public class cudart extends cudart
Modifier and Type | Field and Description |
---|---|
static int |
__CUDART_API_VERSION |
static int |
CU_ACCESS_PROPERTY_NORMAL
enum CUaccessProperty
|
static int |
CU_ACCESS_PROPERTY_PERSISTING
enum CUaccessProperty
|
static int |
CU_ACCESS_PROPERTY_STREAMING
enum CUaccessProperty
|
static int |
CU_AD_FORMAT_BC1_UNORM
enum CUarray_format
|
static int |
CU_AD_FORMAT_BC1_UNORM_SRGB
enum CUarray_format
|
static int |
CU_AD_FORMAT_BC2_UNORM
enum CUarray_format
|
static int |
CU_AD_FORMAT_BC2_UNORM_SRGB
enum CUarray_format
|
static int |
CU_AD_FORMAT_BC3_UNORM
enum CUarray_format
|
static int |
CU_AD_FORMAT_BC3_UNORM_SRGB
enum CUarray_format
|
static int |
CU_AD_FORMAT_BC4_SNORM
enum CUarray_format
|
static int |
CU_AD_FORMAT_BC4_UNORM
enum CUarray_format
|
static int |
CU_AD_FORMAT_BC5_SNORM
enum CUarray_format
|
static int |
CU_AD_FORMAT_BC5_UNORM
enum CUarray_format
|
static int |
CU_AD_FORMAT_BC6H_SF16
enum CUarray_format
|
static int |
CU_AD_FORMAT_BC6H_UF16
enum CUarray_format
|
static int |
CU_AD_FORMAT_BC7_UNORM
enum CUarray_format
|
static int |
CU_AD_FORMAT_BC7_UNORM_SRGB
enum CUarray_format
|
static int |
CU_AD_FORMAT_FLOAT
enum CUarray_format
|
static int |
CU_AD_FORMAT_HALF
enum CUarray_format
|
static int |
CU_AD_FORMAT_NV12
enum CUarray_format
|
static int |
CU_AD_FORMAT_SIGNED_INT16
enum CUarray_format
|
static int |
CU_AD_FORMAT_SIGNED_INT32
enum CUarray_format
|
static int |
CU_AD_FORMAT_SIGNED_INT8
enum CUarray_format
|
static int |
CU_AD_FORMAT_SNORM_INT16X1
enum CUarray_format
|
static int |
CU_AD_FORMAT_SNORM_INT16X2
enum CUarray_format
|
static int |
CU_AD_FORMAT_SNORM_INT16X4
enum CUarray_format
|
static int |
CU_AD_FORMAT_SNORM_INT8X1
enum CUarray_format
|
static int |
CU_AD_FORMAT_SNORM_INT8X2
enum CUarray_format
|
static int |
CU_AD_FORMAT_SNORM_INT8X4
enum CUarray_format
|
static int |
CU_AD_FORMAT_UNORM_INT16X1
enum CUarray_format
|
static int |
CU_AD_FORMAT_UNORM_INT16X2
enum CUarray_format
|
static int |
CU_AD_FORMAT_UNORM_INT16X4
enum CUarray_format
|
static int |
CU_AD_FORMAT_UNORM_INT8X1
enum CUarray_format
|
static int |
CU_AD_FORMAT_UNORM_INT8X2
enum CUarray_format
|
static int |
CU_AD_FORMAT_UNORM_INT8X4
enum CUarray_format
|
static int |
CU_AD_FORMAT_UNSIGNED_INT16
enum CUarray_format
|
static int |
CU_AD_FORMAT_UNSIGNED_INT32
enum CUarray_format
|
static int |
CU_AD_FORMAT_UNSIGNED_INT8
enum CUarray_format
|
static int |
CU_ARRAY_SPARSE_PROPERTIES_SINGLE_MIPTAIL
Indicates that the layered sparse CUDA array or CUDA mipmapped array has a single mip tail region for all layers
|
static int |
CU_ARRAY_SPARSE_SUBRESOURCE_TYPE_MIPTAIL
enum CUarraySparseSubresourceType
|
static int |
CU_ARRAY_SPARSE_SUBRESOURCE_TYPE_SPARSE_LEVEL
enum CUarraySparseSubresourceType
|
static int |
CU_CLUSTER_SCHEDULING_POLICY_DEFAULT
enum CUclusterSchedulingPolicy
|
static int |
CU_CLUSTER_SCHEDULING_POLICY_LOAD_BALANCING
enum CUclusterSchedulingPolicy
|
static int |
CU_CLUSTER_SCHEDULING_POLICY_SPREAD
enum CUclusterSchedulingPolicy
|
static int |
CU_COMPUTE_ACCELERATED_TARGET_BASE |
static int |
CU_COMPUTEMODE_DEFAULT
enum CUcomputemode
|
static int |
CU_COMPUTEMODE_EXCLUSIVE_PROCESS
enum CUcomputemode
|
static int |
CU_COMPUTEMODE_PROHIBITED
enum CUcomputemode
|
static int |
CU_COREDUMP_ENABLE_ON_EXCEPTION
enum CUcoredumpSettings
|
static int |
CU_COREDUMP_ENABLE_USER_TRIGGER
enum CUcoredumpSettings
|
static int |
CU_COREDUMP_FILE
enum CUcoredumpSettings
|
static int |
CU_COREDUMP_LIGHTWEIGHT
enum CUcoredumpSettings
|
static int |
CU_COREDUMP_MAX
enum CUcoredumpSettings
|
static int |
CU_COREDUMP_PIPE
enum CUcoredumpSettings
|
static int |
CU_COREDUMP_TRIGGER_HOST
enum CUcoredumpSettings
|
static int |
CU_CTX_BLOCKING_SYNC
enum CUctx_flags
|
static int |
CU_CTX_COREDUMP_ENABLE
enum CUctx_flags
|
static int |
CU_CTX_FLAGS_MASK
enum CUctx_flags
|
static int |
CU_CTX_LMEM_RESIZE_TO_MAX
enum CUctx_flags
|
static int |
CU_CTX_MAP_HOST
enum CUctx_flags
|
static int |
CU_CTX_SCHED_AUTO
enum CUctx_flags
|
static int |
CU_CTX_SCHED_BLOCKING_SYNC
enum CUctx_flags
|
static int |
CU_CTX_SCHED_MASK
enum CUctx_flags
|
static int |
CU_CTX_SCHED_SPIN
enum CUctx_flags
|
static int |
CU_CTX_SCHED_YIELD
enum CUctx_flags
|
static int |
CU_CTX_SYNC_MEMOPS
enum CUctx_flags
|
static int |
CU_CTX_USER_COREDUMP_ENABLE
enum CUctx_flags
|
static int |
CU_CUBEMAP_FACE_NEGATIVE_X
enum CUarray_cubemap_face
|
static int |
CU_CUBEMAP_FACE_NEGATIVE_Y
enum CUarray_cubemap_face
|
static int |
CU_CUBEMAP_FACE_NEGATIVE_Z
enum CUarray_cubemap_face
|
static int |
CU_CUBEMAP_FACE_POSITIVE_X
enum CUarray_cubemap_face
|
static int |
CU_CUBEMAP_FACE_POSITIVE_Y
enum CUarray_cubemap_face
|
static int |
CU_CUBEMAP_FACE_POSITIVE_Z
enum CUarray_cubemap_face
|
static int |
CU_DEVICE_ATTRIBUTE_ASYNC_ENGINE_COUNT
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_CAN_FLUSH_REMOTE_WRITES
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_CAN_MAP_HOST_MEMORY
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_CAN_TEX2D_GATHER
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_CAN_USE_64_BIT_STREAM_MEM_OPS
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_CAN_USE_64_BIT_STREAM_MEM_OPS_V1
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_CAN_USE_HOST_POINTER_FOR_REGISTERED_MEM
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_CAN_USE_STREAM_MEM_OPS_V1
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_CAN_USE_STREAM_WAIT_VALUE_NOR
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_CAN_USE_STREAM_WAIT_VALUE_NOR_V1
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_CLOCK_RATE
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_CLUSTER_LAUNCH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_COMPUTE_CAPABILITY_MAJOR
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_COMPUTE_CAPABILITY_MINOR
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_COMPUTE_MODE
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_COMPUTE_PREEMPTION_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_CONCURRENT_KERNELS
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_CONCURRENT_MANAGED_ACCESS
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_COOPERATIVE_LAUNCH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_COOPERATIVE_MULTI_DEVICE_LAUNCH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_DEFERRED_MAPPING_CUDA_ARRAY_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_DIRECT_MANAGED_MEM_ACCESS_FROM_HOST
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_DMA_BUF_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_ECC_ENABLED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_GENERIC_COMPRESSION_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_GLOBAL_L1_CACHE_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_GLOBAL_MEMORY_BUS_WIDTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_GPU_DIRECT_RDMA_FLUSH_WRITES_OPTIONS
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_GPU_DIRECT_RDMA_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_GPU_DIRECT_RDMA_WITH_CUDA_VMM_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_GPU_DIRECT_RDMA_WRITES_ORDERING
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_GPU_OVERLAP
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_HANDLE_TYPE_FABRIC_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_HANDLE_TYPE_POSIX_FILE_DESCRIPTOR_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_HANDLE_TYPE_WIN32_HANDLE_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_HANDLE_TYPE_WIN32_KMT_HANDLE_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_HOST_NATIVE_ATOMIC_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_HOST_NUMA_ID
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_HOST_REGISTER_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_INTEGRATED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_IPC_EVENT_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_KERNEL_EXEC_TIMEOUT
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_L2_CACHE_SIZE
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_LOCAL_L1_CACHE_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MANAGED_MEMORY
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAX
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAX_ACCESS_POLICY_WINDOW_SIZE
