struct Stack* parseCmdLine(const char* const clStr)
{
struct CmdLine* cmdLine;
struct Queue* tokens;
char *token, *nextToken, *command;
char **operators, **files;
bool background, pushed, pipedIn, pipeOut;
struct Stack* commandStack;
int i;
struct LinkedList* parameters;
tokens=tokenize(clStr);
commandStack=stackCreate();
pipedIn=false;
pipeOut=false;
while (!queueIsEmpty(tokens))
{
//Get Command
pushed=false;
command=(char *)queueGetFront(tokens);
//Parse Parameters
parameters=listCreate();
if (!queueIsEmpty(tokens))
{
while (!queueIsEmpty(tokens))
{
token=(char *)queuePeek(tokens);
if (!isDelimiter(token))
{
listAppend(parameters, token);
queueGetFront(tokens);
}
else
{
break;
}
}
}
//Check for Redirects
operators=(char **)malloc(sizeof(char*)*REDIRECTS);
files=(char **)malloc(sizeof(char*)*REDIRECTS);
for (i=0; i<REDIRECTS; i++)
{
operators[i]=NULL;
files[i]=NULL;
}
if (!queueIsEmpty(tokens))
{
for (i=0; i<REDIRECTS && !queueIsEmpty(tokens); i++)
{
token=(char *)queuePeek(tokens);
if (strcmp(token, CL_TOKEN_GREATER_THAN)==0
|| strcmp(token, CL_TOKEN_LESS_THAN)==0)
{
operators[i]=token;
queueGetFront(tokens);
token=(char *)queueGetFront(tokens);
files[i]=token;
}
else
{
break;
}
}
}
//Check for background process or pipe
background=false;
pipeOut=false;
if (!queueIsEmpty(tokens))
{
token=(char *)queuePeek(tokens);
if (strcmp(token, CL_TOKEN_AMP)==0)
{
queueGetFront(tokens);
background=true;
}
else if (strcmp(token, CL_TOKEN_PIPE)==0)
{
queueGetFront(tokens);
pipeOut=true;
}
}
//Add Command to Stack of Commands
pushCmdLine(commandStack, command, parameters, operators, files, pipedIn, pipeOut, background);
pushed=true;
pipedIn=pipeOut;
if (background || !pipeOut)
{
break; //Terminated because in background or syntax error
}
}
if (!pushed)
{
free(operators);
free(files);
}
queueDestroy(tokens, false);
return commandStack;
}
/* The SORT command is the most complex command in Redis. Warning: this code
* is optimized for speed and a bit less for readability */
void sortCommand(redisClient *c) {
list *operations;
unsigned int outputlen = 0;
int desc = 0, alpha = 0;
long limit_start = 0, limit_count = -1, start, end;
int j, dontsort = 0, vectorlen;
int getop = 0; /* GET operation counter */
int int_convertion_error = 0;
int syntax_error = 0;
robj *sortval, *sortby = NULL, *storekey = NULL;
redisSortObject *vector; /* Resulting vector to sort */
/* Lookup the key to sort. It must be of the right types */
sortval = lookupKeyRead(c->db,c->argv[1]);
if (sortval && sortval->type != REDIS_SET &&
sortval->type != REDIS_LIST &&
sortval->type != REDIS_ZSET)
{
addReply(c,shared.wrongtypeerr);
return;
}
/* Create a list of operations to perform for every sorted element.
* Operations can be GET */
operations = listCreate();
listSetFreeMethod(operations,zfree);
j = 2; /* options start at argv[2] */
/* Now we need to protect sortval incrementing its count, in the future
* SORT may have options able to overwrite/delete keys during the sorting
* and the sorted key itself may get destroyed */
if (sortval)
incrRefCount(sortval);
else
sortval = createQuicklistObject();
/* The SORT command has an SQL-alike syntax, parse it */
while(j < c->argc) {
int leftargs = c->argc-j-1;
if (!strcasecmp(c->argv[j]->ptr,"asc")) {
desc = 0;
} else if (!strcasecmp(c->argv[j]->ptr,"desc")) {
desc = 1;
} else if (!strcasecmp(c->argv[j]->ptr,"alpha")) {
alpha = 1;
} else if (!strcasecmp(c->argv[j]->ptr,"limit") && leftargs >= 2) {
if ((getLongFromObjectOrReply(c, c->argv[j+1], &limit_start, NULL)
!= REDIS_OK) ||
(getLongFromObjectOrReply(c, c->argv[j+2], &limit_count, NULL)
!= REDIS_OK))
{
syntax_error++;
break;
}
j+=2;
} else if (!strcasecmp(c->argv[j]->ptr,"store") && leftargs >= 1) {
storekey = c->argv[j+1];
j++;
} else if (!strcasecmp(c->argv[j]->ptr,"by") && leftargs >= 1) {
sortby = c->argv[j+1];
/* If the BY pattern does not contain '*', i.e. it is constant,
* we don't need to sort nor to lookup the weight keys. */
if (strchr(c->argv[j+1]->ptr,'*') == NULL) {
dontsort = 1;
} else {
/* If BY is specified with a real patter, we can't accept
* it in cluster mode. */
if (server.cluster_enabled) {
addReplyError(c,"BY option of SORT denied in Cluster mode.");
syntax_error++;
break;
}
}
j++;
} else if (!strcasecmp(c->argv[j]->ptr,"get") && leftargs >= 1) {
if (server.cluster_enabled) {
addReplyError(c,"GET option of SORT denied in Cluster mode.");
syntax_error++;
break;
}
listAddNodeTail(operations,createSortOperation(
REDIS_SORT_GET,c->argv[j+1]));
getop++;
j++;
} else {
addReply(c,shared.syntaxerr);
syntax_error++;
break;
}
j++;
}
/* Handle syntax errors set during options parsing. */
if (syntax_error) {
decrRefCount(sortval);
listRelease(operations);
return;
}
/* When sorting a set with no sort specified, we must sort the output
* so the result is consistent across scripting and replication.
*
* The other types (list, sorted set) will retain their native order
* even if no sort order is requested, so they remain stable across
* scripting and replication. */
if (dontsort &&
sortval->type == REDIS_SET &&
(storekey || c->flags & REDIS_LUA_CLIENT))
{
/* Force ALPHA sorting */
dontsort = 0;
alpha = 1;
sortby = NULL;
}
/* Destructively convert encoded sorted sets for SORT. */
if (sortval->type == REDIS_ZSET)
zsetConvert(sortval, REDIS_ENCODING_SKIPLIST);
/* Objtain the length of the object to sort. */
switch(sortval->type) {
case REDIS_LIST: vectorlen = listTypeLength(sortval); break;
case REDIS_SET: vectorlen = setTypeSize(sortval); break;
case REDIS_ZSET: vectorlen = dictSize(((zset*)sortval->ptr)->dict); break;
default: vectorlen = 0; redisPanic("Bad SORT type"); /* Avoid GCC warning */
}
/* Perform LIMIT start,count sanity checking. */
start = (limit_start < 0) ? 0 : limit_start;
end = (limit_count < 0) ? vectorlen-1 : start+limit_count-1;
if (start >= vectorlen) {
start = vectorlen-1;
end = vectorlen-2;
}
if (end >= vectorlen) end = vectorlen-1;
/* Whenever possible, we load elements into the output array in a more
* direct way. This is possible if:
*
* 1) The object to sort is a sorted set or a list (internally sorted).
* 2) There is nothing to sort as dontsort is true (BY <constant string>).
*
* In this special case, if we have a LIMIT option that actually reduces
* the number of elements to fetch, we also optimize to just load the
* range we are interested in and allocating a vector that is big enough
* for the selected range length. */
if ((sortval->type == REDIS_ZSET || sortval->type == REDIS_LIST) &&
dontsort &&
(start != 0 || end != vectorlen-1))
{
vectorlen = end-start+1;
}
/* Load the sorting vector with all the objects to sort */
vector = zmalloc(sizeof(redisSortObject)*vectorlen);
j = 0;
if (sortval->type == REDIS_LIST && dontsort) {
/* Special handling for a list, if 'dontsort' is true.
* This makes sure we return elements in the list original
* ordering, accordingly to DESC / ASC options.
*
* Note that in this case we also handle LIMIT here in a direct
* way, just getting the required range, as an optimization. */
if (end >= start) {
listTypeIterator *li;
listTypeEntry entry;
li = listTypeInitIterator(sortval,
desc ? (long)(listTypeLength(sortval) - start - 1) : start,
desc ? REDIS_HEAD : REDIS_TAIL);
while(j < vectorlen && listTypeNext(li,&entry)) {
vector[j].obj = listTypeGet(&entry);
vector[j].u.score = 0;
vector[j].u.cmpobj = NULL;
j++;
}
listTypeReleaseIterator(li);
/* Fix start/end: output code is not aware of this optimization. */
end -= start;
start = 0;
}
} else if (sortval->type == REDIS_LIST) {
listTypeIterator *li = listTypeInitIterator(sortval,0,REDIS_TAIL);
listTypeEntry entry;
while(listTypeNext(li,&entry)) {
vector[j].obj = listTypeGet(&entry);
vector[j].u.score = 0;
vector[j].u.cmpobj = NULL;
j++;
}
listTypeReleaseIterator(li);
} else if (sortval->type == REDIS_SET) {
setTypeIterator *si = setTypeInitIterator(sortval);
robj *ele;
while((ele = setTypeNextObject(si)) != NULL) {
vector[j].obj = ele;
vector[j].u.score = 0;
vector[j].u.cmpobj = NULL;
j++;
}
setTypeReleaseIterator(si);
} else if (sortval->type == REDIS_ZSET && dontsort) {
/* Special handling for a sorted set, if 'dontsort' is true.
* This makes sure we return elements in the sorted set original
* ordering, accordingly to DESC / ASC options.
