Package io.github.kirstenali.deepj.tensor

Class GpuBuffer

java.lang.Object

io.github.kirstenali.deepj.tensor.GpuBuffer

public final class GpuBuffer
extends Object

Handle to a GPU-resident float buffer managed by a ComputeGraph.

The buffer persists on the GPU across graph flushes until explicitly released. Not tied to any specific GPU API (Metal, CUDA, etc.).

Field Summary

Fields

Modifier and Type	Field	Description
boolean	allocatedOnGpu	True if the runtime has allocated native memory for this id.
final int	cols
boolean	cpuStale	True if GPU has newer data than Tensor.data (CPU cache is stale).
final int	id
boolean	needsUpload	True if this buffer was just allocated and needs CPU data uploaded before use.
final int	rows

Constructor Summary

Constructors

Constructor	Description
GpuBuffer(int id, int rows, int cols, boolean needsUpload)

Method Summary

Modifier and Type	Method	Description
int	floatCount()

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Details

id

public final int id

rows

public final int rows

cols

public final int cols

needsUpload

public boolean needsUpload

True if this buffer was just allocated and needs CPU data uploaded before use.

cpuStale

public boolean cpuStale

True if GPU has newer data than Tensor.data (CPU cache is stale).

allocatedOnGpu

public boolean allocatedOnGpu

True if the runtime has allocated native memory for this id.

Constructor Details

GpuBuffer

public GpuBuffer(int id,
 int rows,
 int cols,
 boolean needsUpload)

Method Details

floatCount

public int floatCount()