Skip navigation links
OpenJ9 GPU Classes
A C G H I M P R S T V 

A

acquireFreeDevice() - Method in class com.ibm.gpu.CUDAManager
Look for the next free device and mark it as busy.

C

com.ibm.gpu - package com.ibm.gpu
com.ibm.gpu allows the user to perform certain operations using any connected CUDA capable GPU
CUDADevice - Class in com.ibm.gpu
Representation of a real, connected CUDA device.
CUDAManager - Class in com.ibm.gpu
This class contains information important to IBM GPU enabled functions.

G

getCUDADevices() - Method in class com.ibm.gpu.CUDAManager
Use this method to obtain a reference to an ArrayList containing references to all discovered CUDA devices.
getDefaultDevice() - Method in class com.ibm.gpu.CUDAManager
Gets the ID of the default device, set to 0 by default.
getDevice(int) - Method in class com.ibm.gpu.CUDAManager
Get a reference to the CUDA device by means of its index (with 0 being the first).
getDeviceCount() - Method in class com.ibm.gpu.CUDAManager
Identifies the number of available CUDA devices.
getDeviceID() - Method in class com.ibm.gpu.CUDADevice
Returns the device ID of this CUDA device.
getDeviceWithMostAvailableMemory() - Method in class com.ibm.gpu.CUDAManager
Identifies the CUDA device that has the most memory available.
getDoubleThreshold() - Method in class com.ibm.gpu.CUDADevice
Returns the minimum size of a double array that will be sorted using this CUDA device if enabled.
getDoubleThreshold() - Method in class com.ibm.gpu.CUDAManager
Gets the minimum length of a double array that will be sorted using a GPU if enabled.
getEnabledCUDADevices() - Method in class com.ibm.gpu.CUDAManager
Use this method to return an array of enabled CUDA devices.
getFloatThreshold() - Method in class com.ibm.gpu.CUDADevice
Returns the minimum size of a float array that will be sorted using this CUDA device if enabled.
getFloatThreshold() - Method in class com.ibm.gpu.CUDAManager
Gets the minimum length of a float array that will be sorted using a GPU if enabled.
getFreeMemoryForDevice(int) - Method in class com.ibm.gpu.CUDAManager
Get the amount of free memory (in bytes) available for the provided CUDA device.
getInstance() - Static method in class com.ibm.gpu.CUDAManager
Deprecated.
Use instance() instead.
getIntThreshold() - Method in class com.ibm.gpu.CUDADevice
Returns the minimum size of an int array that will be sorted using this CUDA device if enabled.
getIntThreshold() - Method in class com.ibm.gpu.CUDAManager
Gets the minimum length of an int array that will be sorted using a GPU if enabled.
getLongThreshold() - Method in class com.ibm.gpu.CUDADevice
Returns the minimum size of a long array that will be sorted using this CUDA device if enabled.
getLongThreshold() - Method in class com.ibm.gpu.CUDAManager
Gets the minimum length of a long array that will be sorted using a GPU if enabled.
getNextAvailableDevice() - Method in class com.ibm.gpu.CUDAManager
Returns the next CUDA device that is available to run calculations on.
getOutputHeader() - Static method in class com.ibm.gpu.CUDAManager
Get the header used to prefix all IBM GPU related output.
getVerboseGPUOutput() - Method in class com.ibm.gpu.CUDAManager
Get the value of the verboseGPUOutput flag.
getVersion() - Static method in class com.ibm.gpu.CUDAManager
Get the version of this class.
GPUConfigurationException - Exception in com.ibm.gpu
This exception is thrown when GPU operations fail due to configuration or environment issues - for example, an invalid device has been specified or we are running on an unsupported platform.
GPUConfigurationException(String) - Constructor for exception com.ibm.gpu.GPUConfigurationException
Construct a new GPUConfigurationException with the provided message.
GPUConfigurationException(String, Throwable) - Constructor for exception com.ibm.gpu.GPUConfigurationException
Construct a new GPUConfigurationException with the provided message and cause.
GPUPermission - Class in com.ibm.gpu
This class defines GPU permissions as described in the following table.
GPUPermission(String) - Constructor for class com.ibm.gpu.GPUPermission
Create a representation of the named permissions.
GPUPermission(String, String) - Constructor for class com.ibm.gpu.GPUPermission
Create a representation of the named permissions.
GPUSortException - Exception in com.ibm.gpu
This exception is thrown when there is an attempt to perform a sort on GPU and the sort operation fails.
GPUSortException(String) - Constructor for exception com.ibm.gpu.GPUSortException
Creates a new GPUSortException with a provided message.
GPUSortException(String, Throwable) - Constructor for exception com.ibm.gpu.GPUSortException
Creates a new GPUSortException with a provided message and cause.

