Merge TD-Exx into hytd branch

.github/workflows/coverage.yml

@@ -15,10 +15,10 @@ jobs:
         uses: actions/checkout@v4
       - name: Install Requirements for Coverage Testing
         run: |
-          apt update && apt install -y lcov
+          apt update && apt install -y lcov gpg
       - name: Building
         run: |
-          cmake -B build -DENABLE_DEEPKS=ON -DENABLE_LIBXC=ON -DBUILD_TESTING=ON -DENABLE_COVERAGE=ON
+          cmake -B build -DENABLE_COVERAGE=ON -DBUILD_TESTING=ON -DENABLE_DEEPKS=ON -DENABLE_LIBXC=ON -DENABLE_LIBRI=ON -DENABLE_PAW=ON -DENABLE_GOOGLEBENCH=ON -DENABLE_RAPIDJSON=ON
           cmake --build build -j`nproc`
           cmake --install build
       - name: Testing

CMakeLists.txt

@@ -39,6 +39,7 @@ option(ENABLE_RAPIDJSON "Enable rapid-json usage." OFF)
 option(ENABLE_CNPY "Enable cnpy usage." OFF)
 option(ENABLE_PEXSI "Enable support for PEXSI." OFF)
 option(ENABLE_CUSOLVERMP "Enable cusolvermp." OFF)
+option(USE_DSP "Enable DSP usage." OFF)
 
 # enable json support
 if(ENABLE_RAPIDJSON)
@@ -119,6 +120,12 @@ elseif(ENABLE_LCAO AND NOT ENABLE_MPI)
   set(ABACUS_BIN_NAME abacus_serial)
 endif()
 
+if (USE_DSP)
+  set(USE_ELPA OFF)
+  set(ENABLE_LCAO OFF)
+  set(ABACUS_BIN_NAME abacus_dsp)
+endif()
+
 list(APPEND CMAKE_MODULE_PATH ${CMAKE_CURRENT_SOURCE_DIR}/cmake)
 
 if(ENABLE_COVERAGE)
@@ -240,6 +247,11 @@ if(ENABLE_MPI)
   list(APPEND math_libs MPI::MPI_CXX)
 endif()
 
+if (USE_DSP)
+  target_link_libraries(${ABACUS_BIN_NAME} ${DIR_MTBLAS_LIBRARY})
+  add_compile_definitions(__DSP)
+endif()
+
 find_package(Threads REQUIRED)
 target_link_libraries(${ABACUS_BIN_NAME} Threads::Threads)
 
@@ -436,22 +448,6 @@ if(ENABLE_FLOAT_FFTW)
 endif()
 
 if(ENABLE_DEEPKS)
-  # Torch uses outdated components to detect CUDA arch, causing failure on
-  # latest CUDA kits. Set CMake variable TORCH_CUDA_ARCH_LIST in the form of
-  # "major.minor" if required.
-  find_package(Torch REQUIRED)
-  if(NOT Torch_VERSION VERSION_LESS "2.1.0")
-    set_if_higher(CMAKE_CXX_STANDARD 17)
-  elseif(NOT Torch_VERSION VERSION_LESS "1.5.0")
-    set_if_higher(CMAKE_CXX_STANDARD 14)
-  endif()
-  include_directories(${TORCH_INCLUDE_DIRS})
-  if(MKL_FOUND)
-    list(PREPEND math_libs ${TORCH_LIBRARIES})
-  else()
-    list(APPEND math_libs ${TORCH_LIBRARIES})
-  endif()
-  add_compile_options(${TORCH_CXX_FLAGS})
   target_link_libraries(${ABACUS_BIN_NAME} deepks)
 
   find_path(libnpy_SOURCE_DIR npy.hpp HINTS ${libnpy_INCLUDE_DIR})
@@ -470,6 +466,25 @@ if(ENABLE_DEEPKS)
   add_compile_definitions(__DEEPKS)
 endif()
 
