YAKL
Class List
Here are the classes, structs, unions and interfaces with brief descriptions:
[detail level 123]
 ▼Narray_irThe ArrayIR namespace holds the ArrayIR class and memory type constants associated with ArrayIR. This class holds library-agnostic Array metadata to make it easy to transfer array objects between different C++ libraries
 CArrayIRThe ArrayIR class holds library-agnostic Array metadata to make it easy to transfer array objects between different C++ libraries
 ▼Nyakl
 ▼NcContains Bounds class, and parallel_for() routines using C-style indexing and ordering
 CBoundsDescribes a set of C-style tightly-nested loops
 CBounds< N, false >Describes a set of C-style tightly-nested loops where at least one loop has a lower bound other than 0 or a stride other than 1
 CBounds< N, true >Describes a set of C-style tightly-nested loops where all loops have lower bounds of 0 strides of 1
 CLBndDescribes a single C-style loop bound (lower bound default of 0)
 ▼NfortranContains Bounds class, and parallel_for() routines using Fortran-style indexing and ordering
 CBoundsDescribes a set of Fortran-style tightly-nested loops
 CBounds< N, false >Describes a set of Fortran-style tightly-nested loops where at least one loop has a lower bound other than 1 or a stride other than 1
 CBounds< N, true >Describes a set of Fortran-style tightly-nested loops where all loops have lower bounds of 1 strides of 1
 CLBndDescribes a single Fortran-style loop bound (lower bound default of 1)
 ▼NsimdHolds YAKL's Pack class and operators to encourage SIMD vectorization
 CPackThe Pack class performs encourages vectorization by performing operations Packs of known size
 CPackIterConfigInforms iterate_over_pack of the Pack size and whether to apply a SIMD pragma
 CArrayThis declares the yakl::Array class. Please see the yakl::styleC and yakl::styleFortran template specializations for more detailed information about this class
 CArray< T, rank, myMem, styleC >This implements the yakl:Array class with yakl::styleC behavior
 CArray< T, rank, myMem, styleFortran >This implements the yakl:Array class with yakl::styleFortran behavior
 CArrayBaseThis class implements functionality common to both yakl::styleC and yakl::styleFortran Array objects
 CBndDescribes a single bound for creating Fortran-style yakl::Array objects
 CBndsThis class holds Fortran-style dimensions for using in creating yakl::Array objects
 CCSArrayC-style array on the stack similar in nature to, e.g., float arr[ny][nx];
 CDimsThis class holds C-style dimensions for using in yakl::Array objects
 CEventImplements the functionality of an event within a stream. The event is not created until the Event::create() function is called
 CFSArrayFortran-style array on the stack similar in nature to, e.g., float arr[ny][nx];
 ▼CGatorYAKL Pool allocator class
 CWaitEntry
 CInitConfigAn object of this class can optionally be passed to yakl::init() to configure the initialization. IMPORTANT: Creating an InitConfig object pings environment variables, making it quite expensive to create. Please do not create a lot of these
 CInnerHandlerEmptyThis class is necessary for coordination of two-level parallelism
 CLaunchConfigThis class informs YAKL parallel_for and parallel_outer routines how to launch kernels
 CRandomNon-cryptographic pseudo-random number generator with a very small internal state
 CRealFFT1DCompute batched real-to-complex forward and inverse FFTs on yakl::Array objects using vendor libraries
 CSBThis specifies a set of bounds for a dimension when declaring a yakl::FSArray
 CScalarLiveOutClass to handle scalars that exist before kernels, are written to by kernels, and read after the kernel terminates
 CSimpleNetCDFSimple way to write yakl::Array objects to NetCDF files
 CSimplePNetCDFSimple way to write yakl::Array objects to NetCDF files in parallel
 CStreamImplements the functionality of a stream for parallel kernel execution. If the Stream::create() method is not called on this object, then it is the default stream
 CStreamListImplements a list of Stream objects. Needs to store a pointer to avoid construction on the device since Array objects need to store a list of streams on which they depend