diff --git a/src/BulletDynamics/Dynamics/btDiscreteDynamicsWorldMt.cpp b/src/BulletDynamics/Dynamics/btDiscreteDynamicsWorldMt.cpp
index df6c17c5b..1d10bad92 100644
--- a/src/BulletDynamics/Dynamics/btDiscreteDynamicsWorldMt.cpp
+++ b/src/BulletDynamics/Dynamics/btDiscreteDynamicsWorldMt.cpp
@@ -284,26 +284,29 @@ struct UpdaterUnconstrainedMotion : public btIParallelForBody
 void btDiscreteDynamicsWorldMt::predictUnconstraintMotion( btScalar timeStep )
 {
     BT_PROFILE( "predictUnconstraintMotion" );
-    int grainSize = 50;  // num of iterations per task for TBB
-    int bodyCount = m_nonStaticRigidBodies.size();
-    UpdaterUnconstrainedMotion update;
-    update.timeStep = timeStep;
-    update.rigidBodies = bodyCount ? &m_nonStaticRigidBodies[ 0 ] : NULL;
-    btParallelFor( 0, bodyCount, grainSize, update );
+    if ( m_nonStaticRigidBodies.size() > 0 )
+    {
+        UpdaterUnconstrainedMotion update;
+        update.timeStep = timeStep;
+        update.rigidBodies = &m_nonStaticRigidBodies[ 0 ];
+        int grainSize = 50;  // num of iterations per task for task scheduler
+        btParallelFor( 0, m_nonStaticRigidBodies.size(), grainSize, update );
+    }
 }
 
 
 void btDiscreteDynamicsWorldMt::createPredictiveContacts( btScalar timeStep )
 {
+    BT_PROFILE( "createPredictiveContacts" );
     releasePredictiveContacts();
-    int grainSize = 50;  // num of iterations per task for TBB or OPENMP
-    if ( int bodyCount = m_nonStaticRigidBodies.size() )
+    if ( m_nonStaticRigidBodies.size() > 0 )
     {
         UpdaterCreatePredictiveContacts update;
         update.world = this;
         update.timeStep = timeStep;
         update.rigidBodies = &m_nonStaticRigidBodies[ 0 ];
-        btParallelFor( 0, bodyCount, grainSize, update );
+        int grainSize = 50;  // num of iterations per task for task scheduler
+        btParallelFor( 0, m_nonStaticRigidBodies.size(), grainSize, update );
     }
 }
 
@@ -311,14 +314,14 @@ void btDiscreteDynamicsWorldMt::createPredictiveContacts( btScalar timeStep )
 void btDiscreteDynamicsWorldMt::integrateTransforms( btScalar timeStep )
 {
     BT_PROFILE( "integrateTransforms" );
-    int grainSize = 50;  // num of iterations per task for TBB or OPENMP
-    if ( int bodyCount = m_nonStaticRigidBodies.size() )
+    if ( m_nonStaticRigidBodies.size() > 0 )
     {
         UpdaterIntegrateTransforms update;
         update.world = this;
         update.timeStep = timeStep;
         update.rigidBodies = &m_nonStaticRigidBodies[ 0 ];
-        btParallelFor( 0, bodyCount, grainSize, update );
+        int grainSize = 50;  // num of iterations per task for task scheduler
+        btParallelFor( 0, m_nonStaticRigidBodies.size(), grainSize, update );
     }
 }
 
diff --git a/src/BulletDynamics/Dynamics/btSimulationIslandManagerMt.cpp b/src/BulletDynamics/Dynamics/btSimulationIslandManagerMt.cpp
index 6373f60ff..99b34353c 100644
--- a/src/BulletDynamics/Dynamics/btSimulationIslandManagerMt.cpp
+++ b/src/BulletDynamics/Dynamics/btSimulationIslandManagerMt.cpp
@@ -548,6 +548,7 @@ void btSimulationIslandManagerMt::mergeIslands()
 
 void btSimulationIslandManagerMt::serialIslandDispatch( btAlignedObjectArray<Island*>* islandsPtr, IslandCallback* callback )
 {
+    BT_PROFILE( "serialIslandDispatch" );
     // serial dispatch
     btAlignedObjectArray<Island*>& islands = *islandsPtr;
     for ( int i = 0; i < islands.size(); ++i )
@@ -592,6 +593,7 @@ struct UpdateIslandDispatcher : public btIParallelForBody
 
 void btSimulationIslandManagerMt::parallelIslandDispatch( btAlignedObjectArray<Island*>* islandsPtr, IslandCallback* callback )
 {
+    BT_PROFILE( "parallelIslandDispatch" );
     int grainSize = 1;  // iterations per task
     UpdateIslandDispatcher dispatcher;
     dispatcher.islandsPtr = islandsPtr;
diff --git a/src/LinearMath/btThreads.cpp b/src/LinearMath/btThreads.cpp
index 85de7c4d2..876bc4bb1 100644
--- a/src/LinearMath/btThreads.cpp
+++ b/src/LinearMath/btThreads.cpp
@@ -14,6 +14,7 @@ subject to the following restrictions:
 
 
 #include "btThreads.h"
+#include "btQuickprof.h"
 #include <algorithm>  // for min and max
 
 #if BT_THREADSAFE
@@ -114,10 +115,12 @@ public:
     }
     virtual void parallelFor( int iBegin, int iEnd, int grainSize, const btIParallelForBody& body ) BT_OVERRIDE
     {
+        BT_PROFILE( "parallelFor_OpenMP" );
         btPushThreadsAreRunning();
 #pragma omp parallel for schedule( static, 1 )
         for ( int i = iBegin; i < iEnd; i += grainSize )
         {
+            BT_PROFILE( "OpenMP_job" );
             body.forLoop( i, ( std::min )( i + grainSize, iEnd ) );
         }
         btPopThreadsAreRunning();
@@ -174,11 +177,13 @@ public:
 
         void operator()( const tbb::blocked_range<int>& range ) const
         {
+            BT_PROFILE( "TBB_job" );
             mBody->forLoop( range.begin(), range.end() );
         }
     };
     virtual void parallelFor( int iBegin, int iEnd, int grainSize, const btIParallelForBody& body ) BT_OVERRIDE
     {
+        BT_PROFILE( "parallelFor_TBB" );
         // TBB dispatch
         BodyAdapter tbbBody;
         tbbBody.mBody = &body;
@@ -232,11 +237,13 @@ public:
 
         void operator()( int i ) const
         {
+            BT_PROFILE( "PPL_job" );
             mBody->forLoop( i, ( std::min )( i + mGrainSize, mIndexEnd ) );
         }
     };
     virtual void parallelFor( int iBegin, int iEnd, int grainSize, const btIParallelForBody& body ) BT_OVERRIDE
     {
+        BT_PROFILE( "parallelFor_PPL" );
         // PPL dispatch
         BodyAdapter pplBody;
         pplBody.mBody = &body;
@@ -488,6 +495,7 @@ public:
     virtual void setNumThreads( int numThreads ) BT_OVERRIDE {}
     virtual void parallelFor( int iBegin, int iEnd, int grainSize, const btIParallelForBody& body ) BT_OVERRIDE
     {
+        BT_PROFILE( "parallelFor_sequential" );
         body.forLoop( iBegin, iEnd );
     }
 };