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAX_BLOCK_DIM_X
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAX_BLOCK_DIM_Y
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAX_BLOCK_DIM_Z
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAX_BLOCKS_PER_MULTIPROCESSOR
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAX_GRID_DIM_X
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAX_GRID_DIM_Y
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAX_GRID_DIM_Z
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAX_PERSISTING_L2_CACHE_SIZE
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAX_PITCH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAX_REGISTERS_PER_BLOCK
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAX_REGISTERS_PER_MULTIPROCESSOR
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAX_SHARED_MEMORY_PER_BLOCK
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAX_SHARED_MEMORY_PER_BLOCK_OPTIN
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAX_SHARED_MEMORY_PER_MULTIPROCESSOR
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAX_THREADS_PER_BLOCK
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAX_THREADS_PER_MULTIPROCESSOR
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE1D_LAYERED_LAYERS
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE1D_LAYERED_WIDTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE1D_WIDTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE2D_HEIGHT
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE2D_LAYERED_HEIGHT
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE2D_LAYERED_LAYERS
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE2D_LAYERED_WIDTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE2D_WIDTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE3D_DEPTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE3D_HEIGHT
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE3D_WIDTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACECUBEMAP_LAYERED_LAYERS
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACECUBEMAP_LAYERED_WIDTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACECUBEMAP_WIDTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE1D_LAYERED_LAYERS
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE1D_LAYERED_WIDTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE1D_LINEAR_WIDTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE1D_MIPMAPPED_WIDTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE1D_WIDTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_ARRAY_HEIGHT
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_ARRAY_NUMSLICES
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_ARRAY_WIDTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_GATHER_HEIGHT
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_GATHER_WIDTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_HEIGHT
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_LAYERED_HEIGHT
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_LAYERED_LAYERS
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_LAYERED_WIDTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_LINEAR_HEIGHT
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_LINEAR_PITCH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_LINEAR_WIDTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_MIPMAPPED_HEIGHT
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_MIPMAPPED_WIDTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_WIDTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE3D_DEPTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE3D_DEPTH_ALTERNATE
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE3D_HEIGHT
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE3D_HEIGHT_ALTERNATE
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE3D_WIDTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE3D_WIDTH_ALTERNATE
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURECUBEMAP_LAYERED_LAYERS
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURECUBEMAP_LAYERED_WIDTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURECUBEMAP_WIDTH
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MEM_SYNC_DOMAIN_COUNT
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MEMORY_CLOCK_RATE
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MEMORY_POOLS_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MEMPOOL_SUPPORTED_HANDLE_TYPES
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MPS_ENABLED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MULTI_GPU_BOARD
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MULTI_GPU_BOARD_GROUP_ID
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MULTICAST_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_MULTIPROCESSOR_COUNT
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_NUMA_CONFIG
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_NUMA_ID
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_PAGEABLE_MEMORY_ACCESS
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_PAGEABLE_MEMORY_ACCESS_USES_HOST_PAGE_TABLES
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_PCI_BUS_ID
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_PCI_DEVICE_ID
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_PCI_DOMAIN_ID
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_READ_ONLY_HOST_REGISTER_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_REGISTERS_PER_BLOCK
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_RESERVED_SHARED_MEMORY_PER_BLOCK
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_SHARED_MEMORY_PER_BLOCK
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_SINGLE_TO_DOUBLE_PRECISION_PERF_RATIO
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_SPARSE_CUDA_ARRAY_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_STREAM_PRIORITIES_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_SURFACE_ALIGNMENT
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_TCC_DRIVER
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_TENSOR_MAP_ACCESS_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_TEXTURE_ALIGNMENT
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_TEXTURE_PITCH_ALIGNMENT
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_TIMELINE_SEMAPHORE_INTEROP_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_TOTAL_CONSTANT_MEMORY
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_UNIFIED_ADDRESSING
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_UNIFIED_FUNCTION_POINTERS
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_VIRTUAL_ADDRESS_MANAGEMENT_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_VIRTUAL_MEMORY_MANAGEMENT_SUPPORTED
enum CUdevice_attribute
|
static int |
CU_DEVICE_ATTRIBUTE_WARP_SIZE
enum CUdevice_attribute
|
static int |
CU_DEVICE_CPU |
static int |
CU_DEVICE_INVALID |
static int |
CU_DEVICE_NUMA_CONFIG_NONE
enum CUdeviceNumaConfig
|
static int |
CU_DEVICE_NUMA_CONFIG_NUMA_NODE
enum CUdeviceNumaConfig
|
static int |
CU_DEVICE_P2P_ATTRIBUTE_ACCESS_ACCESS_SUPPORTED
enum CUdevice_P2PAttribute
|
static int |
CU_DEVICE_P2P_ATTRIBUTE_ACCESS_SUPPORTED
enum CUdevice_P2PAttribute
|
static int |
CU_DEVICE_P2P_ATTRIBUTE_CUDA_ARRAY_ACCESS_SUPPORTED
enum CUdevice_P2PAttribute
|
static int |
CU_DEVICE_P2P_ATTRIBUTE_NATIVE_ATOMIC_SUPPORTED
enum CUdevice_P2PAttribute
|
static int |
CU_DEVICE_P2P_ATTRIBUTE_PERFORMANCE_RANK
enum CUdevice_P2PAttribute
|
static int |
CU_EVENT_BLOCKING_SYNC
enum CUevent_flags
|
static int |
CU_EVENT_DEFAULT
enum CUevent_flags
|
static int |
CU_EVENT_DISABLE_TIMING
enum CUevent_flags
|
static int |
CU_EVENT_INTERPROCESS
enum CUevent_flags
|
static int |
CU_EVENT_RECORD_DEFAULT
enum CUevent_record_flags
|
static int |
CU_EVENT_RECORD_EXTERNAL
enum CUevent_record_flags
|
static int |
CU_EVENT_SCHED_AUTO
enum CUevent_sched_flags
|
static int |
CU_EVENT_SCHED_BLOCKING_SYNC
enum CUevent_sched_flags
|
static int |
CU_EVENT_SCHED_SPIN
enum CUevent_sched_flags
|
static int |
CU_EVENT_SCHED_YIELD
enum CUevent_sched_flags
|
static int |
CU_EVENT_WAIT_DEFAULT
enum CUevent_wait_flags
|
static int |
CU_EVENT_WAIT_EXTERNAL
enum CUevent_wait_flags
|
static int |
CU_EXEC_AFFINITY_TYPE_MAX
enum CUexecAffinityType
|
static int |
CU_EXEC_AFFINITY_TYPE_SM_COUNT
enum CUexecAffinityType
