diff --git a/src/MANYBODY/pair_tersoff_mod.cpp b/src/MANYBODY/pair_tersoff_mod.cpp
index f29b8c0c59..b452be0fa9 100644
--- a/src/MANYBODY/pair_tersoff_mod.cpp
+++ b/src/MANYBODY/pair_tersoff_mod.cpp
@@ -57,7 +57,7 @@ void PairTersoffMOD::read_file(char *file)
 
   FILE *fp;
   if (comm->me == 0) {
-    fp = fopen(file,"r");
+    fp = open_potential(file);
     if (fp == NULL) {
       char str[128];
       sprintf(str,"Cannot open Tersoff potential file %s",file);
diff --git a/src/MEAM/pair_meam.cpp b/src/MEAM/pair_meam.cpp
index bdfc9ea159..291af44d8f 100644
--- a/src/MEAM/pair_meam.cpp
+++ b/src/MEAM/pair_meam.cpp
@@ -460,7 +460,7 @@ void PairMEAM::read_files(char *globalfile, char *userfile)
 
   FILE *fp;
   if (comm->me == 0) {
-    fp = fopen(globalfile,"r");
+    fp = open_potential(globalfile);
     if (fp == NULL) {
       char str[128];
       sprintf(str,"Cannot open MEAM potential file %s",globalfile);
@@ -641,7 +641,7 @@ void PairMEAM::read_files(char *globalfile, char *userfile)
   // open user param file on proc 0
 
   if (comm->me == 0) {
-    fp = fopen(userfile,"r");
+    fp = open_potential(userfile);
     if (fp == NULL) {
       char str[128];
       sprintf(str,"Cannot open MEAM potential file %s",userfile);
diff --git a/src/MOLECULE/pair_lj_charmm_coul_charmm.cpp b/src/MOLECULE/pair_lj_charmm_coul_charmm.cpp
index 65993c6789..390e21d331 100644
--- a/src/MOLECULE/pair_lj_charmm_coul_charmm.cpp
+++ b/src/MOLECULE/pair_lj_charmm_coul_charmm.cpp
@@ -105,10 +105,10 @@ void PairLJCharmmCoulCharmm::compute(int eflag, int vflag)
 
     for (jj = 0; jj < jnum; jj++) {
       j = jlist[jj];
-      jtype = type[j];
       factor_lj = special_lj[sbmask(j)];
       factor_coul = special_coul[sbmask(j)];
       j &= NEIGHMASK;
+      jtype = type[j];
 
       delx = xtmp - x[j][0];
       dely = ytmp - x[j][1];
diff --git a/src/MPIIO/dump_atom_mpiio.cpp b/src/MPIIO/dump_atom_mpiio.cpp
index e78b5cd444..060d8c0276 100644
--- a/src/MPIIO/dump_atom_mpiio.cpp
+++ b/src/MPIIO/dump_atom_mpiio.cpp
@@ -546,7 +546,7 @@ int DumpAtomMPIIO::convert_string_omp(int n, double *mybuf)
 
 int DumpAtomMPIIO::convert_image_omp(int n, double *mybuf)
 {
-  MPI_Status mpiStatus;
+  double *localbuf = mybuf;
   char **mpifh_buffer_line_per_thread;
   int mpifhStringCount;
   int *mpifhStringCountPerThread, *bufOffset, *bufRange, *bufLength;
@@ -581,7 +581,7 @@ int DumpAtomMPIIO::convert_image_omp(int n, double *mybuf)
     mpifh_buffer_line_per_thread[i] = (char *) malloc(DUMP_BUF_CHUNK_SIZE * sizeof(char));
     mpifh_buffer_line_per_thread[i][0] = '\0';
 
-#pragma omp parallel default(none)
+#pragma omp parallel default(none) shared(localbuf,bufLength,bufOffset,bufRange,mpifhStringCountPerThread,mpifh_buffer_line_per_thread)
     {
       int tid = omp_get_thread_num();
       int m=0;
@@ -593,7 +593,7 @@ int DumpAtomMPIIO::convert_image_omp(int n, double *mybuf)
           bufLength[tid] = (mpifhStringCountPerThread[tid]+DUMP_BUF_CHUNK_SIZE) * sizeof(char);
         }
 
-        mpifhStringCountPerThread[tid] += sprintf(&(mpifh_buffer_line_per_thread[tid][mpifhStringCountPerThread[tid]]),format,static_cast<int> (mybuf[bufOffset[tid]+m]),static_cast<int> (mybuf[bufOffset[tid]+m+1]),mybuf[bufOffset[tid]+m+2],mybuf[bufOffset[tid]+m+3],mybuf[bufOffset[tid]+m+4],static_cast<int> (mybuf[bufOffset[tid]+m+5]),static_cast<int> (mybuf[bufOffset[tid]+m+6]),static_cast<int> (mybuf[bufOffset[tid]+m+7]));
+        mpifhStringCountPerThread[tid] += sprintf(&(mpifh_buffer_line_per_thread[tid][mpifhStringCountPerThread[tid]]),format,static_cast<int> (localbuf[bufOffset[tid]+m]),static_cast<int> (localbuf[bufOffset[tid]+m+1]),localbuf[bufOffset[tid]+m+2],localbuf[bufOffset[tid]+m+3],localbuf[bufOffset[tid]+m+4],static_cast<int> (localbuf[bufOffset[tid]+m+5]),static_cast<int> (localbuf[bufOffset[tid]+m+6]),static_cast<int> (localbuf[bufOffset[tid]+m+7]));
         m += size_one;
       }
     }
@@ -637,7 +637,7 @@ int DumpAtomMPIIO::convert_image_omp(int n, double *mybuf)
 
 int DumpAtomMPIIO::convert_noimage_omp(int n, double *mybuf)
 {
-  MPI_Status mpiStatus;
+  double *localbuf = mybuf;
   char **mpifh_buffer_line_per_thread;
   int mpifhStringCount;
   int *mpifhStringCountPerThread, *bufOffset, *bufRange, *bufLength;
@@ -672,7 +672,7 @@ int DumpAtomMPIIO::convert_noimage_omp(int n, double *mybuf)
     mpifh_buffer_line_per_thread[i] = (char *) malloc(DUMP_BUF_CHUNK_SIZE * sizeof(char));
     mpifh_buffer_line_per_thread[i][0] = '\0';
 
