diff --git a/lib/gpu/lal_answer.cpp b/lib/gpu/lal_answer.cpp
index bd8c7ef843..aa6d33d334 100644
--- a/lib/gpu/lal_answer.cpp
+++ b/lib/gpu/lal_answer.cpp
@@ -15,7 +15,7 @@
 
 #include "lal_answer.h"
 
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define AnswerT Answer<numtyp,acctyp>
 
 template <class numtyp, class acctyp>
@@ -311,4 +311,4 @@ void AnswerT::cq(const int cq_index) {
 }
 
 template class Answer<PRECISION,ACC_PRECISION>;
-
+}
diff --git a/lib/gpu/lal_atom.cpp b/lib/gpu/lal_atom.cpp
index 222ba0525e..bc25c00d93 100644
--- a/lib/gpu/lal_atom.cpp
+++ b/lib/gpu/lal_atom.cpp
@@ -15,7 +15,7 @@
 
 #include "lal_atom.h"
 
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define AtomT Atom<numtyp,acctyp>
 
 template <class numtyp, class acctyp>
@@ -349,4 +349,4 @@ void AtomT::compile_kernels(UCL_Device &dev) {
 #endif
 
 template class Atom<PRECISION,ACC_PRECISION>;
-
+}
diff --git a/lib/gpu/lal_base_atomic.cpp b/lib/gpu/lal_base_atomic.cpp
index da54f1dca3..4aadd3754c 100644
--- a/lib/gpu/lal_base_atomic.cpp
+++ b/lib/gpu/lal_base_atomic.cpp
@@ -14,7 +14,8 @@
  ***************************************************************************/
 
 #include "lal_base_atomic.h"
-using namespace LAMMPS_AL;
+
+namespace LAMMPS_AL {
 #define BaseAtomicT BaseAtomic<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> global_device;
@@ -285,4 +286,4 @@ void BaseAtomicT::compile_kernels(UCL_Device &dev, const void *pair_str,
 }
 
 template class BaseAtomic<PRECISION,ACC_PRECISION>;
-
+}
diff --git a/lib/gpu/lal_base_charge.cpp b/lib/gpu/lal_base_charge.cpp
index a3ec710baa..760e759201 100644
--- a/lib/gpu/lal_base_charge.cpp
+++ b/lib/gpu/lal_base_charge.cpp
@@ -15,7 +15,7 @@
  ***************************************************************************/
 
 #include "lal_base_charge.h"
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define BaseChargeT BaseCharge<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> global_device;
@@ -302,4 +302,4 @@ void BaseChargeT::compile_kernels(UCL_Device &dev, const void *pair_str,
 }
 
 template class BaseCharge<PRECISION,ACC_PRECISION>;
-
+}
diff --git a/lib/gpu/lal_base_dipole.cpp b/lib/gpu/lal_base_dipole.cpp
index 9fc7e1b235..56dcaf8e12 100644
--- a/lib/gpu/lal_base_dipole.cpp
+++ b/lib/gpu/lal_base_dipole.cpp
@@ -15,7 +15,7 @@
  ***************************************************************************/
 
 #include "lal_base_dipole.h"
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define BaseDipoleT BaseDipole<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> global_device;
@@ -311,4 +311,4 @@ void BaseDipoleT::compile_kernels(UCL_Device &dev, const void *pair_str,
 }
 
 template class BaseDipole<PRECISION,ACC_PRECISION>;
-
+}
diff --git a/lib/gpu/lal_base_dpd.cpp b/lib/gpu/lal_base_dpd.cpp
index eb5c2088a6..66c8cf09e9 100644
--- a/lib/gpu/lal_base_dpd.cpp
+++ b/lib/gpu/lal_base_dpd.cpp
@@ -15,7 +15,7 @@
  ***************************************************************************/
 
 #include "lal_base_dpd.h"
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define BaseDPDT BaseDPD<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> global_device;
@@ -308,4 +308,4 @@ void BaseDPDT::compile_kernels(UCL_Device &dev, const void *pair_str,
 }
 
 template class BaseDPD<PRECISION,ACC_PRECISION>;
-
+}
diff --git a/lib/gpu/lal_base_ellipsoid.cpp b/lib/gpu/lal_base_ellipsoid.cpp
index eea5344e33..b8d0b7a666 100644
--- a/lib/gpu/lal_base_ellipsoid.cpp
+++ b/lib/gpu/lal_base_ellipsoid.cpp
@@ -15,7 +15,7 @@
 
 #include "lal_base_ellipsoid.h"
 #include <cstdlib>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 
 #if defined(USE_OPENCL)
 #include "ellipsoid_nbor_cl.h"
@@ -488,4 +488,4 @@ void BaseEllipsoidT::compile_kernels(UCL_Device &dev,
 }
 
 template class BaseEllipsoid<PRECISION,ACC_PRECISION>;
-
+}
diff --git a/lib/gpu/lal_base_three.cpp b/lib/gpu/lal_base_three.cpp
index 0510b84d92..dc5678dd24 100644
--- a/lib/gpu/lal_base_three.cpp
+++ b/lib/gpu/lal_base_three.cpp
@@ -14,7 +14,7 @@
  ***************************************************************************/
 
