Modifier and Type | Method and Description |
---|---|
CudaLinker |
CudaLinker.add(CudaJitInputType type,
byte[] data,
java.lang.String name)
Adds a new code fragment to be linked into the module under construction
using the default options.
|
CudaLinker |
CudaLinker.add(CudaJitInputType type,
byte[] data,
java.lang.String name,
CudaJitOptions options)
Adds a new code fragment to be linked into the module under construction
using the specified options.
|
CudaLinker |
CudaLinker.add(CudaJitInputType type,
java.io.InputStream input,
java.lang.String name)
Adds a new code fragment to be linked into the module under construction
using the default options.
|
CudaLinker |
CudaLinker.add(CudaJitInputType type,
java.io.InputStream input,
java.lang.String name,
CudaJitOptions options)
Adds a new code fragment to be linked into the module under construction
using the default options.
|
void |
CudaDevice.addCallback(java.lang.Runnable callback)
Queues the given
callback to be executed when the associated
device has completed all previous actions in the default stream. |
void |
CudaStream.addCallback(java.lang.Runnable callback)
Enqueues a callback to be run after all previous work on this stream
has been completed.
|
static java.nio.ByteBuffer |
Cuda.allocatePinnedHostBuffer(long capacity)
Allocates a new direct byte buffer, backed by page-locked host memory;
enabling optimal performance of transfers to and from device memory.
|
boolean |
CudaDevice.canAccessPeer(CudaDevice peerDevice)
Returns whether this device can access memory of the specified
peerDevice . |
void |
CudaEvent.close()
Releases resources associated with this event.
|
void |
CudaBuffer.close()
Releases the region of device memory backing this buffer.
|
void |
CudaStream.close()
Closes this stream.
|
byte[] |
CudaLinker.complete()
Completes the module under construction and return an image suitable
for loading.
|
void |
CudaBuffer.copyFrom(byte[] array)
Copies all data from the specified
array (on the Java host) to
this buffer (on the device). |
void |
CudaBuffer.copyFrom(byte[] array,
int fromIndex,
int toIndex)
Copies data from the specified
array (on the Java host) to this
buffer (on the device). |
void |
CudaBuffer.copyFrom(java.nio.ByteBuffer source)
Copies data from the specified
source buffer (on the Java host)
to this buffer (on the device). |
void |
CudaBuffer.copyFrom(char[] array)
Copies all data from the specified
array (on the Java host) to
this buffer (on the device). |
void |
CudaBuffer.copyFrom(char[] array,
int fromIndex,
int toIndex)
Copies data from the specified
array (on the Java host) to this
buffer (on the device). |
void |
CudaBuffer.copyFrom(java.nio.CharBuffer source)
Copies data from the specified
source buffer (on the Java host)
to this buffer (on the device). |
void |
CudaBuffer.copyFrom(CudaBuffer source,
long fromOffset,
long toOffset)
Copies data from the specified
source buffer (on a device)
to this buffer (on the device). |
void |
CudaBuffer.copyFrom(double[] array)
Copies all data from the specified
array (on the Java host) to
this buffer (on the device). |
void |
CudaBuffer.copyFrom(double[] array,
int fromIndex,
int toIndex)
Copies data from the specified
array (on the Java host) to this
buffer (on the device). |
void |
CudaBuffer.copyFrom(java.nio.DoubleBuffer source)
Copies data from the specified
source buffer (on the Java host)
to this buffer (on the device). |
void |
CudaBuffer.copyFrom(float[] array)
Copies all data from the specified
array (on the Java host) to
this buffer (on the device). |
void |
CudaBuffer.copyFrom(float[] array,
int fromIndex,
int toIndex)
Copies data from the specified
array (on the Java host) to this
buffer (on the device). |
void |
CudaBuffer.copyFrom(java.nio.FloatBuffer source)
Copies data from the specified
source buffer (on the Java host)
to this buffer (on the device). |
void |
CudaBuffer.copyFrom(int[] array)
Copies all data from the specified
array (on the Java host) to
this buffer (on the device). |
void |
CudaBuffer.copyFrom(int[] array,
int fromIndex,
int toIndex)
Copies data from the specified
array (on the Java host) to this
buffer (on the device). |
void |
CudaBuffer.copyFrom(java.nio.IntBuffer source)
Copies data from the specified
source buffer (on the Java host)
to this buffer (on the device). |
void |
CudaBuffer.copyFrom(long[] array)
Copies all data from the specified
array (on the Java host) to
this buffer (on the device). |
void |
CudaBuffer.copyFrom(long[] array,
int fromIndex,
int toIndex)
Copies data from the specified
array (on the Java host) to this
buffer (on the device). |
void |
CudaBuffer.copyFrom(java.nio.LongBuffer source)
Copies data from the specified
source buffer (on the Java host)
to this buffer (on the device). |
void |
CudaBuffer.copyFrom(short[] array)
Copies all data from the specified
array (on the Java host) to
this buffer (on the device). |
void |
CudaBuffer.copyFrom(short[] array,
int fromIndex,
int toIndex)