-#pragma omp parallel default(none)
+#pragma omp parallel default(none) shared(localbuf,bufLength,bufOffset,bufRange,mpifhStringCountPerThread,mpifh_buffer_line_per_thread)
     {
       int tid = omp_get_thread_num();
       int m=0;
@@ -684,7 +684,7 @@ int DumpAtomMPIIO::convert_noimage_omp(int n, double *mybuf)
           bufLength[tid] = (mpifhStringCountPerThread[tid]+DUMP_BUF_CHUNK_SIZE) * sizeof(char);
         }
 
-        mpifhStringCountPerThread[tid] += sprintf(&(mpifh_buffer_line_per_thread[tid][mpifhStringCountPerThread[tid]]),format,static_cast<int> (mybuf[bufOffset[tid]+m]),static_cast<int> (mybuf[bufOffset[tid]+m+1]),mybuf[bufOffset[tid]+m+2],mybuf[bufOffset[tid]+m+3],mybuf[bufOffset[tid]+m+4]);
+        mpifhStringCountPerThread[tid] += sprintf(&(mpifh_buffer_line_per_thread[tid][mpifhStringCountPerThread[tid]]),format,static_cast<int> (localbuf[bufOffset[tid]+m]),static_cast<int> (localbuf[bufOffset[tid]+m+1]),localbuf[bufOffset[tid]+m+2],localbuf[bufOffset[tid]+m+3],localbuf[bufOffset[tid]+m+4]);
         m += size_one;
       }
     }
diff --git a/src/MPIIO/dump_custom_mpiio.cpp b/src/MPIIO/dump_custom_mpiio.cpp
index 7d17f4ed9f..461421253e 100644
--- a/src/MPIIO/dump_custom_mpiio.cpp
+++ b/src/MPIIO/dump_custom_mpiio.cpp
@@ -566,7 +566,7 @@ void DumpCustomMPIIO::write_string(int n, double *mybuf)
 
 int DumpCustomMPIIO::convert_string_omp(int n, double *mybuf)
 {
-  MPI_Status mpiStatus;
+  double *localbuf = mybuf;
   char **mpifh_buffer_line_per_thread;
   int mpifhStringCount;
   int *mpifhStringCountPerThread, *bufOffset, *bufRange, *bufLength;
@@ -601,7 +601,7 @@ int DumpCustomMPIIO::convert_string_omp(int n, double *mybuf)
     mpifh_buffer_line_per_thread[i] = (char *) malloc(DUMP_BUF_CHUNK_SIZE * sizeof(char));
     mpifh_buffer_line_per_thread[i][0] = '\0';
 
-#pragma omp parallel default(none)
+#pragma omp parallel default(none) shared(localbuf,bufLength,bufOffset,bufRange,mpifhStringCountPerThread,mpifh_buffer_line_per_thread)
     {
       int tid = omp_get_thread_num();
       int m=0;
@@ -615,11 +615,11 @@ int DumpCustomMPIIO::convert_string_omp(int n, double *mybuf)
         for (int j = 0; j < size_one; j++) {
 
           if (vtype[j] == INT)
-            mpifhStringCountPerThread[tid] += sprintf(&(mpifh_buffer_line_per_thread[tid][mpifhStringCountPerThread[tid]]),vformat[j],static_cast<int> (mybuf[bufOffset[tid]+m]));
+            mpifhStringCountPerThread[tid] += sprintf(&(mpifh_buffer_line_per_thread[tid][mpifhStringCountPerThread[tid]]),vformat[j],static_cast<int> (localbuf[bufOffset[tid]+m]));
           else if (vtype[j] == DOUBLE)
-            mpifhStringCountPerThread[tid] += sprintf(&(mpifh_buffer_line_per_thread[tid][mpifhStringCountPerThread[tid]]),vformat[j],mybuf[bufOffset[tid]+m]);
+            mpifhStringCountPerThread[tid] += sprintf(&(mpifh_buffer_line_per_thread[tid][mpifhStringCountPerThread[tid]]),vformat[j],localbuf[bufOffset[tid]+m]);
           else if (vtype[j] == STRING)
-            mpifhStringCountPerThread[tid] += sprintf(&(mpifh_buffer_line_per_thread[tid][mpifhStringCountPerThread[tid]]),vformat[j],typenames[(int) mybuf[bufOffset[tid]+m]]);
+            mpifhStringCountPerThread[tid] += sprintf(&(mpifh_buffer_line_per_thread[tid][mpifhStringCountPerThread[tid]]),vformat[j],typenames[(int) localbuf[bufOffset[tid]+m]]);
           m ++;
         }
         mpifhStringCountPerThread[tid] += sprintf(&(mpifh_buffer_line_per_thread[tid][mpifhStringCountPerThread[tid]]),"\n");
diff --git a/src/MPIIO/dump_xyz_mpiio.cpp b/src/MPIIO/dump_xyz_mpiio.cpp
index 4457260989..fe1f65fc3c 100644
--- a/src/MPIIO/dump_xyz_mpiio.cpp
+++ b/src/MPIIO/dump_xyz_mpiio.cpp
@@ -308,7 +308,7 @@ void DumpXYZMPIIO::write_string(int n, double *mybuf)
 