 #include "lal_base_three.h"
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define BaseThreeT BaseThree<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> global_device;
@@ -397,4 +397,4 @@ void BaseThreeT::compile_kernels(UCL_Device &dev, const void *pair_str,
 }
 
 template class BaseThree<PRECISION,ACC_PRECISION>;
-
+}
diff --git a/lib/gpu/lal_beck.cpp b/lib/gpu/lal_beck.cpp
index 165a02b71a..be1722c32c 100644
--- a/lib/gpu/lal_beck.cpp
+++ b/lib/gpu/lal_beck.cpp
@@ -23,7 +23,7 @@ const char *beck=0;
 
 #include "lal_beck.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define BeckT Beck<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -150,3 +150,4 @@ void BeckT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class Beck<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_born.cpp b/lib/gpu/lal_born.cpp
index 7c1ed944d3..4a6b789687 100644
--- a/lib/gpu/lal_born.cpp
+++ b/lib/gpu/lal_born.cpp
@@ -23,7 +23,7 @@ const char *born=0;
 
 #include "lal_born.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define BornT Born<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -179,3 +179,4 @@ void BornT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class Born<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_born_coul_long.cpp b/lib/gpu/lal_born_coul_long.cpp
index 116d44d58f..1b147395f6 100644
--- a/lib/gpu/lal_born_coul_long.cpp
+++ b/lib/gpu/lal_born_coul_long.cpp
@@ -23,7 +23,7 @@ const char *born_coul_long=0;
 
 #include "lal_born_coul_long.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define BornCoulLongT BornCoulLong<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -173,3 +173,4 @@ void BornCoulLongT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class BornCoulLong<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_born_coul_long_cs.cpp b/lib/gpu/lal_born_coul_long_cs.cpp
index e7fb946f14..24de1e3a06 100644
--- a/lib/gpu/lal_born_coul_long_cs.cpp
+++ b/lib/gpu/lal_born_coul_long_cs.cpp
@@ -23,7 +23,7 @@ const char *born_coul_long_cs=0;
 
 #include "lal_born_coul_long_cs.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define BornCoulLongCST BornCoulLongCS<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -93,3 +93,4 @@ int BornCoulLongCST::init(const int ntypes, double **host_cutsq, double **host_r
 }
 
 template class BornCoulLongCS<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_born_coul_wolf.cpp b/lib/gpu/lal_born_coul_wolf.cpp
index c44b841463..1624dd9d50 100644
--- a/lib/gpu/lal_born_coul_wolf.cpp
+++ b/lib/gpu/lal_born_coul_wolf.cpp
@@ -23,7 +23,7 @@ const char *born_coul_wolf=0;
 
 #include "lal_born_coul_wolf.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define BornCoulWolfT BornCoulWolf<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -174,3 +174,4 @@ void BornCoulWolfT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class BornCoulWolf<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_born_coul_wolf_cs.cpp b/lib/gpu/lal_born_coul_wolf_cs.cpp
index bdb1c31e55..8deceeb1f4 100644
--- a/lib/gpu/lal_born_coul_wolf_cs.cpp
+++ b/lib/gpu/lal_born_coul_wolf_cs.cpp
@@ -23,7 +23,7 @@ const char *born_coul_wolf_cs=0;
 
 #include "lal_born_coul_wolf_cs.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define BornCoulWolfCST BornCoulWolfCS<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -95,3 +95,4 @@ int BornCoulWolfCST::init(const int ntypes, double **host_cutsq, double **host_r
 }
 
 template class BornCoulWolfCS<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_buck.cpp b/lib/gpu/lal_buck.cpp
index 0da4068d51..5a335a1e51 100644
--- a/lib/gpu/lal_buck.cpp
+++ b/lib/gpu/lal_buck.cpp
@@ -23,7 +23,7 @@ const char *buck=0;
 
 #include "lal_buck.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define BuckT Buck<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -168,3 +168,4 @@ void BuckT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class Buck<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_buck_coul.cpp b/lib/gpu/lal_buck_coul.cpp
index e4f829fc5c..25607eae17 100644
--- a/lib/gpu/lal_buck_coul.cpp
+++ b/lib/gpu/lal_buck_coul.cpp
@@ -23,7 +23,7 @@ const char *buck_coul=0;
 
 #include "lal_buck_coul.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define BuckCoulT BuckCoul<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -161,3 +161,4 @@ void BuckCoulT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class BuckCoul<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_buck_coul_long.cpp b/lib/gpu/lal_buck_coul_long.cpp
index 81faada116..1c0288c2d8 100644
--- a/lib/gpu/lal_buck_coul_long.cpp
+++ b/lib/gpu/lal_buck_coul_long.cpp
@@ -23,7 +23,7 @@ const char *buck_coul_long=0;
 