Copies data from the specified
array (on the Java host) to this
buffer (on the device). |
void |
CudaBuffer.copyFrom(java.nio.ShortBuffer source)
Copies data from the specified
source buffer (on the Java host)
to this buffer (on the device). |
void |
CudaBuffer.copyTo(byte[] array)
Copies data from this buffer (on the device) to the specified
array (on the Java host). |
void |
CudaBuffer.copyTo(byte[] array,
int fromIndex,
int toIndex)
Copies data from this buffer (on the device) to the specified
array (on the Java host). |
void |
CudaBuffer.copyTo(java.nio.ByteBuffer target)
Copies data from this buffer (on the device) to the specified
target buffer (on the Java host). |
void |
CudaBuffer.copyTo(char[] array)
Copies data from this buffer (on the device) to the specified
array (on the Java host). |
void |
CudaBuffer.copyTo(char[] array,
int fromIndex,
int toIndex)
Copies data from this buffer (on the device) to the specified
array (on the Java host). |
void |
CudaBuffer.copyTo(java.nio.CharBuffer target)
Copies data from this buffer (on the device) to the specified
target buffer (on the Java host). |
void |
CudaBuffer.copyTo(double[] array)
Copies data from this buffer (on the device) to the specified
array (on the Java host). |
void |
CudaBuffer.copyTo(double[] array,
int fromIndex,
int toIndex)
Copies data from this buffer (on the device) to the specified
array (on the Java host). |
void |
CudaBuffer.copyTo(java.nio.DoubleBuffer target)
Copies data from this buffer (on the device) to the specified
target buffer (on the Java host). |
void |
CudaBuffer.copyTo(float[] array)
Copies data from this buffer (on the device) to the specified
array (on the Java host). |
void |
CudaBuffer.copyTo(float[] array,
int fromIndex,
int toIndex)
Copies data from this buffer (on the device) to the specified
array (on the Java host). |
void |
CudaBuffer.copyTo(java.nio.FloatBuffer target)
Copies data from this buffer (on the device) to the specified
target buffer (on the Java host). |
void |
CudaBuffer.copyTo(int[] array)
Copies data from this buffer (on the device) to the specified
array (on the Java host). |
void |
CudaBuffer.copyTo(int[] array,
int fromIndex,
int toIndex)
Copies data from this buffer (on the device) to the specified
array (on the Java host). |
void |
CudaBuffer.copyTo(java.nio.IntBuffer target)
Copies data from this buffer (on the device) to the specified
target buffer (on the Java host). |
void |
CudaBuffer.copyTo(long[] array)
Copies data from this buffer (on the device) to the specified
array (on the Java host). |
void |
CudaBuffer.copyTo(long[] array,
int fromIndex,
int toIndex)
Copies data from this buffer (on the device) to the specified
array (on the Java host). |
void |
CudaBuffer.copyTo(java.nio.LongBuffer target)
Copies data from this buffer (on the device) to the specified
target buffer (on the Java host). |
void |
CudaBuffer.copyTo(short[] array)
Copies data from this buffer (on the device) to the specified
array (on the Java host). |
void |
CudaBuffer.copyTo(short[] array,
int fromIndex,
int toIndex)
Copies data from this buffer (on the device) to the specified
array (on the Java host). |
void |
CudaBuffer.copyTo(java.nio.ShortBuffer target)
Copies data from this buffer (on the device) to the specified
target buffer (on the Java host). |
void |
CudaLinker.destroy()
Destroys this linker, releasing associated resources.
|
void |
CudaDevice.disablePeerAccess(CudaDevice peerDevice)
Disable access to memory of
peerDevice by this device. |
float |
CudaEvent.elapsedTimeSince(CudaEvent priorEvent)
Returns the elapsed time (in milliseconds) relative to the specified
priorEvent . |
void |
CudaDevice.enablePeerAccess(CudaDevice peerDevice)
Enable access to memory of
peerDevice by this device. |
void |
CudaBuffer.fillByte(byte value,
long count)
Stores
count copies of value in this buffer. |
void |
CudaBuffer.fillChar(char value,
long count)
Stores
count copies of value in this buffer. |
void |
CudaBuffer.fillFloat(float value,
long count)
Stores
count copies of value in this buffer. |
void |
CudaBuffer.fillInt(int value,
long count)
Stores
count copies of value in this buffer. |
void |
CudaBuffer.fillShort(short value,
long count)
Stores
count copies of value in this buffer. |
int |
CudaDevice.getAttribute(int attribute)
Returns the value of the specified
attribute . |
int |
CudaKernel.getAttribute(int attribute)
Returns the value of the specified @{code attribute} for the
function associated with this kernel. |
int |
CudaFunction.getAttribute(int attribute)
Returns the value of the specified @{code attribute}.
|
CudaDevice.CacheConfig |
CudaDevice.getCacheConfig()
Returns the current cache configuration of this device.
|
static int |
CudaDevice.getCount()
Returns the number of CUDA-capable devices available to the Java host.
|
static int |
Cuda.getDeviceCount()
Returns the number of CUDA-capable devices available to the Java host.
|
static int |
CudaDevice.getDriverVersion()
Deprecated.