|
static int |
CU_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_RESOURCE
enum CUexternalMemoryHandleType
|
static int |
CU_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_RESOURCE_KMT
enum CUexternalMemoryHandleType
|
static int |
CU_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_HEAP
enum CUexternalMemoryHandleType
|
static int |
CU_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_RESOURCE
enum CUexternalMemoryHandleType
|
static int |
CU_EXTERNAL_MEMORY_HANDLE_TYPE_NVSCIBUF
enum CUexternalMemoryHandleType
|
static int |
CU_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_FD
enum CUexternalMemoryHandleType
|
static int |
CU_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32
enum CUexternalMemoryHandleType
|
static int |
CU_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_KMT
enum CUexternalMemoryHandleType
|
static int |
CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D11_FENCE
enum CUexternalSemaphoreHandleType
|
static int |
CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D11_KEYED_MUTEX
enum CUexternalSemaphoreHandleType
|
static int |
CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D11_KEYED_MUTEX_KMT
enum CUexternalSemaphoreHandleType
|
static int |
CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D12_FENCE
enum CUexternalSemaphoreHandleType
|
static int |
CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_NVSCISYNC
enum CUexternalSemaphoreHandleType
|
static int |
CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_FD
enum CUexternalSemaphoreHandleType
|
static int |
CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32
enum CUexternalSemaphoreHandleType
|
static int |
CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_KMT
enum CUexternalSemaphoreHandleType
|
static int |
CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_TIMELINE_SEMAPHORE_FD
enum CUexternalSemaphoreHandleType
|
static int |
CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_TIMELINE_SEMAPHORE_WIN32
enum CUexternalSemaphoreHandleType
|
static int |
CU_FLUSH_GPU_DIRECT_RDMA_WRITES_OPTION_HOST
enum CUflushGPUDirectRDMAWritesOptions
|
static int |
CU_FLUSH_GPU_DIRECT_RDMA_WRITES_OPTION_MEMOPS
enum CUflushGPUDirectRDMAWritesOptions
|
static int |
CU_FLUSH_GPU_DIRECT_RDMA_WRITES_TARGET_CURRENT_CTX
enum CUflushGPUDirectRDMAWritesTarget
|
static int |
CU_FLUSH_GPU_DIRECT_RDMA_WRITES_TO_ALL_DEVICES
enum CUflushGPUDirectRDMAWritesScope
|
static int |
CU_FLUSH_GPU_DIRECT_RDMA_WRITES_TO_OWNER
enum CUflushGPUDirectRDMAWritesScope
|
static int |
CU_FUNC_ATTRIBUTE_BINARY_VERSION
enum CUfunction_attribute
|
static int |
CU_FUNC_ATTRIBUTE_CACHE_MODE_CA
enum CUfunction_attribute
|
static int |
CU_FUNC_ATTRIBUTE_CLUSTER_SCHEDULING_POLICY_PREFERENCE
enum CUfunction_attribute
|
static int |
CU_FUNC_ATTRIBUTE_CLUSTER_SIZE_MUST_BE_SET
enum CUfunction_attribute
|
static int |
CU_FUNC_ATTRIBUTE_CONST_SIZE_BYTES
enum CUfunction_attribute
|
static int |
CU_FUNC_ATTRIBUTE_LOCAL_SIZE_BYTES
enum CUfunction_attribute
|
static int |
CU_FUNC_ATTRIBUTE_MAX
enum CUfunction_attribute
|
static int |
CU_FUNC_ATTRIBUTE_MAX_DYNAMIC_SHARED_SIZE_BYTES
enum CUfunction_attribute
|
static int |
CU_FUNC_ATTRIBUTE_MAX_THREADS_PER_BLOCK
enum CUfunction_attribute
|
static int |
CU_FUNC_ATTRIBUTE_NON_PORTABLE_CLUSTER_SIZE_ALLOWED
enum CUfunction_attribute
|
static int |
CU_FUNC_ATTRIBUTE_NUM_REGS
enum CUfunction_attribute
|
static int |
CU_FUNC_ATTRIBUTE_PREFERRED_SHARED_MEMORY_CARVEOUT
enum CUfunction_attribute
|
static int |
CU_FUNC_ATTRIBUTE_PTX_VERSION
enum CUfunction_attribute
|
static int |
CU_FUNC_ATTRIBUTE_REQUIRED_CLUSTER_DEPTH
enum CUfunction_attribute
|
static int |
CU_FUNC_ATTRIBUTE_REQUIRED_CLUSTER_HEIGHT
enum CUfunction_attribute
|
static int |
CU_FUNC_ATTRIBUTE_REQUIRED_CLUSTER_WIDTH
enum CUfunction_attribute
|
static int |
CU_FUNC_ATTRIBUTE_SHARED_SIZE_BYTES
enum CUfunction_attribute
|
static int |
CU_FUNC_CACHE_PREFER_EQUAL
enum CUfunc_cache
|
static int |
CU_FUNC_CACHE_PREFER_L1
enum CUfunc_cache
|
static int |
CU_FUNC_CACHE_PREFER_NONE
enum CUfunc_cache
|
static int |
CU_FUNC_CACHE_PREFER_SHARED
enum CUfunc_cache
|
static int |
CU_GET_PROC_ADDRESS_DEFAULT
enum CUdriverProcAddress_flags
|
static int |
CU_GET_PROC_ADDRESS_LEGACY_STREAM
enum CUdriverProcAddress_flags
|
static int |
CU_GET_PROC_ADDRESS_PER_THREAD_DEFAULT_STREAM
enum CUdriverProcAddress_flags
|
static int |
CU_GET_PROC_ADDRESS_SUCCESS
enum CUdriverProcAddressQueryResult
|
static int |
CU_GET_PROC_ADDRESS_SYMBOL_NOT_FOUND
enum CUdriverProcAddressQueryResult
|
static int |
CU_GET_PROC_ADDRESS_VERSION_NOT_SUFFICIENT
enum CUdriverProcAddressQueryResult
|
static int |
CU_GL_DEVICE_LIST_ALL
enum CUGLDeviceList
|
static int |
CU_GL_DEVICE_LIST_CURRENT_FRAME
enum CUGLDeviceList
|
static int |
CU_GL_DEVICE_LIST_NEXT_FRAME
enum CUGLDeviceList
|
static int |
CU_GL_MAP_RESOURCE_FLAGS_NONE
enum CUGLmap_flags
|
static int |
CU_GL_MAP_RESOURCE_FLAGS_READ_ONLY
enum CUGLmap_flags
|
static int |
CU_GL_MAP_RESOURCE_FLAGS_WRITE_DISCARD
enum CUGLmap_flags
|
static int |
CU_GPU_DIRECT_RDMA_WRITES_ORDERING_ALL_DEVICES
enum CUGPUDirectRDMAWritesOrdering
|
static int |
CU_GPU_DIRECT_RDMA_WRITES_ORDERING_NONE
enum CUGPUDirectRDMAWritesOrdering
|
static int |
CU_GPU_DIRECT_RDMA_WRITES_ORDERING_OWNER
enum CUGPUDirectRDMAWritesOrdering
|
static int |
CU_GRAPH_COND_ASSIGN_DEFAULT
Default value is applied when graph is launched.
|
static int |
CU_GRAPH_COND_TYPE_IF
enum CUgraphConditionalNodeType
|
static int |
CU_GRAPH_COND_TYPE_WHILE
enum CUgraphConditionalNodeType
|
static int |
CU_GRAPH_DEBUG_DOT_FLAGS_BATCH_MEM_OP_NODE_PARAMS
enum CUgraphDebugDot_flags
|
static int |
CU_GRAPH_DEBUG_DOT_FLAGS_CONDITIONAL_NODE_PARAMS
enum CUgraphDebugDot_flags
|
static int |
CU_GRAPH_DEBUG_DOT_FLAGS_EVENT_NODE_PARAMS
enum CUgraphDebugDot_flags
|
static int |
CU_GRAPH_DEBUG_DOT_FLAGS_EXT_SEMAS_SIGNAL_NODE_PARAMS
enum CUgraphDebugDot_flags
|
static int |
CU_GRAPH_DEBUG_DOT_FLAGS_EXT_SEMAS_WAIT_NODE_PARAMS
enum CUgraphDebugDot_flags
|
static int |
CU_GRAPH_DEBUG_DOT_FLAGS_EXTRA_TOPO_INFO
enum CUgraphDebugDot_flags
|
static int |
CU_GRAPH_DEBUG_DOT_FLAGS_HANDLES
enum CUgraphDebugDot_flags
|
static int |
CU_GRAPH_DEBUG_DOT_FLAGS_HOST_NODE_PARAMS
enum CUgraphDebugDot_flags
|
static int |
CU_GRAPH_DEBUG_DOT_FLAGS_KERNEL_NODE_ATTRIBUTES
enum CUgraphDebugDot_flags
|
static int |
CU_GRAPH_DEBUG_DOT_FLAGS_KERNEL_NODE_PARAMS
enum CUgraphDebugDot_flags
|
static int |
CU_GRAPH_DEBUG_DOT_FLAGS_MEM_ALLOC_NODE_PARAMS
enum CUgraphDebugDot_flags
|
static int |
CU_GRAPH_DEBUG_DOT_FLAGS_MEM_FREE_NODE_PARAMS
enum CUgraphDebugDot_flags
|
static int |
CU_GRAPH_DEBUG_DOT_FLAGS_MEMCPY_NODE_PARAMS
enum CUgraphDebugDot_flags
|
static int |
CU_GRAPH_DEBUG_DOT_FLAGS_MEMSET_NODE_PARAMS
enum CUgraphDebugDot_flags
|
static int |
CU_GRAPH_DEBUG_DOT_FLAGS_RUNTIME_TYPES
enum CUgraphDebugDot_flags
|
static int |
CU_GRAPH_DEBUG_DOT_FLAGS_VERBOSE
enum CUgraphDebugDot_flags
|
static int |
CU_GRAPH_DEPENDENCY_TYPE_DEFAULT
enum CUgraphDependencyType
|
static int |
CU_GRAPH_DEPENDENCY_TYPE_PROGRAMMATIC
enum CUgraphDependencyType
|
static int |
CU_GRAPH_EXEC_UPDATE_ERROR
enum CUgraphExecUpdateResult
|
static int |
CU_GRAPH_EXEC_UPDATE_ERROR_ATTRIBUTES_CHANGED
enum CUgraphExecUpdateResult
|
static int |
CU_GRAPH_EXEC_UPDATE_ERROR_FUNCTION_CHANGED
enum CUgraphExecUpdateResult
|
static int |
CU_GRAPH_EXEC_UPDATE_ERROR_NODE_TYPE_CHANGED
enum CUgraphExecUpdateResult
|
static int |
CU_GRAPH_EXEC_UPDATE_ERROR_NOT_SUPPORTED
enum CUgraphExecUpdateResult
|
static int |
CU_GRAPH_EXEC_UPDATE_ERROR_PARAMETERS_CHANGED
enum CUgraphExecUpdateResult
|
static int |
CU_GRAPH_EXEC_UPDATE_ERROR_TOPOLOGY_CHANGED
enum CUgraphExecUpdateResult
|
static int |
CU_GRAPH_EXEC_UPDATE_ERROR_UNSUPPORTED_FUNCTION_CHANGE
enum CUgraphExecUpdateResult
|
static int |
CU_GRAPH_EXEC_UPDATE_SUCCESS
enum CUgraphExecUpdateResult
|
static int |
CU_GRAPH_KERNEL_NODE_PORT_DEFAULT
This port activates when the kernel has finished executing.
|
static int |
CU_GRAPH_KERNEL_NODE_PORT_LAUNCH_ORDER
This port activates when all blocks of the kernel have begun execution.
|
static int |
CU_GRAPH_KERNEL_NODE_PORT_PROGRAMMATIC
This port activates when all blocks of the kernel have performed cudaTriggerProgrammaticLaunchCompletion()
or have terminated.