*
* Note that in this case we also handle LIMIT here in a direct
* way, just getting the required range, as an optimization. */
zset *zs = sortval->ptr;
zskiplist *zsl = zs->zsl;
zskiplistNode *ln;
robj *ele;
int rangelen = vectorlen;
/* Check if starting point is trivial, before doing log(N) lookup. */
if (desc) {
long zsetlen = dictSize(((zset*)sortval->ptr)->dict);
ln = zsl->tail;
if (start > 0)
ln = zslGetElementByRank(zsl,zsetlen-start);
} else {
ln = zsl->header->level[0].forward;
if (start > 0)
ln = zslGetElementByRank(zsl,start+1);
}
while(rangelen--) {
redisAssertWithInfo(c,sortval,ln != NULL);
ele = ln->obj;
vector[j].obj = ele;
vector[j].u.score = 0;
vector[j].u.cmpobj = NULL;
j++;
ln = desc ? ln->backward : ln->level[0].forward;
}
/* Fix start/end: output code is not aware of this optimization. */
end -= start;
start = 0;
} else if (sortval->type == REDIS_ZSET) {
dict *set = ((zset*)sortval->ptr)->dict;
dictIterator *di;
dictEntry *setele;
di = dictGetIterator(set);
while((setele = dictNext(di)) != NULL) {
vector[j].obj = dictGetKey(setele);
vector[j].u.score = 0;
vector[j].u.cmpobj = NULL;
j++;
}
dictReleaseIterator(di);
} else {
redisPanic("Unknown type");
}
redisAssertWithInfo(c,sortval,j == vectorlen);
/* Now it's time to load the right scores in the sorting vector */
if (dontsort == 0) {
for (j = 0; j < vectorlen; j++) {
robj *byval;
if (sortby) {
/* lookup value to sort by */
byval = lookupKeyByPattern(c->db,sortby,vector[j].obj);
if (!byval) continue;
} else {
/* use object itself to sort by */
byval = vector[j].obj;
}
if (alpha) {
if (sortby) vector[j].u.cmpobj = getDecodedObject(byval);
} else {
if (sdsEncodedObject(byval)) {
char *eptr;
vector[j].u.score = strtod(byval->ptr,&eptr);
if (eptr[0] != '\0' || errno == ERANGE ||
isnan(vector[j].u.score))
{
int_convertion_error = 1;
}
} else if (byval->encoding == REDIS_ENCODING_INT) {
/* Don't need to decode the object if it's
* integer-encoded (the only encoding supported) so
* far. We can just cast it */
vector[j].u.score = (long)byval->ptr;
} else {
redisAssertWithInfo(c,sortval,1 != 1);
}
}
/* when the object was retrieved using lookupKeyByPattern,
* its refcount needs to be decreased. */
if (sortby) {
decrRefCount(byval);
}
}
}
if (dontsort == 0) {
server.sort_desc = desc;
server.sort_alpha = alpha;
server.sort_bypattern = sortby ? 1 : 0;
server.sort_store = storekey ? 1 : 0;
if (sortby && (start != 0 || end != vectorlen-1))
pqsort(vector,vectorlen,sizeof(redisSortObject),sortCompare, start,end);
else
qsort(vector,vectorlen,sizeof(redisSortObject),sortCompare);
}
/* Send command output to the output buffer, performing the specified
* GET/DEL/INCR/DECR operations if any. */
outputlen = getop ? getop*(end-start+1) : end-start+1;
if (int_convertion_error) {
addReplyError(c,"One or more scores can't be converted into double");
} else if (storekey == NULL) {
/* STORE option not specified, sent the sorting result to client */
addReplyMultiBulkLen(c,outputlen);
for (j = start; j <= end; j++) {
listNode *ln;
listIter li;
if (!getop) addReplyBulk(c,vector[j].obj);
listRewind(operations,&li);
while((ln = listNext(&li))) {
redisSortOperation *sop = ln->value;
robj *val = lookupKeyByPattern(c->db,sop->pattern,
vector[j].obj);
if (sop->type == REDIS_SORT_GET) {
if (!val) {
addReply(c,shared.nullbulk);
} else {
addReplyBulk(c,val);
decrRefCount(val);
}
} else {
/* Always fails */
redisAssertWithInfo(c,sortval,sop->type == REDIS_SORT_GET);
}
}
}
} else {
robj *sobj = createQuicklistObject();
/* STORE option specified, set the sorting result as a List object */
for (j = start; j <= end; j++) {
listNode *ln;
listIter li;
if (!getop) {
listTypePush(sobj,vector[j].obj,REDIS_TAIL);
} else {
listRewind(operations,&li);
while((ln = listNext(&li))) {
redisSortOperation *sop = ln->value;
robj *val = lookupKeyByPattern(c->db,sop->pattern,
vector[j].obj);
if (sop->type == REDIS_SORT_GET) {
if (!val) val = createStringObject("",0);
/* listTypePush does an incrRefCount, so we should take care
* care of the incremented refcount caused by either
* lookupKeyByPattern or createStringObject("",0) */
listTypePush(sobj,val,REDIS_TAIL);
decrRefCount(val);
} else {
/* Always fails */
redisAssertWithInfo(c,sortval,sop->type == REDIS_SORT_GET);
}
}
}
}
if (outputlen) {
setKey(c->db,storekey,sobj);
notifyKeyspaceEvent(REDIS_NOTIFY_LIST,"sortstore",storekey,
c->db->id);
server.dirty += outputlen;
} else if (dbDelete(c->db,storekey)) {
signalModifiedKey(c->db,storekey);
notifyKeyspaceEvent(REDIS_NOTIFY_GENERIC,"del",storekey,c->db->id);
server.dirty++;
}
decrRefCount(sobj);
addReplyLongLong(c,outputlen);
}
/* Cleanup */
if (sortval->type == REDIS_LIST || sortval->type == REDIS_SET)
for (j = 0; j < vectorlen; j++)
decrRefCount(vector[j].obj);
decrRefCount(sortval);
listRelease(operations);
for (j = 0; j < vectorlen; j++) {
if (alpha && vector[j].u.cmpobj)
decrRefCount(vector[j].u.cmpobj);
}
zfree(vector);
}
vnode_t *vnode_new(const char *root_dir, uint32_t id, enum eVnodeStorageType storage_type, vnode_write_queue_handle_write_cb vnode_write_queue_handle_write)
{
vnode_t *vnode = (vnode_t*)zmalloc(sizeof(vnode_t));
memset(vnode, 0, sizeof(vnode_t));
vnode->id = id;
vnode->storage_type = storage_type;
/*vnode->max_dbsize = 1024L * 1024L * 1024L * 4L;*/
vnode->max_dbsize = 1024L * 1024L * 500L;
/* Create vnode root dir */
sprintf(vnode->root_dir, "%s/%04d", root_dir, id);
if ( mkdir_if_not_exist(vnode->root_dir) != 0 ){
error_log("Cann't create vnode(%d) dir:%s", id, vnode->root_dir);
zfree(vnode);
return NULL;
}
/* datazones */
int i;
for ( i = 0 ; i < MAX_DATAZONES ; i++ ){
vnode->datazones[i] = (datazone_t*)zmalloc(sizeof(datazone_t));
if ( datazone_init(vnode->datazones[i], vnode, i) != 0 ){
zfree(vnode);
return NULL;
}
}
/* Slices DB */
if ( vnode->storage_type >= STORAGE_KVDB ){
// Create Metadata DB.
const char *metadata_dbname = "metadata";
kvdb_t *kvdb_metadata = vnode_open_kvdb(vnode, metadata_dbname);
if ( kvdb_metadata == NULL ){
error_log("MetadataDB create failed. dbname:%s", metadata_dbname);
zfree(vnode);
return NULL;
}
vnode->kvdb_metadata = kvdb_metadata;
uint32_t active_slicedb_id = 0;
if ( kvdb_get_uint32(kvdb_metadata, "active_slicedb_id", &active_slicedb_id) != 0 ){
active_slicedb_id = 0;
}
notice_log("vnode active_slicedb_id:%d", active_slicedb_id);
// Create Slice DB.
for ( int db_id = 0 ; db_id <= active_slicedb_id ; db_id++ ){
slicedb_t *slicedb = vnode_open_slicedb(vnode, db_id);
if ( slicedb != NULL ){
} else {
/*char dbname[NAME_MAX];*/
/*sprintf(dbname, "slice-%03d", db_id);*/
/*kvdb_t *kvdb = vnode_open_kvdb(vnode, dbname);*/
/*if ( kvdb != NULL ){*/
/*vnode->slicedbs[db_id] = slicedb_new(db_id, kvdb, vnode->max_dbsize);*/
/*} else {*/
/*error_log("SliceDB create failed. dbname:%s", dbname);*/
for ( int n = 0 ; n < db_id ; n++ ){
slicedb_free(vnode->slicedbs[n]);
vnode->slicedbs[n] = NULL;
}
zfree(vnode);
return NULL;
}
}
vnode->active_slicedb = vnode->slicedbs[active_slicedb_id];
/*vnode->kvdb = vnode->active_slicedb->kvdb;*/
}
vnode->caching_objects = object_queue_new(object_compare_md5_func);
vnode->received_objects = listCreate();
vnode->received_object_size = 0;
vnode->standby_objects = listCreate();
vnode->standby_object_size = 0;
vnode->write_queue = init_work_queue(vnode_write_queue_handle_write, VNODE_WRITE_QUEUE_INTERVAL);
return vnode;
}
int main(int argc, char **argv)
{
int i, maxnconn;
pthread_t tid;
struct file *fptr;
int current;
listNode *listNode;
for (i = 0; i < MAXFILES; i++) {
file[i].f_flags = 0;
}
context = connectRedis(localhost,port);
queue = listCreate();
home_page(HOST, "/");
nlefttoread = nlefttoconn = queue->len;
nconn = 0;
while (nlefttoconn > 0 || nconn > 0) {
printf("lefttoconn: %d, conn:%d\n",nlefttoconn,nconn);
current = 0;
while (nconn < MAXFILES && nlefttoconn > 0) {
for (i = current ; i < MAXFILES; i++)
if (file[i].f_flags == 0)
{
current = i;
break;
}
pthread_mutex_lock(&queue_mutex);
listNode = queue->head;
if (listNode == NULL)
{
printf("have some but no!\n");
exit(1);
}
file[i].f_flags = F_CONNECTING;
setFileNameAndHost(listNode->value,file[i].f_name,file[i].f_host);
listDelNodeHead(queue);
pthread_create(&tid, NULL, &do_get_read, &file[i]);
file[i].f_tid = tid;
nconn++;
nlefttoconn = queue->len;
pthread_mutex_unlock(&queue_mutex);
}
pthread_mutex_lock(&ndone_mutex);
while (ndone == 0)
pthread_cond_wait(&ndone_cond, &ndone_mutex);
for (i = 0; i < MAXFILES; i++) {
if (file[i].f_flags & F_DONE) {
pthread_join(file[i].f_tid, (void **) &fptr);
if (&file[i] != fptr)
{
printf("file[i]!=ptr\n");
exit(1);
}
fptr->f_flags = 0; /* clears F_DONE */
ndone--;
nconn--;
printf("thread %d for name:%s host:%s done\n", fptr->f_tid, fptr->f_name,fptr->f_host);
}
}
pthread_mutex_unlock(&ndone_mutex);
}
listRelease(queue);
disconnectRedis(context);
exit(0);
}
/* Watch for the specified key
*
* 让客户端 c 监视给定的键 key
*/
void watchForKey(redisClient *c, robj *key) {
list *clients = NULL;
listIter li;
listNode *ln;
watchedKey *wk;
/* Check if we are already watching for this key */
// 检查 key 是否已经保存在 watched_keys 链表中，
// 如果是的话，直接返回
listRewind(c->watched_keys,&li);
while((ln = listNext(&li))) {
wk = listNodeValue(ln);
if (wk->db == c->db && equalStringObjects(key,wk->key))
return; /* Key already watched */
}
// 键不存在于 watched_keys ，添加它
// 以下是一个 key 不存在于字典的例子：
// before :
// {
// 'key-1' : [c1, c2, c3],
// 'key-2' : [c1, c2],
// }
// after c-10086 WATCH key-1 and key-3:
// {
// 'key-1' : [c1, c2, c3, c-10086],
// 'key-2' : [c1, c2],
// 'key-3' : [c-10086]
// }
/* This key is not already watched in this DB. Let's add it */
// 检查 key 是否存在于数据库的 watched_keys 字典中
clients = dictFetchValue(c->db->watched_keys,key);
// 如果不存在的话，添加它
if (!clients) {
// 值为链表
clients = listCreate();
// 关联键值对到字典
dictAdd(c->db->watched_keys,key,clients);
incrRefCount(key);
}
// 将客户端添加到链表的末尾
listAddNodeTail(clients,c);
/* Add the new key to the list of keys watched by this client */
// 将新 watchedKey 结构添加到客户端 watched_keys 链表的表尾
// 以下是一个添加 watchedKey 结构的例子
// before:
// [
// {
// 'key': 'key-1',
// 'db' : 0
// }
// ]
// after client watch key-123321 in db 0:
// [
// {
// 'key': 'key-1',
// 'db' : 0
// }
// ,
// {
// 'key': 'key-123321',
// 'db': 0
// }
// ]
wk = zmalloc(sizeof(*wk));
wk->key = key;
wk->db = c->db;
incrRefCount(key);
listAddNodeTail(c->watched_keys,wk);
}
void vmInit(void) {
off totsize;
int pipefds[2];
size_t stacksize;
#ifndef _WIN32
struct flock fl;
#endif
if (server.vm_max_threads != 0)
zmalloc_enable_thread_safeness(); /* we need thread safe zmalloc() */
redisLog(REDIS_NOTICE,"Using '%s' as swap file",server.vm_swap_file);
/* Try to open the old swap file, otherwise create it */
if ((server.vm_fp = fopen(server.vm_swap_file,"r+b")) == NULL) {
server.vm_fp = fopen(server.vm_swap_file,"w+b");
}
if (server.vm_fp == NULL) {
redisLog(REDIS_WARNING,
"Can't open the swap file: %s. Exiting.",
strerror(errno));
exit(1);
}
server.vm_fd = fileno(server.vm_fp);
/* Lock the swap file for writing, this is useful in order to avoid
* another instance to use the same swap file for a config error. */
#ifdef _WIN32
// LockFile(pFile->h, RESERVED_BYTE, 0, 1, 0)
if(!LockFile((HANDLE) _get_osfhandle(server.vm_fd), (0x40000001), 0, 1, 0) ) {
redisLog(REDIS_WARNING,
"Can't lock the swap file at '%s': %s. Make sure it is not used by another Redis instance.", server.vm_swap_file, strerror(errno));
WSACleanup();
exit(1);
}
#else
fl.l_type = F_WRLCK;
fl.l_whence = SEEK_SET;
fl.l_start = fl.l_len = 0;
if (fcntl(server.vm_fd,F_SETLK,&fl) == -1) {
redisLog(REDIS_WARNING,
"Can't lock the swap file at '%s': %s. Make sure it is not used by another Redis instance.", server.vm_swap_file, strerror(errno));
exit(1);
}
#endif
/* Initialize */
server.vm_next_page = 0;
server.vm_near_pages = 0;
server.vm_stats_used_pages = 0;
server.vm_stats_swapped_objects = 0;
server.vm_stats_swapouts = 0;
server.vm_stats_swapins = 0;
totsize = server.vm_pages*server.vm_page_size;
#ifdef _WIN32
redisLog(REDIS_NOTICE,"Allocating %lld bytes of swap file",(long long) totsize);
#else
redisLog(REDIS_NOTICE,"Allocating %lld bytes of swap file",totsize);
#endif
if (ftruncate(server.vm_fd,totsize) == -1) {
redisLog(REDIS_WARNING,"Can't ftruncate swap file: %s. Exiting.",
strerror(errno));
exit(1);
} else {
redisLog(REDIS_NOTICE,"Swap file allocated with success");
}
server.vm_bitmap = zcalloc((server.vm_pages+7)/8);
redisLog(REDIS_VERBOSE,"Allocated %lld bytes page table for %lld pages",
(long long) (server.vm_pages+7)/8, server.vm_pages);
/* Initialize threaded I/O (used by Virtual Memory) */
server.io_newjobs = listCreate();
server.io_processing = listCreate();
server.io_processed = listCreate();
server.io_ready_clients = listCreate();
#ifndef _WIN32
/* moved to InitSharedObjecsts since they are first time used there */
pthread_mutex_init(&server.io_mutex,NULL);
pthread_mutex_init(&server.io_swapfile_mutex,NULL);
#endif
server.io_active_threads = 0;
if (pipe(pipefds) == -1) {
redisLog(REDIS_WARNING,"Unable to intialized VM: pipe(2): %s. Exiting."
