stat_fork_time taken at wrong place

src/ae.cpp

@@ -716,11 +716,6 @@ int aeProcessEvents(aeEventLoop *eventLoop, int flags)
             }
         }
 
-        if (eventLoop->flags & AE_DONT_WAIT) {
-            tv.tv_sec = tv.tv_usec = 0;
-            tvp = &tv;
-        }
-
         if (eventLoop->beforesleep != NULL && flags & AE_CALL_BEFORE_SLEEP) {
             std::unique_lock<decltype(g_lock)> ulock(g_lock, std::defer_lock);
             if (!(eventLoop->beforesleepFlags & AE_SLEEP_THREADSAFE)) {
@@ -731,6 +726,11 @@ int aeProcessEvents(aeEventLoop *eventLoop, int flags)
             eventLoop->beforesleep(eventLoop);
         }
 
+        if (eventLoop->flags & AE_DONT_WAIT) {
+            tv.tv_sec = tv.tv_usec = 0;
+            tvp = &tv;
+        }
+
         /* Call the multiplexing API, will return only on timeout or when
          * some event fires. */
         numevents = aeApiPoll(eventLoop, tvp);

src/aof.cpp

@@ -285,7 +285,7 @@ int startAppendOnly(void) {
             strerror(errno));
         return C_ERR;
     }
-    if (hasActiveChildProcess() && g_pserver->child_type != CHILD_TYPE_AOF) {
+    if (hasActiveChildProcessOrBGSave() && g_pserver->child_type != CHILD_TYPE_AOF) {
         g_pserver->aof_rewrite_scheduled = 1;
         serverLog(LL_WARNING,"AOF was enabled but there is already another background operation. An AOF background was scheduled to start when possible.");
     } else {
@@ -438,7 +438,7 @@ void flushAppendOnlyFile(int force) {
      * useful for graphing / monitoring purposes. */
     if (sync_in_progress) {
         latencyAddSampleIfNeeded("aof-write-pending-fsync",latency);
-    } else if (hasActiveChildProcess()) {
+    } else if (hasActiveChildProcessOrBGSave()) {
         latencyAddSampleIfNeeded("aof-write-active-child",latency);
     } else {
         latencyAddSampleIfNeeded("aof-write-alone",latency);
@@ -535,7 +535,7 @@ void flushAppendOnlyFile(int force) {
 try_fsync:
     /* Don't fsync if no-appendfsync-on-rewrite is set to yes and there are
      * children doing I/O in the background. */
-    if (g_pserver->aof_no_fsync_on_rewrite && hasActiveChildProcess())
+    if (g_pserver->aof_no_fsync_on_rewrite && hasActiveChildProcessOrBGSave())
         return;
 
     /* Perform the fsync if needed. */
@@ -1887,7 +1887,7 @@ void aofClosePipes(void) {
 int rewriteAppendOnlyFileBackground(void) {
     pid_t childpid;
 
-    if (hasActiveChildProcess()) return C_ERR;
+    if (hasActiveChildProcessOrBGSave()) return C_ERR;
     if (aofCreatePipes() != C_OK) return C_ERR;
     if ((childpid = redisFork(CHILD_TYPE_AOF)) == 0) {
         char tmpfile[256];
@@ -1930,7 +1930,7 @@ int rewriteAppendOnlyFileBackground(void) {
 void bgrewriteaofCommand(client *c) {
     if (g_pserver->child_type == CHILD_TYPE_AOF) {
         addReplyError(c,"Background append only file rewriting already in progress");
-    } else if (hasActiveChildProcess()) {
+    } else if (hasActiveChildProcessOrBGSave()) {
         g_pserver->aof_rewrite_scheduled = 1;
         addReplyStatus(c,"Background append only file rewriting scheduled");
     } else if (rewriteAppendOnlyFileBackground() == C_OK) {

src/config.cpp

@@ -40,6 +40,7 @@
 #endif
 
 const char *KEYDB_SET_VERSION = KEYDB_REAL_VERSION;
+size_t g_semiOrderedSetTargetBucketSize = 0;    // Its a header only class so nowhere else for this to go
 
 /*-----------------------------------------------------------------------------
  * Config file name-value maps.
@@ -2910,6 +2911,7 @@ standardConfig configs[] = {
     createBoolConfig("allow-write-during-load", NULL, MODIFIABLE_CONFIG, g_pserver->fWriteDuringActiveLoad, 0, NULL, NULL),
     createBoolConfig("force-backlog-disk-reserve", NULL, MODIFIABLE_CONFIG, cserver.force_backlog_disk, 0, NULL, NULL),
     createBoolConfig("soft-shutdown", NULL, MODIFIABLE_CONFIG, g_pserver->config_soft_shutdown, 0, NULL, NULL),
+    createSizeTConfig("semi-ordered-set-bucket-size", NULL, MODIFIABLE_CONFIG, 0, 1024, g_semiOrderedSetTargetBucketSize, 0, INTEGER_CONFIG, NULL, NULL),
 