 int DumpXYZMPIIO::convert_string_omp(int n, double *mybuf)
 {
-  MPI_Status mpiStatus;
+  double *localbuf = mybuf;
   char **mpifh_buffer_line_per_thread;
   int mpifhStringCount;
   int *mpifhStringCountPerThread, *bufOffset, *bufRange, *bufLength;
@@ -343,7 +343,7 @@ int DumpXYZMPIIO::convert_string_omp(int n, double *mybuf)
     mpifh_buffer_line_per_thread[i] = (char *) malloc(DUMP_BUF_CHUNK_SIZE * sizeof(char));
     mpifh_buffer_line_per_thread[i][0] = '\0';
 
-#pragma omp parallel default(none)
+#pragma omp parallel default(none) shared(localbuf,bufLength,bufOffset,bufRange,mpifhStringCountPerThread,mpifh_buffer_line_per_thread)
     {
       int tid = omp_get_thread_num();
       int m=0;
@@ -355,7 +355,7 @@ int DumpXYZMPIIO::convert_string_omp(int n, double *mybuf)
           bufLength[tid] = (mpifhStringCountPerThread[tid]+DUMP_BUF_CHUNK_SIZE) * sizeof(char);
         }
 
-        mpifhStringCountPerThread[tid] += sprintf(&(mpifh_buffer_line_per_thread[tid][mpifhStringCountPerThread[tid]]),format,typenames[static_cast<int> (mybuf[bufOffset[tid]+m+1])],mybuf[bufOffset[tid]+m+2],mybuf[bufOffset[tid]+m+3],mybuf[bufOffset[tid]+m+4]);
+        mpifhStringCountPerThread[tid] += sprintf(&(mpifh_buffer_line_per_thread[tid][mpifhStringCountPerThread[tid]]),format,typenames[static_cast<int> (localbuf[bufOffset[tid]+m+1])],localbuf[bufOffset[tid]+m+2],localbuf[bufOffset[tid]+m+3],localbuf[bufOffset[tid]+m+4]);
           m += size_one;
 
       }
diff --git a/src/Package.sh b/src/Package.sh
index 839481c1b5..80677665eb 100644
--- a/src/Package.sh
+++ b/src/Package.sh
@@ -78,9 +78,9 @@ elif (test $2 = "diff") then
         echo "  src/$file does not exist"
       elif (! cmp -s $file ../$file) then
         echo "************************************************"
-        echo "diff $1/$file src/$file "
+        echo "diff -u $1/$file src/$file "
         echo "************************************************"
-	diff $file  ../$file 
+	diff -u $file  ../$file 
       fi
     done
   fi
diff --git a/src/Purge.list b/src/Purge.list
index 7f53e9c67a..992f1719ae 100644
--- a/src/Purge.list
+++ b/src/Purge.list
@@ -24,6 +24,8 @@ pair_dipole_sf.h
 pair_dipole_sf.cpp
 pair_dipole_sf_omp.h
 pair_dipole_sf_omp.cpp
+pair_dipole_sf_gpu.h
+pair_dipole_sf_gpu.cpp
 # deleted on Wed May 8 15:24:36 2013 +0000
 compute_spec_atom.cpp
 compute_spec_atom.h
diff --git a/src/RIGID/fix_shake.cpp b/src/RIGID/fix_shake.cpp
index 2c6ca9f7b1..8741cf464a 100644
--- a/src/RIGID/fix_shake.cpp
+++ b/src/RIGID/fix_shake.cpp
@@ -649,8 +649,6 @@ void FixShake::find_clusters()
   tagint tagprev;
   double massone;
   tagint *buf;
-  MPI_Request request;
-  MPI_Status status;
 
   if (me == 0 && screen) fprintf(screen,"Finding SHAKE clusters ...\n");
 
diff --git a/src/SHOCK/fix_append_atoms.cpp b/src/SHOCK/fix_append_atoms.cpp
index 295337e6f0..5e7af01f6e 100644
--- a/src/SHOCK/fix_append_atoms.cpp
+++ b/src/SHOCK/fix_append_atoms.cpp
@@ -315,7 +315,8 @@ int FixAppendAtoms::get_spatial()
 
     if (domain->boxhi[2] - shockfront_loc < spatlead) advance = 1;
 
-    delete [] pos,val;
+    delete [] pos;
+    delete [] val;
   }
 
   advance_sum = 0;
@@ -453,7 +454,6 @@ void FixAppendAtoms::pre_exchange()
       double x[3];
       double *sublo = domain->sublo;
       double *subhi = domain->subhi;
-      double *mass = atom->mass;
 
       int i,j,k,m;
       for (k = klo; k <= khi; k++) {
diff --git a/src/USER-ATC/fix_atc.cpp b/src/USER-ATC/fix_atc.cpp
index 2092340677..7e873b64d5 100644
--- a/src/USER-ATC/fix_atc.cpp
+++ b/src/USER-ATC/fix_atc.cpp
@@ -38,6 +38,7 @@
 #include "stdio.h"
 #include "string.h"
 #include <sstream>
+#include <string>
 
 using namespace LAMMPS_NS; 
 using namespace FixConst;
@@ -288,7 +289,7 @@ FixATC::FixATC(LAMMPS *lmp, int narg, char **arg) : Fix(lmp, narg, arg),
 
   int me = ATC::LammpsInterface::instance()->comm_rank();
 