,strerror(errno));
exit(1);
}
server.io_ready_pipe_read = pipefds[0];
server.io_ready_pipe_write = pipefds[1];
#ifndef _WIN32
/* Windows distincts pipe and socket destriptors */
/* We will use blocking pipe, with peek before blocking read */
redisAssert(anetNonBlock(NULL,server.io_ready_pipe_read) != ANET_ERR);
#endif
/* LZF requires a lot of stack */
pthread_attr_init(&server.io_threads_attr);
pthread_attr_getstacksize(&server.io_threads_attr, &stacksize);
/* Solaris may report a stacksize of 0, let's set it to 1 otherwise
* multiplying it by 2 in the while loop later will not really help ;) */
if (!stacksize) stacksize = 1;
while (stacksize < REDIS_THREAD_STACK_SIZE) stacksize *= 2;
pthread_attr_setstacksize(&server.io_threads_attr, stacksize);
/* Listen for events in the threaded I/O pipe */
#ifdef _WIN32
/* Windows fix: need to pass flag that notifies Api that file descriptor is pipe */
if (aeCreateFileEvent(server.el, server.io_ready_pipe_read, AE_READABLE | AE_PIPE,
vmThreadedIOCompletedJob, NULL) == AE_ERR)
oom("creating file event");
#else
if (aeCreateFileEvent(server.el, server.io_ready_pipe_read, AE_READABLE,
vmThreadedIOCompletedJob, NULL) == AE_ERR)
oom("creating file event");
#endif
}
Q330_CFG *q330ReadCfg(char *name)
{
int i, status, ntoken;
FILE *fp = NULL;
Q330_CFG *cfg = NULL;
QDP_TYPE_C1_QCAL DefaultCalib = Q330_DEFAULT_CALIB;
char *token[MAX_TOKEN];
char input[MAXLINELEN];
LNKLST *addr = NULL, *detector = NULL;
static char *DefaultName = Q330_DEFAULT_CFG_PATH;
/* Locate and open the file */
if (name == NULL) name = getenv(Q330_CFG_ENV_STRING);
if (name == NULL) name = DefaultName;
if (!utilFileExists(name)) return NULL;
if ((fp = fopen(name, "r")) == NULL) return ReadCfgCleanReturn(fp, cfg, addr, NULL);
/* Create the config structure and save the name */
if ((cfg = CreateCfg(name)) == NULL) return ReadCfgCleanReturn(fp, cfg, addr, NULL);
cfg->calib = DefaultCalib;
/* Initialize the temp lists */
if ((addr = listCreate()) == NULL) return ReadCfgCleanReturn(fp, cfg, addr, NULL);
if ((detector = listCreate()) == NULL) return ReadCfgCleanReturn(fp, cfg, addr, NULL);
/* Read the contents of the file into the list */
while ((status = utilGetLine(fp, input, MAXLINELEN, COMMENT, NULL)) == 0) {
ntoken = utilParse(input, token, " \t", MAX_TOKEN, QUOTE);
/* "q330" entries define Q330_ADDR structures */
if (strcasecmp(token[0], "q330") == 0) {
if (!AddAddr(addr, token, ntoken)) return ReadCfgCleanReturn(fp, cfg, addr, NULL);
/* "detector" entries define Q330_ADDR structures */
} else if (strcasecmp(token[0], "detector") == 0) {
if (!AddDetector(detector, token, ntoken)) return ReadCfgCleanReturn(fp, cfg, addr, NULL);
/* "calib" entries define QDP_TYPE_C1_QCAL structures */
} else if (strcasecmp(token[0], "calib") == 0) {
if (!AddCalib(&cfg->calib, token, ntoken)) return ReadCfgCleanReturn(fp, cfg, addr, NULL);
/* anything else is not supported */
} else {
errno = EINVAL;
return ReadCfgCleanReturn(fp, cfg, addr, NULL);
}
}
if (status != 1) return ReadCfgCleanReturn(fp, cfg, addr, NULL);
/* Copy the addresses into the output structure */
if (!listSetArrayView(addr)) return ReadCfgCleanReturn(fp, cfg, addr, NULL);
cfg->naddr = addr->count;
if ((cfg->addr = (Q330_ADDR *) malloc(sizeof(Q330_ADDR) * cfg->naddr)) == NULL) return ReadCfgCleanReturn(fp, cfg, addr, NULL);
for (i = 0; i < addr->count; i++) cfg->addr[i] = *((Q330_ADDR *) addr->array[i]);
/* Copy the detectors into the output structure */
if (!listSetArrayView(detector)) return ReadCfgCleanReturn(fp, cfg, addr, NULL);
cfg->ndetector = detector->count;
if ((cfg->detector = (Q330_DETECTOR *) malloc(sizeof(Q330_DETECTOR) * cfg->ndetector)) == NULL) return ReadCfgCleanReturn(fp, cfg, detector, NULL);
for (i = 0; i < detector->count; i++) cfg->detector[i] = *((Q330_DETECTOR *) detector->array[i]);
/* All done, return the newly created structure */
return ReadCfgCleanReturn(fp, NULL, addr, cfg);
}
/* *
* slotscheck
* */
void
slotscheckCommand(redisClient *c) {
sds bug = NULL;
int i;
for (i = 0; i < HASH_SLOTS_SIZE && bug == NULL; i ++) {
dict *d = c->db->hash_slots[i];
if (dictSize(d) == 0) {
continue;
}
list *l = listCreate();
listSetFreeMethod(l, decrRefCountVoid);
unsigned long cursor = 0;
do {
cursor = dictScan(d, cursor, slotsScanSdsKeyCallback, l);
while (1) {
listNode *head = listFirst(l);
if (head == NULL) {
break;
}
robj *key = listNodeValue(head);
if (lookupKey(c->db, key) == NULL) {
if (bug == NULL) {
bug = sdsdup(key->ptr);
}
}
listDelNode(l, head);
}
} while (cursor != 0 && bug == NULL);
listRelease(l);
}
if (bug != NULL) {
addReplyErrorFormat(c, "step 1, miss = '%s'", bug);
sdsfree(bug);
return;
}
do {
dict *d = c->db->dict;
if (dictSize(d) == 0) {
break;
}
list *l = listCreate();
listSetFreeMethod(l, decrRefCountVoid);
unsigned long cursor = 0;
do {
cursor = dictScan(d, cursor, slotsScanSdsKeyCallback, l);
while (1) {
listNode *head = listFirst(l);
if (head == NULL) {
break;
}
robj *key = listNodeValue(head);
int slot = slots_num(key->ptr, NULL);
if (dictFind(c->db->hash_slots[slot], key->ptr) == NULL) {
if (bug == NULL) {
bug = sdsdup(key->ptr);
}
}
listDelNode(l, head);
}
} while (cursor != 0 && bug == NULL);
listRelease(l);
} while (0);
if (bug != NULL) {
addReplyErrorFormat(c, "step 2, miss = '%s'", bug);
sdsfree(bug);
return;
}
addReply(c, shared.ok);
}
list *createClientlist( ) {
list *clients = listCreate();
listSetFreeMethod(clients, listDeleteClientObjects);
listSetMatchMethod(clients, listMatchClientObjects);
return clients;
}
int main(int argc, char **argv) {
int i = 0;
signal(SIGHUP, SIG_IGN);
signal(SIGPIPE, SIG_IGN);
config.el = aeCreateEventLoop();
config.debug = 0;
config.done = 0;
config.clients = listCreate();
config.num_clients = 50;
config.num_requests = 10000;
config.issued_requests = 0;
config.keepalive = 1;
config.set_perc = 50;
config.del_perc = 0;
config.swapin_perc = 0;
config.lpush_perc = 0;
config.lpop_perc = 0;
config.hset_perc = 0;
config.hget_perc = 0;
config.hgetall_perc = 0;
config.datasize_min = 1;
config.datasize_max = 64;
config.keyspace = DEFAULT_KEYSPACE; /* 100k */
config.hashkeyspace = DEFAULT_HASHKEYSPACE; /* 1k */
config.check = 0;
config.rand = 0;
config.longtail = 0;
config.quiet = 0;
config.loop = 0;
config.idlemode = 0;
config.latency = NULL;
config.latency = zmalloc(sizeof(int)*(MAX_LATENCY+1));
config.ctrlc = 0;
config.prngseed = (unsigned int) (microseconds()^getpid());
config.hostip = "127.0.0.1";
config.hostport = 6379;
parseOptions(argc,argv);
config.databuf = zmalloc(config.datasize_max);
if (config.keepalive == 0) {
printf("WARNING: keepalive disabled, you probably need 'echo 1 > /proc/sys/net/ipv4/tcp_tw_reuse' for Linux and 'sudo sysctl -w net.inet.tcp.msl=1000' for Mac OS X in order to use a lot of clients/requests\n");
}
if (config.idlemode) {
printf("Creating %d idle connections and waiting forever (Ctrl+C when done)\n", config.num_clients);
memset(config.optab,REDIS_IDLE,100);
} else {
/* Setup the operation table. Start with a table with just GET
* operations and overwrite it with others as needed. */
memset(config.optab,REDIS_GET,100);
fillOpTab(&i,REDIS_SET,config.set_perc);
fillOpTab(&i,REDIS_DEL,config.del_perc);
fillOpTab(&i,REDIS_LPUSH,config.lpush_perc);
fillOpTab(&i,REDIS_LPOP,config.lpop_perc);
fillOpTab(&i,REDIS_HSET,config.hset_perc);
fillOpTab(&i,REDIS_HGET,config.hget_perc);
fillOpTab(&i,REDIS_HGETALL,config.hgetall_perc);
fillOpTab(&i,REDIS_SWAPIN,config.swapin_perc);
}
signal(SIGINT,ctrlc);
srandom(config.prngseed);
printf("PRNG seed is: %u - use the 'seed' option to reproduce the same sequence\n", config.prngseed);
do {
prepareForBenchmark();
createMissingClients();
aeMain(config.el);
endBenchmark();
} while(config.loop);
return 0;
}
/*
* Entry point of Multi-core Insense runtime.