Use Cuda.getDriverVersion() instead.
|
static int |
Cuda.getDriverVersion()
Returns a number identifying the driver version.
|
int |
CudaStream.getFlags()
Returns the flags of this stream.
|
long |
CudaDevice.getFreeMemory()
Returns the amount of free device memory in bytes.
|
CudaFunction |
CudaModule.getFunction(java.lang.String name)
Returns the function of the specified name from this module.
|
CudaGlobal |
CudaModule.getGlobal(java.lang.String name)
Returns the global variable of the specified name from this module.
|
int |
CudaDevice.getGreatestStreamPriority()
Returns the greatest possible priority of a stream on this device.
|
int |
CudaDevice.getLeastStreamPriority()
Returns the least possible priority of a stream on this device.
|
long |
CudaDevice.getLimit(CudaDevice.Limit limit)
Returns the value of the specified
limit . |
java.lang.String |
CudaDevice.getName()
Returns the name of this device.
|
int |
CudaStream.getPriority()
Returns the priority of this stream.
|
static int |
CudaDevice.getRuntimeVersion()
Deprecated.
Use Cuda.getRuntimeVersion() instead.
|
static int |
Cuda.getRuntimeVersion()
Returns a number identifying the runtime version.
|
CudaDevice.SharedMemConfig |
CudaDevice.getSharedMemConfig()
Returns the current shared memory configuration of this device.
|
CudaSurface |
CudaModule.getSurface(java.lang.String name)
Returns the surface of the specified name from this module.
|
CudaTexture |
CudaModule.getTexture(java.lang.String name)
Returns the texture of the specified name from this module.
|
long |
CudaDevice.getTotalMemory()
Returns the total amount of memory on this device in bytes.
|
void |
CudaKernel.launch(CudaGrid grid,
CudaKernel.Parameters parameters)
Launches this kernel.
|
void |
CudaKernel.launch(CudaGrid grid,
java.lang.Object... parameters)
Launches this kernel.
|
void |
CudaEvent.record(CudaDevice device)
Records this event on the default stream of the specified device.
|
void |
CudaEvent.record(CudaStream stream)
Records this event on the specified stream.
|
void |
CudaDevice.setCacheConfig(CudaDevice.CacheConfig config)
Configures the cache of this device.
|
void |
CudaKernel.setCacheConfig(CudaDevice.CacheConfig config)
Configures the cache for the
function associated
with this kernel. |
void |
CudaFunction.setCacheConfig(CudaDevice.CacheConfig config)
Configures the cache for this function.
|
void |
CudaDevice.setLimit(CudaDevice.Limit limit,
long value)
Configures the specified
limit . |
void |
CudaDevice.setSharedMemConfig(CudaDevice.SharedMemConfig config)
Configures the shared memory of this device.
|
void |
CudaKernel.setSharedMemConfig(CudaDevice.SharedMemConfig config)
Configures the shared memory of the
function
associated with this kernel. |
void |
CudaFunction.setSharedMemConfig(CudaDevice.SharedMemConfig config)
Configures the shared memory of this function.
|
void |
CudaEvent.synchronize()
Synchronizes on this event.
|
void |
CudaDevice.synchronize()
Synchronizes on this device.
|
void |
CudaStream.synchronize()
Synchronizes with this stream.
|
void |
CudaModule.unload()
Unloads this module from the associated device.
|
void |
CudaStream.waitFor(CudaEvent event)
Makes all future work submitted to this stream wait for the specified
event to occur.
|
Constructor and Description |
---|
CudaBuffer(CudaDevice device,
long byteCount)
Allocates a new region on the specified
device of size byteCount bytes. |
CudaEvent(CudaDevice device)
Creates a new event on the specified device with default flags.
|
CudaEvent(CudaDevice device,
int flags)
Creates a new event on the specified device with the specified
flags . |
CudaKernel(CudaModule module,
java.lang.String functionName)
Creates a new kernel object in the given module whose entry point
is the function with the specified name.
|
CudaLinker(CudaDevice device)
Creates a new linker for the specified
device
using default options. |
CudaLinker(CudaDevice device,
CudaJitOptions options)
Creates a new linker for the specified
device
using the specified options . |
CudaModule(CudaDevice device,
byte[] image)
Loads a module on the specified device, using the given image and the
default options.
|
CudaModule(CudaDevice device,
byte[] image,
CudaJitOptions options)
Loads a module on the specified device, using the given image and the
given options.
|
CudaModule(CudaDevice device,
java.io.InputStream input)
Loads a module on the specified device from the given input stream using
the default options.
|
CudaModule(CudaDevice device,
java.io.InputStream input,
CudaJitOptions options)
Loads a module on the specified device from the given input stream using
the specified options.
|
CudaStream(CudaDevice device)
Creates a new stream on the specified device, with the default flags
and the default priority.
|
CudaStream(CudaDevice device,
int flags,
int priority)
Creates a new stream on the specified device, with the specified flags
and priority.
|
Eclipse OpenJ9 website.
To raise a bug report or suggest an improvement create an Eclipse OpenJ9 issue.
Copyright © 2013, 2025 IBM Corp. and others.