 #ifdef USE_OPENSSL
     createIntConfig("tls-port", NULL, MODIFIABLE_CONFIG, 0, 65535, g_pserver->tls_port, 0, INTEGER_CONFIG, NULL, updateTLSPort), /* TCP port. */

src/db.cpp

@@ -93,7 +93,7 @@ static void lookupKeyUpdateObj(robj *val, int flags)
     /* Update the access time for the ageing algorithm.
      * Don't do it if we have a saving child, as this will trigger
      * a copy on write madness. */
-    if (!hasActiveChildProcess() && !(flags & LOOKUP_NOTOUCH))
+    if (!hasActiveChildProcessOrBGSave() && !(flags & LOOKUP_NOTOUCH))
     {
         if (g_pserver->maxmemory_policy & MAXMEMORY_FLAG_LFU) {
             updateLFU(val);

src/rdb.cpp

@@ -1679,17 +1679,15 @@ int rdbSaveBackground(rdbSaveInfo *rsi) {
     pthread_t child;
     long long start;
 
-    if (hasActiveChildProcess()) return C_ERR;
+    if (hasActiveChildProcessOrBGSave()) return C_ERR;
 
     g_pserver->dirty_before_bgsave = g_pserver->dirty;
     g_pserver->lastbgsave_try = time(NULL);
     openChildInfoPipe();
 
     start = ustime();
+    latencyStartMonitor(g_pserver->rdb_save_latency);
 
-    
-    g_pserver->stat_fork_time = ustime()-start;
-    g_pserver->stat_fork_rate = (double) zmalloc_used_memory() * 1000000 / g_pserver->stat_fork_time / (1024*1024*1024); /* GB per second. */
     if (launchRdbSaveThread(child, rsi) != C_OK) {
         closeChildInfoPipe();
         g_pserver->lastbgsave_status = C_ERR;
@@ -1697,6 +1695,9 @@ int rdbSaveBackground(rdbSaveInfo *rsi) {
             strerror(errno));
         return C_ERR;
     }
+
+    g_pserver->stat_fork_time = ustime()-start;
+    g_pserver->stat_fork_rate = (double) zmalloc_used_memory() * 1000000 / g_pserver->stat_fork_time / (1024*1024*1024); /* GB per second. */
     latencyAddSampleIfNeeded("fork",g_pserver->stat_fork_time/1000);
     serverLog(LL_NOTICE,"Background saving started");
     g_pserver->rdb_save_time_start = time(NULL);
@@ -3601,6 +3602,8 @@ static void backgroundSaveDoneHandlerDisk(int exitcode, bool fCancelled) {
         g_pserver->dirty = g_pserver->dirty - g_pserver->dirty_before_bgsave;
         g_pserver->lastsave = time(NULL);
         g_pserver->lastbgsave_status = C_OK;
+        latencyEndMonitor(g_pserver->rdb_save_latency);
+        latencyAddSampleIfNeeded("rdb-save",g_pserver->rdb_save_latency);
     } else if (!fCancelled && exitcode != 0) {
         serverLog(LL_WARNING, "Background saving error");
         g_pserver->lastbgsave_status = C_ERR;
@@ -3782,7 +3785,7 @@ int rdbSaveToSlavesSockets(rdbSaveInfo *rsi) {
     int pipefds[2];
     rdbSaveSocketThreadArgs *args = nullptr;
 
-    if (hasActiveChildProcess()) return C_ERR;
+    if (hasActiveChildProcessOrBGSave()) return C_ERR;
 
     /* Even if the previous fork child exited, don't start a new one until we
      * drained the pipe. */

src/replication.cpp

@@ -1538,7 +1538,7 @@ void syncCommand(client *c) {
         } else {
             /* We don't have a BGSAVE in progress, let's start one. Diskless
              * or disk-based mode is determined by replica's capacity. */
-            if (!hasActiveChildProcess()) {
+            if (!hasActiveChildProcessOrBGSave()) {
                 startBgsaveForReplication(c->slave_capa);
             } else {
                 serverLog(LL_NOTICE,
@@ -4984,7 +4984,7 @@ void replicationStartPendingFork(void) {
      * In case of diskless replication, we make sure to wait the specified
      * number of seconds (according to configuration) so that other slaves
      * have the time to arrive before we start streaming. */
-    if (!hasActiveChildProcess()) {
+    if (!hasActiveChildProcessOrBGSave()) {
         time_t idle, max_idle = 0;
         int slaves_waiting = 0;
         int mincapa = -1;

src/semiorderedset.h

@@ -27,6 +27,7 @@ namespace keydbutils
     template<>
     size_t hash(const sdsview &);
 }
+extern size_t g_semiOrderedSetTargetBucketSize;
 
 template<typename T, typename T_KEY = T, bool MEMMOVE_SAFE = false>
 class semiorderedset
@@ -41,11 +42,14 @@ class semiorderedset
     size_t idxRehash = (1ULL << bits_min); 
     int cfPauseRehash = 0;
 