*/
int main(int argc, char* argv[]) {
PRINTFMC("Cache line size: %dB\n", cache_line_size());
PRINTFMC("Main thread: %u\n", (unsigned) pthread_self());
errval_t err;
coreid_t mycore = disp_get_core_id();
if (argc == 2) {
num_to_span = atoi(argv[1]);
if(num_to_span==0)
all_spanned = true;
debug_printf("Spanning onto %d cores\n", num_to_span);
for (int i = 1; i < num_to_span; i++) {
err = domain_new_dispatcher(mycore + i, span_cb, NULL);
if (err_is_fail(err)) {
DEBUG_ERR(err, "failed span %d", i);
}
}
} else {
debug_printf("ERROR: Must specify number of cores to span\n");
return EXIT_FAILURE;
}
posixcompat_pthread_set_placement_fn(rrPlacement);
while (!all_spanned) {
thread_yield();
}
my_mutex_init(&shared_heap_mutex);
#if HEAPS == HEAP_PRIVATE // Private heaps
// Initialize mutex
if (pthread_mutex_init(&thread_lock, NULL ) != 0) {
PRINTF("Mutex initialization failed.\n");
return -1;
}
#endif
mainThread = pthread_self(); // Note the ID of the main thread.
// Create a list for storing references to p-threads
threadList = listCreate();
// Create map used to store memory locations of small heaps (using Thread safe list)
SHList = listCreate();
// Create map used to store memory locations what is allocated using malloc
mallocList = listCreate();
// Start recording execution time
#if TIMING
// CPU time
uint64_t start, end;
uint64_t tsc_per_ms = 0;
sys_debug_get_tsc_per_ms(&tsc_per_ms);
start = rdtsc();
#endif
// Call primordial_main.
primordial_main(NULL );
// Join all p-threads
if (threadList != NULL ) {
listJoinThreads(threadList);
}
// Stop recording execution time
#if TIMING
end = rdtsc();
uint64_t diff = (end - start) / tsc_per_ms;
float elapsed = (diff / 1000) + ((diff % 1000) / 1000.0);
printf("CPU: %f seconds elapsed\n", elapsed);
#endif
// Destroy lists and free memory
listDestroy(threadList);
listDestroy(SHList);
listDestroy(mallocList);
#if HEAPS == HEAP_PRIVATE
pthread_mutex_destroy(&thread_lock); // Destroy mutex lock used with pthreads
#endif
return 0;
}
static void *check_thread_run(void *args)
{
int ret;
thread_data *cdata = args;
redis_group *srgroup = cdata->srgroup;
redis_group *trgroup = cdata->trgroup;
dict *nodes;
dictEntry *de;
dictIterator *di;
redis_node *rnode;
struct mbuf *mbuf;
nodes = srgroup->nodes;
di = dictGetIterator(nodes);
while ((de = dictNext(di)) != NULL) {
rnode = dictGetVal(de);
rnode->write_data = cdata;
/* remove the not used part for source redis node */
if (rnode->rdb != NULL) {
redis_rdb_deinit(rnode->rdb);
rmt_free(rnode->rdb);
rnode->rdb = NULL;
}
if (rnode->cmd_data != NULL) {
while (!mttlist_empty(rnode->cmd_data)) {
mbuf = mttlist_pop(rnode->cmd_data);
mbuf_put(mbuf);
}
mttlist_destroy(rnode->cmd_data);
rnode->cmd_data = NULL;
}
if (rnode->sockpairfds[0] > 0) {
close(rnode->sockpairfds[0]);
rnode->sockpairfds[0] = -1;
}
if (rnode->sockpairfds[1] > 0) {
close(rnode->sockpairfds[1]);
rnode->sockpairfds[1] = -1;
}
if (rnode->rr != NULL) {
redis_replication_deinit(rnode->rr);
rmt_free(rnode->rr);
rnode->rr = NULL;
}
if (rnode->piece_data != NULL) {
while ((mbuf = listPop(rnode->piece_data)) != NULL) {
mbuf_put(mbuf);
}
listRelease(rnode->piece_data);
rnode->piece_data = NULL;
}
/* add the used part for source redis node */
if (rnode->send_data == NULL) {
rnode->send_data = listCreate();
if (rnode->send_data == NULL) {
log_error("ERROR: Create msg list failed: out of memory");
return 0;
}
}
if (rnode->sent_data == NULL) {
rnode->sent_data = listCreate();
if (rnode->sent_data == NULL) {
log_error("ERROR: Create msg list failed: out of memory");
return 0;
}
}
if (rnode->sk_event < 0) {
rnode->sk_event = socket(AF_INET, SOCK_STREAM, 0);
if(rnode->sk_event < 0){
log_error("ERROR: Create sk_event for node[%s] failed: %s",
rnode->addr, strerror(errno));
return 0;
}
}
ret = aeCreateFileEvent(cdata->loop, rnode->sk_event,
AE_WRITABLE, check_begin, rnode);
if (ret != AE_OK) {
log_error("ERROR: send_data event create %ld failed: %s",
cdata->thread_id, strerror(errno));
return 0;
}
}
dictReleaseIterator(di);
nodes = trgroup->nodes;
di = dictGetIterator(nodes);
while ((de = dictNext(di)) != NULL) {
rnode = dictGetVal(de);
rnode->write_data = cdata;
}
dictReleaseIterator(di);
aeMain(cdata->loop);
return 0;
}
/* This function should be called by Redis every time a single command,
* a MULTI/EXEC block, or a Lua script, terminated its execution after
* being called by a client.
*
* All the keys with at least one client blocked that received at least
* one new element via some PUSH/XADD operation are accumulated into
* the server.ready_keys list. This function will run the list and will
* serve clients accordingly. Note that the function will iterate again and
* again as a result of serving BRPOPLPUSH we can have new blocking clients
* to serve because of the PUSH side of BRPOPLPUSH. */
void handleClientsBlockedOnKeys(void) {
while(listLength(server.ready_keys) != 0) {
list *l;
/* Point server.ready_keys to a fresh list and save the current one
* locally. This way as we run the old list we are free to call
* signalKeyAsReady() that may push new elements in server.ready_keys
* when handling clients blocked into BRPOPLPUSH. */
l = server.ready_keys;
server.ready_keys = listCreate();
while(listLength(l) != 0) {
listNode *ln = listFirst(l);
readyList *rl = ln->value;
/* First of all remove this key from db->ready_keys so that
* we can safely call signalKeyAsReady() against this key. */
dictDelete(rl->db->ready_keys,rl->key);
/* Serve clients blocked on list key. */
robj *o = lookupKeyWrite(rl->db,rl->key);
if (o != NULL && o->type == OBJ_LIST) {
dictEntry *de;
/* We serve clients in the same order they blocked for
* this key, from the first blocked to the last. */
de = dictFind(rl->db->blocking_keys,rl->key);
if (de) {
list *clients = dictGetVal(de);
int numclients = listLength(clients);
while(numclients--) {
listNode *clientnode = listFirst(clients);
client *receiver = clientnode->value;
if (receiver->btype != BLOCKED_LIST) {
/* Put on the tail, so that at the next call
* we'll not run into it again. */
listDelNode(clients,clientnode);
listAddNodeTail(clients,receiver);
continue;
}
robj *dstkey = receiver->bpop.target;
int where = (receiver->lastcmd &&
receiver->lastcmd->proc == blpopCommand) ?
