Impose a minimum tcache count for small size classes.

Now that small allocation runs have fewer regions due to run metadata
residing in chunk headers, an explicit minimum tcache count is needed to
make sure that tcache adequately amortizes synchronization overhead.

include/jemalloc/internal/tcache.h

@@ -16,6 +16,11 @@ typedef struct tcaches_s tcaches_t;
 #define	TCACHE_STATE_PURGATORY		((tcache_t *)(uintptr_t)3)
 #define	TCACHE_STATE_MAX		TCACHE_STATE_PURGATORY
 
+/*
+ * Absolute minimum number of cache slots for each small bin.
+ */
+#define	TCACHE_NSLOTS_SMALL_MIN		20
+
 /*
  * Absolute maximum number of cache slots for each small bin in the thread
  * cache.  This is an additional constraint beyond that imposed as: twice the