|
static int |
CU_GRAPH_MEM_ATTR_RESERVED_MEM_CURRENT
enum CUgraphMem_attribute
|
static int |
CU_GRAPH_MEM_ATTR_RESERVED_MEM_HIGH
enum CUgraphMem_attribute
|
static int |
CU_GRAPH_MEM_ATTR_USED_MEM_CURRENT
enum CUgraphMem_attribute
|
static int |
CU_GRAPH_MEM_ATTR_USED_MEM_HIGH
enum CUgraphMem_attribute
|
static int |
CU_GRAPH_NODE_TYPE_BATCH_MEM_OP
enum CUgraphNodeType
|
static int |
CU_GRAPH_NODE_TYPE_CONDITIONAL
enum CUgraphNodeType
|
static int |
CU_GRAPH_NODE_TYPE_EMPTY
enum CUgraphNodeType
|
static int |
CU_GRAPH_NODE_TYPE_EVENT_RECORD
enum CUgraphNodeType
|
static int |
CU_GRAPH_NODE_TYPE_EXT_SEMAS_SIGNAL
enum CUgraphNodeType
|
static int |
CU_GRAPH_NODE_TYPE_EXT_SEMAS_WAIT
enum CUgraphNodeType
|
static int |
CU_GRAPH_NODE_TYPE_GRAPH
enum CUgraphNodeType
|
static int |
CU_GRAPH_NODE_TYPE_HOST
enum CUgraphNodeType
|
static int |
CU_GRAPH_NODE_TYPE_KERNEL
enum CUgraphNodeType
|
static int |
CU_GRAPH_NODE_TYPE_MEM_ALLOC
enum CUgraphNodeType
|
static int |
CU_GRAPH_NODE_TYPE_MEM_FREE
enum CUgraphNodeType
|
static int |
CU_GRAPH_NODE_TYPE_MEMCPY
enum CUgraphNodeType
|
static int |
CU_GRAPH_NODE_TYPE_MEMSET
enum CUgraphNodeType
|
static int |
CU_GRAPH_NODE_TYPE_WAIT_EVENT
enum CUgraphNodeType
|
static int |
CU_GRAPH_USER_OBJECT_MOVE
enum CUuserObjectRetain_flags
|
static int |
CU_GRAPHICS_MAP_RESOURCE_FLAGS_NONE
enum CUgraphicsMapResourceFlags
|
static int |
CU_GRAPHICS_MAP_RESOURCE_FLAGS_READ_ONLY
enum CUgraphicsMapResourceFlags
|
static int |
CU_GRAPHICS_MAP_RESOURCE_FLAGS_WRITE_DISCARD
enum CUgraphicsMapResourceFlags
|
static int |
CU_GRAPHICS_REGISTER_FLAGS_NONE
enum CUgraphicsRegisterFlags
|
static int |
CU_GRAPHICS_REGISTER_FLAGS_READ_ONLY
enum CUgraphicsRegisterFlags
|
static int |
CU_GRAPHICS_REGISTER_FLAGS_SURFACE_LDST
enum CUgraphicsRegisterFlags
|
static int |
CU_GRAPHICS_REGISTER_FLAGS_TEXTURE_GATHER
enum CUgraphicsRegisterFlags
|
static int |
CU_GRAPHICS_REGISTER_FLAGS_WRITE_DISCARD
enum CUgraphicsRegisterFlags
|
static int |
CU_IPC_HANDLE_SIZE
CUDA IPC handle size
|
static int |
CU_IPC_MEM_LAZY_ENABLE_PEER_ACCESS
enum CUipcMem_flags
|
static int |
CU_JIT_CACHE_MODE
enum CUjit_option
|
static int |
CU_JIT_CACHE_OPTION_CA
enum CUjit_cacheMode
|
static int |
CU_JIT_CACHE_OPTION_CG
enum CUjit_cacheMode
|
static int |
CU_JIT_CACHE_OPTION_NONE
enum CUjit_cacheMode
|
static int |
CU_JIT_ERROR_LOG_BUFFER
enum CUjit_option
|
static int |
CU_JIT_ERROR_LOG_BUFFER_SIZE_BYTES
enum CUjit_option
|
static int |
CU_JIT_FALLBACK_STRATEGY
enum CUjit_option
|
static int |
CU_JIT_FAST_COMPILE
enum CUjit_option
|
static int |
CU_JIT_FMA
enum CUjit_option
|
static int |
CU_JIT_FTZ
enum CUjit_option
|
static int |
CU_JIT_GENERATE_DEBUG_INFO
enum CUjit_option
|
static int |
CU_JIT_GENERATE_LINE_INFO
enum CUjit_option
|
static int |
CU_JIT_GLOBAL_SYMBOL_ADDRESSES
enum CUjit_option
|
static int |
CU_JIT_GLOBAL_SYMBOL_COUNT
enum CUjit_option
|
static int |
CU_JIT_GLOBAL_SYMBOL_NAMES
enum CUjit_option
|
static int |
CU_JIT_INFO_LOG_BUFFER
enum CUjit_option
|
static int |
CU_JIT_INFO_LOG_BUFFER_SIZE_BYTES
enum CUjit_option
|
static int |
CU_JIT_INPUT_CUBIN
enum CUjitInputType
|
static int |
CU_JIT_INPUT_FATBINARY
enum CUjitInputType
|
static int |
CU_JIT_INPUT_LIBRARY
enum CUjitInputType
|
static int |
CU_JIT_INPUT_NVVM
enum CUjitInputType
|
static int |
CU_JIT_INPUT_OBJECT
enum CUjitInputType
|
static int |
CU_JIT_INPUT_PTX
enum CUjitInputType
|
static int |
CU_JIT_LOG_VERBOSE
enum CUjit_option
|
static int |
CU_JIT_LTO
enum CUjit_option
|
static int |
CU_JIT_MAX_REGISTERS
enum CUjit_option
|
static int |
CU_JIT_MIN_CTA_PER_SM
enum CUjit_option
|
static int |
CU_JIT_NEW_SM3X_OPT
enum CUjit_option
|
static int |
CU_JIT_NUM_INPUT_TYPES
enum CUjitInputType
|
static int |
CU_JIT_NUM_OPTIONS
enum CUjit_option
|
static int |
CU_JIT_OPTIMIZATION_LEVEL
enum CUjit_option
|
static int |
CU_JIT_OPTIMIZE_UNUSED_DEVICE_VARIABLES
enum CUjit_option
|
static int |
CU_JIT_POSITION_INDEPENDENT_CODE
enum CUjit_option
|
static int |
CU_JIT_PREC_DIV
enum CUjit_option
|
static int |
CU_JIT_PREC_SQRT
enum CUjit_option
|
static int |
CU_JIT_REFERENCED_KERNEL_COUNT
enum CUjit_option
|
static int |
CU_JIT_REFERENCED_KERNEL_NAMES
enum CUjit_option
|
static int |
CU_JIT_REFERENCED_VARIABLE_COUNT
enum CUjit_option
|
static int |
CU_JIT_REFERENCED_VARIABLE_NAMES
enum CUjit_option
|
static int |
CU_JIT_TARGET
enum CUjit_option
|
static int |
CU_JIT_TARGET_FROM_CUCONTEXT
enum CUjit_option
|
static int |
CU_JIT_THREADS_PER_BLOCK
enum CUjit_option
|
static int |
CU_JIT_WALL_TIME
enum CUjit_option
|
static int |
CU_KERNEL_NODE_ATTRIBUTE_ACCESS_POLICY_WINDOW |
static int |
CU_KERNEL_NODE_ATTRIBUTE_CLUSTER_DIMENSION |
static int |
CU_KERNEL_NODE_ATTRIBUTE_CLUSTER_SCHEDULING_POLICY_PREFERENCE |
static int |
CU_KERNEL_NODE_ATTRIBUTE_COOPERATIVE |
static int |
CU_KERNEL_NODE_ATTRIBUTE_MEM_SYNC_DOMAIN |
static int |
CU_KERNEL_NODE_ATTRIBUTE_MEM_SYNC_DOMAIN_MAP |
static int |
CU_KERNEL_NODE_ATTRIBUTE_PRIORITY |
static int |
CU_LAUNCH_ATTRIBUTE_ACCESS_POLICY_WINDOW
enum CUlaunchAttributeID
|
static int |
CU_LAUNCH_ATTRIBUTE_CLUSTER_DIMENSION
enum CUlaunchAttributeID
|
static int |
CU_LAUNCH_ATTRIBUTE_CLUSTER_SCHEDULING_POLICY_PREFERENCE
enum CUlaunchAttributeID
|
static int |
CU_LAUNCH_ATTRIBUTE_COOPERATIVE
enum CUlaunchAttributeID
|
static int |
CU_LAUNCH_ATTRIBUTE_IGNORE
enum CUlaunchAttributeID
|
static int |
CU_LAUNCH_ATTRIBUTE_LAUNCH_COMPLETION_EVENT
enum CUlaunchAttributeID
|
static int |
CU_LAUNCH_ATTRIBUTE_MAX
enum CUlaunchAttributeID
|
static int |
CU_LAUNCH_ATTRIBUTE_MEM_SYNC_DOMAIN
enum CUlaunchAttributeID
|
static int |
CU_LAUNCH_ATTRIBUTE_MEM_SYNC_DOMAIN_MAP
enum CUlaunchAttributeID
|
static int |
CU_LAUNCH_ATTRIBUTE_PRIORITY
enum CUlaunchAttributeID
|
static int |
CU_LAUNCH_ATTRIBUTE_PROGRAMMATIC_EVENT
enum CUlaunchAttributeID
|
static int |
CU_LAUNCH_ATTRIBUTE_PROGRAMMATIC_STREAM_SERIALIZATION
enum CUlaunchAttributeID
|
static int |
CU_LAUNCH_ATTRIBUTE_SYNCHRONIZATION_POLICY
enum CUlaunchAttributeID
|
static int |
CU_LAUNCH_MEM_SYNC_DOMAIN_DEFAULT
enum CUlaunchMemSyncDomain
|
static int |
CU_LAUNCH_MEM_SYNC_DOMAIN_REMOTE
enum CUlaunchMemSyncDomain
|
static Pointer |
CU_LAUNCH_PARAM_BUFFER_POINTER |
static int |
CU_LAUNCH_PARAM_BUFFER_POINTER_AS_INT
C++ compile time constant for CU_LAUNCH_PARAM_BUFFER_POINTER
|
static Pointer |
CU_LAUNCH_PARAM_BUFFER_SIZE |
static int |
CU_LAUNCH_PARAM_BUFFER_SIZE_AS_INT
C++ compile time constant for CU_LAUNCH_PARAM_BUFFER_SIZE
|
static Pointer |
CU_LAUNCH_PARAM_END |
static int |
CU_LAUNCH_PARAM_END_AS_INT