LIST_HEAD : LIST_TAIL;
robj *value = listTypePop(o,where);
if (value) {
/* Protect receiver->bpop.target, that will be
* freed by the next unblockClient()
* call. */
if (dstkey) incrRefCount(dstkey);
unblockClient(receiver);
if (serveClientBlockedOnList(receiver,
rl->key,dstkey,rl->db,value,
where) == C_ERR)
{
/* If we failed serving the client we need
* to also undo the POP operation. */
listTypePush(o,value,where);
}
if (dstkey) decrRefCount(dstkey);
decrRefCount(value);
} else {
break;
}
}
}
if (listTypeLength(o) == 0) {
dbDelete(rl->db,rl->key);
notifyKeyspaceEvent(NOTIFY_GENERIC,"del",rl->key,rl->db->id);
}
/* We don't call signalModifiedKey() as it was already called
* when an element was pushed on the list. */
}
/* Serve clients blocked on stream key. */
else if (o != NULL && o->type == OBJ_STREAM) {
dictEntry *de = dictFind(rl->db->blocking_keys,rl->key);
stream *s = o->ptr;
/* We need to provide the new data arrived on the stream
* to all the clients that are waiting for an offset smaller
* than the current top item. */
if (de) {
list *clients = dictGetVal(de);
listNode *ln;
listIter li;
listRewind(clients,&li);
while((ln = listNext(&li))) {
client *receiver = listNodeValue(ln);
if (receiver->btype != BLOCKED_STREAM) continue;
streamID *gt = dictFetchValue(receiver->bpop.keys,
rl->key);
if (s->last_id.ms > gt->ms ||
(s->last_id.ms == gt->ms &&
s->last_id.seq > gt->seq))
{
streamID start = *gt;
start.seq++; /* Can't overflow, it's an uint64_t */
/* If we blocked in the context of a consumer
* group, we need to resolve the group and
* consumer here. */
streamCG *group = NULL;
streamConsumer *consumer = NULL;
if (receiver->bpop.xread_group) {
group = streamLookupCG(s,
receiver->bpop.xread_group->ptr);
/* In theory if the group is not found we
* just perform the read without the group,
* but actually when the group, or the key
* itself is deleted (triggering the removal
* of the group), we check for blocked clients
* and send them an error. */
}
if (group) {
consumer = streamLookupConsumer(group,
receiver->bpop.xread_consumer->ptr,
1);
}
/* Note that after we unblock the client, 'gt'
* and other receiver->bpop stuff are no longer
* valid, so we must do the setup above before
* this call. */
unblockClient(receiver);
/* Emit the two elements sub-array consisting of
* the name of the stream and the data we
* extracted from it. Wrapped in a single-item
* array, since we have just one key. */
addReplyMultiBulkLen(receiver,1);
addReplyMultiBulkLen(receiver,2);
addReplyBulk(receiver,rl->key);
streamPropInfo pi = {
rl->key,
receiver->bpop.xread_group
};
streamReplyWithRange(receiver,s,&start,NULL,
receiver->bpop.xread_count,
0, group, consumer, 0, &pi);
}
}
}
}
/* Free this item. */
decrRefCount(rl->key);
zfree(rl);
listDelNode(l,ln);
}
listRelease(l); /* We have the new list on place at this point. */
}
}
void
init_res (void)
{
static char fname[] = "init_res";
int i, maxfds;
if (logclass & (LC_TRACE | LC_HANG))
ls_syslog (LOG_DEBUG, "%s: Entering this routine...", fname);
if (!sbdMode)
{
if (!debug)
{
if (geteuid () || getuid ())
{
fprintf (stderr, "RES should be run as root.\n");
fflush (stderr);
resExit_ (1);
}
chdir ("/tmp");
}
if (debug <= 1)
{
daemonize_ ();
ls_openlog ("res", resParams[LSF_LOGDIR].paramValue, 0,
resParams[LSF_LOG_MASK].paramValue);
umask (0);
nice (NICE_LEAST);
}
}
if ((Myhost = ls_getmyhostname ()) == NULL)
{
ls_syslog (LOG_ERR, I18N_FUNC_FAIL_MM, fname, "ls_getmyhostname");
resExit_ (-1);
}
if (isatty (0))
{
tcgetattr (0, &defaultTty.attr);
defaultTty.ws.ws_row = 24;
defaultTty.ws.ws_col = 80;
defaultTty.ws.ws_xpixel = defaultTty.ws.ws_ypixel = 0;
}
else
{
defaultTty.ws.ws_row = 24;
defaultTty.ws.ws_col = 80;
defaultTty.ws.ws_xpixel = defaultTty.ws.ws_ypixel = 0;
}
if (!sbdMode)
{
init_AcceptSock ();
}
client_cnt = child_cnt = 0;
for (i = 0; i < MAXCLIENTS_HIGHWATER_MARK + 1; i++)
{
clients[i] = NULL;
}
children =
(struct child **) calloc (sysconf (_SC_OPEN_MAX),
sizeof (struct children *));
if (!children)
{
ls_syslog (LOG_ERR, I18N_FUNC_FAIL_M, fname, "calloc");
resExit_ (-1);
}
maxfds = sysconf (_SC_OPEN_MAX);
for (i = 0; i < maxfds; i++)
{
children[i] = NULL;
}
initConn2NIOS ();
resNotifyList = listCreate ("resNotifyList");
if (!resNotifyList)
{
ls_syslog (LOG_ERR, I18N_FUNC_FAIL_M, fname, "listCreate");
resExit_ (-1);
}
ls_syslog (LOG_INFO, (_i18n_msg_get (ls_catd, NL_SETN, 5346, "Daemon started"))); /* catgets 5346 */
initResLog ();
}
rstatus_t
init_server(struct instance *nci)
{
rstatus_t status;
uint32_t i;
redisDb *db;
vr_conf *conf;
conf = conf_create(nci->conf_filename);
server.pid = getpid();
server.configfile = getAbsolutePath(nci->conf_filename);
server.hz = 10;
server.dbnum = 16;
array_init(&server.dbs, server.dbnum, sizeof(redisDb));
server.pidfile = nci->pid_filename;
server.executable = NULL;
server.activerehashing = CONFIG_DEFAULT_ACTIVE_REHASHING;
get_random_hex_chars(server.runid, CONFIG_RUN_ID_SIZE);
server.arch_bits = (sizeof(long) == 8) ? 64 : 32;
server.requirepass = NULL;
server.starttime = time(NULL);
server.maxclients = conf->max_clients;
server.maxmemory = conf->maxmemory == CONF_UNSET_NUM ? 0 : conf->maxmemory;
server.maxmemory_policy = CONFIG_DEFAULT_MAXMEMORY_POLICY;
server.client_max_querybuf_len = PROTO_MAX_QUERYBUF_LEN;
server.commands = dictCreate(&commandTableDictType,NULL);
populateCommandTable();
server.delCommand = lookupCommandByCString("del");
server.multiCommand = lookupCommandByCString("multi");
server.lpushCommand = lookupCommandByCString("lpush");
server.lpopCommand = lookupCommandByCString("lpop");
server.rpopCommand = lookupCommandByCString("rpop");
server.sremCommand = lookupCommandByCString("srem");
server.execCommand = lookupCommandByCString("exec");
for (i = 0; i < server.dbnum; i ++) {
db = array_push(&server.dbs);
redisDbInit(db);
}
server.monitors = listCreate();
server.loading = 0;
server.lua_timedout = 0;
server.aof_state = AOF_OFF;
server.stop_writes_on_bgsave_err = 0;
server.ready_keys = listCreate();
server.slowlog = listCreate();
server.slowlog_entry_id = 0;
server.slowlog_log_slower_than = -1;
server.slowlog_max_len = CONFIG_DEFAULT_SLOWLOG_MAX_LEN;
server.stat_peak_memory = 0;
server.system_memory_size = zmalloc_get_memory_size();
server.rdb_child_pid = -1;
server.aof_child_pid = -1;
server.hash_max_ziplist_entries = OBJ_HASH_MAX_ZIPLIST_ENTRIES;
server.hash_max_ziplist_value = OBJ_HASH_MAX_ZIPLIST_VALUE;
server.list_max_ziplist_size = OBJ_LIST_MAX_ZIPLIST_SIZE;
server.list_compress_depth = OBJ_LIST_COMPRESS_DEPTH;
server.set_max_intset_entries = OBJ_SET_MAX_INTSET_ENTRIES;
server.zset_max_ziplist_entries = OBJ_ZSET_MAX_ZIPLIST_ENTRIES;
server.zset_max_ziplist_value = OBJ_ZSET_MAX_ZIPLIST_VALUE;
server.hll_sparse_max_bytes = CONFIG_DEFAULT_HLL_SPARSE_MAX_BYTES;
vr_replication_init();
createSharedObjects();
status = master_init(conf);
if (status != VR_OK) {
log_error("init master thread failed");
return VR_ERROR;
}
server.port = master.listen->port;
status = workers_init(nci->thread_num);
if (status != VR_OK) {
log_error("init worker threads failed");
return VR_ERROR;
}
log_debug(LOG_NOTICE, "mem_alloc_lock_type: %s", malloc_lock_type());
log_debug(LOG_NOTICE, "malloc lib: %s", VR_MALLOC_LIB);
return VR_OK;
}
/* This function should be called by Redis every time a single command,
* a MULTI/EXEC block, or a Lua script, terminated its execution after
* being called by a client.
*
* All the keys with at least one client blocked that received at least
* one new element via some PUSH operation are accumulated into
* the server.ready_keys list. This function will run the list and will
* serve clients accordingly. Note that the function will iterate again and
* again as a result of serving BRPOPLPUSH we can have new blocking clients
* to serve because of the PUSH side of BRPOPLPUSH. */
void handleClientsBlockedOnLists(void) {
while(listLength(server.ready_keys) != 0) {
list *l;
/* Point server.ready_keys to a fresh list and save the current one
* locally. This way as we run the old list we are free to call
* signalListAsReady() that may push new elements in server.ready_keys
* when handling clients blocked into BRPOPLPUSH. */
l = server.ready_keys;
server.ready_keys = listCreate();
while(listLength(l) != 0) {
listNode *ln = listFirst(l);
readyList *rl = ln->value;
/* First of all remove this key from db->ready_keys so that
* we can safely call signalListAsReady() against this key. */
dictDelete(rl->db->ready_keys,rl->key);
/* If the key exists and it's a list, serve blocked clients
* with data. */
robj *o = lookupKeyWrite(rl->db,rl->key);
if (o != NULL && o->type == REDIS_LIST) {
dictEntry *de;
/* We serve clients in the same order they blocked for
* this key, from the first blocked to the last. */
de = dictFind(rl->db->blocking_keys,rl->key);
if (de) {
list *clients = dictGetVal(de);
int numclients = listLength(clients);
while(numclients--) {
listNode *clientnode = listFirst(clients);
redisClient *receiver = clientnode->value;
robj *dstkey = receiver->bpop.target;
int where = (receiver->lastcmd &&
receiver->lastcmd->proc == blpopCommand) ?
REDIS_HEAD : REDIS_TAIL;
robj *value = listTypePop(o,where);
if (value) {
/* Protect receiver->bpop.target, that will be
* freed by the next unblockClientWaitingData()
* call. */
if (dstkey) incrRefCount(dstkey);
unblockClientWaitingData(receiver);
if (serveClientBlockedOnList(receiver,
rl->key,dstkey,rl->db,value,
where) == REDIS_ERR)
{
/* If we failed serving the client we need
* to also undo the POP operation. */
listTypePush(o,value,where);
}
if (dstkey) decrRefCount(dstkey);
decrRefCount(value);
} else {
break;
}
}
}
if (listTypeLength(o) == 0) dbDelete(rl->db,rl->key);
/* We don't call signalModifiedKey() as it was already called
* when an element was pushed on the list. */
}
/* Free this item. */
decrRefCount(rl->key);
zfree(rl);
listDelNode(l,ln);
}
listRelease(l); /* We have the new list on place at this point. */
}
}
static list *index_list_create() {
list *idx = listCreate();
listSetMatchMethod(idx, fts_doc_match);
listSetFreeMethod(idx, index_item_free);
return idx;
}
/* *
* do migrate mutli key-value(s) for {slotsmgrt/slotsmgrtone}with tag commands
* return value:
* -1 - error happens
* >=0 - # of success migration
* */
static int
slotsmgrttag_command(redisClient *c, sds host, sds port, int timeout, robj *key) {
int taglen;
void *tag = slots_tag(key->ptr, &taglen);
if (tag == NULL) {
return slotsmgrtone_command(c, host, port, timeout, key);
}
int fd = slotsmgrt_get_socket(c, host, port, timeout);
if (fd == -1) {
return -1;
}
list *l = listCreate();
listSetFreeMethod(l, decrRefCountVoid);
do {
uint32_t crc;
int slot = slots_num(key->ptr, &crc);
dict *d = c->db->hash_slots[slot];
long long cursor = 0;
void *args[] = {l, tag, &taglen, (void *)(long)crc};
do {
cursor = dictScan(d, cursor, slotsScanSdsKeyTagCallback, args);
} while (cursor != 0);
} while (0);
int max = listLength(l);
if (max == 0) {
listRelease(l);
return 0;
}
robj **keys = zmalloc(sizeof(robj *) * max);
robj **vals = zmalloc(sizeof(robj *) * max);
int n = 0;
for (int i = 0; i < max; i ++) {
listNode *head = listFirst(l);
robj *key = listNodeValue(head);
robj *val = lookupKeyWrite(c->db, key);
if (val != NULL) {
keys[n] = key;
vals[n] = val;
n ++;
incrRefCount(key);
}
listDelNode(l, head);
}
int ret = 0;
if (n != 0) {
if (slotsmgrt(c, host, port, fd, c->db->id, timeout, keys, vals, n) != 0) {
slotsmgrt_close_socket(host, port);
ret = -1;
} else {
slotsremove(c, keys, n, 1);
ret = n;
}
}
listRelease(l);
for (int i = 0; i < n; i ++) {
decrRefCount(keys[i]);
}
zfree(keys);
zfree(vals);
return ret;
}
//函数会在redis每次执行完单个命令，事务块或lua脚本之后被调用
//对于所有被阻塞在client的key来说，只要key被执行了PUSH，那么这个key会被加入到server.ready_keys中
//处理client的阻塞状态
void handleClientsBlockedOnLists(void) {
//只要server.ready_keys还有要解除阻塞的key，就循环遍历server.ready_keys链表
while(listLength(server.ready_keys) != 0) {
list *l;
/* Point server.ready_keys to a fresh list and save the current one
* locally. This way as we run the old list we are free to call
* signalListAsReady() that may push new elements in server.ready_keys
* when handling clients blocked into BRPOPLPUSH. */
//备份一个server.ready_keys链表
l = server.ready_keys;
//生成一个新的空链表
server.ready_keys = listCreate();
//只要链表中还有就绪的key
while(listLength(l) != 0) {
listNode *ln = listFirst(l); //链表头结点地址
readyList *rl = ln->value; //保存链表节点的值，每个值都是readyList结构
/* First of all remove this key from db->ready_keys so that
* we can safely call signalListAsReady() against this key. */
//从rl->db->ready_keys中删除就绪的key
dictDelete(rl->db->ready_keys,rl->key);
/* If the key exists and it's a list, serve blocked clients
* with data. */
//以读操作将就绪key的值读出来
robj *o = lookupKeyWrite(rl->db,rl->key);
//读出的value对象必须是列表类型
if (o != NULL && o->type == OBJ_LIST) {
dictEntry *de;
/* We serve clients in the same order they blocked for
* this key, from the first blocked to the last. */
// blocking_keys是一个字典，字典的键是造成client阻塞的键，字典的值是链表，保存被阻塞的client
// 根据key取出被阻塞的client
de = dictFind(rl->db->blocking_keys,rl->key);
// 链表非空
if (de) {
// 获取de节点的值
list *clients = dictGetVal(de);
// 获取链表的长度
int numclients = listLength(clients);
//遍历链表的所有节点
while(numclients--) {
// 第一个client节点地址
listNode *clientnode = listFirst(clients);
// 取出节点的值，是一个client类型
client *receiver = clientnode->value;
// 从client类型中的target获得要PUSH出的dstkey，该键保存在target中
robj *dstkey = receiver->bpop.target;
// 获取弹出的位置，根据BRPOPLPUSH命令
int where = (receiver->lastcmd &&
receiver->lastcmd->proc == blpopCommand) ?
