Deprecated API

Contents

• Deprecated Classes
• Deprecated Enums
• Deprecated Methods
• Deprecated Enum Constants

Deprecated Classes

Class and Description
org.bytedeco.tensorrt.nvinfer.IPluginV2 Deprecated in TensorRT 8.5. Implement IPluginV3 instead.
org.bytedeco.tensorrt.nvinfer.IPluginV2DynamicExt Deprecated in TensorRT 10.0. Please implement IPluginV3 instead.
org.bytedeco.tensorrt.nvinfer.IPluginV2Ext Deprecated in TensorRT 8.5. Implement IPluginV3 instead.
org.bytedeco.tensorrt.nvinfer.IPluginV2IOExt Deprecated in TensorRT 10.0. Implement IPluginV3 instead.
org.bytedeco.tensorrt.nvinfer.IPluginV2Layer Deprecated in TensorRT 10.8. Superseded by IPluginV3Layer.
org.bytedeco.tensorrt.nvinfer.IStreamReader Deprecated in TensorRT 11.0. Superseded by IStreamReaderV2.

Deprecated Enums

Enum and Description
org.bytedeco.tensorrt.global.nvinfer.PluginCreatorVersion Deprecated in TensorRT 10.10. PluginCreatorVersion is used only in relation to plugin creators based off IPluginCreator, which is deprecated.
org.bytedeco.tensorrt.global.nvinfer.PluginVersion Deprecated in TensorRT 10.10. PluginVersion is used only in relation to IPluginV2-descendent plugin interfaces, which are all deprecated.