-  string groupName(arg[1]);
+  std::string groupName(arg[1]);
   int igroup = group->find(groupName.c_str());
   int atomCount = group->count(igroup);
 
diff --git a/src/USER-LB/fix_lb_rigid_pc_sphere.cpp b/src/USER-LB/fix_lb_rigid_pc_sphere.cpp
index 8031bbbf98..abc104cb1f 100755
--- a/src/USER-LB/fix_lb_rigid_pc_sphere.cpp
+++ b/src/USER-LB/fix_lb_rigid_pc_sphere.cpp
@@ -1332,7 +1332,7 @@ void FixLbRigidPCSphere::set_xv()
 
 void FixLbRigidPCSphere::pre_neighbor()
 {
-  int original,oldimage,newimage;
+  imageint original,oldimage,newimage;
 
   for (int ibody = 0; ibody < nbody; ibody++) {
     original = imagebody[ibody];
@@ -1355,10 +1355,11 @@ void FixLbRigidPCSphere::pre_neighbor()
 
   // adjust image flags of any atom in a rigid body whose xcm was remapped
 
-  int *atomimage = atom->image;
+  imageint *atomimage = atom->image;
   int nlocal = atom->nlocal;
 
-  int ibody,idim,otherdims;
+  int ibody;
+  imageint idim,otherdims;
 
   for (int i = 0; i < nlocal; i++) {
     if (body[i] == -1) continue;
diff --git a/src/USER-LB/fix_lb_rigid_pc_sphere.h b/src/USER-LB/fix_lb_rigid_pc_sphere.h
index 069f38c35d..90ed979e28 100755
--- a/src/USER-LB/fix_lb_rigid_pc_sphere.h
+++ b/src/USER-LB/fix_lb_rigid_pc_sphere.h
@@ -78,7 +78,7 @@ class FixLbRigidPCSphere : public Fix {
   double **torque_fluid;
   double **torque_fluid_old;
   double **rotate;
-  int *imagebody;               // image flags of xcm of each rigid body
+  imageint *imagebody;               // image flags of xcm of each rigid body
   double **fflag;           // flag for on/off of center-of-mass force
   double **tflag;           // flag for on/off of center-of-mass torque
 
diff --git a/src/USER-MISC/angle_cosine_shift_exp.cpp b/src/USER-MISC/angle_cosine_shift_exp.cpp
index ee4c014b10..2941293be7 100644
--- a/src/USER-MISC/angle_cosine_shift_exp.cpp
+++ b/src/USER-MISC/angle_cosine_shift_exp.cpp
@@ -303,7 +303,6 @@ double AngleCosineShiftExp::single(int type, int i1, int i2, int i3)
   double s=sqrt(1.0-c*c);
 
   double cccpsss=c*cost[type]+s*sint[type];
-  double cssmscc=c*sint[type]-s*cost[type];
 
   if (doExpansion[type])
        {
diff --git a/src/USER-REAXC/reaxc_ffield.cpp b/src/USER-REAXC/reaxc_ffield.cpp
index 7d587ddb27..a0a23777e5 100644
--- a/src/USER-REAXC/reaxc_ffield.cpp
+++ b/src/USER-REAXC/reaxc_ffield.cpp
@@ -52,7 +52,7 @@ char Read_Force_Field( char *ffield_file, reax_interaction *reax,
 
   /* open force field file */
   if ( (fp = fopen( ffield_file, "r" ) ) == NULL ) {
-    fprintf( stderr, "error opening the force filed file! terminating...\n" );
+    fprintf( stderr, "error opening the force field file! terminating...\n" );
     MPI_Abort( comm, FILE_NOT_FOUND );
   }
 
diff --git a/src/USER-REAXC/reaxc_tool_box.cpp b/src/USER-REAXC/reaxc_tool_box.cpp
index bee7bb6efe..5f18a9800c 100644
--- a/src/USER-REAXC/reaxc_tool_box.cpp
+++ b/src/USER-REAXC/reaxc_tool_box.cpp
@@ -388,7 +388,7 @@ int Allocate_Tokenizer_Space( char **line, char **backup, char ***tokens )
 int Tokenize( char* s, char*** tok )
 {
   char test[MAX_LINE];
-  char *sep = "\t \n!=";
+  const char *sep = (const char *)"\t \n!=";
   char *word;
   int count=0;
 
@@ -405,7 +405,7 @@ int Tokenize( char* s, char*** tok )
 
 /***************** taken from lammps ************************/
 /* safe malloc */
-void *smalloc( long n, char *name, MPI_Comm comm )
+void *smalloc( long n, const char *name, MPI_Comm comm )
 {
   void *ptr;
 
@@ -428,7 +428,7 @@ void *smalloc( long n, char *name, MPI_Comm comm )
 
 
 /* safe calloc */
-void *scalloc( int n, int size, char *name, MPI_Comm comm )
+void *scalloc( int n, int size, const char *name, MPI_Comm comm )
 {
   void *ptr;
 
@@ -458,7 +458,7 @@ void *scalloc( int n, int size, char *name, MPI_Comm comm )
 