LIST_HEAD : LIST_TAIL;
// 从列表中弹出元素
robj *value = listTypePop(o,where);
// 弹出成功
if (value) {
/* Protect receiver->bpop.target, that will be
* freed by the next unblockClient()
* call. */
//增加dstkey的引用计数，保护该键，在unblockClient函数中释放
if (dstkey) incrRefCount(dstkey);
//取消client的阻塞状态
unblockClient(receiver);
// 将value推入造成client阻塞的键上，
if (serveClientBlockedOnList(receiver,
rl->key,dstkey,rl->db,value,
where) == C_ERR)
{
/* If we failed serving the client we need
* to also undo the POP operation. */
// 如果推入失败，则需要键弹出的value还原回去
listTypePush(o,value,where);
}
// 释放dstkey和value
if (dstkey) decrRefCount(dstkey);
decrRefCount(value);
} else {
break;
}
}
}
// 如果弹出了所有元素，将key从数据库中删除
if (listTypeLength(o) == 0) {
dbDelete(rl->db,rl->key);
}
/* We don't call signalModifiedKey() as it was already called
* when an element was pushed on the list. */
}
/* Free this item. */
//释放所有空间
decrRefCount(rl->key);
zfree(rl);
listDelNode(l,ln);
}
//释放原来的ready_keys，因为之前创建了新的链表
listRelease(l); /* We have the new list on place at this point. */
}
}
/* This function makes the clinet 'c' waiting for the key 'key' to be loaded.
* If there is not already a job loading the key, it is craeted.
* The key is added to the io_keys list in the client structure, and also
* in the hash table mapping swapped keys to waiting clients, that is,
* server.io_waited_keys. */
int waitForSwappedKey(redisClient *c, robj *key) {
struct dictEntry *de;
robj *o;
list *l;
/* If the key does not exist or is already in RAM we don't need to
* block the client at all. */
de = dictFind(c->db->dict,key->ptr);
if (de == NULL) return 0;
o = dictGetEntryVal(de);
if (o->storage == REDIS_VM_MEMORY) {
return 0;
} else if (o->storage == REDIS_VM_SWAPPING) {
/* We were swapping the key, undo it! */
vmCancelThreadedIOJob(o);
return 0;
}
/* OK: the key is either swapped, or being loaded just now. */
/* Add the key to the list of keys this client is waiting for.
* This maps clients to keys they are waiting for. */
listAddNodeTail(c->io_keys,key);
incrRefCount(key);
/* Add the client to the swapped keys => clients waiting map. */
de = dictFind(c->db->io_keys,key);
if (de == NULL) {
int retval;
/* For every key we take a list of clients blocked for it */
l = listCreate();
retval = dictAdd(c->db->io_keys,key,l);
incrRefCount(key);
redisAssert(retval == DICT_OK);
} else {
l = dictGetEntryVal(de);
}
listAddNodeTail(l,c);
/* Are we already loading the key from disk? If not create a job */
if (o->storage == REDIS_VM_SWAPPED) {
iojob *j;
vmpointer *vp = (vmpointer*)o;
o->storage = REDIS_VM_LOADING;
j = zmalloc(sizeof(*j));
j->type = REDIS_IOJOB_LOAD;
j->db = c->db;
j->id = (robj*)vp;
j->key = key;
incrRefCount(key);
j->page = vp->page;
j->val = NULL;
j->canceled = 0;
j->thread = (pthread_t) -1;
lockThreadedIO();
queueIOJob(j);
unlockThreadedIO();
}
return 1;
}
void vmInit(void) {
off_t totsize;
int pipefds[2];
size_t stacksize;
struct flock fl;
if (server.vm_max_threads != 0)
zmalloc_enable_thread_safeness(); /* we need thread safe zmalloc() */
redisLog(REDIS_NOTICE,"Using '%s' as swap file",server.vm_swap_file);
/* Try to open the old swap file, otherwise create it */
if ((server.vm_fp = fopen(server.vm_swap_file,"r+b")) == NULL) {
server.vm_fp = fopen(server.vm_swap_file,"w+b");
}
if (server.vm_fp == NULL) {
redisLog(REDIS_WARNING,
"Can't open the swap file: %s. Exiting.",
strerror(errno));
exit(1);
}
server.vm_fd = fileno(server.vm_fp);
/* Lock the swap file for writing, this is useful in order to avoid
* another instance to use the same swap file for a config error. */
fl.l_type = F_WRLCK;
fl.l_whence = SEEK_SET;
fl.l_start = fl.l_len = 0;
if (fcntl(server.vm_fd,F_SETLK,&fl) == -1) {
redisLog(REDIS_WARNING,
"Can't lock the swap file at '%s': %s. Make sure it is not used by another Redis instance.", server.vm_swap_file, strerror(errno));
exit(1);
}
/* Initialize */
server.vm_next_page = 0;
server.vm_near_pages = 0;
server.vm_stats_used_pages = 0;
server.vm_stats_swapped_objects = 0;
server.vm_stats_swapouts = 0;
server.vm_stats_swapins = 0;
totsize = server.vm_pages*server.vm_page_size;
redisLog(REDIS_NOTICE,"Allocating %lld bytes of swap file",totsize);
if (ftruncate(server.vm_fd,totsize) == -1) {
redisLog(REDIS_WARNING,"Can't ftruncate swap file: %s. Exiting.",
strerror(errno));
exit(1);
} else {
redisLog(REDIS_NOTICE,"Swap file allocated with success");
}
server.vm_bitmap = zcalloc((server.vm_pages+7)/8);
redisLog(REDIS_VERBOSE,"Allocated %lld bytes page table for %lld pages",
(long long) (server.vm_pages+7)/8, server.vm_pages);
/* Initialize threaded I/O (used by Virtual Memory) */
server.io_newjobs = listCreate();
server.io_processing = listCreate();
server.io_processed = listCreate();
server.io_ready_clients = listCreate();
pthread_mutex_init(&server.io_mutex,NULL);
pthread_mutex_init(&server.io_swapfile_mutex,NULL);
server.io_active_threads = 0;
if (pipe(pipefds) == -1) {
redisLog(REDIS_WARNING,"Unable to intialized VM: pipe(2): %s. Exiting."
,strerror(errno));
exit(1);
}
server.io_ready_pipe_read = pipefds[0];
server.io_ready_pipe_write = pipefds[1];
redisAssert(anetNonBlock(NULL,server.io_ready_pipe_read) != ANET_ERR);
/* LZF requires a lot of stack */
pthread_attr_init(&server.io_threads_attr);
pthread_attr_getstacksize(&server.io_threads_attr, &stacksize);
/* Solaris may report a stacksize of 0, let's set it to 1 otherwise
* multiplying it by 2 in the while loop later will not really help ;) */
if (!stacksize) stacksize = 1;
while (stacksize < REDIS_THREAD_STACK_SIZE) stacksize *= 2;
pthread_attr_setstacksize(&server.io_threads_attr, stacksize);
/* Listen for events in the threaded I/O pipe */
if (aeCreateFileEvent(server.el, server.io_ready_pipe_read, AE_READABLE,
vmThreadedIOCompletedJob, NULL) == AE_ERR)
oom("creating file event");
}
int main(int argc, const char **argv) {
int i;
char *data, *cmd;
int len;
client c;
#ifdef _WIN32
w32initWinSock();
#endif
signal(SIGHUP, SIG_IGN);
signal(SIGPIPE, SIG_IGN);
config.numclients = 50;
config.requests = 10000;
config.liveclients = 0;
config.el = aeCreateEventLoop();
aeCreateTimeEvent(config.el,1,showThroughput,NULL,NULL);
config.keepalive = 1;
config.datasize = 3;
config.randomkeys = 0;
config.randomkeys_keyspacelen = 0;
config.quiet = 0;
config.loop = 0;
config.idlemode = 0;
config.latency = NULL;
config.clients = listCreate();
config.hostip = "127.0.0.1";
config.hostport = 6379;
config.hostsocket = NULL;
i = parseOptions(argc,argv);
argc -= i;
argv += i;
config.latency = zmalloc(sizeof(long long)*config.requests);
if (config.keepalive == 0) {
printf("WARNING: keepalive disabled, you probably need 'echo 1 > /proc/sys/net/ipv4/tcp_tw_reuse' for Linux and 'sudo sysctl -w net.inet.tcp.msl=1000' for Mac OS X in order to use a lot of clients/requests\n");
}
if (config.idlemode) {
printf("Creating %d idle connections and waiting forever (Ctrl+C when done)\n", config.numclients);
c = createClient("",0); /* will never receive a reply */
createMissingClients(c);
aeMain(config.el);
/* and will wait for every */
}
/* Run benchmark with command in the remainder of the arguments. */
if (argc) {
sds title = sdsnew(argv[0]);
for (i = 1; i < argc; i++) {
title = sdscatlen(title, " ", 1);
title = sdscatlen(title, (char*)argv[i], strlen(argv[i]));
}
do {
len = redisFormatCommandArgv(&cmd,argc,argv,NULL);
benchmark(title,cmd,len);
free(cmd);
} while(config.loop);
return 0;
}
/* Run default benchmark suite. */
do {
data = zmalloc(config.datasize+1);
memset(data,'x',config.datasize);
data[config.datasize] = '\0';
benchmark("PING (inline)","PING\r\n",6);
len = redisFormatCommand(&cmd,"PING");
benchmark("PING",cmd,len);
free(cmd);
const char *argv[21];
argv[0] = "MSET";
for (i = 1; i < 21; i += 2) {
argv[i] = "foo:rand:000000000000";
argv[i+1] = data;
}
len = redisFormatCommandArgv(&cmd,21,argv,NULL);
benchmark("MSET (10 keys)",cmd,len);
free(cmd);
len = redisFormatCommand(&cmd,"SET foo:rand:000000000000 %s",data);
benchmark("SET",cmd,len);
free(cmd);
len = redisFormatCommand(&cmd,"GET foo:rand:000000000000");
benchmark("GET",cmd,len);
free(cmd);
len = redisFormatCommand(&cmd,"INCR counter:rand:000000000000");
benchmark("INCR",cmd,len);
free(cmd);
len = redisFormatCommand(&cmd,"LPUSH mylist %s",data);
benchmark("LPUSH",cmd,len);
free(cmd);
len = redisFormatCommand(&cmd,"LPOP mylist");
benchmark("LPOP",cmd,len);
free(cmd);
len = redisFormatCommand(&cmd,"SADD myset counter:rand:000000000000");
benchmark("SADD",cmd,len);
free(cmd);
len = redisFormatCommand(&cmd,"SPOP myset");
benchmark("SPOP",cmd,len);
free(cmd);
len = redisFormatCommand(&cmd,"LPUSH mylist %s",data);
benchmark("LPUSH (again, in order to bench LRANGE)",cmd,len);
free(cmd);
len = redisFormatCommand(&cmd,"LRANGE mylist 0 99");
benchmark("LRANGE (first 100 elements)",cmd,len);
free(cmd);
len = redisFormatCommand(&cmd,"LRANGE mylist 0 299");
benchmark("LRANGE (first 300 elements)",cmd,len);
free(cmd);
len = redisFormatCommand(&cmd,"LRANGE mylist 0 449");
benchmark("LRANGE (first 450 elements)",cmd,len);
free(cmd);
len = redisFormatCommand(&cmd,"LRANGE mylist 0 599");
benchmark("LRANGE (first 600 elements)",cmd,len);
free(cmd);
printf("\n");
} while(config.loop);
#ifdef _WIN32
WSACleanup();
#endif
return 0;
}
/* The SORT command is the most complex command in Redis. Warning: this code
* is optimized for speed and a bit less for readability */
void sortCommand(redisClient *c) {
list *operations;
unsigned int outputlen = 0;
int desc = 0, alpha = 0;
int limit_start = 0, limit_count = -1, start, end;
int j, dontsort = 0, vectorlen;
int getop = 0; /* GET operation counter */
robj *sortval, *sortby = NULL, *storekey = NULL;
redisSortObject *vector; /* Resulting vector to sort */
/* Lookup the key to sort. It must be of the right types */
sortval = lookupKeyRead(c->db,c->argv[1]);
if (sortval == NULL) {
addReply(c,shared.emptymultibulk);
return;
}
if (sortval->type != REDIS_SET && sortval->type != REDIS_LIST &&
sortval->type != REDIS_ZSET)
{
addReply(c,shared.wrongtypeerr);
return;
}
/* Create a list of operations to perform for every sorted element.