 #include "lal_buck_coul_long.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define BuckCoulLongT BuckCoulLong<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -166,3 +166,4 @@ void BuckCoulLongT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class BuckCoulLong<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_charmm_long.cpp b/lib/gpu/lal_charmm_long.cpp
index 9cd032b3c6..a78996a7d5 100644
--- a/lib/gpu/lal_charmm_long.cpp
+++ b/lib/gpu/lal_charmm_long.cpp
@@ -23,7 +23,7 @@ const char *charmm_long=0;
 
 #include "lal_charmm_long.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define CHARMMLongT CHARMMLong<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -174,3 +174,4 @@ void CHARMMLongT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class CHARMMLong<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_colloid.cpp b/lib/gpu/lal_colloid.cpp
index fb2b643e5e..c441d50968 100644
--- a/lib/gpu/lal_colloid.cpp
+++ b/lib/gpu/lal_colloid.cpp
@@ -23,7 +23,7 @@ const char *colloid=0;
 
 #include "lal_colloid.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define ColloidT Colloid<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -179,3 +179,4 @@ void ColloidT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class Colloid<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_coul.cpp b/lib/gpu/lal_coul.cpp
index a06a29e610..3e29215c91 100644
--- a/lib/gpu/lal_coul.cpp
+++ b/lib/gpu/lal_coul.cpp
@@ -23,7 +23,7 @@ const char *coul=0;
 
 #include "lal_coul.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define CoulT Coul<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -164,3 +164,4 @@ void CoulT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class Coul<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_coul_debye.cpp b/lib/gpu/lal_coul_debye.cpp
index 9098aeacb1..08ceb99300 100644
--- a/lib/gpu/lal_coul_debye.cpp
+++ b/lib/gpu/lal_coul_debye.cpp
@@ -23,7 +23,7 @@ const char *coul_debye=0;
 
 #include "lal_coul_debye.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define CoulDebyeT CoulDebye<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -165,3 +165,4 @@ void CoulDebyeT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class CoulDebye<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_coul_dsf.cpp b/lib/gpu/lal_coul_dsf.cpp
index 32c4342fbe..fe1fbfede7 100644
--- a/lib/gpu/lal_coul_dsf.cpp
+++ b/lib/gpu/lal_coul_dsf.cpp
@@ -23,7 +23,7 @@ const char *coul_dsf=0;
 
 #include "lal_coul_dsf.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define CoulDSFT CoulDSF<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -151,3 +151,4 @@ void CoulDSFT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class CoulDSF<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_coul_long.cpp b/lib/gpu/lal_coul_long.cpp
index b4c6a44d2f..02097a2c61 100644
--- a/lib/gpu/lal_coul_long.cpp
+++ b/lib/gpu/lal_coul_long.cpp
@@ -23,7 +23,7 @@ const char *coul_long=0;
 
 #include "lal_coul_long.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define CoulLongT CoulLong<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> pair_gpu_device;
@@ -156,3 +156,4 @@ void CoulLongT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class CoulLong<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_coul_long_cs.cpp b/lib/gpu/lal_coul_long_cs.cpp
index 7afa0ae5d2..32b5691610 100644
--- a/lib/gpu/lal_coul_long_cs.cpp
+++ b/lib/gpu/lal_coul_long_cs.cpp
@@ -23,7 +23,7 @@ const char *coul_long_cs=0;
 
 #include "lal_coul_long_cs.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define CoulLongCST CoulLongCS<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> pair_gpu_device;
@@ -76,3 +76,4 @@ int CoulLongCST::init(const int ntypes, double **host_scale,
 }
 
 template class CoulLongCS<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_device.cpp b/lib/gpu/lal_device.cpp
index 9410cc5250..5bd306ea5b 100644
--- a/lib/gpu/lal_device.cpp
+++ b/lib/gpu/lal_device.cpp
@@ -30,7 +30,7 @@ const char *device=0;
 #include "device_cubin.h"
 #endif
 
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define DeviceT Device<numtyp, acctyp>
 
 template <class numtyp, class acctyp>
@@ -762,7 +762,9 @@ double DeviceT::host_memory_usage() const {
 
 template class Device<PRECISION,ACC_PRECISION>;
 Device<PRECISION,ACC_PRECISION> global_device;
+}
 
+using namespace LAMMPS_AL;
 int lmp_init_device(MPI_Comm world, MPI_Comm replica, const int first_gpu,
                     const int last_gpu, const int gpu_mode,
                     const double particle_split, const int nthreads,
@@ -781,4 +783,3 @@ double lmp_gpu_forces(double **f, double **tor, double *eatom,
                       double **vatom, double *virial, double &ecoul) {
   return global_device.fix_gpu(f,tor,eatom,vatom,virial,ecoul);
 }
-
diff --git a/lib/gpu/lal_dipole_lj.cpp b/lib/gpu/lal_dipole_lj.cpp
index c97b76c820..b0929e2ffb 100644
--- a/lib/gpu/lal_dipole_lj.cpp
+++ b/lib/gpu/lal_dipole_lj.cpp
@@ -23,7 +23,7 @@ const char *dipole_lj=0;
 
