gpu_types_common.h File Reference

#include "config.h"
#include "gromacs/gpu_utils/hostallocator.h"
#include "gromacs/mdtypes/interaction_const.h"
#include "gromacs/mdtypes/locality.h"
#include "gromacs/nbnxm/nbnxm_enums.h"
#include "gromacs/utility/enumerationhelpers.h"
#include "nbnxm.h"
#include "pairlist.h"

Include dependency graph for gpu_types_common.h:

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

Description

Implements common internal types for different NBNXN GPU implementations.

Author

Szilárd Páll pall..nosp@m.szil.nosp@m.ard@g.nosp@m.mail.nosp@m..com

Classes
struct	gmx::NBStagingData
	Staging area for temporary data downloaded from the GPU. More...
struct	gmx::NBAtomDataGpu
	Nonbonded atom data - both inputs and outputs. More...
struct	gmx::NBParamGpu
	Parameters required for the GPU nonbonded calculations. More...
struct	gmx::GpuTimers
	GPU region timers used for timing GPU kernels and H2D/D2H transfers. More...
struct	gmx::GpuTimers::XFTransfers
	Timers for local or non-local coordinate/force transfers. More...
struct	gmx::GpuTimers::Interaction
	Timers for local or non-local interaction related operations. More...
class	gmx::GpuPairlistSorting
	Sorted pair list on GPU and data required for performing the sorting. More...
class	gmx::GpuPairlist
	GPU pair list structure. More...
struct	gmx::EnergyFunctionProperties< elecType, vdwType >
	Set of boolean constants mimicking preprocessor macros. More...

Macros
#define	GMX_NBNXN_PRUNE_KERNEL_JPACKED_CONCURRENCY   4
	Macro definining default for the prune kernel's jPacked processing concurrency. More...

Variables
static constexpr int	gmx::c_sciHistogramSize = 8192
	Number of separate bins used during sorting of plist on gpu. More...
static constexpr int	gmx::c_sciSortingThreadsPerBlock = 256
	Number of threads per block used by the gpu sorting kernel. More...
static constexpr int	gmx::c_pruneKernelJPackedConcurrency = 4
	Default for the prune kernel's jPacked processing concurrency.
template<enum VdwType vdwType>
constexpr bool	gmx::ljComb = EnergyFunctionProperties<ElecType::Count, vdwType>().vdwComb
	Templated constants to shorten kernel function declaration.
template<enum ElecType elecType>
constexpr bool	gmx::elecEwald = EnergyFunctionProperties<elecType, VdwType::Count>().elecEwald
template<enum ElecType elecType>
constexpr bool	gmx::elecEwaldTab = EnergyFunctionProperties<elecType, VdwType::Count>().elecEwaldTab
template<enum VdwType vdwType>
constexpr bool	gmx::ljEwald = EnergyFunctionProperties<ElecType::Count, vdwType>().vdwEwald

Macro Definition Documentation

#define GMX_NBNXN_PRUNE_KERNEL_JPACKED_CONCURRENCY   4

Macro definining default for the prune kernel's jPacked processing concurrency.

The GMX_NBNXN_PRUNE_KERNEL_JPACKED_CONCURRENCY macro allows compile-time override with the default value of 4.