* Operations can be GET/DEL/INCR/DECR */
operations = listCreate();
listSetFreeMethod(operations,zfree);
j = 2;
/* Now we need to protect sortval incrementing its count, in the future
* SORT may have options able to overwrite/delete keys during the sorting
* and the sorted key itself may get destroied */
incrRefCount(sortval);
/* The SORT command has an SQL-alike syntax, parse it */
while(j < c->argc) {
int leftargs = c->argc-j-1;
if (!strcasecmp(c->argv[j]->ptr,"asc")) {
desc = 0;
} else if (!strcasecmp(c->argv[j]->ptr,"desc")) {
desc = 1;
} else if (!strcasecmp(c->argv[j]->ptr,"alpha")) {
alpha = 1;
} else if (!strcasecmp(c->argv[j]->ptr,"limit") && leftargs >= 2) {
limit_start = atoi(c->argv[j+1]->ptr);
limit_count = atoi(c->argv[j+2]->ptr);
j+=2;
} else if (!strcasecmp(c->argv[j]->ptr,"store") && leftargs >= 1) {
storekey = c->argv[j+1];
j++;
} else if (!strcasecmp(c->argv[j]->ptr,"by") && leftargs >= 1) {
sortby = c->argv[j+1];
/* If the BY pattern does not contain '*', i.e. it is constant,
* we don't need to sort nor to lookup the weight keys. */
if (strchr(c->argv[j+1]->ptr,'*') == NULL) dontsort = 1;
j++;
} else if (!strcasecmp(c->argv[j]->ptr,"get") && leftargs >= 1) {
listAddNodeTail(operations,createSortOperation(
REDIS_SORT_GET,c->argv[j+1]));
getop++;
j++;
} else {
decrRefCount(sortval);
listRelease(operations);
addReply(c,shared.syntaxerr);
return;
}
j++;
}
/* Destructively convert encoded sorted sets for SORT. */
if (sortval->type == REDIS_ZSET) zsetConvert(sortval, REDIS_ENCODING_SKIPLIST);
/* Load the sorting vector with all the objects to sort */
switch(sortval->type) {
case REDIS_LIST: vectorlen = listTypeLength(sortval); break;
case REDIS_SET: vectorlen = setTypeSize(sortval); break;
case REDIS_ZSET: vectorlen = dictSize(((zset*)sortval->ptr)->dict); break;
default: vectorlen = 0; redisPanic("Bad SORT type"); /* Avoid GCC warning */
}
vector = zmalloc(sizeof(redisSortObject)*vectorlen);
j = 0;
if (sortval->type == REDIS_LIST) {
listTypeIterator *li = listTypeInitIterator(sortval,0,REDIS_TAIL);
listTypeEntry entry;
while(listTypeNext(li,&entry)) {
vector[j].obj = listTypeGet(&entry);
vector[j].u.score = 0;
vector[j].u.cmpobj = NULL;
j++;
}
listTypeReleaseIterator(li);
} else if (sortval->type == REDIS_SET) {
setTypeIterator *si = setTypeInitIterator(sortval);
robj *ele;
while((ele = setTypeNextObject(si)) != NULL) {
vector[j].obj = ele;
vector[j].u.score = 0;
vector[j].u.cmpobj = NULL;
j++;
}
setTypeReleaseIterator(si);
} else if (sortval->type == REDIS_ZSET) {
dict *set = ((zset*)sortval->ptr)->dict;
dictIterator *di;
dictEntry *setele;
di = dictGetIterator(set);
while((setele = dictNext(di)) != NULL) {
vector[j].obj = dictGetEntryKey(setele);
vector[j].u.score = 0;
vector[j].u.cmpobj = NULL;
j++;
}
dictReleaseIterator(di);
} else {
redisPanic("Unknown type");
}
redisAssert(j == vectorlen);
/* Now it's time to load the right scores in the sorting vector */
if (dontsort == 0) {
for (j = 0; j < vectorlen; j++) {
robj *byval;
if (sortby) {
/* lookup value to sort by */
byval = lookupKeyByPattern(c->db,sortby,vector[j].obj);
if (!byval) continue;
} else {
/* use object itself to sort by */
byval = vector[j].obj;
}
if (alpha) {
if (sortby) vector[j].u.cmpobj = getDecodedObject(byval);
} else {
if (byval->encoding == REDIS_ENCODING_RAW) {
vector[j].u.score = strtod(byval->ptr,NULL);
} else if (byval->encoding == REDIS_ENCODING_INT) {
/* Don't need to decode the object if it's
* integer-encoded (the only encoding supported) so
* far. We can just cast it */
vector[j].u.score = (long)byval->ptr;
} else {
redisAssert(1 != 1);
}
}
/* when the object was retrieved using lookupKeyByPattern,
* its refcount needs to be decreased. */
if (sortby) {
decrRefCount(byval);
}
}
}
/* We are ready to sort the vector... perform a bit of sanity check
* on the LIMIT option too. We'll use a partial version of quicksort. */
start = (limit_start < 0) ? 0 : limit_start;
end = (limit_count < 0) ? vectorlen-1 : start+limit_count-1;
if (start >= vectorlen) {
start = vectorlen-1;
end = vectorlen-2;
}
if (end >= vectorlen) end = vectorlen-1;
if (dontsort == 0) {
server.sort_desc = desc;
server.sort_alpha = alpha;
server.sort_bypattern = sortby ? 1 : 0;
if (sortby && (start != 0 || end != vectorlen-1))
pqsort(vector,vectorlen,sizeof(redisSortObject),sortCompare, start,end);
else
qsort(vector,vectorlen,sizeof(redisSortObject),sortCompare);
}
/* Send command output to the output buffer, performing the specified
* GET/DEL/INCR/DECR operations if any. */
outputlen = getop ? getop*(end-start+1) : end-start+1;
if (storekey == NULL) {
/* STORE option not specified, sent the sorting result to client */
addReplyMultiBulkLen(c,outputlen);
for (j = start; j <= end; j++) {
listNode *ln;
listIter li;
if (!getop) addReplyBulk(c,vector[j].obj);
listRewind(operations,&li);
while((ln = listNext(&li))) {
redisSortOperation *sop = ln->value;
robj *val = lookupKeyByPattern(c->db,sop->pattern,
vector[j].obj);
if (sop->type == REDIS_SORT_GET) {
if (!val) {
addReply(c,shared.nullbulk);
} else {
addReplyBulk(c,val);
decrRefCount(val);
}
} else {
redisAssert(sop->type == REDIS_SORT_GET); /* always fails */
}
}
}
} else {
robj *sobj = createZiplistObject();
/* STORE option specified, set the sorting result as a List object */
for (j = start; j <= end; j++) {
listNode *ln;
listIter li;
if (!getop) {
listTypePush(sobj,vector[j].obj,REDIS_TAIL);
} else {
listRewind(operations,&li);
while((ln = listNext(&li))) {
redisSortOperation *sop = ln->value;
robj *val = lookupKeyByPattern(c->db,sop->pattern,
vector[j].obj);
if (sop->type == REDIS_SORT_GET) {
if (!val) val = createStringObject("",0);
/* listTypePush does an incrRefCount, so we should take care
* care of the incremented refcount caused by either
* lookupKeyByPattern or createStringObject("",0) */
listTypePush(sobj,val,REDIS_TAIL);
decrRefCount(val);
} else {
/* always fails */
redisAssert(sop->type == REDIS_SORT_GET);
}
}
}
}
setKey(c->db,storekey,sobj);
decrRefCount(sobj);
/* Note: we add 1 because the DB is dirty anyway since even if the
* SORT result is empty a new key is set and maybe the old content
* replaced. */
server.dirty += 1+outputlen;
addReplyLongLong(c,outputlen);
}
/* Cleanup */
if (sortval->type == REDIS_LIST || sortval->type == REDIS_SET)
for (j = 0; j < vectorlen; j++)
decrRefCount(vector[j].obj);
decrRefCount(sortval);
listRelease(operations);
for (j = 0; j < vectorlen; j++) {
if (alpha && vector[j].u.cmpobj)
decrRefCount(vector[j].u.cmpobj);
}
zfree(vector);
}
/* This command implements SCAN, HSCAN and SSCAN commands.
* If object 'o' is passed, then it must be a Hash or Set object, otherwise
* if 'o' is NULL the command will operate on the dictionary associated with
* the current database.
*
* When 'o' is not NULL the function assumes that the first argument in
* the client arguments vector is a key so it skips it before iterating
* in order to parse options.
*
* In the case of a Hash object the function returns both the field and value
* of every element on the Hash. */
void scanGenericCommand(client *c, robj *o, unsigned long cursor) {
int i, j;
list *keys = listCreate();
listNode *node, *nextnode;
long count = 10;
sds pat = NULL;
int patlen = 0, use_pattern = 0;
dict *ht;
/* Object must be NULL (to iterate keys names), or the type of the object
* must be Set, Sorted Set, or Hash. */
serverAssert(o == NULL || o->type == OBJ_SET || o->type == OBJ_HASH ||
o->type == OBJ_ZSET);
/* Set i to the first option argument. The previous one is the cursor. */
i = (o == NULL) ? 2 : 3; /* Skip the key argument if needed. */
/* Step 1: Parse options. */
while (i < c->argc) {
j = c->argc - i;
if (!strcasecmp(c->argv[i]->ptr, "count") && j >= 2) {
if (getLongFromObjectOrReply(c, c->argv[i+1], &count, NULL)
!= C_OK)
{
goto cleanup;
}
if (count < 1) {
addReply(c,shared.syntaxerr);
goto cleanup;
}
i += 2;
} else if (!strcasecmp(c->argv[i]->ptr, "match") && j >= 2) {
pat = c->argv[i+1]->ptr;
patlen = sdslen(pat);
/* The pattern always matches if it is exactly "*", so it is
* equivalent to disabling it. */
use_pattern = !(pat[0] == '*' && patlen == 1);
i += 2;
} else {
addReply(c,shared.syntaxerr);
goto cleanup;
}
}
/* Step 2: Iterate the collection.