 #include "lal_dipole_lj.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define DipoleLJT DipoleLJ<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -168,3 +168,4 @@ void DipoleLJT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class DipoleLJ<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_dipole_lj_sf.cpp b/lib/gpu/lal_dipole_lj_sf.cpp
index a33f38084f..dcf95bb126 100644
--- a/lib/gpu/lal_dipole_lj_sf.cpp
+++ b/lib/gpu/lal_dipole_lj_sf.cpp
@@ -23,7 +23,7 @@ const char *dipole_lj_sf=0;
 
 #include "lal_dipole_lj_sf.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define DipoleLJSFT DipoleLJSF<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -168,3 +168,4 @@ void DipoleLJSFT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class DipoleLJSF<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_dipole_long_lj.cpp b/lib/gpu/lal_dipole_long_lj.cpp
index 251e1def92..9648e9b15e 100644
--- a/lib/gpu/lal_dipole_long_lj.cpp
+++ b/lib/gpu/lal_dipole_long_lj.cpp
@@ -23,7 +23,7 @@ const char *dipole_long_lj=0;
 
 #include "lal_dipole_long_lj.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define DipoleLongLJT DipoleLongLJ<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -171,3 +171,4 @@ void DipoleLongLJT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class DipoleLongLJ<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_dpd.cpp b/lib/gpu/lal_dpd.cpp
index 4f6f2d641f..c5cbc7eb53 100644
--- a/lib/gpu/lal_dpd.cpp
+++ b/lib/gpu/lal_dpd.cpp
@@ -23,7 +23,7 @@ const char *dpd=0;
 
 #include "lal_dpd.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define DPDT DPD<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -168,3 +168,4 @@ void DPDT::update_coeff(int ntypes, double **host_a0, double **host_gamma,
 }
 
 template class DPD<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_eam.cpp b/lib/gpu/lal_eam.cpp
index b83972f4db..8c81353f36 100644
--- a/lib/gpu/lal_eam.cpp
+++ b/lib/gpu/lal_eam.cpp
@@ -23,7 +23,7 @@ const char *eam=0;
 
 #include "lal_eam.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define EAMT EAM<numtyp, acctyp>
 
 
@@ -531,3 +531,4 @@ void EAMT::loop2(const bool _eflag, const bool _vflag) {
 }
 
 template class EAM<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_gauss.cpp b/lib/gpu/lal_gauss.cpp
index 1ef215d7ff..2f965758eb 100644
--- a/lib/gpu/lal_gauss.cpp
+++ b/lib/gpu/lal_gauss.cpp
@@ -23,7 +23,7 @@ const char *gauss=0;
 
 #include "lal_gauss.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define GaussT Gauss<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -159,3 +159,4 @@ void GaussT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class Gauss<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_gayberne.cpp b/lib/gpu/lal_gayberne.cpp
index ba15af672e..f17fc50f5f 100644
--- a/lib/gpu/lal_gayberne.cpp
+++ b/lib/gpu/lal_gayberne.cpp
@@ -26,7 +26,7 @@ const char *gayberne_lj=0;
 
 #include "lal_gayberne.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 
 #define GayBerneT GayBerne<numtyp, acctyp>
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -315,4 +315,4 @@ void GayBerneT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class GayBerne<PRECISION,ACC_PRECISION>;
-
+}
diff --git a/lib/gpu/lal_lj.cpp b/lib/gpu/lal_lj.cpp
index 978b33e5d7..5bd015e364 100644
--- a/lib/gpu/lal_lj.cpp
+++ b/lib/gpu/lal_lj.cpp
@@ -23,7 +23,7 @@ const char *lj=0;
 
 #include "lal_lj.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define LJT LJ<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -168,3 +168,4 @@ void LJT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class LJ<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_lj96.cpp b/lib/gpu/lal_lj96.cpp
index 191f211ae4..6f74cd0f19 100644
--- a/lib/gpu/lal_lj96.cpp
+++ b/lib/gpu/lal_lj96.cpp
@@ -23,7 +23,7 @@ const char *lj96=0;
 
 #include "lal_lj96.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define LJ96T LJ96<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -152,3 +152,4 @@ void LJ96T::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class LJ96<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_lj_class2_long.cpp b/lib/gpu/lal_lj_class2_long.cpp
index 497e5989ad..24b07212ed 100644
--- a/lib/gpu/lal_lj_class2_long.cpp
+++ b/lib/gpu/lal_lj_class2_long.cpp
@@ -23,7 +23,7 @@ const char *lj_class2_long=0;
 
 #include "lal_lj_class2_long.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 
 #define LJClass2LongT LJClass2Long<numtyp, acctyp>
 
@@ -164,4 +164,4 @@ void LJClass2LongT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class LJClass2Long<PRECISION,ACC_PRECISION>;
-
+}
diff --git a/lib/gpu/lal_lj_coul.cpp b/lib/gpu/lal_lj_coul.cpp
index a8255318bd..59ce9c5e61 100644
--- a/lib/gpu/lal_lj_coul.cpp
+++ b/lib/gpu/lal_lj_coul.cpp
@@ -23,7 +23,7 @@ const char *lj_coul=0;
 
