {
            signalModifiedKey(c->db,c->argv[j]);
            notifyKeyspaceEvent(NOTIFY_GENERIC,
                "del",c->argv[j],c->db->id);
            server.dirty++;
            numdel++;
        }
    }
    // 响应删除数据量, 粒度到 key 级别
    addReplyLongLong(c,numdel);
}

// db.c, unlink 删除处理
void unlinkCommand(client *c) {
    // 与 del 一致，只是 lazy 标识不一样
    delGenericCommand(c,1);
}

// db.c, 同步删除数据
/* Delete a key, value, and associated expiration entry if any, from the DB */
int dbSyncDelete(redisDb *db, robj *key) {
    /* Deleting an entry from the expires dict will not free the sds of
     * the key, because it is shared with the main dictionary. */
    // 首先从 expires 队列删除，然后再从 db->dict 中删除
    if (dictSize(db->expires) > 0) dictDelete(db->expires,key->ptr);
    if (dictDelete(db->dict,key->ptr) == DICT_OK) {
        if (server.cluster_enabled) slotToKeyDel(key);
        return 1;
    } else {
        return 0;
    }
}
// dict.c, 如上, 仅仅是 dictDelete() 就可以了，所以真正的删除动作是在 dict 中实现的。
int dictDelete(dict *ht, const void *key) {
    // nofree: 0, 即要求释放内存
    return dictGenericDelete(ht,key,0);
}
// dict.c, nofree: 0:要释放相应的val内存, 1:不释放相应val内存只删除key
/* Search and remove an element */
static int dictGenericDelete(dict *d, const void *key, int nofree)
{
    unsigned int h, idx;
    dictEntry *he, *prevHe;
    int table;

    if (d->ht[0].size == 0) return DICT_ERR; /* d->ht[0].table is NULL */
    if (dictIsRehashing(d)) _dictRehashStep(d);
    h = dictHashKey(d, key);
    // ht[0] 和 ht[1] 如有可能都进行扫描
    for (table = 0; table <= 1; table++) {
        idx = h & d->ht[table].sizemask;
        he = d->ht[table].table[idx];
        prevHe = NULL;
        while(he) {
            if (dictCompareKeys(d, key, he->key)) {
                /* Unlink the element from the list */
                if (prevHe)
                    prevHe->next = he->next;
                else
                    d->ht[table].table[idx] = he->next;
                // no nofree, 就是要 free 内存咯
                if (!nofree) {
                    // 看起来 key/value 需要单独释放内存哦
                    dictFreeKey(d, he);
                    dictFreeVal(d, he);
                }
                zfree(he);
                d->ht[table].used--;
                return DICT_OK;
            }
            prevHe = he;
            he = he->next;
        }
        // 如果没有进行 rehashing, 只需扫描0就行了
        if (!dictIsRehashing(d)) break;
    }
    return DICT_ERR; /* not found */
}

// dict.h, 释放key, value 的逻辑也是非常简单，用一个宏就定义好了
// 释放依赖于 keyDestructor, valDestructor
#define dictFreeKey(d, entry) \
    if ((d)->type->keyDestructor) \
        (d)->type->keyDestructor((d)->privdata, (entry)->key)
#define dictFreeVal(d, entry) \
    if ((d)->type->valDestructor) \
        (d)->type->valDestructor((d)->privdata, (entry)->v.val)
// 所以，我们有必要回去看看 key,value 的析构方法
// 而这，又依赖于具体的数据类型，也就是你在 setXXX 的时候用到的数据类型
// 我们看一下这个 keyDestructor,valDestructor 初始化的样子
// server.c  kv的析构函数定义
/* Db->dict, keys are sds strings, vals are Redis objects. */
dictType dbDictType = {
    dictSdsHash,                /* hash function */
    NULL,                       /* key dup */
    NULL,                       /* val dup */
    dictSdsKeyCompare,          /* key compare */
    dictSdsDestructor,          /* key destructor */
    dictObjectDestructor   /* val destructor */
};

// 1. 先看看 key destructor, key 的释放
// server.c, 直接调用 sds 提供的服务即可
void dictSdsDestructor(void *privdata, void *val)
{
    DICT_NOTUSED(privdata);
    // sds 直接释放key就行了
    sdsfree(val);
}
// sds.c, 真正释放 value 内存
/* Free an sds string. No operation is performed if 's' is NULL. */
void sdsfree(sds s) {
    if (s == NULL) return;
    // zfree, 确实很简单嘛, 因为 sds 是连续的内存空间，直接使用系统提供的方法即可删除
    s_free((char*)s-sdsHdrSize(s[-1]));
}