-    constexpr size_t targetElementsPerBucket()
+    inline size_t targetElementsPerBucket()
     {
         // Aim for roughly 4 cache lines per bucket (determined by imperical testing)
         //  lower values are faster but use more memory
-        return std::max((64/sizeof(T))*8, (size_t)2);
+        if (g_semiOrderedSetTargetBucketSize == 0)
+            return std::max((64/sizeof(T))*8, (size_t)2);
+        else
+            return g_semiOrderedSetTargetBucketSize;
     }
 
 public:

src/server.cpp

@@ -1706,7 +1706,7 @@ int redisDbPersistentData::incrementallyRehash() {
  * for dict.c to resize the hash tables accordingly to the fact we have an
  * active fork child running. */
 void updateDictResizePolicy(void) {
-    if (!hasActiveChildProcess() || (g_pserver->FRdbSaveInProgress() && !cserver.fForkBgSave))
+    if (!hasActiveChildProcess())
         dictEnableResize();
     else
         dictDisableResize();
@@ -1725,7 +1725,11 @@ const char *strChildType(int type) {
 /* Return true if there are active children processes doing RDB saving,
  * AOF rewriting, or some side process spawned by a loaded module. */
 int hasActiveChildProcess() {
-    return g_pserver->FRdbSaveInProgress() || g_pserver->child_pid != -1;
+    return g_pserver->child_pid != -1;
+}
+
+int hasActiveChildProcessOrBGSave() {
+    return g_pserver->FRdbSaveInProgress() || hasActiveChildProcess();
 }
 
 void resetChildState() {
@@ -2074,7 +2078,7 @@ void databasesCron(bool fMainThread) {
     /* Perform hash tables rehashing if needed, but only if there are no
      * other processes saving the DB on disk. Otherwise rehashing is bad
      * as will cause a lot of copy-on-write of memory pages. */
-    if (!hasActiveChildProcess() || g_pserver->FRdbSaveInProgress()) {
+    if (!hasActiveChildProcess()) {
         /* We use global counters so if we stop the computation at a given
          * DB we'll be able to start from the successive in the next
          * cron loop iteration. */
@@ -2468,14 +2472,14 @@ int serverCron(struct aeEventLoop *eventLoop, long long id, void *clientData) {
 
     /* Start a scheduled AOF rewrite if this was requested by the user while
      * a BGSAVE was in progress. */
-    if (!hasActiveChildProcess() &&
+    if (!hasActiveChildProcessOrBGSave() &&
         g_pserver->aof_rewrite_scheduled)
     {
         rewriteAppendOnlyFileBackground();
     }
 
     /* Check if a background saving or AOF rewrite in progress terminated. */
-    if (hasActiveChildProcess() || ldbPendingChildren())
+    if (hasActiveChildProcessOrBGSave() || ldbPendingChildren())
     {
         run_with_period(1000) receiveChildInfo();
         checkChildrenDone();
@@ -2517,7 +2521,7 @@ int serverCron(struct aeEventLoop *eventLoop, long long id, void *clientData) {
 
         /* Trigger an AOF rewrite if needed. */
         if (g_pserver->aof_state == AOF_ON &&
-            !hasActiveChildProcess() &&
+            !hasActiveChildProcessOrBGSave() &&
             g_pserver->aof_rewrite_perc &&
             g_pserver->aof_current_size > g_pserver->aof_rewrite_min_size)
         {
@@ -2601,7 +2605,7 @@ int serverCron(struct aeEventLoop *eventLoop, long long id, void *clientData) {
      * Note: this code must be after the replicationCron() call above so
      * make sure when refactoring this file to keep this order. This is useful
      * because we want to give priority to RDB savings for replication. */
-    if (!hasActiveChildProcess() &&
+    if (!hasActiveChildProcessOrBGSave() &&
         g_pserver->rdb_bgsave_scheduled &&
         (g_pserver->unixtime-g_pserver->lastbgsave_try > CONFIG_BGSAVE_RETRY_DELAY ||
          g_pserver->lastbgsave_status == C_OK))

src/server.h

@@ -2470,7 +2470,8 @@ struct redisServer {
     time_t lastbgsave_try;          /* Unix time of last attempted bgsave */
     time_t rdb_save_time_last;      /* Time used by last RDB save run. */
     time_t rdb_save_time_start;     /* Current RDB save start time. */
-    pid_t rdb_child_pid = -1;            /* Used only during fork bgsave */
+    mstime_t rdb_save_latency;      /* Used to track end to end latency of rdb save*/
+    pid_t rdb_child_pid = -1;       /* Used only during fork bgsave */
     int rdb_bgsave_scheduled;       /* BGSAVE when possible if true. */
     int rdb_child_type;             /* Type of save by active child. */
     int lastbgsave_status;          /* C_OK or C_ERR */
@@ -3254,6 +3255,7 @@ void receiveChildInfo(void);
 void executeWithoutGlobalLock(std::function<void()> func);
 int redisFork(int type);
 int hasActiveChildProcess();
+int hasActiveChildProcessOrBGSave();
 void resetChildState();
 int isMutuallyExclusiveChildType(int type);