C++ compile time constant for CU_LAUNCH_PARAM_END
|
static int |
CU_LIBRARY_BINARY_IS_PRESERVED
enum CUlibraryOption
|
static int |
CU_LIBRARY_HOST_UNIVERSAL_FUNCTION_AND_DATA_TABLE
enum CUlibraryOption
|
static int |
CU_LIBRARY_NUM_OPTIONS
enum CUlibraryOption
|
static int |
CU_LIMIT_DEV_RUNTIME_PENDING_LAUNCH_COUNT
enum CUlimit
|
static int |
CU_LIMIT_DEV_RUNTIME_SYNC_DEPTH
enum CUlimit
|
static int |
CU_LIMIT_MALLOC_HEAP_SIZE
enum CUlimit
|
static int |
CU_LIMIT_MAX
enum CUlimit
|
static int |
CU_LIMIT_MAX_L2_FETCH_GRANULARITY
enum CUlimit
|
static int |
CU_LIMIT_PERSISTING_L2_CACHE_SIZE
enum CUlimit
|
static int |
CU_LIMIT_PRINTF_FIFO_SIZE
enum CUlimit
|
static int |
CU_LIMIT_STACK_SIZE
enum CUlimit
|
static int |
CU_MEM_ACCESS_FLAGS_PROT_MAX
enum CUmemAccess_flags
|
static int |
CU_MEM_ACCESS_FLAGS_PROT_NONE
enum CUmemAccess_flags
|
static int |
CU_MEM_ACCESS_FLAGS_PROT_READ
enum CUmemAccess_flags
|
static int |
CU_MEM_ACCESS_FLAGS_PROT_READWRITE
enum CUmemAccess_flags
|
static int |
CU_MEM_ADVISE_SET_ACCESSED_BY
enum CUmem_advise
|
static int |
CU_MEM_ADVISE_SET_PREFERRED_LOCATION
enum CUmem_advise
|
static int |
CU_MEM_ADVISE_SET_READ_MOSTLY
enum CUmem_advise
|
static int |
CU_MEM_ADVISE_UNSET_ACCESSED_BY
enum CUmem_advise
|
static int |
CU_MEM_ADVISE_UNSET_PREFERRED_LOCATION
enum CUmem_advise
|
static int |
CU_MEM_ADVISE_UNSET_READ_MOSTLY
enum CUmem_advise
|
static int |
CU_MEM_ALLOC_GRANULARITY_MINIMUM
enum CUmemAllocationGranularity_flags
|
static int |
CU_MEM_ALLOC_GRANULARITY_RECOMMENDED
enum CUmemAllocationGranularity_flags
|
static int |
CU_MEM_ALLOCATION_COMP_GENERIC
enum CUmemAllocationCompType
|
static int |
CU_MEM_ALLOCATION_COMP_NONE
enum CUmemAllocationCompType
|
static int |
CU_MEM_ALLOCATION_TYPE_INVALID
enum CUmemAllocationType
|
static int |
CU_MEM_ALLOCATION_TYPE_MAX
enum CUmemAllocationType
|
static int |
CU_MEM_ALLOCATION_TYPE_PINNED
enum CUmemAllocationType
|
static int |
CU_MEM_ATTACH_GLOBAL
enum CUmemAttach_flags
|
static int |
CU_MEM_ATTACH_HOST
enum CUmemAttach_flags
|
static int |
CU_MEM_ATTACH_SINGLE
enum CUmemAttach_flags
|
static int |
CU_MEM_CREATE_USAGE_TILE_POOL
This flag if set indicates that the memory will be used as a tile pool.
|
static int |
CU_MEM_HANDLE_TYPE_FABRIC
enum CUmemAllocationHandleType
|
static int |
CU_MEM_HANDLE_TYPE_GENERIC
enum CUmemHandleType
|
static int |
CU_MEM_HANDLE_TYPE_MAX
enum CUmemAllocationHandleType
|
static int |
CU_MEM_HANDLE_TYPE_NONE
enum CUmemAllocationHandleType
|
static int |
CU_MEM_HANDLE_TYPE_POSIX_FILE_DESCRIPTOR
enum CUmemAllocationHandleType
|
static int |
CU_MEM_HANDLE_TYPE_WIN32
enum CUmemAllocationHandleType
|
static int |
CU_MEM_HANDLE_TYPE_WIN32_KMT
enum CUmemAllocationHandleType
|
static int |
CU_MEM_LOCATION_TYPE_DEVICE
enum CUmemLocationType
|
static int |
CU_MEM_LOCATION_TYPE_HOST
enum CUmemLocationType
|
static int |
CU_MEM_LOCATION_TYPE_HOST_NUMA
enum CUmemLocationType
|
static int |
CU_MEM_LOCATION_TYPE_HOST_NUMA_CURRENT
enum CUmemLocationType
|
static int |
CU_MEM_LOCATION_TYPE_INVALID
enum CUmemLocationType
|
static int |
CU_MEM_LOCATION_TYPE_MAX
enum CUmemLocationType
|
static int |
CU_MEM_OPERATION_TYPE_MAP
enum CUmemOperationType
|
static int |
CU_MEM_OPERATION_TYPE_UNMAP
enum CUmemOperationType
|
static int |
CU_MEM_RANGE_ATTRIBUTE_ACCESSED_BY
enum CUmem_range_attribute
|
static int |
CU_MEM_RANGE_ATTRIBUTE_LAST_PREFETCH_LOCATION
enum CUmem_range_attribute
|
static int |
CU_MEM_RANGE_ATTRIBUTE_LAST_PREFETCH_LOCATION_ID
enum CUmem_range_attribute
|
static int |
CU_MEM_RANGE_ATTRIBUTE_LAST_PREFETCH_LOCATION_TYPE
enum CUmem_range_attribute
|
static int |
CU_MEM_RANGE_ATTRIBUTE_PREFERRED_LOCATION
enum CUmem_range_attribute
|
static int |
CU_MEM_RANGE_ATTRIBUTE_PREFERRED_LOCATION_ID
enum CUmem_range_attribute
|
static int |
CU_MEM_RANGE_ATTRIBUTE_PREFERRED_LOCATION_TYPE
enum CUmem_range_attribute
|
static int |
CU_MEM_RANGE_ATTRIBUTE_READ_MOSTLY
enum CUmem_range_attribute
|
static int |
CU_MEM_RANGE_HANDLE_TYPE_DMA_BUF_FD
enum CUmemRangeHandleType
|
static int |
CU_MEM_RANGE_HANDLE_TYPE_MAX
enum CUmemRangeHandleType
|
static int |
CU_MEMHOSTALLOC_DEVICEMAP
If set, host memory is mapped into CUDA address space and
::cuMemHostGetDevicePointer() may be called on the host pointer.
|
static int |
CU_MEMHOSTALLOC_PORTABLE
If set, host memory is portable between CUDA contexts.
|
static int |
CU_MEMHOSTALLOC_WRITECOMBINED
If set, host memory is allocated as write-combined - fast to write,
faster to DMA, slow to read except via SSE4 streaming load instruction
(MOVNTDQA).
|
static int |
CU_MEMHOSTREGISTER_DEVICEMAP
If set, host memory is mapped into CUDA address space and
::cuMemHostGetDevicePointer() may be called on the host pointer.
|
static int |
CU_MEMHOSTREGISTER_IOMEMORY
If set, the passed memory pointer is treated as pointing to some
memory-mapped I/O space, e.g.
|
static int |
CU_MEMHOSTREGISTER_PORTABLE
If set, host memory is portable between CUDA contexts.
|
static int |
CU_MEMHOSTREGISTER_READ_ONLY
If set, the passed memory pointer is treated as pointing to memory that is
considered read-only by the device.
|
static int |
CU_MEMORYTYPE_ARRAY
enum CUmemorytype
|
static int |
CU_MEMORYTYPE_DEVICE
enum CUmemorytype
|
static int |
CU_MEMORYTYPE_HOST
enum CUmemorytype
|
static int |
CU_MEMORYTYPE_UNIFIED
enum CUmemorytype
|
static int |
CU_MEMPOOL_ATTR_RELEASE_THRESHOLD
enum CUmemPool_attribute
|
static int |
CU_MEMPOOL_ATTR_RESERVED_MEM_CURRENT
enum CUmemPool_attribute
|
static int |
CU_MEMPOOL_ATTR_RESERVED_MEM_HIGH
enum CUmemPool_attribute
|
static int |
CU_MEMPOOL_ATTR_REUSE_ALLOW_INTERNAL_DEPENDENCIES
enum CUmemPool_attribute
|
static int |
CU_MEMPOOL_ATTR_REUSE_ALLOW_OPPORTUNISTIC
enum CUmemPool_attribute
|
static int |
CU_MEMPOOL_ATTR_REUSE_FOLLOW_EVENT_DEPENDENCIES
enum CUmemPool_attribute
|
static int |
CU_MEMPOOL_ATTR_USED_MEM_CURRENT
enum CUmemPool_attribute
|
static int |
CU_MEMPOOL_ATTR_USED_MEM_HIGH
enum CUmemPool_attribute
|
static int |
CU_MODULE_EAGER_LOADING
enum CUmoduleLoadingMode
|
static int |
CU_MODULE_LAZY_LOADING
enum CUmoduleLoadingMode
|
static int |
CU_MULTICAST_GRANULARITY_MINIMUM
enum CUmulticastGranularity_flags
|
static int |
CU_MULTICAST_GRANULARITY_RECOMMENDED
enum CUmulticastGranularity_flags
|
static int |
CU_OCCUPANCY_DEFAULT
enum CUoccupancy_flags
|
static int |
CU_OCCUPANCY_DISABLE_CACHING_OVERRIDE
enum CUoccupancy_flags
|
static int |
CU_PARAM_TR_DEFAULT
For texture references loaded into the module, use default texunit from
texture reference.
