Merge 'origin/nonhydro/homme/thetah+pese+master' (PR #1824)

-reorg HOMME dycore: improvements to multi-model framework to support original and nonhydrostatic core

-Bring in theta-formulation nonhydrostatic dycore as unsupported option

-Add nonhydrostatic test cases (disabled by default)

[BFB]

config/acme/machines/Depends.cetus

@@ -13,8 +13,8 @@ shr_reprosum_mod.o: shr_reprosum_mod.F90
 
 # These routines benefit from -qnostrict without violating the bfb test
 PERFOBJS=\
-prim_advection_mod_base.o \
-vertremap_mod_base.o \
+prim_advection_base.o \
+vertremap_base.o \
 edge_mod_base.o \
 derivative_mod_base.o \
 bndry_mod_base.o \

config/acme/machines/Depends.intel

@@ -1,11 +1,16 @@
 #
 PERFOBJS=\
-prim_advection_mod_base.o \
-vertremap_mod_base.o \
+prim_advection_base.o \
+vertremap_base.o \
 edge_mod_base.o \
 derivative_mod_base.o \
 bndry_mod_base.o \
 prim_advance_mod.o \
+viscosity_preqx_base.o \
+viscosity_base.o \
+viscosity_theta.o \
+eos.o \
+hevi_mod.o \
 uwshcu.o
 
 # CLM's SatellitePhenologyMod is compiled incorrectly with intel 15.0.0 at -O2

config/acme/machines/Depends.intel18

@@ -1,13 +1,19 @@
 #
 PERFOBJS=\
-prim_advection_mod_base.o \
-vertremap_mod_base.o \
+prim_advection_base.o \
+vertremap_base.o \
 edge_mod_base.o \
 derivative_mod_base.o \
 bndry_mod_base.o \
 prim_advance_mod.o \
+viscosity_preqx_base.o \
+viscosity_base.o \
+viscosity_theta.o \
+eos.o \
+hevi_mod.o \
 uwshcu.o
 
+
 # shr_wv_sat_mod does not need to have better than ~0.1% precision, and benefits
 # enormously from a lower precision in the vector functions.
 REDUCED_PRECISION_OBJS=\

config/acme/machines/Depends.mira

@@ -13,8 +13,8 @@ shr_reprosum_mod.o: shr_reprosum_mod.F90
 
 # These routines benefit from -qnostrict without violating the bfb test
 PERFOBJS=\
-prim_advection_mod_base.o \
-vertremap_mod_base.o \
+prim_advection_base.o \
+vertremap_base.o \
 edge_mod_base.o \
 derivative_mod_base.o \
 bndry_mod_base.o \

config/acme/machines/Depends.summitdev.ibm

@@ -15,8 +15,8 @@ shr_reprosum_mod.o: shr_reprosum_mod.F90
 
 # These routines benefit from -qnostrict without violating the bfb test
 PERFOBJS=\
-prim_advection_mod_base.o \
-vertremap_mod_base.o \
+prim_advection_base.o \
+vertremap_base.o \
 edge_mod_base.o \
 derivative_mod_base.o \
 bndry_mod_base.o \

config/acme/machines/Depends.titan.pgiacc

@@ -18,6 +18,9 @@ edge_mod.o: edge_mod.F90
 element_mod.o: element_mod.F90
 	$(FC) -c $(INCLDIR) $(INCS) $(FFLAGS) -DUSE_OPENACC=1 -acc -ta=tesla,pinned,cc35,cuda7.5,ptxinfo -Minfo=accel $(FREEFLAGS) $<
 
+element_state.o: element_state.F90
+	$(FC) -c $(INCLDIR) $(INCS) $(FFLAGS) -DUSE_OPENACC=1 -acc -ta=tesla,pinned,cc35,cuda7.5,ptxinfo -Minfo=accel $(FREEFLAGS) $<
+
 openacc_utils_mod.o: openacc_utils_mod.F90
 	$(FC) -c $(INCLDIR) $(INCS) $(FFLAGS) -DUSE_OPENACC=1 -acc -ta=tesla,pinned,cc35,cuda7.5,ptxinfo -Minfo=accel $(FREEFLAGS) $<
 
@@ -30,7 +33,7 @@ prim_advection_mod.o: prim_advection_mod.F90
 prim_si_mod.o: prim_si_mod.F90
 	$(FC) -c $(INCLDIR) $(INCS) $(FFLAGS) -DUSE_OPENACC=1 -acc -ta=tesla,pinned,cc35,cuda7.5,ptxinfo -Minfo=accel $(FREEFLAGS) $<
 
-solver_init_mod.o: solver_init_mod.F90
+model_init_mod.o: model_init_mod.F90
 	$(FC) -c $(INCLDIR) $(INCS) $(FFLAGS) -DUSE_OPENACC=1 -acc -ta=tesla,pinned,cc35,cuda7.5,ptxinfo -Minfo=accel $(FREEFLAGS) $<
 
 vertremap_mod.o: vertremap_mod.F90
@@ -42,6 +45,9 @@ viscosity_mod.o: viscosity_mod.F90
 prim_driver_mod.o: prim_driver_mod.F90
 	$(FC) -c $(INCLDIR) $(INCS) $(FFLAGS) -DUSE_OPENACC=1 -acc -ta=tesla,pinned,cc35,cuda7.5,ptxinfo -Minfo=accel $(FREEFLAGS) $<
 
+prim_driver_base.o: prim_driver_base.F90
+	$(FC) -c $(INCLDIR) $(INCS) $(FFLAGS) -DUSE_OPENACC=1 -acc -ta=tesla,pinned,cc35,cuda7.5,ptxinfo -Minfo=accel $(FREEFLAGS) $<
+
 physics_mod.o: physics_mod.F90
 	$(FC) -c $(INCLDIR) $(INCS) $(FFLAGS) -DUSE_OPENACC=1 -acc -ta=tesla,pinned,cc35,cuda7.5,ptxinfo -Minfo=accel $(FREEFLAGS) $<