 #include "lal_lj_coul.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define LJCoulT LJCoul<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -164,3 +164,4 @@ void LJCoulT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class LJCoul<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_lj_coul_debye.cpp b/lib/gpu/lal_lj_coul_debye.cpp
index 92167f314f..556a0a5cd3 100644
--- a/lib/gpu/lal_lj_coul_debye.cpp
+++ b/lib/gpu/lal_lj_coul_debye.cpp
@@ -23,7 +23,7 @@ const char *lj_coul_debye=0;
 
 #include "lal_lj_coul_debye.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define LJCoulDebyeT LJCoulDebye<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -166,3 +166,4 @@ void LJCoulDebyeT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class LJCoulDebye<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_lj_coul_long.cpp b/lib/gpu/lal_lj_coul_long.cpp
index 29d648bed2..66897a4aa7 100644
--- a/lib/gpu/lal_lj_coul_long.cpp
+++ b/lib/gpu/lal_lj_coul_long.cpp
@@ -23,7 +23,7 @@ const char *lj_coul_long=0;
 
 #include "lal_lj_coul_long.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define LJCoulLongT LJCoulLong<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -181,3 +181,4 @@ void LJCoulLongT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class LJCoulLong<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_lj_coul_msm.cpp b/lib/gpu/lal_lj_coul_msm.cpp
index 1358de9ee1..9a17d068ec 100644
--- a/lib/gpu/lal_lj_coul_msm.cpp
+++ b/lib/gpu/lal_lj_coul_msm.cpp
@@ -23,7 +23,7 @@ const char *lj_coul_msm=0;
 
 #include "lal_lj_coul_msm.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define LJCoulMSMT LJCoulMSM<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -198,3 +198,4 @@ void LJCoulMSMT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class LJCoulMSM<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_lj_cubic.cpp b/lib/gpu/lal_lj_cubic.cpp
index 21ea22845c..f8200ec037 100644
--- a/lib/gpu/lal_lj_cubic.cpp
+++ b/lib/gpu/lal_lj_cubic.cpp
@@ -23,7 +23,7 @@ const char *lj_cubic=0;
 
 #include "lal_lj_cubic.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define LJCubicT LJCubic<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -157,3 +157,4 @@ void LJCubicT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class LJCubic<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_lj_dsf.cpp b/lib/gpu/lal_lj_dsf.cpp
index 1efac3e821..b888f33f00 100644
--- a/lib/gpu/lal_lj_dsf.cpp
+++ b/lib/gpu/lal_lj_dsf.cpp
@@ -23,7 +23,7 @@ const char *lj_dsf=0;
 
 #include "lal_lj_dsf.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define LJDSFT LJDSF<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -166,3 +166,4 @@ void LJDSFT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class LJDSF<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_lj_expand.cpp b/lib/gpu/lal_lj_expand.cpp
index 34a4d71c0b..1c58cecfae 100644
--- a/lib/gpu/lal_lj_expand.cpp
+++ b/lib/gpu/lal_lj_expand.cpp
@@ -23,7 +23,7 @@ const char *lj_expand=0;
 
 #include "lal_lj_expand.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define LJExpandT LJExpand<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -171,3 +171,4 @@ void LJExpandT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class LJExpand<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_lj_expand_coul_long.cpp b/lib/gpu/lal_lj_expand_coul_long.cpp
index 32a4f8120e..3e5e00ef6a 100644
--- a/lib/gpu/lal_lj_expand_coul_long.cpp
+++ b/lib/gpu/lal_lj_expand_coul_long.cpp
@@ -23,7 +23,7 @@ const char *lj_expand_coul_long=0;
 
 #include "lal_lj_expand_coul_long.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define LJExpandCoulLongT LJExpandCoulLong<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -181,3 +181,4 @@ void LJExpandCoulLongT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class LJExpandCoulLong<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_lj_gromacs.cpp b/lib/gpu/lal_lj_gromacs.cpp
index 75f5a41917..0563151ddd 100644
--- a/lib/gpu/lal_lj_gromacs.cpp
+++ b/lib/gpu/lal_lj_gromacs.cpp
@@ -23,7 +23,7 @@ const char *lj_gromacs=0;
 
 #include "lal_lj_gromacs.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define LJGROMACST LJGROMACS<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -162,3 +162,4 @@ void LJGROMACST::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class LJGROMACS<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_lj_sdk.cpp b/lib/gpu/lal_lj_sdk.cpp
index 618555e38a..c6a282576c 100644
--- a/lib/gpu/lal_lj_sdk.cpp
+++ b/lib/gpu/lal_lj_sdk.cpp
@@ -23,7 +23,7 @@ const char *lj_sdk=0;
 
 #include "lal_lj_sdk.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define CGCMMT CGCMM<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -152,3 +152,4 @@ void CGCMMT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class CGCMM<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_lj_sdk_long.cpp b/lib/gpu/lal_lj_sdk_long.cpp
index 46caf6bd36..74dbfc40e3 100644
--- a/lib/gpu/lal_lj_sdk_long.cpp
+++ b/lib/gpu/lal_lj_sdk_long.cpp
@@ -23,7 +23,7 @@ const char *lj_sdk_long=0;
 