|
static int |
CU_POINTER_ATTRIBUTE_ACCESS_FLAG_NONE
enum CUDA_POINTER_ATTRIBUTE_ACCESS_FLAGS
|
static int |
CU_POINTER_ATTRIBUTE_ACCESS_FLAG_READ
enum CUDA_POINTER_ATTRIBUTE_ACCESS_FLAGS
|
static int |
CU_POINTER_ATTRIBUTE_ACCESS_FLAG_READWRITE
enum CUDA_POINTER_ATTRIBUTE_ACCESS_FLAGS
|
static int |
CU_POINTER_ATTRIBUTE_ACCESS_FLAGS
enum CUpointer_attribute
|
static int |
CU_POINTER_ATTRIBUTE_ALLOWED_HANDLE_TYPES
enum CUpointer_attribute
|
static int |
CU_POINTER_ATTRIBUTE_BUFFER_ID
enum CUpointer_attribute
|
static int |
CU_POINTER_ATTRIBUTE_CONTEXT
enum CUpointer_attribute
|
static int |
CU_POINTER_ATTRIBUTE_DEVICE_ORDINAL
enum CUpointer_attribute
|
static int |
CU_POINTER_ATTRIBUTE_DEVICE_POINTER
enum CUpointer_attribute
|
static int |
CU_POINTER_ATTRIBUTE_HOST_POINTER
enum CUpointer_attribute
|
static int |
CU_POINTER_ATTRIBUTE_IS_GPU_DIRECT_RDMA_CAPABLE
enum CUpointer_attribute
|
static int |
CU_POINTER_ATTRIBUTE_IS_LEGACY_CUDA_IPC_CAPABLE
enum CUpointer_attribute
|
static int |
CU_POINTER_ATTRIBUTE_IS_MANAGED
enum CUpointer_attribute
|
static int |
CU_POINTER_ATTRIBUTE_MAPPED
enum CUpointer_attribute
|
static int |
CU_POINTER_ATTRIBUTE_MAPPING_BASE_ADDR
enum CUpointer_attribute
|
static int |
CU_POINTER_ATTRIBUTE_MAPPING_SIZE
enum CUpointer_attribute
|
static int |
CU_POINTER_ATTRIBUTE_MEMORY_BLOCK_ID
enum CUpointer_attribute
|
static int |
CU_POINTER_ATTRIBUTE_MEMORY_TYPE
enum CUpointer_attribute
|
static int |
CU_POINTER_ATTRIBUTE_MEMPOOL_HANDLE
enum CUpointer_attribute
|
static int |
CU_POINTER_ATTRIBUTE_P2P_TOKENS
enum CUpointer_attribute
|
static int |
CU_POINTER_ATTRIBUTE_RANGE_SIZE
enum CUpointer_attribute
|
static int |
CU_POINTER_ATTRIBUTE_RANGE_START_ADDR
enum CUpointer_attribute
|
static int |
CU_POINTER_ATTRIBUTE_SYNC_MEMOPS
enum CUpointer_attribute
|
static int |
CU_PREFER_BINARY
enum CUjit_fallback
|
static int |
CU_PREFER_PTX
enum CUjit_fallback
|
static int |
CU_RES_VIEW_FORMAT_FLOAT_1X16
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_FLOAT_1X32
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_FLOAT_2X16
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_FLOAT_2X32
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_FLOAT_4X16
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_FLOAT_4X32
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_NONE
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_SIGNED_BC4
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_SIGNED_BC5
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_SIGNED_BC6H
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_SINT_1X16
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_SINT_1X32
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_SINT_1X8
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_SINT_2X16
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_SINT_2X32
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_SINT_2X8
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_SINT_4X16
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_SINT_4X32
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_SINT_4X8
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_UINT_1X16
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_UINT_1X32
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_UINT_1X8
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_UINT_2X16
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_UINT_2X32
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_UINT_2X8
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_UINT_4X16
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_UINT_4X32
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_UINT_4X8
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_UNSIGNED_BC1
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_UNSIGNED_BC2
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_UNSIGNED_BC3
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_UNSIGNED_BC4
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_UNSIGNED_BC5
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_UNSIGNED_BC6H
enum CUresourceViewFormat
|
static int |
CU_RES_VIEW_FORMAT_UNSIGNED_BC7
enum CUresourceViewFormat
|
static int |
CU_RESOURCE_TYPE_ARRAY
enum CUresourcetype
|
static int |
CU_RESOURCE_TYPE_LINEAR
enum CUresourcetype
|
static int |
CU_RESOURCE_TYPE_MIPMAPPED_ARRAY
enum CUresourcetype
|
static int |
CU_RESOURCE_TYPE_PITCH2D
enum CUresourcetype
|
static int |
CU_SHARED_MEM_CONFIG_DEFAULT_BANK_SIZE
enum CUsharedconfig
|
static int |
CU_SHARED_MEM_CONFIG_EIGHT_BYTE_BANK_SIZE
enum CUsharedconfig
|
static int |
CU_SHARED_MEM_CONFIG_FOUR_BYTE_BANK_SIZE
enum CUsharedconfig
|
static int |
CU_SHAREDMEM_CARVEOUT_DEFAULT
enum CUshared_carveout
|
static int |
CU_SHAREDMEM_CARVEOUT_MAX_L1
enum CUshared_carveout
|
static int |
CU_SHAREDMEM_CARVEOUT_MAX_SHARED
enum CUshared_carveout
|
static int |
CU_STREAM_ADD_CAPTURE_DEPENDENCIES
enum CUstreamUpdateCaptureDependencies_flags
|
static int |
CU_STREAM_ATTRIBUTE_ACCESS_POLICY_WINDOW |
static int |
CU_STREAM_ATTRIBUTE_MEM_SYNC_DOMAIN |
static int |
CU_STREAM_ATTRIBUTE_MEM_SYNC_DOMAIN_MAP |
static int |
CU_STREAM_ATTRIBUTE_PRIORITY |
static int |
CU_STREAM_ATTRIBUTE_SYNCHRONIZATION_POLICY |
static int |
CU_STREAM_CAPTURE_MODE_GLOBAL
enum CUstreamCaptureMode
|
static int |
CU_STREAM_CAPTURE_MODE_RELAXED
enum CUstreamCaptureMode
|
static int |
CU_STREAM_CAPTURE_MODE_THREAD_LOCAL
enum CUstreamCaptureMode
|
static int |
CU_STREAM_CAPTURE_STATUS_ACTIVE
enum CUstreamCaptureStatus
|
static int |
CU_STREAM_CAPTURE_STATUS_INVALIDATED
enum CUstreamCaptureStatus
|
static int |
CU_STREAM_CAPTURE_STATUS_NONE
enum CUstreamCaptureStatus
|
static int |
CU_STREAM_DEFAULT
enum CUstream_flags
|
static CUstream_st |
CU_STREAM_LEGACY |
static int |
CU_STREAM_MEM_OP_BARRIER
enum CUstreamBatchMemOpType
|
static int |
CU_STREAM_MEM_OP_FLUSH_REMOTE_WRITES
enum CUstreamBatchMemOpType
|
static int |
CU_STREAM_MEM_OP_WAIT_VALUE_32
enum CUstreamBatchMemOpType
|
static int |
CU_STREAM_MEM_OP_WAIT_VALUE_64
enum CUstreamBatchMemOpType
|
static int |
CU_STREAM_MEM_OP_WRITE_VALUE_32
enum CUstreamBatchMemOpType
|
static int |
CU_STREAM_MEM_OP_WRITE_VALUE_64
enum CUstreamBatchMemOpType
|
static int |
CU_STREAM_MEMORY_BARRIER_TYPE_GPU
enum CUstreamMemoryBarrier_flags
|
static int |
CU_STREAM_MEMORY_BARRIER_TYPE_SYS
enum CUstreamMemoryBarrier_flags
|
static int |
CU_STREAM_NON_BLOCKING
enum CUstream_flags
|
static CUstream_st |
CU_STREAM_PER_THREAD |
static int |
CU_STREAM_SET_CAPTURE_DEPENDENCIES
enum CUstreamUpdateCaptureDependencies_flags
|
static int |
CU_STREAM_WAIT_VALUE_AND
enum CUstreamWaitValue_flags
|
static int |
CU_STREAM_WAIT_VALUE_EQ
enum CUstreamWaitValue_flags
|
static int |
CU_STREAM_WAIT_VALUE_FLUSH
enum CUstreamWaitValue_flags
|
static int |
CU_STREAM_WAIT_VALUE_GEQ
enum CUstreamWaitValue_flags
|
static int |
CU_STREAM_WAIT_VALUE_NOR
enum CUstreamWaitValue_flags