H

hasCUDASupport() - Method in class com.ibm.gpu.CUDAManager
Use this method to identify if CUDA is supported on this machine and within this environment: returns true if the number of CUDA devices detected is greater than 0.

I

instance() - Static method in class com.ibm.gpu.CUDAManager
Return a CUDAManager instance.
isSortEnabledOnGPU() - Method in class com.ibm.gpu.CUDAManager
This method provides a means to determine if sort is enabled to be used by any available CUDA device.
isSortEnforcedOnGPU() - Method in class com.ibm.gpu.CUDAManager
This method provides a means to determine if sort is forced to be used by any available CUDA device.

M

main(String[]) - Static method in class com.ibm.gpu.Version
Prints the current level of ibmgpu to stdout.
Maths - Class in com.ibm.gpu
This class is used to perform sorting operations of primitive arrays of type int, long, float, double on any connected CUDA GPU.
Maths() - Constructor for class com.ibm.gpu.Maths
 

P

printAllDeviceInfo() - Method in class com.ibm.gpu.CUDAManager
Print information for each detected CUDA device.

R

releaseDevice(int) - Method in class com.ibm.gpu.CUDAManager
Mark a device as being free; must be in a try finally block as we MUST release the handle regardless of whether or not a sort was successful.

S

setDefaultDevice(int) - Method in class com.ibm.gpu.CUDAManager
Sets the default device to the given device ID.
setDevice(int) - Method in class com.ibm.gpu.CUDAManager
Use this method to set the device to use for subsequent calls.
setVerboseGPU(boolean) - Method in class com.ibm.gpu.CUDAManager
Set the value of the verboseGPUOutput flag.
sortArray(double[]) - Static method in class com.ibm.gpu.Maths
Sort the given array of doubles into ascending order, using the default CUDA device.
sortArray(double[], int, int) - Static method in class com.ibm.gpu.Maths
Sort the specified range of the array of doubles into ascending order, using the default CUDA device.
sortArray(float[]) - Static method in class com.ibm.gpu.Maths
Sort the given array of floats into ascending order, using the default CUDA device.
sortArray(float[], int, int) - Static method in class com.ibm.gpu.Maths
Sort the specified range of the array of floats into ascending order, using the default CUDA device.
sortArray(int, double[]) - Static method in class com.ibm.gpu.Maths
Sort the given array of doubles into ascending order, using the specified CUDA device.
sortArray(int, double[], int, int) - Static method in class com.ibm.gpu.Maths
Sort the specified range of the array of doubles into ascending order, using the specified CUDA device.
sortArray(int, float[]) - Static method in class com.ibm.gpu.Maths
Sort the given array of floats into ascending order, using the specified CUDA device.
sortArray(int, float[], int, int) - Static method in class com.ibm.gpu.Maths
Sort the specified range of the array of floats into ascending order, using the specified CUDA device.
sortArray(int, int[]) - Static method in class com.ibm.gpu.Maths
Sort the given array of integers into ascending order, using the specified CUDA device.
sortArray(int, int[], int, int) - Static method in class com.ibm.gpu.Maths
Sort the specified range of the array of integers into ascending order, using the specified CUDA device.
sortArray(int, long[]) - Static method in class com.ibm.gpu.Maths
Sort the given array of longs into ascending order, using the specified CUDA device.
sortArray(int, long[], int, int) - Static method in class com.ibm.gpu.Maths
Sort the specified range of the array of longs into ascending order, using the specified CUDA device.
sortArray(int[]) - Static method in class com.ibm.gpu.Maths
Sort the given array of integers into ascending order, using the default CUDA device.
sortArray(int[], int, int) - Static method in class com.ibm.gpu.Maths
Sort the specified range of the array of integers into ascending order, using the default CUDA device.
sortArray(long[]) - Static method in class com.ibm.gpu.Maths
Sort the given array of longs into ascending order, using the default CUDA device.
sortArray(long[], int, int) - Static method in class com.ibm.gpu.Maths
Sort the specified range of the array of longs into ascending order, using the default CUDA device.

T

tearDown() - Static method in class com.ibm.gpu.CUDAManager
Deprecated.
This method has no effect; it will be removed in a future version.
toString() - Method in class com.ibm.gpu.CUDADevice
Answers a string containing a concise, human-readable description of the receiver.

V

Version - Class in com.ibm.gpu
Acts as an accessor to the current build level of the ibmgpu library.
Version() - Constructor for class com.ibm.gpu.Version
 
VERSION - Static variable in class com.ibm.gpu.Version
The current build level of this package.
A C G H I M P R S T V 
Skip navigation links
OpenJ9 GPU Classes

Eclipse OpenJ9 website.
To raise a bug report or suggest an improvement create an Eclipse OpenJ9 issue.
Copyright © 2014, 2023 IBM Corp. and others.