 #include "lal_lj_sdk_long.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define CGCMMLongT CGCMMLong<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -164,3 +164,4 @@ void CGCMMLongT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class CGCMMLong<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_mie.cpp b/lib/gpu/lal_mie.cpp
index 1510275047..394d1f8a2f 100644
--- a/lib/gpu/lal_mie.cpp
+++ b/lib/gpu/lal_mie.cpp
@@ -23,7 +23,7 @@ const char *mie=0;
 
 #include "lal_mie.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define MieT Mie<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -150,3 +150,4 @@ void MieT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class Mie<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_morse.cpp b/lib/gpu/lal_morse.cpp
index cbdf928863..09da65d252 100644
--- a/lib/gpu/lal_morse.cpp
+++ b/lib/gpu/lal_morse.cpp
@@ -23,7 +23,7 @@ const char *morse=0;
 
 #include "lal_morse.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define MorseT Morse<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -150,4 +150,4 @@ void MorseT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class Morse<PRECISION,ACC_PRECISION>;
-
+}
diff --git a/lib/gpu/lal_pppm.cpp b/lib/gpu/lal_pppm.cpp
index fefa1172ab..8b5012f312 100644
--- a/lib/gpu/lal_pppm.cpp
+++ b/lib/gpu/lal_pppm.cpp
@@ -25,7 +25,7 @@ const char *pppm_d=0;
 #include "lal_pppm.h"
 #include <cassert>
 
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define PPPMT PPPM<numtyp, acctyp, grdtyp, grdtyp4>
 
 extern Device<PRECISION,ACC_PRECISION> global_device;
@@ -402,3 +402,4 @@ void PPPMT::compile_kernels(UCL_Device &dev) {
 
 template class PPPM<PRECISION,ACC_PRECISION,float,_lgpu_float4>;
 template class PPPM<PRECISION,ACC_PRECISION,double,_lgpu_double4>;
+}
diff --git a/lib/gpu/lal_re_squared.cpp b/lib/gpu/lal_re_squared.cpp
index 9513f5a633..81dc3b13a4 100644
--- a/lib/gpu/lal_re_squared.cpp
+++ b/lib/gpu/lal_re_squared.cpp
@@ -26,7 +26,7 @@ const char *re_squared_lj=0;
 
 #include "lal_re_squared.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 
 #define RESquaredT RESquared<numtyp, acctyp>
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -315,4 +315,4 @@ void RESquaredT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class RESquared<PRECISION,ACC_PRECISION>;
-
+}
diff --git a/lib/gpu/lal_soft.cpp b/lib/gpu/lal_soft.cpp
index 727b112ea5..8e944fa0a5 100644
--- a/lib/gpu/lal_soft.cpp
+++ b/lib/gpu/lal_soft.cpp
@@ -23,7 +23,7 @@ const char *soft=0;
 
 #include "lal_soft.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define SoftT Soft<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -158,3 +158,4 @@ void SoftT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class Soft<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_sw.cpp b/lib/gpu/lal_sw.cpp
index 46b6382a60..5c7bd45c76 100644
--- a/lib/gpu/lal_sw.cpp
+++ b/lib/gpu/lal_sw.cpp
@@ -23,7 +23,7 @@ const char *sw=0;
 
 #include "lal_sw.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define SWT SW<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -262,4 +262,4 @@ void SWT::loop(const bool _eflag, const bool _vflag, const int evatom) {
 }
 
 template class SW<PRECISION,ACC_PRECISION>;
-
+}
diff --git a/lib/gpu/lal_table.cpp b/lib/gpu/lal_table.cpp
index 0de59c84b2..d07b2716e4 100644
--- a/lib/gpu/lal_table.cpp
+++ b/lib/gpu/lal_table.cpp
@@ -23,7 +23,7 @@ const char *table=0;
 
 #include "lal_table.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define TableT Table<numtyp, acctyp>
 
 #define LOOKUP 0
@@ -337,3 +337,4 @@ void TableT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class Table<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_tersoff.cpp b/lib/gpu/lal_tersoff.cpp
index ef55b98a2d..27143acb0c 100644
--- a/lib/gpu/lal_tersoff.cpp
+++ b/lib/gpu/lal_tersoff.cpp
@@ -23,7 +23,7 @@ const char *tersoff=0;
 
 #include "lal_tersoff.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define TersoffT Tersoff<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -329,4 +329,4 @@ void TersoffT::loop(const bool _eflag, const bool _vflag, const int evatom) {
 }
 
 template class Tersoff<PRECISION,ACC_PRECISION>;
-
+}
diff --git a/lib/gpu/lal_tersoff_mod.cpp b/lib/gpu/lal_tersoff_mod.cpp
index 3cbb488cab..182859bdd4 100644
--- a/lib/gpu/lal_tersoff_mod.cpp
+++ b/lib/gpu/lal_tersoff_mod.cpp
@@ -23,7 +23,7 @@ const char *tersoff_mod=0;
 
