diff --git a/Makefile b/Makefile
index 55d1f33d4ef5e43584a47494cd48e5f12e55f801..75e3f48f3063f1dc7085ea214cbb7f9b69f9e54f 100755
--- a/Makefile
+++ b/Makefile
@@ -42,10 +42,10 @@ FFTWLIBS =
 OPT += -DUSE_FFTW
 
 # use omp-ized version of fftw routines
-#OPT += -DHYBRID_FFTW
+OPT += -DHYBRID_FFTW
 
 # switch on the OpenMP parallelization
-#OPT += -DUSE_OMP
+OPT += -DUSE_OMP
 
 # ========================================================
 
@@ -92,7 +92,7 @@ OPT += -DGAUSS_HI_PRECISION
 #OPT += -DCUFFTMP
 
 # FULL NVIDIA GPU SUPPORT - Recommended for full NVIDIA GPU code execution
-OPT += -DFULL_NVIDIA
+#OPT += -DFULL_NVIDIA
 ifeq (FULL_NVIDIA,$(findstring FULL_NVIDIA,$(OPT)))
 OPT += -DCUDACC -DNCCL_REDUCE -DCUFFTMP
 endif
diff --git a/cd b/cd
deleted file mode 100644
index e69de29bb2d1d6434b8b29ae775ad8c2e48c5391..0000000000000000000000000000000000000000
diff --git a/gridding_cpu.c b/gridding_cpu.c
index 40278537b04c8f7c99953cc09172dfeb399c68c1..ab8ea3716edba298ce80dc99ee718c34f6fad2a7 100755
--- a/gridding_cpu.c
+++ b/gridding_cpu.c
@@ -67,12 +67,12 @@ void gridding_data()
       
     }   // closes reduce_method == REDUCE_RING
 
-  
-  timing_wt.kernel     = 0.0;
-  timing_wt.reduce     = 0.0;
-  timing_wt.reduce_mpi = 0.0;
-  timing_wt.reduce_sh  = 0.0;
-  timing_wt.compose    = 0.0;
+  //CLAAAA
+  //timing_wt.kernel     = 0.0;
+  //timing_wt.reduce     = 0.0;
+  //timing_wt.reduce_mpi = 0.0;
+  //timing_wt.reduce_sh  = 0.0;
+  //timing_wt.compose    = 0.0;
 
   // calculate the resolution in radians
   resolution = 1.0/MAX(fabs(metaData.uvmin),fabs(metaData.uvmax));
@@ -142,6 +142,7 @@ void gridding_data()
       double uumax = -1e20;
       double vvmax = -1e20;
 
+     /*
      #pragma omp parallel reduction( min: uumin, vvmin) reduction( max: uumax, vvmax) num_threads(param.num_threads)
       {
 	double my_uumin = 1e20;
@@ -165,7 +166,7 @@ void gridding_data()
       }
 
       //printf("UU, VV, min, max = %f %f %f %f\n", uumin, uumax, vvmin, vvmax);
-      
+      */
 
       timing_wt.compose += CPU_TIME_wt - start;
       
diff --git a/gridding_nccl.cu b/gridding_nccl.cu
index 36b1d2cd4322b12c77efc4d59ac90ee144b4440f..92fa1f900b13aeb2a8ac69467c890c403d5c6cec 100755
--- a/gridding_nccl.cu
+++ b/gridding_nccl.cu
@@ -52,11 +52,12 @@ void gridding_data(){
 
   double shift = (double)(dx*yaxis);
 
-  timing_wt.kernel     = 0.0;
-  timing_wt.reduce     = 0.0;
-  timing_wt.reduce_mpi = 0.0;
-  timing_wt.reduce_sh  = 0.0;
-  timing_wt.compose    = 0.0;
+  // CLAAAA
+  //timing_wt.kernel     = 0.0;
+  //timing_wt.reduce     = 0.0;
+  //timing_wt.reduce_mpi = 0.0;
+  //timing_wt.reduce_sh  = 0.0;
+  //timing_wt.compose    = 0.0;
   
   // calculate the resolution in radians
   resolution = 1.0/MAX(fabs(metaData.uvmin),fabs(metaData.uvmax));
diff --git a/main.c b/main.c
index 5f5bd5d590e7d72dfc9c3ffaa272eeabcaf6111a..6b0ee85ba721104f59d36c798ef9f5d9a3e74e3d 100755
--- a/main.c
+++ b/main.c
@@ -33,6 +33,13 @@ int main(int argc, char * argv[])
 {
 
 
+//CLAAAA
+  timing_wt.kernel     = 0.0;
+  timing_wt.reduce     = 0.0;
+  timing_wt.reduce_mpi = 0.0;
+  timing_wt.reduce_sh  = 0.0;
+  timing_wt.compose    = 0.0;
+
   if(argc > 1)
     {
       strcpy(in.paramfile, argv[1]);
diff --git a/timing.h b/timing.h
index eb2570b2bdd410f00f794da8d399846745158df4..71e778ee8f66d1409b765e98878370264e7ddcc6 100755
--- a/timing.h
+++ b/timing.h
@@ -77,7 +77,6 @@ typedef struct {
 
 extern timing_t timing_wt;      // wall-clock process timing, at Task 0
 
-
 extern double start_tot;
 extern double reduce_shmem_time;
 extern double reduce_mpi_time;