 
 /* safe free */
-void sfree( void *ptr, char *name )
+void sfree( void *ptr, const char *name )
 {
   if( ptr == NULL ) {
     fprintf( stderr, "WARNING: trying to free the already NULL pointer %s!\n",
diff --git a/src/USER-REAXC/reaxc_tool_box.h b/src/USER-REAXC/reaxc_tool_box.h
index b9e1f4a8db..2ef27dd6da 100644
--- a/src/USER-REAXC/reaxc_tool_box.h
+++ b/src/USER-REAXC/reaxc_tool_box.h
@@ -67,8 +67,8 @@ int   Allocate_Tokenizer_Space( char**, char**, char*** );
 int   Tokenize( char*, char*** );
 
 /* from lammps */
-void *smalloc( long, char*, MPI_Comm );
-void *scalloc( int, int, char*, MPI_Comm );
-void sfree( void*, char* );
+void *smalloc( long, const char*, MPI_Comm );
+void *scalloc( int, int, const char*, MPI_Comm );
+void sfree( void*, const char* );
 
 #endif
diff --git a/src/balance.cpp b/src/balance.cpp
index b315a4454b..e1bcf36924 100644
--- a/src/balance.cpp
+++ b/src/balance.cpp
@@ -721,116 +721,6 @@ int Balance::adjust(int n, double *split)
   return change;
 }
 
-/* ----------------------------------------------------------------------
-   OLD code: for local diffusion method that didn't work as well as RCB
-   adjust cuts between N slices in a dim via diffusive method
-   count = atoms per slice
-   split = current N+1 cuts, with 0.0 and 1.0 at end points
-   overwrite split with new cuts
-   diffusion means slices with more atoms than their neighbors "send" atoms,
-     by moving cut closer to sender, further from receiver
-------------------------------------------------------------------------- */
-
-void Balance::old_adjust(int iter, int n, bigint *count, double *split)
-{
-  // need to allocate this if start using it again
-
-  double *cuts;
-
-  // damping factor
-
-  double damp = 0.5;
-
-  // loop over slices
-  // cut I is between 2 slices (I-1 and I) with counts
-  // cut I+1 is between 2 slices (I and I+1) with counts
-  // for a cut between 2 slices, only slice with larger count adjusts it
-  // special treatment of end slices with only 1 neighbor
-
-  bigint leftcount,mycount,rightcount;
-  double rho,target,targetleft,targetright;
-
-  for (int i = 0; i < n; i++) {
-    if (i == 0) leftcount = MAXBIGINT;
-    else leftcount = count[i-1];
-    mycount = count[i];
-    if (i == n-1) rightcount = MAXBIGINT;
-    else rightcount = count[i+1];
-
-    // middle slice is <= both left and right, so do nothing
-    // special case if 2 slices both have count = 0 -> no change in cut
-
-    if (mycount <= leftcount && mycount <= rightcount) {
-      if (leftcount == 0) cuts[i] = split[i];
-      if (rightcount == 0) cuts[i+1] = split[i+1];
-      continue;
-    }
-
-    // rho = density of atoms in the slice
-
-    rho = mycount / (split[i+1] - split[i]);
-
-    // middle slice has more atoms than left or right slice
-    // send atoms in that dir
-
-    if (mycount > leftcount) {
-      target = damp * 0.5*(mycount-leftcount);
-      cuts[i] = split[i] + target/rho;
-    }
-    if (mycount > rightcount) {
-      target = damp * 0.5*(mycount-rightcount);
-      cuts[i+1] = split[i+1] - target/rho;
-    }
-
-    /*
-    // middle slice has more atoms then left or right slice
-    // if delta from middle to top slice > delta between top and bottom slice
-    //   then send atoms both dirs to bring all 3 slices to same count
-    // else bottom slice is very low, so send atoms only in that dir
-
-    if (mycount > leftcount && mycount > rightcount) {
-      if (mycount-MAX(leftcount,rightcount) >= fabs(leftcount-rightcount)) {
-        if (leftcount <= rightcount) {
-          targetleft = damp *
-            (rightcount-leftcount + (mycount-rightcount)/3.0);
-          targetright = damp * (mycount-rightcount)/3.0;
-          cuts[i] = split[i] + targetleft/rho;
-          cuts[i+1] = split[i+1] - targetright/rho;
-        } else {
-          targetleft = damp * (mycount-leftcount)/3.0;
-          targetright = damp *
-            (leftcount-rightcount + (mycount-leftcount)/3.0);
-          cuts[i] = split[i] + targetleft/rho;
-          cuts[i+1] = split[i+1] - targetright/rho;
-        }
-      } else if (leftcount < rightcount) {
-        target = damp * 0.5*(mycount-leftcount);
-        cuts[i] = split[i] + target/rho;
-        cuts[i+1] = split[i+1];
-      } else if (rightcount < leftcount) {
-        target = damp * 0.5*(mycount-rightcount);
-        cuts[i+1] = split[i+1] - target/rho;
-        cuts[i] = split[i];
-      }
-
-    // middle slice has more atoms than only left or right slice
-    // send atoms only in that dir
-
-    } else if (mycount > leftcount) {
-      target = damp * 0.5*(mycount-leftcount);
-      cuts[i] = split[i] + target/rho;
-    } else if (mycount > rightcount) {
-      target = damp * 0.5*(mycount-rightcount);
-      cuts[i+1] = split[i+1] - target/rho;
-    }
-    */
-  }
-
-  // overwrite adjustable splits with new cuts
-
-  for (int i = 1; i < n; i++) split[i] = cuts[i];
-}
-
 /* ----------------------------------------------------------------------
    binary search for where value falls in N-length vec
    note that vec actually has N+1 values, but ignore last one
diff --git a/src/balance.h b/src/balance.h
index fda682cc98..ee1d171546 100644
--- a/src/balance.h
+++ b/src/balance.h
@@ -68,7 +68,6 @@ class Balance : protected Pointers {
   double imbalance_splits(int &);
   void tally(int, int, double *);
   int adjust(int, double *);
-  void old_adjust(int, int, bigint *, double *);
   int binary(double, int, double *);
   void debug_output(int, int, int, double *);
 };
diff --git a/src/body.h b/src/body.h
index 606615e56e..4e36a86155 100644
--- a/src/body.h
+++ b/src/body.h
@@ -35,10 +35,10 @@ class Body : protected Pointers {
 