 #include "lal_tersoff_mod.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define TersoffMT TersoffMod<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -329,4 +329,4 @@ void TersoffMT::loop(const bool _eflag, const bool _vflag, const int evatom) {
 }
 
 template class TersoffMod<PRECISION,ACC_PRECISION>;
-
+}
diff --git a/lib/gpu/lal_tersoff_zbl.cpp b/lib/gpu/lal_tersoff_zbl.cpp
index ebf67285ed..92db59679e 100644
--- a/lib/gpu/lal_tersoff_zbl.cpp
+++ b/lib/gpu/lal_tersoff_zbl.cpp
@@ -23,7 +23,7 @@ const char *tersoff_zbl=0;
 
 #include "lal_tersoff_zbl.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define TersoffZT TersoffZBL<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -355,4 +355,4 @@ void TersoffZT::loop(const bool _eflag, const bool _vflag, const int evatom) {
 }
 
 template class TersoffZBL<PRECISION,ACC_PRECISION>;
-
+}
diff --git a/lib/gpu/lal_ufm.cpp b/lib/gpu/lal_ufm.cpp
index c7aa2cca39..9b7d42dec9 100644
--- a/lib/gpu/lal_ufm.cpp
+++ b/lib/gpu/lal_ufm.cpp
@@ -25,7 +25,7 @@ const char *ufm=0;
 
 #include "lal_ufm.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define UFMT UFM<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -170,3 +170,4 @@ void UFMT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class UFM<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_vashishta.cpp b/lib/gpu/lal_vashishta.cpp
index 5a01a9bd46..350ae8e8fd 100644
--- a/lib/gpu/lal_vashishta.cpp
+++ b/lib/gpu/lal_vashishta.cpp
@@ -23,7 +23,7 @@ const char *vashishta=0;
 
 #include "lal_vashishta.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define VashishtaT Vashishta<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -295,4 +295,4 @@ void VashishtaT::loop(const bool _eflag, const bool _vflag, const int evatom) {
 }
 
 template class Vashishta<PRECISION,ACC_PRECISION>;
-
+}
diff --git a/lib/gpu/lal_yukawa.cpp b/lib/gpu/lal_yukawa.cpp
index a316d195ac..453139e537 100644
--- a/lib/gpu/lal_yukawa.cpp
+++ b/lib/gpu/lal_yukawa.cpp
@@ -23,7 +23,7 @@ const char *yukawa=0;
 
 #include "lal_yukawa.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define YukawaT Yukawa<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -147,3 +147,4 @@ void YukawaT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class Yukawa<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_yukawa_colloid.cpp b/lib/gpu/lal_yukawa_colloid.cpp
index af29938a68..e71e962ffd 100644
--- a/lib/gpu/lal_yukawa_colloid.cpp
+++ b/lib/gpu/lal_yukawa_colloid.cpp
@@ -23,7 +23,7 @@ const char *yukawa_colloid=0;
 
 #include "lal_yukawa_colloid.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define YukawaColloidT YukawaColloid<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -289,3 +289,4 @@ void YukawaColloidT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class YukawaColloid<PRECISION,ACC_PRECISION>;
+}
diff --git a/lib/gpu/lal_zbl.cpp b/lib/gpu/lal_zbl.cpp
index 77e55a62f9..2bf3369174 100644
--- a/lib/gpu/lal_zbl.cpp
+++ b/lib/gpu/lal_zbl.cpp
@@ -23,7 +23,7 @@ const char *zbl=0;
 
 #include "lal_zbl.h"
 #include <cassert>
-using namespace LAMMPS_AL;
+namespace LAMMPS_AL {
 #define ZBLT ZBL<numtyp, acctyp>
 
 extern Device<PRECISION,ACC_PRECISION> device;
@@ -157,3 +157,4 @@ void ZBLT::loop(const bool _eflag, const bool _vflag) {
 }
 
 template class ZBL<PRECISION,ACC_PRECISION>;
+}
diff --git a/src/KOKKOS/neigh_list_kokkos.cpp b/src/KOKKOS/neigh_list_kokkos.cpp
index 1c78fe3a68..2b9c5ef645 100644
--- a/src/KOKKOS/neigh_list_kokkos.cpp
+++ b/src/KOKKOS/neigh_list_kokkos.cpp
@@ -35,8 +35,9 @@ template<class Device>
 void NeighListKokkos<Device>::grow(int nmax)
 {
   // skip if this list is already long enough to store nmax atoms
+  //  and maxneighs neighbors
 
