Merged in memorySpaceEigVecs (pull request #527)

FEBasisOperations Class

Approved-by: Sambit Das
Approved-by: Phani Motamarri

CMakeLists.txt

@@ -47,7 +47,7 @@ SET(TARGET_SRC
  ./src/dft/dft.cc
  ./src/dft/vselfBinsManager.cc
  ./src/dft/energyCalculator.cc
- ./src/dft/densityCalculatorCPU.cc
+ ./src/dft/densityCalculator.cc
  ./src/dft/densityFirstOrderResponseCalculatorCPU.cc
  ./src/excManager/excDensityBaseClass.cpp
  ./src/excManager/excDensityLDAClass.cpp
@@ -166,7 +166,7 @@ SET(DEVICE_SRC
  ./utils/MemoryTransferKernelsDevice.cc
  ./utils/DeviceKernelsGeneric.cc
  ./utils/DeviceDirectCCLWrapper.cc
- ./src/dft/densityCalculatorDevice.cc
+ ./src/dft/densityCalculatorDeviceKernels.cc
  ./src/dft/densityFirstOrderResponseCalculatorDevice.cc
  ./src/dftOperator/operatorDevice.cc
  ./src/dftOperator/kohnShamDFTOperatorDevice.cc
@@ -185,6 +185,7 @@ SET(DEVICE_SRC
  ./src/solvers/linearSolverProblemDevice.cc
  ./src/poisson/poissonSolverProblemDevice.cc
  ./src/helmholtz/kerkerSolverProblemDevice.cc
+ ./utils/FEBasisOperationsKernelsDevice.cc
  )
 
 ELSEIF ("${GPU_LANG}" STREQUAL "hip")
@@ -193,7 +194,7 @@ SET(DEVICE_SRC
  ./utils/MemoryTransferKernelsDevice.cc
  ./utils/DeviceKernelsGeneric.cc
  ./utils/DeviceDirectCCLWrapper.cc
- ./src/dft/densityCalculatorDevice.cc
+ ./src/dft/densityCalculatorDeviceKernels.cc
  ./src/dft/densityFirstOrderResponseCalculatorDevice.cc
  ./src/dftOperator/operatorDevice.cc
  ./src/dftOperator/kohnShamDFTOperatorDevice.cc
@@ -212,6 +213,7 @@ SET(DEVICE_SRC
  ./src/solvers/linearSolverProblemDevice.cc
  ./src/poisson/poissonSolverProblemDevice.cc
  ./src/helmholtz/kerkerSolverProblemDevice.cc
+ ./utils/FEBasisOperationsKernelsDevice.cc
  )
 
 ENDIF()
@@ -222,7 +224,11 @@ IF (WITH_GPU)
  set_source_files_properties(${DEVICE_SRC} PROPERTIES LANGUAGE CUDA)
  ELSEIF ("${GPU_LANG}" STREQUAL "hip")
  set_source_files_properties(${DEVICE_SRC} PROPERTIES LANGUAGE HIP)
- ADD_DEFINITIONS(-D__HIP_PLATFORM_AMD__)
+ IF ("${GPU_VENDOR}" STREQUAL "amd")
+ ADD_DEFINITIONS(-D__HIP_PLATFORM_AMD__)
+ ELSEIF ("${GPU_VENDOR}" STREQUAL "nvidia")
+ ADD_DEFINITIONS(-D__HIP_PLATFORM_NVIDIA__)
+ ENDIF()
  ENDIF()
 ENDIF()
 IF (WITH_GPU)

include/DeviceBlasWrapper.h

@@ -260,6 +260,62 @@ namespace dftfe
  long long int strideC,
  int batchCount);
 
+ deviceBlasStatus_t
+ gemv(deviceBlasHandle_t handle,
+ deviceBlasOperation_t trans,
+ int m,
+ int n,
+ const double * alpha,
+ const double * A,
+ int lda,
+ const double * x,
+ int incx,
+ const double * beta,
+ double * y,
+ int incy);
+
+ deviceBlasStatus_t
+ gemv(deviceBlasHandle_t handle,
+ deviceBlasOperation_t trans,
+ int m,
+ int n,
+ const float * alpha,
+ const float * A,
+ int lda,
+ const float * x,
+ int incx,
+ const float * beta,
+ float * y,
+ int incy);
+
+ deviceBlasStatus_t
+ gemv(deviceBlasHandle_t handle,
+ deviceBlasOperation_t trans,
+ int m,
+ int n,
+ const std::complex<double> *alpha,
+ const std::complex<double> *A,
+ int lda,
+ const std::complex<double> *x,
+ int incx,
+ const std::complex<double> *beta,
+ std::complex<double> * y,
+ int incy);
+
+ deviceBlasStatus_t
+ gemv(deviceBlasHandle_t handle,
+ deviceBlasOperation_t trans,
+ int m,
+ int n,
+ const std::complex<float> *alpha,
+ const std::complex<float> *A,
+ int lda,
+ const std::complex<float> *x,
+ int incx,
+ const std::complex<float> *beta,
+ std::complex<float> * y,
+ int incy);
+
 
  } // namespace deviceBlasWrapper
  } // namespace utils