   // methods implemented by child classes
 
-  virtual int pack_comm_body(class AtomVecBody::Bonus *, double *) {return 0;}
-  virtual int unpack_comm_body(class AtomVecBody::Bonus *, double *) {return 0;}
-  virtual int pack_border_body(class AtomVecBody::Bonus *, double *) {return 0;}
-  virtual int unpack_border_body(class AtomVecBody::Bonus *, 
+  virtual int pack_comm_body(struct AtomVecBody::Bonus *, double *) {return 0;}
+  virtual int unpack_comm_body(struct AtomVecBody::Bonus *, double *) {return 0;}
+  virtual int pack_border_body(struct AtomVecBody::Bonus *, double *) {return 0;}
+  virtual int unpack_border_body(struct AtomVecBody::Bonus *, 
                                  double *) {return 0;}
 
   virtual void data_body(int, int, int, char **, char **) = 0;
diff --git a/src/dump_cfg.cpp b/src/dump_cfg.cpp
index a6f05dcfcf..558512ace7 100755
--- a/src/dump_cfg.cpp
+++ b/src/dump_cfg.cpp
@@ -273,9 +273,9 @@ void DumpCFG::write_lines(int n, double *mybuf)
     for (i = 0; i < n; i++) {
       for (j = 0; j < size_one; j++) {
         if (j == 0) {
-	  fprintf(fp,"%f \n",mybuf[m]);
+          fprintf(fp,"%f \n",mybuf[m]);
         } else if (j == 1) {
-	  fprintf(fp,"%s \n",typenames[(int) mybuf[m]]);
+          fprintf(fp,"%s \n",typenames[(int) mybuf[m]]);
         } else if (j >= 2) {
           if (vtype[j] == INT) 
             fprintf(fp,vformat[j],static_cast<int> (mybuf[m]));
@@ -296,9 +296,9 @@ void DumpCFG::write_lines(int n, double *mybuf)
     for (i = 0; i < n; i++) {
       for (j = 0; j < size_one; j++) {
         if (j == 0) {
-	  fprintf(fp,"%f \n",mybuf[m]);
+          fprintf(fp,"%f \n",mybuf[m]);
         } else if (j == 1) {
-	  fprintf(fp,"%s \n",typenames[(int) mybuf[m]]);
+          fprintf(fp,"%s \n",typenames[(int) mybuf[m]]);
         } else if (j >= 2 && j <= 4) {
           unwrap_coord = (mybuf[m] - 0.5)/UNWRAPEXPAND + 0.5;
           fprintf(fp,vformat[j],unwrap_coord);
diff --git a/src/dump_dcd.h b/src/dump_dcd.h
index 47b6afe0be..c04c760a01 100644
--- a/src/dump_dcd.h
+++ b/src/dump_dcd.h
@@ -27,7 +27,7 @@ namespace LAMMPS_NS {
 class DumpDCD : public Dump {
  public:
   DumpDCD(LAMMPS *, int, char**);
-  ~DumpDCD();
+  virtual ~DumpDCD();
 
  private:
   int natoms,ntotal;
diff --git a/src/dump_local.h b/src/dump_local.h
index dec5e27834..3e06906e9a 100644
--- a/src/dump_local.h
+++ b/src/dump_local.h
@@ -27,7 +27,7 @@ namespace LAMMPS_NS {
 class DumpLocal : public Dump {
  public:
   DumpLocal(LAMMPS *, int, char **);
-  ~DumpLocal();
+  virtual ~DumpLocal();
 
  private:
   int nevery;                // dump frequency to check Fix against
diff --git a/src/dump_xyz.h b/src/dump_xyz.h
index 879b46e051..17503d53f1 100644
--- a/src/dump_xyz.h
+++ b/src/dump_xyz.h
@@ -27,7 +27,7 @@ namespace LAMMPS_NS {
 class DumpXYZ : public Dump {
  public:
   DumpXYZ(class LAMMPS *, int, char**);
-  ~DumpXYZ();
+  virtual ~DumpXYZ();
 
  protected:
   int ntypes;
diff --git a/src/fix_adapt.cpp b/src/fix_adapt.cpp
index 2f1793dfba..3f04cb8254 100644
--- a/src/fix_adapt.cpp
+++ b/src/fix_adapt.cpp
@@ -200,8 +200,16 @@ void FixAdapt::init()
 
     if (ad->which == PAIR) {
       anypair = 1;
+      Pair *pair = NULL;
 