*
* Note that if the object is encoded with a ziplist, intset, or any other
* representation that is not a hash table, we are sure that it is also
* composed of a small number of elements. So to avoid taking state we
* just return everything inside the object in a single call, setting the
* cursor to zero to signal the end of the iteration. */
/* Handle the case of a hash table. */
ht = NULL;
if (o == NULL) {
ht = c->db->dict;
} else if (o->type == OBJ_SET && o->encoding == OBJ_ENCODING_HT) {
ht = o->ptr;
} else if (o->type == OBJ_HASH && o->encoding == OBJ_ENCODING_HT) {
ht = o->ptr;
count *= 2; /* We return key / value for this type. */
} else if (o->type == OBJ_ZSET && o->encoding == OBJ_ENCODING_SKIPLIST) {
zset *zs = o->ptr;
ht = zs->dict;
count *= 2; /* We return key / value for this type. */
}
if (ht) {
void *privdata[2];
/* We set the max number of iterations to ten times the specified
* COUNT, so if the hash table is in a pathological state (very
* sparsely populated) we avoid to block too much time at the cost
* of returning no or very few elements. */
long maxiterations = count*10;
/* We pass two pointers to the callback: the list to which it will
* add new elements, and the object containing the dictionary so that
* it is possible to fetch more data in a type-dependent way. */
privdata[0] = keys;
privdata[1] = o;
do {
cursor = dictScan(ht, cursor, scanCallback, privdata);
} while (cursor &&
maxiterations-- &&
listLength(keys) < (unsigned long)count);
} else if (o->type == OBJ_SET) {
int pos = 0;
int64_t ll;
while(intsetGet(o->ptr,pos++,&ll))
listAddNodeTail(keys,createStringObjectFromLongLong(ll));
cursor = 0;
} else if (o->type == OBJ_HASH || o->type == OBJ_ZSET) {
unsigned char *p = ziplistIndex(o->ptr,0);
unsigned char *vstr;
unsigned int vlen;
long long vll;
while(p) {
ziplistGet(p,&vstr,&vlen,&vll);
listAddNodeTail(keys,
(vstr != NULL) ? createStringObject((char*)vstr,vlen) :
createStringObjectFromLongLong(vll));
p = ziplistNext(o->ptr,p);
}
cursor = 0;
} else {
serverPanic("Not handled encoding in SCAN.");
}
/* Step 3: Filter elements. */
node = listFirst(keys);
while (node) {
robj *kobj = listNodeValue(node);
nextnode = listNextNode(node);
int filter = 0;
/* Filter element if it does not match the pattern. */
if (!filter && use_pattern) {
if (sdsEncodedObject(kobj)) {
if (!stringmatchlen(pat, patlen, kobj->ptr, sdslen(kobj->ptr), 0))
filter = 1;
} else {
char buf[LONG_STR_SIZE];
int len;
serverAssert(kobj->encoding == OBJ_ENCODING_INT);
len = ll2string(buf,sizeof(buf),(long)kobj->ptr);
if (!stringmatchlen(pat, patlen, buf, len, 0)) filter = 1;
}
}
/* Filter element if it is an expired key. */
if (!filter && o == NULL && expireIfNeeded(c->db, kobj)) filter = 1;
/* Remove the element and its associted value if needed. */
if (filter) {
decrRefCount(kobj);
listDelNode(keys, node);
}
/* If this is a hash or a sorted set, we have a flat list of
* key-value elements, so if this element was filtered, remove the
* value, or skip it if it was not filtered: we only match keys. */
if (o && (o->type == OBJ_ZSET || o->type == OBJ_HASH)) {
node = nextnode;
nextnode = listNextNode(node);
if (filter) {
kobj = listNodeValue(node);
decrRefCount(kobj);
listDelNode(keys, node);
}
}
node = nextnode;
}
/* Step 4: Reply to the client. */
addReplyMultiBulkLen(c, 2);
addReplyBulkLongLong(c,cursor);
addReplyMultiBulkLen(c, listLength(keys));
while ((node = listFirst(keys)) != NULL) {
robj *kobj = listNodeValue(node);
addReplyBulk(c, kobj);
decrRefCount(kobj);
listDelNode(keys, node);
}
cleanup:
listSetFreeMethod(keys,decrRefCountVoid);
listRelease(keys);
}
/* Initialize the slow log. This function should be called a single time
* at server startup. */
void slowlogInit(void) {
server.slowlog = listCreate();
server.slowlog_entry_id = 0;
listSetFreeMethod(server.slowlog,slowlogFreeEntry);
}
int main(int argc, const char **argv) {
int i;
char *data, *cmd;
int len;
client c;
srandom(time(NULL));
signal(SIGHUP, SIG_IGN);
signal(SIGPIPE, SIG_IGN);
config.numclients = 50;
config.requests = 100000;
config.liveclients = 0;
config.el = aeCreateEventLoop(1024*10);
aeCreateTimeEvent(config.el,1,showThroughput,NULL,NULL);
config.keepalive = 1;
config.datasize = 3;
config.pipeline = 1;
config.randomkeys = 0;
config.randomkeys_keyspacelen = 0;
config.quiet = 0;
config.csv = 0;
config.loop = 0;
config.idlemode = 0;
config.latency = NULL;
config.clients = listCreate();
config.hostip = "127.0.0.1";
config.hostport = 6379;
config.hostsocket = NULL;
config.tests = NULL;
config.dbnum = 0;
config.auth = NULL;
i = parseOptions(argc,argv);
argc -= i;
argv += i;
config.latency = zmalloc(sizeof(long long)*config.requests);
if (config.keepalive == 0) {
printf("WARNING: keepalive disabled, you probably need 'echo 1 > /proc/sys/net/ipv4/tcp_tw_reuse' for Linux and 'sudo sysctl -w net.inet.tcp.msl=1000' for Mac OS X in order to use a lot of clients/requests\n");
}
if (config.idlemode) {
printf("Creating %d idle connections and waiting forever (Ctrl+C when done)\n", config.numclients);
c = createClient("",0,NULL); /* will never receive a reply */
createMissingClients(c);
aeMain(config.el);
/* and will wait for every */
}
/* Run benchmark with command in the remainder of the arguments. */
if (argc) {
sds title = sdsnew(argv[0]);
for (i = 1; i < argc; i++) {
title = sdscatlen(title, " ", 1);
title = sdscatlen(title, (char*)argv[i], strlen(argv[i]));
}
do {
len = redisFormatCommandArgv(&cmd,argc,argv,NULL);
benchmark(title,cmd,len);
free(cmd);
} while(config.loop);
return 0;
}
/* Run default benchmark suite. */
data = zmalloc(config.datasize+1);
do {
memset(data,'x',config.datasize);
data[config.datasize] = '\0';
if (test_is_selected("ping_inline") || test_is_selected("ping"))
benchmark("PING_INLINE","PING\r\n",6);
if (test_is_selected("ping_mbulk") || test_is_selected("ping")) {
len = redisFormatCommand(&cmd,"PING");
benchmark("PING_BULK",cmd,len);
free(cmd);
}
if (test_is_selected("set")) {
len = redisFormatCommand(&cmd,"SET key:__rand_int__ %s",data);
benchmark("SET",cmd,len);
free(cmd);
}
if (test_is_selected("get")) {
len = redisFormatCommand(&cmd,"GET key:__rand_int__");
benchmark("GET",cmd,len);
free(cmd);
}
if (test_is_selected("incr")) {
len = redisFormatCommand(&cmd,"INCR counter:__rand_int__");
benchmark("INCR",cmd,len);
free(cmd);
}
if (test_is_selected("lpush")) {
len = redisFormatCommand(&cmd,"LPUSH mylist %s",data);
benchmark("LPUSH",cmd,len);
free(cmd);
}
if (test_is_selected("lpop")) {
len = redisFormatCommand(&cmd,"LPOP mylist");
benchmark("LPOP",cmd,len);
free(cmd);
}
if (test_is_selected("sadd")) {
len = redisFormatCommand(&cmd,
"SADD myset element:__rand_int__");
benchmark("SADD",cmd,len);
free(cmd);
}
if (test_is_selected("spop")) {
len = redisFormatCommand(&cmd,"SPOP myset");
benchmark("SPOP",cmd,len);
free(cmd);
}
if (test_is_selected("lrange") ||
test_is_selected("lrange_100") ||
test_is_selected("lrange_300") ||
test_is_selected("lrange_500") ||
test_is_selected("lrange_600"))
{
len = redisFormatCommand(&cmd,"LPUSH mylist %s",data);
benchmark("LPUSH (needed to benchmark LRANGE)",cmd,len);
free(cmd);
}
if (test_is_selected("lrange") || test_is_selected("lrange_100")) {
len = redisFormatCommand(&cmd,"LRANGE mylist 0 99");
benchmark("LRANGE_100 (first 100 elements)",cmd,len);
free(cmd);
}
if (test_is_selected("lrange") || test_is_selected("lrange_300")) {
len = redisFormatCommand(&cmd,"LRANGE mylist 0 299");
benchmark("LRANGE_300 (first 300 elements)",cmd,len);
free(cmd);
}
if (test_is_selected("lrange") || test_is_selected("lrange_500")) {
len = redisFormatCommand(&cmd,"LRANGE mylist 0 449");
benchmark("LRANGE_500 (first 450 elements)",cmd,len);
free(cmd);
}
if (test_is_selected("lrange") || test_is_selected("lrange_600")) {
len = redisFormatCommand(&cmd,"LRANGE mylist 0 599");
benchmark("LRANGE_600 (first 600 elements)",cmd,len);
free(cmd);
}
if (test_is_selected("mset")) {
const char *argv[21];
argv[0] = "MSET";
for (i = 1; i < 21; i += 2) {
argv[i] = "key:__rand_int__";
argv[i+1] = data;
}
len = redisFormatCommandArgv(&cmd,21,argv,NULL);
benchmark("MSET (10 keys)",cmd,len);
free(cmd);
}
if (!config.csv) printf("\n");
} while(config.loop);
return 0;
}
int main(void)
{
const int N = 10;
int i, isFound, action, pos, arg;
List list;
Iterator it1, it2;
listCreate(&list, N);
do
{
printf("Меню:\n");
printf("1) Вставить элемент\n");
printf("2) Удалить элемент\n");
printf("3) Печать списка\n");
printf("4) Подсчет длины списка\n");
printf("5) Выполнить задание над списком\n");
printf("6) Выход\n");
printf("Выберите действие: ");
scanf("%d", &action);
switch (action)
{
case 1:
{
printf("Введите позицию элемента: ");
scanf("%d", &pos);
printf("Введите значение элемента (1 - true, 0 - false): ");
scanf("%d", &arg);
if (arg != 0 && arg != 1)
printf("Ошибка. Введено недопустимое значение\n");
else
listInsert(&list, pos - 1, arg);
break;
}
case 2:
{
printf("Введите номер элемента: ");
scanf("%d", &pos);
listRemove(&list, pos - 1);
break;
}
case 3:
{
if (listEmpty(&list))
printf("Список пуст\n");
else
listPrint(&list);
break;
}
case 4:
{
printf("Длина списка: %d\n", listSize(&list));
break;
}
case 5:
{
it1 = itFirst(&list);
it2 = it1;
itNext(&it2);
for (i = 0; i < listSize(&list) / 2; i++)
{
listSwapElems(&it1, &it2);
itNext(&it1);
itNext(&it1);
itNext(&it2);
itNext(&it2);
}
break;
}
case 6: break;
default:
{
printf("Ошибка. Такого пункта меню не существует\n");
break;
}
}
}
while (action != 6);
listDestroy(&list);
return 0;
}