-  if (nmax <= maxatoms) return;
+  if (nmax <= maxatoms && d_neighbors.extent(1) >= maxneighs) return;
   maxatoms = nmax;
 
   k_ilist =
diff --git a/src/USER-INTEL/fix_nh_intel.cpp b/src/USER-INTEL/fix_nh_intel.cpp
index bf4764f104..a072e823a4 100644
--- a/src/USER-INTEL/fix_nh_intel.cpp
+++ b/src/USER-INTEL/fix_nh_intel.cpp
@@ -335,7 +335,7 @@ void FixNHIntel::reset_dt()
   if (nlocal > _nlocal_max) {
     if (_nlocal_max) memory->destroy(_dtfm);
     _nlocal_max = static_cast<int>(1.20 * nlocal);
-    memory->create(_dtfm, _nlocal_max * 3, "fix_nve_intel:dtfm");
+    memory->create(_dtfm, _nlocal_max * 3, "fix_nh_intel:dtfm");
   }
 
   _nlocal3 = nlocal * 3;
diff --git a/src/USER-INTEL/fix_nve_intel.cpp b/src/USER-INTEL/fix_nve_intel.cpp
index 7dee31d2f4..f5d0513b65 100644
--- a/src/USER-INTEL/fix_nve_intel.cpp
+++ b/src/USER-INTEL/fix_nve_intel.cpp
@@ -75,6 +75,7 @@ void FixNVEIntel::initial_integrate(int /*vflag*/)
       x[i] += dtv * v[i];
     }
   } else if (igroup == 0) {
+    if (neighbor->ago == 0) reset_dt();
     #if defined(LMP_SIMD_COMPILER)
     #pragma vector aligned
     #pragma simd
@@ -84,6 +85,7 @@ void FixNVEIntel::initial_integrate(int /*vflag*/)
       x[i] += dtv * v[i];
     }
   } else {
+    if (neighbor->ago == 0) reset_dt();
     #if defined(LMP_SIMD_COMPILER)
     #pragma vector aligned
     #pragma simd
@@ -114,6 +116,15 @@ void FixNVEIntel::final_integrate()
     #endif
     for (int i = 0; i < _nlocal3; i++)
       v[i] += dtfm * f[i];
+  } else if (igroup == 0) {
+    if (neighbor->ago == 0) reset_dt();
+    #if defined(LMP_SIMD_COMPILER)
+    #pragma vector aligned
+    #pragma simd
+    #endif
+    for (int i = 0; i < _nlocal3; i++) {
+      v[i] += _dtfm[i] * f[i];
+    }
   } else {
     if (neighbor->ago == 0) reset_dt();
     #if defined(LMP_SIMD_COMPILER)
diff --git a/src/fix_langevin.cpp b/src/fix_langevin.cpp
index f8fea76c94..a050e5a13a 100644
--- a/src/fix_langevin.cpp
+++ b/src/fix_langevin.cpp
@@ -301,7 +301,6 @@ void FixLangevin::post_force(int /*vflag*/)
   // this avoids testing them inside inner loop:
   // TSTYLEATOM, GJF, TALLY, BIAS, RMASS, ZERO
 
-#ifdef TEMPLATED_FIX_LANGEVIN
   if (tstyle == ATOM)
     if (gjfflag)
       if (tallyflag)
@@ -428,10 +427,6 @@ void FixLangevin::post_force(int /*vflag*/)
           else
             if (zeroflag) post_force_templated<0,0,0,0,0,1>();
             else          post_force_templated<0,0,0,0,0,0>();
-#else
-  post_force_untemplated(int(tstyle==ATOM), gjfflag, tallyflag,
-                         int(tbiasflag==BIAS), int(rmass!=NULL), zeroflag);
-#endif
 }
 
 /* ---------------------------------------------------------------------- */
@@ -445,15 +440,9 @@ void FixLangevin::post_force_respa(int vflag, int ilevel, int /*iloop*/)
    modify forces using one of the many Langevin styles
 ------------------------------------------------------------------------- */
 
-#ifdef TEMPLATED_FIX_LANGEVIN
 template < int Tp_TSTYLEATOM, int Tp_GJF, int Tp_TALLY,
            int Tp_BIAS, int Tp_RMASS, int Tp_ZERO >
 void FixLangevin::post_force_templated()
-#else
-void FixLangevin::post_force_untemplated
-  (int Tp_TSTYLEATOM, int Tp_GJF, int Tp_TALLY,
-   int Tp_BIAS, int Tp_RMASS, int Tp_ZERO)
-#endif
 {
   double gamma1,gamma2;
 
diff --git a/src/fix_langevin.h b/src/fix_langevin.h
index 024e7a9539..4b5570ac2e 100644
--- a/src/fix_langevin.h
+++ b/src/fix_langevin.h
@@ -72,16 +72,10 @@ class FixLangevin : public Fix {
   class RanMars *random;
   int seed;
 
-  // comment next line to turn off templating
-#define TEMPLATED_FIX_LANGEVIN
-#ifdef TEMPLATED_FIX_LANGEVIN
   template < int Tp_TSTYLEATOM, int Tp_GJF, int Tp_TALLY,
              int Tp_BIAS, int Tp_RMASS, int Tp_ZERO >
   void post_force_templated();
-#else
-  void post_force_untemplated(int, int, int,
-                              int, int, int);
-#endif
+
   void omega_thermostat();
   void angmom_thermostat();
   void compute_target();