-      Pair *pair = force->pair_match(ad->pstyle,1);
+      if (lmp->suffix_enable) {
+        char psuffix[128];
+        strcpy(psuffix,ad->pstyle);
+        strcat(psuffix,"/");
+        strcat(psuffix,lmp->suffix);
+        pair = force->pair_match(psuffix,1);
+      }
+      if (pair == NULL) pair = force->pair_match(ad->pstyle,1);
       if (pair == NULL) error->all(FLERR,"Fix adapt pair style does not exist");
       void *ptr = pair->extract(ad->pparam,ad->pdim);
       if (ptr == NULL) 
diff --git a/src/fix_wall_harmonic.h b/src/fix_wall_harmonic.h
index d8d7b309d6..b9efea7400 100644
--- a/src/fix_wall_harmonic.h
+++ b/src/fix_wall_harmonic.h
@@ -1,4 +1,4 @@
-/* ----------------------------------------------------------------------
+/* -*- c++ -*- ----------------------------------------------------------
    LAMMPS - Large-scale Atomic/Molecular Massively Parallel Simulator
    http://lammps.sandia.gov, Sandia National Laboratories
    Steve Plimpton, sjplimp@sandia.gov
@@ -29,9 +29,6 @@ class FixWallHarmonic : public FixWall {
   FixWallHarmonic(class LAMMPS *, int, char **);
   void precompute(int) {}
   void wall_particle(int, int, double);
-
- private:
-  double offset[6];
 };
 
 }
diff --git a/src/lmpwindows.h b/src/lmpwindows.h
index 9d3cbc6151..50f64eafb2 100644
--- a/src/lmpwindows.h
+++ b/src/lmpwindows.h
@@ -37,9 +37,16 @@ inline double trunc(double x) {
 }
 
 // Windows version of mkdir function does not have permission flags
+#ifndef S_IRWXU
 # define S_IRWXU 0
+#endif
+#ifndef S_IRGRP
 # define S_IRGRP 0
+#endif
+#ifndef S_IXGRP
 # define S_IXGRP 0
+#endif
 inline int mkdir(const char *path, int){
   return _mkdir(path);
 }
+
diff --git a/src/main.cpp b/src/main.cpp
index 471be9aa46..fd8a4ebe3f 100644
--- a/src/main.cpp
+++ b/src/main.cpp
@@ -27,8 +27,10 @@ int main(int argc, char **argv)
   MPI_Init(&argc,&argv);
 
   LAMMPS *lammps = new LAMMPS(argc,argv,MPI_COMM_WORLD);
+
   lammps->input->file();
   delete lammps;
 
+  MPI_Barrier(MPI_COMM_WORLD);
   MPI_Finalize();
 }
diff --git a/src/neighbor.cpp b/src/neighbor.cpp
index 3a05f7e85b..731278a87e 100644
--- a/src/neighbor.cpp
+++ b/src/neighbor.cpp
@@ -320,7 +320,9 @@ void Neighbor::init()
     special_flag[3] = 1;
   else special_flag[3] = 2;
 
-  if (force->kspace) special_flag[1] = special_flag[2] = special_flag[3] = 2;
+  if (force->kspace || force->pair_match("coul/wolf",0)
+      || force->pair_match("coul/dsf",0))
+     special_flag[1] = special_flag[2] = special_flag[3] = 2;
 
   // maxwt = max multiplicative factor on atom indices stored in neigh list
 
diff --git a/src/pair_born_coul_wolf.cpp b/src/pair_born_coul_wolf.cpp
index fc2feeaa56..4ecaa88340 100644
--- a/src/pair_born_coul_wolf.cpp
+++ b/src/pair_born_coul_wolf.cpp
@@ -37,6 +37,7 @@ using namespace MathConst;
 PairBornCoulWolf::PairBornCoulWolf(LAMMPS *lmp) : Pair(lmp)
 {
   writedata = 1;
+  single_enable = 0;
 }
 
 /* ---------------------------------------------------------------------- */
diff --git a/src/pair_coul_dsf.cpp b/src/pair_coul_dsf.cpp
index 12c3f58ea5..09550cae11 100644
--- a/src/pair_coul_dsf.cpp
+++ b/src/pair_coul_dsf.cpp
@@ -43,7 +43,10 @@ using namespace MathConst;
 
 /* ---------------------------------------------------------------------- */
 
-PairCoulDSF::PairCoulDSF(LAMMPS *lmp) : Pair(lmp) {}
+PairCoulDSF::PairCoulDSF(LAMMPS *lmp) : Pair(lmp)
+{
+  single_enable = 0;
+}
 
 /* ---------------------------------------------------------------------- */
 
@@ -59,7 +62,7 @@ PairCoulDSF::~PairCoulDSF()
 
 void PairCoulDSF::compute(int eflag, int vflag)
 {
-  int i,j,ii,jj,inum,jnum,itype,jtype;
+  int i,j,ii,jj,inum,jnum;
   double qtmp,xtmp,ytmp,ztmp,delx,dely,delz,ecoul,fpair;
   double r,rsq,r2inv,forcecoul,factor_coul;
   double prefactor,erfcc,erfcd,e_self,t;
@@ -72,7 +75,6 @@ void PairCoulDSF::compute(int eflag, int vflag)
   double **x = atom->x;
   double **f = atom->f;
   double *q = atom->q;
-  int *type = atom->type;
   int nlocal = atom->nlocal;
   double *special_coul = force->special_coul;
   int newton_pair = force->newton_pair;
@@ -91,11 +93,10 @@ void PairCoulDSF::compute(int eflag, int vflag)
     xtmp = x[i][0];
     ytmp = x[i][1];
     ztmp = x[i][2];
-    itype = type[i];
     jlist = firstneigh[i];
     jnum = numneigh[i];
     
-    if (evflag) {
+    if (eflag) {
       e_self = -(e_shift/2.0 + alpha/MY_PIS) * qtmp*qtmp*qqrd2e;
       ev_tally(i,i,nlocal,0,0.0,e_self,0.0,0.0,0.0,0.0);
     }
@@ -109,20 +110,17 @@ void PairCoulDSF::compute(int eflag, int vflag)
       dely = ytmp - x[j][1];
       delz = ztmp - x[j][2];
       rsq = delx*delx + dely*dely + delz*delz;
-      jtype = type[j];
 