// 2. value destructor 对value的释放, 如果说 key 一定是string格式的话，value可主不一定了，因为 redis提供丰富的数据类型呢
// server.c
void dictObjectDestructor(void *privdata, void *val)
{
    DICT_NOTUSED(privdata);

    if (val == NULL) return; /* Lazy freeing will set value to NULL. */
    decrRefCount(val);
}
// 减少 value 的引用计数
void decrRefCount(robj *o) {
    if (o->refcount == 1) {
        switch(o->type) {
            // string 类型
            case OBJ_STRING: freeStringObject(o); break;
            // list 类型
            case OBJ_LIST: freeListObject(o); break;
            // set 类型
            case OBJ_SET: freeSetObject(o); break;
            // zset 类型
            case OBJ_ZSET: freeZsetObject(o); break;
            // hash 类型
            case OBJ_HASH: freeHashObject(o); break;
            default: serverPanic("Unknown object type"); break;
        }
        zfree(o);
    } else {
        if (o->refcount <= 0) serverPanic("decrRefCount against refcount <= 0");
        if (o->refcount != OBJ_SHARED_REFCOUNT) o->refcount--;
    }
}

// 在介绍不同数据类型的内存释放前，我们可以先来看下每个元素的数据结构
// dict.h
typedef struct dictEntry {
    // 存储 key 字段内容
    void *key;
    // 用一个联合体存储value
    union {
        // 存储数据时使用 *val 存储
        void *val;
        uint64_t u64;
        // 存储过期时间时使用该字段
        int64_t s64;
        // 存储 score 时使用
        double d;
    } v;
    // 存在hash冲突时，作链表使用
    struct dictEntry *next;
} dictEntry;

// 1. string 类型的释放
// object.c
void freeStringObject(robj *o) {
    // 直接调用 sds服务释放
    if (o->encoding == OBJ_ENCODING_RAW) {
        sdsfree(o->ptr);
    }
}

// 2. list 类型的释放
// object.c
void freeListObject(robj *o) {
    switch (o->encoding) {
    case OBJ_ENCODING_QUICKLIST:
        quicklistRelease(o->ptr);
        break;
    default:
        serverPanic("Unknown list encoding type");
    }
}
// quicklist.c
/* Free entire quicklist. */
void quicklistRelease(quicklist *quicklist) {
    unsigned long len;
    quicklistNode *current, *next;

    current = quicklist->head;
    len = quicklist->len;
    // 链表依次迭代就可以释放完成了
    while (len--) {
        next = current->next;
        // 释放list具体值
        zfree(current->zl);
        quicklist->count -= current->count;
        // 释放list对象
        zfree(current);

        quicklist->len--;
        current = next;
    }
    zfree(quicklist);
}

// 3. set 类型的释放
// object.c, set 分两种类型, ht, intset
void freeSetObject(robj *o) {
    switch (o->encoding) {
    case OBJ_ENCODING_HT:
        // hash 类型则需要删除每个 hash 的 kv
        dictRelease((dict*) o->ptr);
        break;
    case OBJ_ENCODING_INTSET:
        // intset 直接释放
        zfree(o->ptr);
        break;
    default:
        serverPanic("Unknown set encoding type");
    }
}
// dict.c, 
/* Clear & Release the hash table */
void dictRelease(dict *d)
{
    // ht[0],ht[1] 依次清理
    _dictClear(d,&d->ht[0],NULL);
    _dictClear(d,&d->ht[1],NULL);
    zfree(d);
}
// dict.c, 
/* Destroy an entire dictionary */
int _dictClear(dict *d, dictht *ht, void(callback)(void *)) {
    unsigned long i;

