Modified thread count for several kernels

plugins/amoeba/platforms/cuda/src/kernels/amoebaCudaGpu.cpp

@@ -209,6 +209,9 @@ void gpuPrintCudaAmoebaGmxSimulation(amoebaGpuContext amoebaGpu, FILE* log )
     (void) fprintf( log, "     device                             %u\n",      gpu->device );
     (void) fprintf( log, "     sharedMemoryPerBlock               %u\n",      gpu->sharedMemoryPerBlock );
     (void) fprintf( log, "     bOutputBufferPerWarp               %d\n",      gpu->bOutputBufferPerWarp );
+    (void) fprintf( log, "     blocks                             %u\n",      gpu->sim.blocks );
+    (void) fprintf( log, "     threads_per_block                  %u\n",      gpu->sim.threads_per_block);
+    (void) fprintf( log, "     update_threads_per_block           %u\n",      gpu->sim.update_threads_per_block);
     (void) fprintf( log, "     nonbondBlocks                      %u\n",      gpu->sim.nonbond_blocks );
     (void) fprintf( log, "     nonbondThreadsPerBlock             %u\n",      gpu->sim.nonbond_threads_per_block);
     (void) fprintf( log, "     nonbondOutputBuffers               %u\n",      gpu->sim.nonbondOutputBuffers );

plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaMapTorques.cu

@@ -359,7 +359,7 @@ void cudaComputeAmoebaMapTorqueAndAddToForce( amoebaGpuContext amoebaGpu, CUDASt
 {
   
     gpuContext gpu    = amoebaGpu->gpuContext;
-    amoebaMapTorqueToForce_kernel<<< gpu->sim.blocks, gpu->sim.update_threads_per_block>>> ( psTorque->_pDevData );
+    amoebaMapTorqueToForce_kernel<<< gpu->sim.blocks, gpu->sim.threads_per_block>>> ( psTorque->_pDevData );
     LAUNCHERROR("amoebaMapTorqueToForce");
 
 }

plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaPmeDirectElectrostatic.cu

@@ -1965,7 +1965,7 @@ void cudaComputeAmoebaPmeDirectElectrostatic( amoebaGpuContext amoebaGpu )
 
     kClearFields_3( amoebaGpu, 1 );
 
-//#ifdef AMOEBA_DEBUG
+#ifdef AMOEBA_DEBUG
     if( amoebaGpu->log ){
         (void) fprintf( amoebaGpu->log, "kCalculateAmoebaPmeDirectElectrostaticCutoffForces: numBlocks=%u numThreads=%u bufferPerWarp=%u atm=%lu shrd=%lu ixnCt=%lu workUnits=%u maxL1=%d\n",
                         gpu->sim.nonbond_blocks, threadsPerBlock, gpu->bOutputBufferPerWarp,
@@ -1973,7 +1973,7 @@ void cudaComputeAmoebaPmeDirectElectrostatic( amoebaGpuContext amoebaGpu )
                         (*gpu->psInteractionCount)[0], gpu->sim.workUnits, maxL1 );
         (void) fflush( amoebaGpu->log );
     }   
-//#endif
+#endif
 
     if (gpu->bOutputBufferPerWarp){
 

plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaRotateFrame.cu

@@ -388,7 +388,7 @@ void cudaComputeAmoebaLabFrameMoments( amoebaGpuContext amoebaGpu )
     gpuContext gpu    = amoebaGpu->gpuContext;
 
     int numBlocks     = gpu->sim.blocks;
-    int numThreads    = gpu->sim.update_threads_per_block;
+    int numThreads    = gpu->sim.threads_per_block;
 
 //#define AMOEBA_DEBUG  
 #ifdef AMOEBA_DEBUG