-      if (rsq < cutsq[itype][jtype]) {
+      if (rsq < cut_coulsq) {
         r2inv = 1.0/rsq;
 
-        if (rsq < cut_coulsq) {
-          r = sqrt(rsq);
-          prefactor = factor_coul * qqrd2e*qtmp*q[j]/r;
-          erfcd = exp(-alpha*alpha*rsq);
-          t = 1.0 / (1.0 + EWALD_P*alpha*r);
-          erfcc = t * (A1+t*(A2+t*(A3+t*(A4+t*A5)))) * erfcd;
-          forcecoul = prefactor * (erfcc/r + 2.0*alpha/MY_PIS * erfcd + 
-            r*f_shift) * r;
-        }
+        r = sqrt(rsq);
+        prefactor = factor_coul * qqrd2e*qtmp*q[j]/r;
+        erfcd = exp(-alpha*alpha*rsq);
+        t = 1.0 / (1.0 + EWALD_P*alpha*r);
+        erfcc = t * (A1+t*(A2+t*(A3+t*(A4+t*A5)))) * erfcd;
+        forcecoul = prefactor * (erfcc/r + 2.0*alpha/MY_PIS * erfcd + 
+                                 r*f_shift) * r;
 
         fpair = forcecoul * r2inv;
         f[i][0] += delx*fpair;
@@ -135,10 +133,8 @@ void PairCoulDSF::compute(int eflag, int vflag)
         }
 
         if (eflag) {
-          if (rsq < cut_coulsq) {
-            ecoul = prefactor * (erfcc - r*e_shift - rsq*f_shift);
-          } else ecoul = 0.0;
-        }
+          ecoul = prefactor * (erfcc - r*e_shift - rsq*f_shift);
+        } else ecoul = 0.0;
 
         if (evflag) ev_tally(i,j,nlocal,newton_pair,
                              0.0,ecoul,fpair,delx,dely,delz);
diff --git a/src/pair_coul_wolf.cpp b/src/pair_coul_wolf.cpp
index 5c6b32013c..973d1e2c2d 100644
--- a/src/pair_coul_wolf.cpp
+++ b/src/pair_coul_wolf.cpp
@@ -34,7 +34,10 @@ using namespace MathConst;
 
 /* ---------------------------------------------------------------------- */
 
-PairCoulWolf::PairCoulWolf(LAMMPS *lmp) : Pair(lmp) {}
+PairCoulWolf::PairCoulWolf(LAMMPS *lmp) : Pair(lmp)
+{
+  single_enable = 0;
+}
 
 /* ---------------------------------------------------------------------- */
 
diff --git a/src/pair_lj_cut_coul_dsf.cpp b/src/pair_lj_cut_coul_dsf.cpp
index 63ec7b002b..a18e9ccf6b 100644
--- a/src/pair_lj_cut_coul_dsf.cpp
+++ b/src/pair_lj_cut_coul_dsf.cpp
@@ -43,7 +43,10 @@ using namespace MathConst;
 
 /* ---------------------------------------------------------------------- */
 
-PairLJCutCoulDSF::PairLJCutCoulDSF(LAMMPS *lmp) : Pair(lmp) {}
+PairLJCutCoulDSF::PairLJCutCoulDSF(LAMMPS *lmp) : Pair(lmp)
+{
+  single_enable = 0;
+}
 
 /* ---------------------------------------------------------------------- */
 
@@ -72,7 +75,7 @@ void PairLJCutCoulDSF::compute(int eflag, int vflag)
   int i,j,ii,jj,inum,jnum,itype,jtype;
   double qtmp,xtmp,ytmp,ztmp,delx,dely,delz,evdwl,ecoul,fpair;
   double r,rsq,r2inv,r6inv,forcecoul,forcelj,factor_coul,factor_lj;
-  double prefactor,erfcc,erfcd,e_self,t;
+  double prefactor,erfcc,erfcd,t;
   int *ilist,*jlist,*numneigh,**firstneigh;
   
   evdwl = ecoul = 0.0;
@@ -84,8 +87,8 @@ void PairLJCutCoulDSF::compute(int eflag, int vflag)
   double *q = atom->q;
   int *type = atom->type;
   int nlocal = atom->nlocal;
-  double *special_coul = force->special_coul;
   double *special_lj = force->special_lj;
+  double *special_coul = force->special_coul;
   int newton_pair = force->newton_pair;
   double qqrd2e = force->qqrd2e;
 
@@ -106,8 +109,8 @@ void PairLJCutCoulDSF::compute(int eflag, int vflag)
     jlist = firstneigh[i];
     jnum = numneigh[i];
     
-    if (evflag) {
-      e_self = -(e_shift/2.0 + alpha/MY_PIS) * qtmp*qtmp*qqrd2e;
+    if (eflag) {
+      double e_self = -(e_shift/2.0 + alpha/MY_PIS) * qtmp*qtmp*qqrd2e;
       ev_tally(i,i,nlocal,0,0.0,e_self,0.0,0.0,0.0,0.0);
     }
 
@@ -139,7 +142,7 @@ void PairLJCutCoulDSF::compute(int eflag, int vflag)
           erfcc = t * (A1+t*(A2+t*(A3+t*(A4+t*A5)))) * erfcd;
           forcecoul = prefactor * (erfcc/r + 2.0*alpha/MY_PIS * erfcd + 
             r*f_shift) * r;
-        }
+        } else forcecoul = 0.0;
 
         fpair = (forcecoul + factor_lj*forcelj) * r2inv;
         f[i][0] += delx*fpair;