Deprecated Methods

Method and Description
org.bytedeco.tensorrt.nvinfer.IGpuAsyncAllocator._allocate(long, long, int) Deprecated in TensorRT 10.0. Superseded by allocateAsync
org.bytedeco.tensorrt.nvinfer.IGpuAllocator._allocate(long, long, int) Deprecated in TensorRT 10.0. Superseded by allocateAsync
org.bytedeco.tensorrt.nvinfer.IGpuAsyncAllocator._deallocate(Pointer) Deprecated in TensorRT 10.0. Superseded by deallocateAsync
org.bytedeco.tensorrt.nvinfer.IGpuAllocator._deallocate(Pointer) Deprecated in TensorRT 10.0. Superseded by deallocateAsync
org.bytedeco.tensorrt.nvinfer.VNetworkDefinition.addAttention(ITensor, ITensor, ITensor, int, boolean)
org.bytedeco.tensorrt.nvinfer.INetworkDefinition.addAttention(ITensor, ITensor, ITensor, int, boolean)
org.bytedeco.tensorrt.nvinfer.VNetworkDefinition.addAttention(ITensor, ITensor, ITensor, nvinfer.AttentionNormalizationOp, boolean)
org.bytedeco.tensorrt.nvinfer.INetworkDefinition.addAttention(ITensor, ITensor, ITensor, nvinfer.AttentionNormalizationOp, boolean) Deprecated in TensorRT 10.16. Superseded by addAttentionV2 with CausalMaskKind parameter.
org.bytedeco.tensorrt.nvinfer.INetworkDefinition.addDynamicQuantize(ITensor, int, int, int, int)
org.bytedeco.tensorrt.nvinfer.INetworkDefinition.addDynamicQuantize(ITensor, int, int, nvinfer.DataType, nvinfer.DataType)
org.bytedeco.tensorrt.nvinfer.INetworkDefinition.addNMS(ITensor, ITensor, ITensor) Deprecated in TensorRT 10.14. Superseded by four-argument addNMS.
org.bytedeco.tensorrt.nvinfer.INetworkDefinition.addNonZero(ITensor) Deprecated in TensorRT 10.14. Superseded by two-argument addNonZero.
org.bytedeco.tensorrt.nvinfer.INetworkDefinition.addNormalization(ITensor, ITensor, ITensor, int) Deprecated in TensorRT 10.15. Superseded by addNormalizationV2().
org.bytedeco.tensorrt.nvinfer.INetworkDefinition.addPluginV2(ITensor, int, IPluginV2)
org.bytedeco.tensorrt.nvinfer.INetworkDefinition.addPluginV2(PointerPointer, int, IPluginV2) Deprecated in TensorRT 10.8. Superseded by addPluginV3.
org.bytedeco.tensorrt.nvinfer.INetworkDefinition.addTopK(ITensor, int, int, int)
org.bytedeco.tensorrt.nvinfer.INetworkDefinition.addTopK(ITensor, nvinfer.TopKOperation, int, int) Deprecated in TensorRT 10.14. Superseded by five-argument addTopK.
org.bytedeco.tensorrt.nvinfer.VDynamicQuantizeLayer.getAxis()
org.bytedeco.tensorrt.nvinfer.IDynamicQuantizeLayer.getAxis()
org.bytedeco.tensorrt.nvinfer.VDynamicQuantizeLayer.getBlockSize()
org.bytedeco.tensorrt.nvinfer.IDynamicQuantizeLayer.getBlockSize()
org.bytedeco.tensorrt.nvinfer.ITensor.getBroadcastAcrossBatch() Deprecated in TensorRT 10.0. Implicit batch is not supported since TensorRT 10.0.
org.bytedeco.tensorrt.global.nvinfer.getBuilderSafePluginRegistry(int)
org.bytedeco.tensorrt.global.nvinfer.getBuilderSafePluginRegistry(nvinfer.EngineCapability)
org.bytedeco.tensorrt.nvinfer.VAttention.getCausal()
org.bytedeco.tensorrt.nvinfer.IAttention.getCausal() Deprecated in TensorRT 10.16. Superseded by getCausalKind.
org.bytedeco.tensorrt.nvinfer.INetworkDefinition.hasImplicitBatchDimension() Deprecated in TensorRT 10.0. Implicit batch is not supported since TensorRT 10.0.
org.bytedeco.tensorrt.nvinfer.IOutputAllocator.reallocateOutput(BytePointer, Pointer, long, long)
org.bytedeco.tensorrt.nvinfer.IOutputAllocator.reallocateOutput(String, Pointer, long, long) Deprecated in TensorRT 10.0. Superseded by reallocateOutputAsync with cudaStream_t argument
org.bytedeco.tensorrt.nvinfer.VDynamicQuantizeLayer.setAxis(int)
org.bytedeco.tensorrt.nvinfer.IDynamicQuantizeLayer.setAxis(int)
org.bytedeco.tensorrt.nvinfer.VDynamicQuantizeLayer.setBlockSize(int)
org.bytedeco.tensorrt.nvinfer.IDynamicQuantizeLayer.setBlockSize(int)
org.bytedeco.tensorrt.nvinfer.ITensor.setBroadcastAcrossBatch(boolean) Deprecated in TensorRT 10.0. Implicit batch is not supported since TensorRT 10.0.
org.bytedeco.tensorrt.nvinfer.VAttention.setCausal(boolean)
org.bytedeco.tensorrt.nvinfer.IAttention.setCausal(boolean) Deprecated in TensorRT 10.16. Superseded by setCausalKind.
org.bytedeco.tensorrt.nvinfer.IExecutionContext.setDeviceMemory(Pointer) Deprecated in TensorRT 10.1. Superseded by setDeviceMemoryV2(). \warning Weight streaming related scratch memory will be allocated by TensorRT if the memory is set by this API. Please use setDeviceMemoryV2() instead.
org.bytedeco.tensorrt.nvinfer.ITensor.setLocation(int)
org.bytedeco.tensorrt.nvinfer.ITensor.setLocation(nvinfer.TensorLocation) Deprecated in TensorRT 10.0. RNNv2 is not supported and the location must always be TensorLocation::kDEVICE since TensorRT 10.0.

Deprecated Enum Constants

Enum Constant and Description
org.bytedeco.tensorrt.global.nvinfer.BuilderFlag.kDIRECT_IO Deprecated in TensorRT 10.7. Unneeded API.
org.bytedeco.tensorrt.global.nvinfer.BuilderFlag.kSAFETY_SCOPE Deprecated in TensorRT 10.16. In EngineCapability::kSTANDARD flow, safety restrictions are no longer supported. In EngineCapability::kSAFETY and EngineCapability::kDLA_STANDALONE flows, restrictions are enforced natively. This flag is retained for API compatibility but is ignored.
org.bytedeco.tensorrt.global.nvinfer.NetworkDefinitionCreationFlag.kSTRONGLY_TYPED Deprecated in TensorRT 11.0. Strongly typed mode is always enabled. This flag is retained for API compatibility but is ignored.