+# Torch uses outdated components to detect CUDA arch, causing failure on
+# latest CUDA kits. Set CMake variable TORCH_CUDA_ARCH_LIST in the form of
+# "major.minor" if required.
+if(ENABLE_DEEPKS OR DEFINED Torch_DIR)
+  find_package(Torch REQUIRED)
+  if(NOT Torch_VERSION VERSION_LESS "2.1.0")
+    set_if_higher(CMAKE_CXX_STANDARD 17)
+  elseif(NOT Torch_VERSION VERSION_LESS "1.5.0")
+    set_if_higher(CMAKE_CXX_STANDARD 14)
+  endif()
+  include_directories(${TORCH_INCLUDE_DIRS})
+  if(MKL_FOUND)
+    list(PREPEND math_libs ${TORCH_LIBRARIES})
+  else()
+    list(APPEND math_libs ${TORCH_LIBRARIES})
+  endif()
+  add_compile_options(${TORCH_CXX_FLAGS})
+endif()
+
 if (ENABLE_CNPY)
   find_path(cnpy_SOURCE_DIR
     cnpy.h
@@ -590,13 +605,14 @@ if(DEFINED DeePMD_DIR)
     add_compile_definitions(__DPMDC)
   else()
     target_link_libraries(${ABACUS_BIN_NAME} DeePMD::deepmd_cc)
-    if(NOT DEFINED TensorFlow_DIR)
-      set(TensorFlow_DIR ${DeePMD_DIR})
-    endif()
-    find_package(TensorFlow REQUIRED)
-    if(TensorFlow_FOUND)
-      target_link_libraries(${ABACUS_BIN_NAME} TensorFlow::tensorflow_cc)
-    endif()
+  endif()
+endif()
+
+if(DEFINED TensorFlow_DIR)
+  find_package(TensorFlow REQUIRED)
+  include_directories(${TensorFlow_DIR}/include)
+  if(TensorFlow_FOUND)
+    target_link_libraries(${ABACUS_BIN_NAME} TensorFlow::tensorflow_cc)
   endif()
 endif()
 

README.md

@@ -12,7 +12,9 @@
 
 # About ABACUS
 
-ABACUS (Atomic-orbital Based Ab-initio Computation at UStc) is an open-source package based on density functional theory (DFT). The package utilizes both plane wave and numerical atomic basis sets with the usage of norm-conserving pseudopotentials to describe the interactions between nuclear ions and valence electrons. ABACUS supports LDA, GGA, meta-GGA, and hybrid functionals. Apart from single-point calculations, the package allows geometry optimizations and ab-initio molecular dynamics with various ensembles. The package also provides a variety of advanced functionalities for simulating materials, including the DFT+U, VdW corrections, and implicit solvation model, etc. In addition, ABACUS strives to provide a general infrastructure to facilitate the developments and applications of novel machine-learning-assisted DFT methods (DeePKS, DP-GEN, DeepH, etc.) in molecular and material simulations.
+ABACUS (Atomic-orbital Based Ab-initio Computation at UStc) is an open-source package based on density functional theory (DFT). The package utilizes both plane wave and numerical atomic basis sets with the usage of norm-conserving pseudopotentials to describe the interactions between nuclear ions and valence electrons. ABACUS supports LDA, GGA, meta-GGA, and hybrid functionals. Apart from single-point calculations, the package allows geometry optimizations and ab-initio molecular dynamics with various ensembles. The package also provides a variety of advanced functionalities for simulating materials, including the DFT+U, VdW corrections, and implicit solvation model, etc. In addition, ABACUS strives to provide a general infrastructure to facilitate the developments and applications of novel machine-learning-assisted DFT methods (DeePKS, DP-GEN, DeepH, DeePTB etc.) in molecular and material simulations.
 
 # Online Documentation
 For detailed documentation, please refer to [our documentation website](https://abacus.deepmodeling.com/).
+
+See our [Github Pages](https://mcresearch.github.io/abacus-user-guide/) for more tutorials and developer guides.

docs/CITATIONS.md

@@ -26,7 +26,7 @@ The following references are required to be cited when using ABACUS. Specificall
 
 - **If DeePKS is used:**
 
-    Wenfei Li, Qi Ou, et al. "DeePKS+ABACUS as a Bridge between Expensive Quantum Mechanical Models and Machine Learning Potentials." <https://arxiv.org/abs/2206.10093>.
+    Wenfei Li, Qi Ou, et al. "DeePKS+ABACUS as a Bridge between Expensive Quantum Mechanical Models and Machine Learning Potentials." J. Phys. Chem. A 126.49 (2022): 9154-9164.
 
