Tidy-up CUDA implementation of IProductWRTBase and add CUDA kernels with additional parallelism

See merge request !73

Tidy-up CUDA implementation of IProductWRTDerivBase and add CUDA kernels with additional parallelism

See merge request !72

Jacques Xing authored 1 year ago and Chris Cantwell committed 1 year ago

Tidy-up CUDA implementation of IProductWRTDerivBase and add CUDA kernels with additional parallelism

Tidy-up CUDA implementation of PhysDeriv and add CUDA kernels with additional parallelism

See merge request !71

Implement ConjGrad, FwdTrans, and HelmSove operator for CUDA

See merge request !61

Implemented IProductWRTBaseSumFac for simplexes

See merge request !69

Fix CMake file to solve compilation error when CUDA is disable

See merge request !70

Implement various vector math operators on CUDA

See merge request !68

Fix various CUDA memory leaks

See merge request !67

Tidy-up AssmbScatr operators implementation

See merge request !66

Some tidy-up for the Mass and Matrix operators

See merge request !65

Specify MemoryRegionCUDA template parameter in test to avoid back storage warning message when using CUDA

See merge request !64

Specify MemoryRegionCUDA template parameter in test to avoid back storage warning message when using CUDA

Simplify Identity operator implementation

See merge request !63

Implement NeuBndCond operator for CUDA

See merge request !59

Implement DiagPrecon operator for CUDA

See merge request !60

Fix Field member function to avoid back storage warning message when using CUDA

See merge request !62

Implement DirBndCond operator for CUDA

See merge request !58

Implement NullPrecond operator for CUDA

See merge request !57

Turn on compilation warnings

See merge request !56

Fix various operators for padding when using SIMD/AVX

See merge request !55

Implement AssmbScatr operator for CUDA

See merge request !53