|
static int |
CU_STREAM_WRITE_VALUE_DEFAULT
enum CUstreamWriteValue_flags
|
static int |
CU_STREAM_WRITE_VALUE_NO_MEMORY_BARRIER
enum CUstreamWriteValue_flags
|
static int |
CU_SYNC_POLICY_AUTO
enum CUsynchronizationPolicy
|
static int |
CU_SYNC_POLICY_BLOCKING_SYNC
enum CUsynchronizationPolicy
|
static int |
CU_SYNC_POLICY_SPIN
enum CUsynchronizationPolicy
|
static int |
CU_SYNC_POLICY_YIELD
enum CUsynchronizationPolicy
|
static int |
CU_TARGET_COMPUTE_30
enum CUjit_target
|
static int |
CU_TARGET_COMPUTE_32
enum CUjit_target
|
static int |
CU_TARGET_COMPUTE_35
enum CUjit_target
|
static int |
CU_TARGET_COMPUTE_37
enum CUjit_target
|
static int |
CU_TARGET_COMPUTE_50
enum CUjit_target
|
static int |
CU_TARGET_COMPUTE_52
enum CUjit_target
|
static int |
CU_TARGET_COMPUTE_53
enum CUjit_target
|
static int |
CU_TARGET_COMPUTE_60
enum CUjit_target
|
static int |
CU_TARGET_COMPUTE_61
enum CUjit_target
|
static int |
CU_TARGET_COMPUTE_62
enum CUjit_target
|
static int |
CU_TARGET_COMPUTE_70
enum CUjit_target
|
static int |
CU_TARGET_COMPUTE_72
enum CUjit_target
|
static int |
CU_TARGET_COMPUTE_75
enum CUjit_target
|
static int |
CU_TARGET_COMPUTE_80
enum CUjit_target
|
static int |
CU_TARGET_COMPUTE_86
enum CUjit_target
|
static int |
CU_TARGET_COMPUTE_87
enum CUjit_target
|
static int |
CU_TARGET_COMPUTE_89
enum CUjit_target
|
static int |
CU_TARGET_COMPUTE_90
enum CUjit_target
|
static int |
CU_TARGET_COMPUTE_90A
enum CUjit_target
|
static int |
CU_TENSOR_MAP_DATA_TYPE_BFLOAT16
enum CUtensorMapDataType
|
static int |
CU_TENSOR_MAP_DATA_TYPE_FLOAT16
enum CUtensorMapDataType
|
static int |
CU_TENSOR_MAP_DATA_TYPE_FLOAT32
enum CUtensorMapDataType
|
static int |
CU_TENSOR_MAP_DATA_TYPE_FLOAT32_FTZ
enum CUtensorMapDataType
|
static int |
CU_TENSOR_MAP_DATA_TYPE_FLOAT64
enum CUtensorMapDataType
|
static int |
CU_TENSOR_MAP_DATA_TYPE_INT32
enum CUtensorMapDataType
|
static int |
CU_TENSOR_MAP_DATA_TYPE_INT64
enum CUtensorMapDataType
|
static int |
CU_TENSOR_MAP_DATA_TYPE_TFLOAT32
enum CUtensorMapDataType
|
static int |
CU_TENSOR_MAP_DATA_TYPE_TFLOAT32_FTZ
enum CUtensorMapDataType
|
static int |
CU_TENSOR_MAP_DATA_TYPE_UINT16
enum CUtensorMapDataType
|
static int |
CU_TENSOR_MAP_DATA_TYPE_UINT32
enum CUtensorMapDataType
|
static int |
CU_TENSOR_MAP_DATA_TYPE_UINT64
enum CUtensorMapDataType
|
static int |
CU_TENSOR_MAP_DATA_TYPE_UINT8
enum CUtensorMapDataType
|
static int |
CU_TENSOR_MAP_FLOAT_OOB_FILL_NAN_REQUEST_ZERO_FMA
enum CUtensorMapFloatOOBfill
|
static int |
CU_TENSOR_MAP_FLOAT_OOB_FILL_NONE
enum CUtensorMapFloatOOBfill
|
static int |
CU_TENSOR_MAP_INTERLEAVE_16B
enum CUtensorMapInterleave
|
static int |
CU_TENSOR_MAP_INTERLEAVE_32B
enum CUtensorMapInterleave
|
static int |
CU_TENSOR_MAP_INTERLEAVE_NONE
enum CUtensorMapInterleave
|
static int |
CU_TENSOR_MAP_L2_PROMOTION_L2_128B
enum CUtensorMapL2promotion
|
static int |
CU_TENSOR_MAP_L2_PROMOTION_L2_256B
enum CUtensorMapL2promotion
|
static int |
CU_TENSOR_MAP_L2_PROMOTION_L2_64B
enum CUtensorMapL2promotion
|
static int |
CU_TENSOR_MAP_L2_PROMOTION_NONE
enum CUtensorMapL2promotion
|
static int |
CU_TENSOR_MAP_NUM_QWORDS
Size of tensor map descriptor
|
static int |
CU_TENSOR_MAP_SWIZZLE_128B
enum CUtensorMapSwizzle
|
static int |
CU_TENSOR_MAP_SWIZZLE_32B
enum CUtensorMapSwizzle
|
static int |
CU_TENSOR_MAP_SWIZZLE_64B
enum CUtensorMapSwizzle
|
static int |
CU_TENSOR_MAP_SWIZZLE_NONE
enum CUtensorMapSwizzle
|
static int |
CU_TR_ADDRESS_MODE_BORDER
enum CUaddress_mode
|
static int |
CU_TR_ADDRESS_MODE_CLAMP
enum CUaddress_mode
|
static int |
CU_TR_ADDRESS_MODE_MIRROR
enum CUaddress_mode
|
static int |
CU_TR_ADDRESS_MODE_WRAP
enum CUaddress_mode
|
static int |
CU_TR_FILTER_MODE_LINEAR
enum CUfilter_mode
|
static int |
CU_TR_FILTER_MODE_POINT
enum CUfilter_mode
|
static int |
CU_TRSA_OVERRIDE_FORMAT
Override the texref format with a format inferred from the array.
|
static int |
CU_TRSF_DISABLE_TRILINEAR_OPTIMIZATION
Disable any trilinear filtering optimizations.
|
static int |
CU_TRSF_NORMALIZED_COORDINATES
Use normalized texture coordinates in the range [0,1) instead of [0,dim).
|
static int |
CU_TRSF_READ_AS_INTEGER
Read the texture as integers rather than promoting the values to floats
in the range [0,1].
|
static int |
CU_TRSF_SEAMLESS_CUBEMAP
Enable seamless cube map filtering.
|
static int |
CU_TRSF_SRGB
Perform sRGB->linear conversion during texture read.
|
static int |
CU_USER_OBJECT_NO_DESTRUCTOR_SYNC
enum CUuserObject_flags
|
static int |
CUDA_ARRAY3D_2DARRAY
Deprecated, use CUDA_ARRAY3D_LAYERED
|
static int |
CUDA_ARRAY3D_COLOR_ATTACHMENT
This flag indicates that the CUDA array may be bound as a color target
in an external graphics API
|
static int |
CUDA_ARRAY3D_CUBEMAP
If set, the CUDA array is a collection of six 2D arrays, representing faces of a cube.
|
static int |
CUDA_ARRAY3D_DEFERRED_MAPPING
This flag if set indicates that the CUDA array or CUDA mipmapped array
will allow deferred memory mapping
|
static int |
CUDA_ARRAY3D_DEPTH_TEXTURE
This flag if set indicates that the CUDA
array is a DEPTH_TEXTURE.
|
static int |
CUDA_ARRAY3D_LAYERED
If set, the CUDA array is a collection of layers, where each layer is either a 1D
or a 2D array and the Depth member of CUDA_ARRAY3D_DESCRIPTOR specifies the number
of layers, not the depth of a 3D array.
|
static int |
CUDA_ARRAY3D_SPARSE
This flag if set indicates that the CUDA array or CUDA mipmapped array
is a sparse CUDA array or CUDA mipmapped array respectively
|
static int |
CUDA_ARRAY3D_SURFACE_LDST
This flag must be set in order to bind a surface reference
to the CUDA array
|
static int |
CUDA_ARRAY3D_TEXTURE_GATHER
This flag must be set in order to perform texture gather operations
on a CUDA array.
|
static int |
CUDA_C_16BF
enum cudaDataType
|
static int |
CUDA_C_16F
enum cudaDataType
|
static int |
CUDA_C_16I
enum cudaDataType
|
static int |
CUDA_C_16U
enum cudaDataType
|
static int |
CUDA_C_32F
enum cudaDataType
|
static int |
CUDA_C_32I
enum cudaDataType
|
static int |
CUDA_C_32U
enum cudaDataType
|
static int |
CUDA_C_4I
enum cudaDataType
|
static int |
CUDA_C_4U
enum cudaDataType
|
static int |
CUDA_C_64F
enum cudaDataType
|
static int |
CUDA_C_64I
enum cudaDataType
|
static int |
CUDA_C_64U
enum cudaDataType
|
static int |
CUDA_C_8I
enum cudaDataType
|
static int |
CUDA_C_8U
enum cudaDataType
|
static int |
CUDA_COOPERATIVE_LAUNCH_MULTI_DEVICE_NO_POST_LAUNCH_SYNC
If set, any subsequent work pushed in a stream that participated in a call to
::cuLaunchCooperativeKernelMultiDevice will only wait for the kernel launched on
the GPU corresponding to that stream to complete before it begins execution.
|
static int |
CUDA_COOPERATIVE_LAUNCH_MULTI_DEVICE_NO_PRE_LAUNCH_SYNC
If set, each kernel launched as part of ::cuLaunchCooperativeKernelMultiDevice only
waits for prior work in the stream corresponding to that GPU to complete before the
kernel begins execution.
