oclutils.h File Reference

#include <string>
#include "gromacs/gpu_utils/device_context.h"
#include "gromacs/gpu_utils/device_stream.h"
#include "gromacs/gpu_utils/gmxopencl.h"
#include "gromacs/gpu_utils/gputraits_ocl.h"
#include "gromacs/utility/exceptions.h"
#include "gromacs/utility/fatalerror.h"
#include "gromacs/utility/gmxassert.h"
#include "gromacs/utility/stringutil.h"

Include dependency graph for oclutils.h:

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

Description

Declare utility routines for OpenCL.

Author

Anca Hamuraru anca@.nosp@m.stre.nosp@m.amcom.nosp@m.puti.nosp@m.ng.eu

Functions
std::string	ocl_get_error_string (cl_int error)
	Convert error code to diagnostic string.
static bool	haveStreamTasksCompleted (const DeviceStream &)
	Pretend to synchronize an OpenCL stream (dummy implementation). More...
void	prepareGpuKernelArgument (cl_kernel kernel, const KernelLaunchConfig &config, size_t argIndex)
	A function for setting up a single OpenCL kernel argument. This is the tail of the compile-time recursive function below. It has to be seen by the compiler first. As NB kernels might be using dynamic local memory as the last argument, this function also manages that, using sharedMemorySize from config. More...
template<typename CurrentArg , typename... RemainingArgs>
void	prepareGpuKernelArgument (cl_kernel kernel, const KernelLaunchConfig &config, size_t argIndex, const CurrentArg *argPtr, const RemainingArgs *...otherArgsPtrs)
	Compile-time recursive function for setting up a single OpenCL kernel argument. This function uses one kernel argument pointer argPtr to call clSetKernelArg(), and calls itself on the next argument, eventually calling the tail function above. More...
template<typename... Args>
void *	prepareGpuKernelArguments (cl_kernel kernel, const KernelLaunchConfig &config, const Args *...argsPtrs)
	A wrapper function for setting up all the OpenCL kernel arguments. Calls the recursive functions above. More...
void	launchGpuKernel (cl_kernel kernel, const KernelLaunchConfig &config, const DeviceStream &deviceStream, CommandEvent *timingEvent, const char *kernelName, const void *)
	Launches the OpenCL kernel and handles the errors. More...

Function Documentation

static bool haveStreamTasksCompleted ( const DeviceStream &  )

inlinestatic

Pretend to synchronize an OpenCL stream (dummy implementation).

Returns

Not implemented in OpenCL.

void launchGpuKernel ( cl_kernel  kernel, const KernelLaunchConfig &  config, const DeviceStream &  deviceStream, CommandEvent *  timingEvent, const char *  kernelName, const void *    )

inline

Launches the OpenCL kernel and handles the errors.

Parameters

[in]	kernel	Kernel function handle
[in]	config	Kernel configuration for launching
[in]	deviceStream	GPU stream to launch kernel in
[in]	timingEvent	Timing event, fetched from GpuRegionTimer
[in]	kernelName	Human readable kernel description, for error handling only

Exceptions

gmx::InternalError

on kernel launch failure

void prepareGpuKernelArgument ( cl_kernel  kernel, const KernelLaunchConfig &  config, size_t  argIndex  )

inline

A function for setting up a single OpenCL kernel argument. This is the tail of the compile-time recursive function below. It has to be seen by the compiler first. As NB kernels might be using dynamic local memory as the last argument, this function also manages that, using sharedMemorySize from config.

Parameters

[in]	kernel	Kernel function handle
[in]	config	Kernel configuration for launching
[in]	argIndex	Index of the current argument

template<typename CurrentArg , typename... RemainingArgs>

void prepareGpuKernelArgument	(	cl_kernel	kernel,
		const KernelLaunchConfig &	config,
		size_t	argIndex,
		const CurrentArg *	argPtr,
		const RemainingArgs *...	otherArgsPtrs
	)

Compile-time recursive function for setting up a single OpenCL kernel argument. This function uses one kernel argument pointer argPtr to call clSetKernelArg(), and calls itself on the next argument, eventually calling the tail function above.

Template Parameters

CurrentArg	Type of the current argument
RemainingArgs	Types of remaining arguments after the current one

Parameters

[in]	kernel	Kernel function handle
[in]	config	Kernel configuration for launching
[in]	argIndex	Index of the current argument
[in]	argPtr	Pointer to the current argument
[in]	otherArgsPtrs	Pack of pointers to arguments remaining to process after the current one

template<typename... Args>

void* prepareGpuKernelArguments	(	cl_kernel	kernel,
		const KernelLaunchConfig &	config,
		const Args *...	argsPtrs
	)

A wrapper function for setting up all the OpenCL kernel arguments. Calls the recursive functions above.

Template Parameters

Args	Types of all the kernel arguments

Parameters

[in]	kernel	Kernel function handle
[in]	config	Kernel configuration for launching
[in]	argsPtrs	Pointers to all the kernel arguments

Returns

A handle for the prepared parameter pack to be used with launchGpuKernel() as the last argument

• currently always nullptr for OpenCL, as it manages kernel/arguments association by itself.