    /* Free all the elements */
    for (i = 0; i < ht->size && ht->used > 0; i++) {
        dictEntry *he, *nextHe;

        if (callback && (i & 65535) == 0) callback(d->privdata);
        // 元素为空，hash未命中，但只要 used > 0, 代表就还有需要删除的元素存在
        // 其实对于只有少数几个元素的情况下，这个效率就呵呵了
        if ((he = ht->table[i]) == NULL) continue;
        while(he) {
            nextHe = he->next;
            // 这里的释放 kv 逻辑和前面是一致的
            // 看起来像是递归，其实不然，因为redis不存在数据类型嵌套问题，比如 hash下存储hash, 所以不会存在递归
            // 具体结构会在后续解读到
            dictFreeKey(d, he);
            dictFreeVal(d, he);
            zfree(he);
            ht->used--;
            he = nextHe;
        }
    }
    /* Free the table and the allocated cache structure */
    zfree(ht->table);
    /* Re-initialize the table */
    _dictReset(ht);
    return DICT_OK; /* never fails */
}

// 4. hash 类型的释放
// object.c, hash和set其实是很相似的，代码也做了大量的复用
void freeHashObject(robj *o) {
    switch (o->encoding) {
    case OBJ_ENCODING_HT:
        // ht 形式与set一致
        dictRelease((dict*) o->ptr);
        break;
    case OBJ_ENCODING_ZIPLIST:
        // ziplist 直接释放
        zfree(o->ptr);
        break;
    default:
        serverPanic("Unknown hash encoding type");
        break;
    }
}

// 5. zset 类型的释放
// object.c, zset 的存储形式与其他几个
void freeZsetObject(robj *o) {
    zset *zs;
    switch (o->encoding) {
    case OBJ_ENCODING_SKIPLIST:
        zs = o->ptr;
        // 释放dict 数据, ht 0,1 的释放
        dictRelease(zs->dict);
        // 释放skiplist 数据, 主要看下这个
        zslFree(zs->zsl);
        zfree(zs);
        break;
    case OBJ_ENCODING_ZIPLIST:
        zfree(o->ptr);
        break;
    default:
        serverPanic("Unknown sorted set encoding");
    }
}
// t_zset.c, 释放跳表数据
/* Free a whole skiplist. */
void zslFree(zskiplist *zsl) {
    zskiplistNode *node = zsl->header->level[0].forward, *next;

    zfree(zsl->header);
    while(node) {
        // 基于第0层数据释放，也基于第0层做迭代，直到删除完成
        // 因为其他层数据都是引用的第0层的数据，所以释放时无需关注
        next = node->level[0].forward;
        zslFreeNode(node);
        node = next;
    }
    zfree(zsl);
}
// t_zset 也很简单，只是把 node.ele 释放掉，再把自身释放到即可
// 这样的删除方式依赖于其存储结构，咱们后续再聊
/* Free the specified skiplist node. The referenced SDS string representation
 * of the element is freed too, unless node->ele is set to NULL before calling
 * this function. */
void zslFreeNode(zskiplistNode *node) {
    sdsfree(node->ele);
    zfree(node);
}

// lazyfree.c, 
int dbAsyncDelete(redisDb *db, robj *key) {
    /* Deleting an entry from the expires dict will not free the sds of
     * the key, because it is shared with the main dictionary. */
    if (dictSize(db->expires) > 0) dictDelete(db->expires,key->ptr);

    /* If the value is composed of a few allocations, to free in a lazy way
     * is actually just slower... So under a certain limit we just free
     * the object synchronously. */
    dictEntry *de = dictFind(db->dict,key->ptr);
    if (de) {
        robj *val = dictGetVal(de);
        size_t free_effort = lazyfreeGetFreeEffort(val);

        /* If releasing the object is too much work, let's put it into the
         * lazy free list. */
        // 其实异步方法与同步方法的差别在这，即要求 删除的元素影响须大于某阀值(64)
        // 否则按照同步方式直接删除，因为那样代价更小
        if (free_effort > LAZYFREE_THRESHOLD) {
            // 异步释放+1，原子操作
            atomicIncr(lazyfree_objects,1,&lazyfree_objects_mutex);
            // 将 value 的释放添加到异步线程队列中去，后台处理, 任务类型为 异步释放内存
            bioCreateBackgroundJob(BIO_LAZY_FREE,val,NULL,NULL);
            // 设置val为NULL, 以便在外部进行删除时忽略释放value相关内存
            dictSetVal(db->dict,de,NULL);
        }
    }

