ComputeFunctions.cxx File Reference

#include "RooBatchCompute.h"
#include "RooNaNPacker.h"
#include "RooVDTHeaders.h"
#include "Batches.h"
#include <TMath.h>
#include <RooHeterogeneousMath.h>
#include <vector>

Include dependency graph for ComputeFunctions.cxx:

This graph shows which files directly or indirectly include this file:

Namespaces
namespace	RooBatchCompute
	Namespace for dispatching RooFit computations to various backends.
namespace	RooBatchCompute::CUDA

Macros
#define	BEGIN   0
#define	STEP   1

Functions
__rooglobal__ void	RooBatchCompute::CUDA::computeAddPdf (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeArgusBG (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeBernstein (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeBifurGauss (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeBMixDecay (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeBreitWigner (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeBukin (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeCBShape (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeChebychev (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeChiSquare (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeDeltaFunction (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeDstD0BG (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeExponential (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeExponentialNeg (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeExpPoly (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeGamma (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeGaussian (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeGaussModelExpBasis (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeIdentity (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeJohnson (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeLandau (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeLognormal (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeLognormalStandard (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeNegativeLogarithms (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeNormalizedPdf (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeNovosibirsk (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computePoisson (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computePolynomial (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computePower (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeProdPdf (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeRatio (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeTruthModelCosBasis (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeTruthModelCoshBasis (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeTruthModelExpBasis (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeTruthModelLinBasis (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeTruthModelQuadBasis (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeTruthModelSinBasis (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeTruthModelSinhBasis (Batches &batches)
__rooglobal__ void	RooBatchCompute::CUDA::computeVoigtian (Batches &batches)
std::vector< void(*)(Batches &)>	RooBatchCompute::CUDA::getFunctions ()
	Returns a std::vector of pointers to the compute functions in this file.

Detailed Description

This file contains vectorizable computation functions for PDFs and other Roofit objects.

The same source file can also be compiled with nvcc. All functions have a single Batches object as an argument passed by value, which contains all the information necessary for the computation. In case of cuda computations, the loops have a step (stride) the size of the grid which allows for reusing the same code as the cpu implementations, easier debugging and in terms of performance, maximum memory coalescing. For more details, see https://developer.nvidia.com/blog/cuda-pro-tip-write-flexible-kernels-grid-stride-loops/

Definition in file ComputeFunctions.cxx.

Macro Definition Documentation

BEGIN

#define BEGIN   0

Definition at line 41 of file ComputeFunctions.cxx.

STEP

#define STEP   1

Definition at line 42 of file ComputeFunctions.cxx.