refactor: esolver to ctrl_scf_lcao

Author: dyzheng

source/source_esolver/esolver_ks_lcao.cpp

@@ -20,7 +20,6 @@
 #include "source_io/print_info.h"
 #include "source_lcao/rho_tau_lcao.h" // mohan add 20251024
 #include "source_lcao/LCAO_set.h" // mohan add 20251111
-#include "source_lcao/module_operator_lcao/overlap_new.h"
 
 namespace ModuleESolver
 {
@@ -555,37 +554,10 @@ void ESolver_KS_LCAO<TK, TR>::after_scf(UnitCell& ucell, const int istep, const
             PARAM.inp, this->kv, this->pelec, this->dmat.dm, this->pv,
             this->gd, this->psi, hamilt_lcao, this->dftu, this->two_center_bundle_,
             this->orb_, this->pw_wfc, this->pw_rho, this->pw_big, this->sf,
-            this->rdmft_solver, this->deepks, this->exx_nao, 
+            this->rdmft_solver, this->deepks, this->exx_nao,
             this->conv_esolver, this->scf_nmax_flag, istep);
 
-    //! 3) Calculate and output asynchronous overlap matrix for Hefei-NAMD
-    if (PARAM.inp.cal_syns && (istep > 0 || PARAM.inp.init_vel))
-    {
-        // Create a new OverlapNew instance specifically for SR_async calculation
-        // This allows SR_async to be initialized with velocity-shifted dtau
-        hamilt::OverlapNew<hamilt::OperatorLCAO<TK, TR>>* overlap_async =
-            new hamilt::OverlapNew<hamilt::OperatorLCAO<TK, TR>>(
-                nullptr,  // hsk_in: not needed for SR_async calculation
-                this->kv.kvec_d,
-                nullptr,  // hR_in: not needed for SR_async calculation
-                nullptr,  // SR_in: not needed for SR_async calculation
-                &ucell,
-                this->orb_.cutoffs(),
-                &this->gd,
-                this->two_center_bundle_.overlap_orb.get());
-
-        // Use same precision as DMR output (default 8 if not specified)
-        const int precision = PARAM.inp.out_dmr[0] > 0 ? PARAM.inp.out_dmr[1] : 8;
-        const Parallel_Orbitals* paraV = hamilt_lcao->getSR()->get_paraV();
-        hamilt::HContainer<TR>* SR_async = overlap_async->calculate_SR_async(ucell, PARAM.mdp.md_dt, paraV);
-        overlap_async->output_SR_async_csr(istep, SR_async, precision);
-
-        // Clean up
-        delete SR_async;
-        delete overlap_async;
-    }
-
-    //! 4) Clean up RA, which is used to serach for adjacent atoms
+    //! 3) Clean up RA, which is used to serach for adjacent atoms
     if (!PARAM.inp.cal_force && !PARAM.inp.cal_stress)
     {
         this->RA.delete_grid();

source/source_io/ctrl_scf_lcao.cpp

@@ -31,6 +31,7 @@
 #include "source_io/to_qo.h"                // use toQO
 #include "source_lcao/module_rdmft/rdmft.h" // use RDMFT codes
 #include "source_lcao/rho_tau_lcao.h"       // mohan add 2025-10-24
+#include "source_lcao/module_operator_lcao/overlap_new.h" // use hamilt::OverlapNew for NAMD
 
 template <typename TK, typename TR>
 void ModuleIO::ctrl_scf_lcao(UnitCell& ucell,
@@ -459,6 +460,40 @@ void ModuleIO::ctrl_scf_lcao(UnitCell& ucell,
         tqo.calculate();
     }
 
+    //------------------------------------------------------------------
+    //! 18) Calculate and output asynchronous overlap matrix for Hefei-NAMD
+    //------------------------------------------------------------------
+    if (inp.cal_syns && (istep > 0 || inp.init_vel))
+    {
+        ModuleBase::TITLE("ModuleIO", "output_namd_async_overlap");
+        ModuleBase::timer::tick("ModuleIO", "output_namd_async_overlap");
+
+        // Create a new OverlapNew instance specifically for SR_async calculation
+        // This allows SR_async to be initialized with velocity-shifted dtau
+        hamilt::OverlapNew<hamilt::OperatorLCAO<TK, TR>>* overlap_async =
+            new hamilt::OverlapNew<hamilt::OperatorLCAO<TK, TR>>(
+                nullptr,  // hsk_in: not needed for SR_async calculation
+                kv.kvec_d,
+                nullptr,  // hR_in: not needed for SR_async calculation
+                nullptr,  // SR_in: not needed for SR_async calculation
+                &ucell,
+                orb.cutoffs(),
+                &gd,
+                two_center_bundle.overlap_orb.get());
+
+        // Use same precision as DMR output (default 8 if not specified)
+        const int precision = inp.out_dmr[0] > 0 ? inp.out_dmr[1] : 8;
+        const Parallel_Orbitals* paraV = p_hamilt->getSR()->get_paraV();
+        hamilt::HContainer<TR>* SR_async = overlap_async->calculate_SR_async(ucell, PARAM.mdp.md_dt, paraV);
+        overlap_async->output_SR_async_csr(istep, SR_async, precision);
+
+        // Clean up
+        delete SR_async;
+        delete overlap_async;
+
+        ModuleBase::timer::tick("ModuleIO", "output_namd_async_overlap");
+    }
+
     ModuleBase::timer::tick("ModuleIO", "ctrl_scf_lcao");
 }