Optimize memory allocation

A number of memory allocation optimizations have been implemented. Most
optimizations reduce contention caused by synchronization between
threads during allocation and deallocation of memory. Most notably:
* Synchronization of memory management in scheduler specific allocator
  instances has been rewritten to use lock-free synchronization.
* Synchronization of memory management in scheduler specific
  pre-allocators has been rewritten to use lock-free synchronization.
* The 'mseg_alloc' memory segment allocator now use scheduler specific
  instances instead of one instance. Apart from reducing contention
  this also ensures that memory allocators always create memory
  segments on the local NUMA node on a NUMA system.

1 files changed, 16 insertions, 1 deletions

diff --git a/erts/emulator/beam/erl_process_lock.h b/erts/emulator/beam/erl_process_lock.h
index cd3b2182fd..97f250138e 100644
--- a/erts/emulator/beam/erl_process_lock.h
+++ b/erts/emulator/beam/erl_process_lock.h
@@ -651,7 +651,7 @@ ERTS_GLB_INLINE int erts_smp_proc_trylock(Process *, ErtsProcLocks);
 
 ERTS_GLB_INLINE void erts_smp_proc_inc_refc(Process *);
 ERTS_GLB_INLINE void erts_smp_proc_dec_refc(Process *);
-
+ERTS_GLB_INLINE void erts_smp_proc_add_refc(Process *, Sint32);
 
 #if ERTS_GLB_INLINE_INCL_FUNC_DEF
 
@@ -737,6 +737,21 @@ ERTS_GLB_INLINE void erts_smp_proc_dec_refc(Process *p)
 #endif
 }
 
+ERTS_GLB_INLINE void erts_smp_proc_add_refc(Process *p, Sint32 refc)
+{
+#ifdef ERTS_SMP
+    Process *fp;
+    erts_pix_lock_t *pixlck = ERTS_PID2PIXLOCK(p->id);
+    erts_pix_lock(pixlck);
+    ERTS_LC_ASSERT(p->lock.refc > 0);
+    p->lock.refc += refc;
+    fp = p->lock.refc == 0 ? p : NULL; 
+    erts_pix_unlock(pixlck);
+    if (fp)
+	erts_free_proc(fp);
+#endif
+}
+
 #endif /* #if ERTS_GLB_INLINE_INCL_FUNC_DEF */
 
 #ifdef ERTS_SMP