 - **If hybrid functional is used:**
 

docs/advanced/acceleration/cuda.md

@@ -29,12 +29,14 @@ To compile and use ABACUS in CUDA mode, you currently need to have an NVIDIA GPU
 
 Check the [Advanced Installation Options](https://abacus-rtd.readthedocs.io/en/latest/advanced/install.html#build-with-cuda-support) for the installation of CUDA version support.
 
-When the compilation parameter USE_ELPA is ON (which is the default value) and USE_CUDA is also set to ON, the ELPA library needs to [enable GPU support](https://github.com/marekandreas/elpa/blob/master/documentation/INSTALL.md) at compile time.
+Setting both USE_ELPA and USE_CUDA to ON does not automatically enable ELPA to run on GPUs. ELPA support for GPUs needs to be enabled when ELPA is compiled. [enable GPU support](https://github.com/marekandreas/elpa/blob/master/documentation/INSTALL.md).
+
+The ABACUS program will automatically determine whether the current ELPA supports GPU based on the elpa/elpa_configured_options.h header file. Users can also check this header file to determine the GPU support of ELPA in their environment. ELPA introduced a new API elpa_setup_gpu in version 2023.11.001. So if you want to enable ELPA GPU in ABACUS, the ELPA version must be greater than or equal to 2023.11.001.
 
 ## Run with the GPU support by editing the INPUT script:
 
 In `INPUT` file we need to set the input parameter [device](../input_files/input-main.md#device) to `gpu`. If this parameter is not set, ABACUS will try to determine if there are available GPUs.
-- Set `ks_solver`: For the PW basis, CG, BPCG and Davidson methods are supported on GPU; set the input parameter [ks_solver](../input_files/input-main.md#ks_solver) to `cg`, `bpcg` or `dav`. For the LCAO basis, `cusolver` and `elpa` is supported on GPU.
+- Set `ks_solver`: For the PW basis, CG, BPCG and Davidson methods are supported on GPU; set the input parameter [ks_solver](../input_files/input-main.md#ks_solver) to `cg`, `bpcg` or `dav`. For the LCAO basis, `cusolver`, `cusolvermp` and `elpa` is supported on GPU.
 - **multi-card**: ABACUS allows for multi-GPU acceleration. If you have multiple GPU cards, you can run ABACUS with several MPI processes, and each process will utilize one GPU card. For example, the command `mpirun -n 2 abacus` will by default launch two GPUs for computation. If you only have one card, this command will only start one GPU. 
 
 ## Examples

docs/advanced/elec_properties/Berry_phase.md

@@ -23,7 +23,7 @@ pseudo_dir      ../../../tests/PP_ORB  //the path to locate the pesudopotential
 orbital_dir     ../../../tests/PP_ORB  //the path to locate the numerical orbital files
 ntype         3
 ecutwfc       50 // Ry
-symmetry      0 // turn off symmetry
+symmetry      -1 // turn off symmetry
 calculation   nscf // non-self-consistent calculation
 basis_type    lcao // atomic basis
 init_chg  file // read charge from files
@@ -70,4 +70,6 @@ The results are shown as follows:
  P =    0.8906925  (mod    2.1748536)  (   0.0000000,   0.0000000,   0.8906925) C/m^2
 ```
 
-The electric polarization **P** is multivalued, which modulo a quantum e**R**/V~cell~. Note: the values in parentheses are the components of the **P** along the c axis in the x, y, z Cartesian coordinates when set gdir = 3 in INPUT file.
+The electric polarization **P** is multivalued, which modulo a quantum e**R**/V~cell~. 
+
+Note: The vectors R1, R2, and R3 refer to the three lattice vectors of the unit cell. When gdir=3, the calculated polarization is along the R3 direction. The three values in parentheses represent the re-projection of the polarization along the R3 direction onto the Cartesian coordinate system (i.e., the xyz coordinate system). To obtain the full polarization components in the Cartesian system, you need to calculate the polarization for R1, R2, and R3 separately, and then sum their respective x, y, and z components.