    /* Release the key-val pair, or just the key if we set the val
     * field to NULL in order to lazy free it later. */
    if (dictDelete(db->dict,key->ptr) == DICT_OK) {
        if (server.cluster_enabled) slotToKeyDel(key);
        return 1;
    } else {
        return 0;
    }
}
// bio.c, 添加异步任务到线程中, 类型由type决定，线程安全地添加
// 然后嘛，后台线程就不会停地运行了任务了
void bioCreateBackgroundJob(int type, void *arg1, void *arg2, void *arg3) {
    struct bio_job *job = zmalloc(sizeof(*job));

    job->time = time(NULL);
    job->arg1 = arg1;
    job->arg2 = arg2;
    job->arg3 = arg3;
    // 上锁操作
    pthread_mutex_lock(&bio_mutex[type]);
    listAddNodeTail(bio_jobs[type],job);
    bio_pending[type]++;
    // 唤醒任务线程
    pthread_cond_signal(&bio_newjob_cond[type]);
    pthread_mutex_unlock(&bio_mutex[type]);
}
// bio.c, 后台线程任务框架，总之还是有事情可做了。
void *bioProcessBackgroundJobs(void *arg) {
    struct bio_job *job;
    unsigned long type = (unsigned long) arg;
    sigset_t sigset;

    /* Check that the type is within the right interval. */
    if (type >= BIO_NUM_OPS) {
        serverLog(LL_WARNING,
            "Warning: bio thread started with wrong type %lu",type);
        return NULL;
    }

    /* Make the thread killable at any time, so that bioKillThreads()
     * can work reliably. */
    pthread_setcancelstate(PTHREAD_CANCEL_ENABLE, NULL);
    pthread_setcanceltype(PTHREAD_CANCEL_ASYNCHRONOUS, NULL);

    pthread_mutex_lock(&bio_mutex[type]);
    /* Block SIGALRM so we are sure that only the main thread will
     * receive the watchdog signal. */
    sigemptyset(&sigset);
    sigaddset(&sigset, SIGALRM);
    if (pthread_sigmask(SIG_BLOCK, &sigset, NULL))
        serverLog(LL_WARNING,
            "Warning: can't mask SIGALRM in bio.c thread: %s", strerror(errno));
    // 任务一直运行
    while(1) {
        listNode *ln;

        /* The loop always starts with the lock hold. */
        if (listLength(bio_jobs[type]) == 0) {
            // 注意此处将会释放锁哟，以便外部可以添加任务进来
            pthread_cond_wait(&bio_newjob_cond[type],&bio_mutex[type]);
            continue;
        }
        /* Pop the job from the queue. */
        ln = listFirst(bio_jobs[type]);
        job = ln->value;
        /* It is now possible to unlock the background system as we know have
         * a stand alone job structure to process.*/
        pthread_mutex_unlock(&bio_mutex[type]);

        /* Process the job accordingly to its type. */
        if (type == BIO_CLOSE_FILE) {
            close((long)job->arg1);
        } else if (type == BIO_AOF_FSYNC) {
            aof_fsync((long)job->arg1);
        } 
        // 也就是这玩意了，会去处理提交过来的任务
        else if (type == BIO_LAZY_FREE) {
            /* What we free changes depending on what arguments are set:
             * arg1 -> free the object at pointer.
             * arg2 & arg3 -> free two dictionaries (a Redis DB).
             * only arg3 -> free the skiplist. */
            // 本文介绍的删除value形式，用第一种情况
            if (job->arg1)
                lazyfreeFreeObjectFromBioThread(job->arg1);
            else if (job->arg2 && job->arg3)
                lazyfreeFreeDatabaseFromBioThread(job->arg2,job->arg3);
            else if (job->arg3)
                lazyfreeFreeSlotsMapFromBioThread(job->arg3);
        } else {
            serverPanic("Wrong job type in bioProcessBackgroundJobs().");
        }
        zfree(job);

        /* Unblock threads blocked on bioWaitStepOfType() if any. */
        // 唤醒所有相关等待线程
        pthread_cond_broadcast(&bio_step_cond[type]);

        /* Lock again before reiterating the loop, if there are no longer
         * jobs to process we'll block again in pthread_cond_wait(). */
        pthread_mutex_lock(&bio_mutex[type]);
        listDelNode(bio_jobs[type],ln);
        bio_pending[type]--;
    }
}

// lazyfree.c, 和同步删除一致了
/* Release objects from the lazyfree thread. It's just decrRefCount()
 * updating the count of objects to release. */
void lazyfreeFreeObjectFromBioThread(robj *o) {
    decrRefCount(o);
    atomicDecr(lazyfree_objects,1,&lazyfree_objects_mutex);
}