|
static int |
CUDA_ERROR_ALREADY_ACQUIRED
enum CUresult
|
static int |
CUDA_ERROR_ALREADY_MAPPED
enum CUresult
|
static int |
CUDA_ERROR_ARRAY_IS_MAPPED
enum CUresult
|
static int |
CUDA_ERROR_ASSERT
enum CUresult
|
static int |
CUDA_ERROR_CAPTURED_EVENT
enum CUresult
|
static int |
CUDA_ERROR_CDP_NOT_SUPPORTED
enum CUresult
|
static int |
CUDA_ERROR_CDP_VERSION_MISMATCH
enum CUresult
|
static int |
CUDA_ERROR_COMPAT_NOT_SUPPORTED_ON_DEVICE
enum CUresult
|
static int |
CUDA_ERROR_CONTEXT_ALREADY_CURRENT
enum CUresult
|
static int |
CUDA_ERROR_CONTEXT_ALREADY_IN_USE
enum CUresult
|
static int |
CUDA_ERROR_CONTEXT_IS_DESTROYED
enum CUresult
|
static int |
CUDA_ERROR_COOPERATIVE_LAUNCH_TOO_LARGE
enum CUresult
|
static int |
CUDA_ERROR_DEINITIALIZED
enum CUresult
|
static int |
CUDA_ERROR_DEVICE_NOT_LICENSED
enum CUresult
|
static int |
CUDA_ERROR_DEVICE_UNAVAILABLE
enum CUresult
|
static int |
CUDA_ERROR_ECC_UNCORRECTABLE
enum CUresult
|
static int |
CUDA_ERROR_EXTERNAL_DEVICE
enum CUresult
|
static int |
CUDA_ERROR_FILE_NOT_FOUND
enum CUresult
|
static int |
CUDA_ERROR_GRAPH_EXEC_UPDATE_FAILURE
enum CUresult
|
static int |
CUDA_ERROR_HARDWARE_STACK_ERROR
enum CUresult
|
static int |
CUDA_ERROR_HOST_MEMORY_ALREADY_REGISTERED
enum CUresult
|
static int |
CUDA_ERROR_HOST_MEMORY_NOT_REGISTERED
enum CUresult
|
static int |
CUDA_ERROR_ILLEGAL_ADDRESS
enum CUresult
|
static int |
CUDA_ERROR_ILLEGAL_INSTRUCTION
enum CUresult
|
static int |
CUDA_ERROR_ILLEGAL_STATE
enum CUresult
|
static int |
CUDA_ERROR_INVALID_ADDRESS_SPACE
enum CUresult
|
static int |
CUDA_ERROR_INVALID_CLUSTER_SIZE
enum CUresult
|
static int |
CUDA_ERROR_INVALID_CONTEXT
enum CUresult
|
static int |
CUDA_ERROR_INVALID_DEVICE
enum CUresult
|
static int |
CUDA_ERROR_INVALID_GRAPHICS_CONTEXT
enum CUresult
|
static int |
CUDA_ERROR_INVALID_HANDLE
enum CUresult
|
static int |
CUDA_ERROR_INVALID_IMAGE
enum CUresult
|
static int |
CUDA_ERROR_INVALID_PC
enum CUresult
|
static int |
CUDA_ERROR_INVALID_PTX
enum CUresult
|
static int |
CUDA_ERROR_INVALID_SOURCE
enum CUresult
|
static int |
CUDA_ERROR_INVALID_VALUE
enum CUresult
|
static int |
CUDA_ERROR_JIT_COMPILATION_DISABLED
enum CUresult
|
static int |
CUDA_ERROR_JIT_COMPILER_NOT_FOUND
enum CUresult
|
static int |
CUDA_ERROR_LAUNCH_FAILED
enum CUresult
|
static int |
CUDA_ERROR_LAUNCH_INCOMPATIBLE_TEXTURING
enum CUresult
|
static int |
CUDA_ERROR_LAUNCH_OUT_OF_RESOURCES
enum CUresult
|
static int |
CUDA_ERROR_LAUNCH_TIMEOUT
enum CUresult
|
static int |
CUDA_ERROR_LOSSY_QUERY
enum CUresult
|
static int |
CUDA_ERROR_MAP_FAILED
enum CUresult
|
static int |
CUDA_ERROR_MISALIGNED_ADDRESS
enum CUresult
|
static int |
CUDA_ERROR_MPS_CLIENT_TERMINATED
enum CUresult
|
static int |
CUDA_ERROR_MPS_CONNECTION_FAILED
enum CUresult
|
static int |
CUDA_ERROR_MPS_MAX_CLIENTS_REACHED
enum CUresult
|
static int |
CUDA_ERROR_MPS_MAX_CONNECTIONS_REACHED
enum CUresult
|
static int |
CUDA_ERROR_MPS_RPC_FAILURE
enum CUresult
|
static int |
CUDA_ERROR_MPS_SERVER_NOT_READY
enum CUresult
|
static int |
CUDA_ERROR_NO_BINARY_FOR_GPU
enum CUresult
|
static int |
CUDA_ERROR_NO_DEVICE
enum CUresult
|
static int |
CUDA_ERROR_NOT_FOUND
enum CUresult
|
static int |
CUDA_ERROR_NOT_INITIALIZED
enum CUresult
|
static int |
CUDA_ERROR_NOT_MAPPED
enum CUresult
|
static int |
CUDA_ERROR_NOT_MAPPED_AS_ARRAY
enum CUresult
|
static int |
CUDA_ERROR_NOT_MAPPED_AS_POINTER
enum CUresult
|
static int |
CUDA_ERROR_NOT_PERMITTED
enum CUresult
|
static int |
CUDA_ERROR_NOT_READY
enum CUresult
|
static int |
CUDA_ERROR_NOT_SUPPORTED
enum CUresult
|
static int |
CUDA_ERROR_NVLINK_UNCORRECTABLE
enum CUresult
|
static int |
CUDA_ERROR_OPERATING_SYSTEM
enum CUresult
|
static int |
CUDA_ERROR_OUT_OF_MEMORY
enum CUresult
|
static int |
CUDA_ERROR_PEER_ACCESS_ALREADY_ENABLED
enum CUresult
|
static int |
CUDA_ERROR_PEER_ACCESS_NOT_ENABLED
enum CUresult
|
static int |
CUDA_ERROR_PEER_ACCESS_UNSUPPORTED
enum CUresult
|
static int |
CUDA_ERROR_PRIMARY_CONTEXT_ACTIVE
enum CUresult
|
static int |
CUDA_ERROR_PROFILER_ALREADY_STARTED
enum CUresult
|
static int |
CUDA_ERROR_PROFILER_ALREADY_STOPPED
enum CUresult
|
static int |
CUDA_ERROR_PROFILER_DISABLED
enum CUresult
|
static int |
CUDA_ERROR_PROFILER_NOT_INITIALIZED
enum CUresult
|
static int |
CUDA_ERROR_SHARED_OBJECT_INIT_FAILED
enum CUresult
|
static int |
CUDA_ERROR_SHARED_OBJECT_SYMBOL_NOT_FOUND
enum CUresult
|
static int |
CUDA_ERROR_STREAM_CAPTURE_IMPLICIT
enum CUresult
|
static int |
CUDA_ERROR_STREAM_CAPTURE_INVALIDATED
enum CUresult
|
static int |
CUDA_ERROR_STREAM_CAPTURE_ISOLATION
enum CUresult
|
static int |
CUDA_ERROR_STREAM_CAPTURE_MERGE
enum CUresult
|
static int |
CUDA_ERROR_STREAM_CAPTURE_UNJOINED
enum CUresult
|
static int |
CUDA_ERROR_STREAM_CAPTURE_UNMATCHED
enum CUresult
|
static int |
CUDA_ERROR_STREAM_CAPTURE_UNSUPPORTED
enum CUresult
|
static int |
CUDA_ERROR_STREAM_CAPTURE_WRONG_THREAD
enum CUresult
|
static int |
CUDA_ERROR_STUB_LIBRARY
enum CUresult
|
static int |
CUDA_ERROR_SYSTEM_DRIVER_MISMATCH
enum CUresult
|
static int |
CUDA_ERROR_SYSTEM_NOT_READY
enum CUresult
|
static int |
CUDA_ERROR_TIMEOUT
enum CUresult
|
static int |
CUDA_ERROR_TOO_MANY_PEERS
enum CUresult
|
static int |
CUDA_ERROR_UNKNOWN
enum CUresult
|
static int |
CUDA_ERROR_UNMAP_FAILED
enum CUresult
|
static int |
CUDA_ERROR_UNSUPPORTED_DEVSIDE_SYNC
enum CUresult
|
static int |
CUDA_ERROR_UNSUPPORTED_EXEC_AFFINITY
enum CUresult
|
static int |
CUDA_ERROR_UNSUPPORTED_LIMIT
enum CUresult
|
static int |
CUDA_ERROR_UNSUPPORTED_PTX_VERSION
enum CUresult
|
static int |
CUDA_EXTERNAL_MEMORY_DEDICATED
Indicates that the external memory object is a dedicated resource
|
static int |
CUDA_EXTERNAL_SEMAPHORE_SIGNAL_SKIP_NVSCIBUF_MEMSYNC
When the \p flags parameter of ::CUDA_EXTERNAL_SEMAPHORE_SIGNAL_PARAMS
contains this flag, it indicates that signaling an external semaphore object
should skip performing appropriate memory synchronization operations over all
the external memory objects that are imported as ::CU_EXTERNAL_MEMORY_HANDLE_TYPE_NVSCIBUF,
which otherwise are performed by default to ensure data coherency with other
importers of the same NvSciBuf memory objects.
|
static int |
CUDA_EXTERNAL_SEMAPHORE_WAIT_SKIP_NVSCIBUF_MEMSYNC
When the \p flags parameter of ::CUDA_EXTERNAL_SEMAPHORE_WAIT_PARAMS
contains this flag, it indicates that waiting on an external semaphore object
should skip performing appropriate memory synchronization operations over all
the external memory objects that are imported as ::CU_EXTERNAL_MEMORY_HANDLE_TYPE_NVSCIBUF,
which otherwise are performed by default to ensure data coherency with other
importers of the same NvSciBuf memory objects.
|
static int |
CUDA_GRAPH_INSTANTIATE_ERROR
enum CUgraphInstantiateResult
|
static int |
CUDA_GRAPH_INSTANTIATE_FLAG_AUTO_FREE_ON_LAUNCH
enum CUgraphInstantiate_flags
|
static int |
CUDA_GRAPH_INSTANTIATE_FLAG_DEVICE_LAUNCH
enum CUgraphInstantiate_flags
|
static int |
CUDA_GRAPH_INSTANTIATE_FLAG_UPLOAD
enum CUgraphInstantiate_flags
|
static int |
CUDA_GRAPH_INSTANTIATE_FLAG_USE_NODE_PRIORITY
enum CUgraphInstantiate_flags
|
static int |
CUDA_GRAPH_INSTANTIATE_INVALID_STRUCTURE
enum CUgraphInstantiateResult
|
static int |
CUDA_GRAPH_INSTANTIATE_MULTIPLE_CTXS_NOT_SUPPORTED
enum CUgraphInstantiateResult
|
static int |
CUDA_GRAPH_INSTANTIATE_NODE_OPERATION_NOT_SUPPORTED
enum CUgraphInstantiateResult
|
static int |
CUDA_GRAPH_INSTANTIATE_SUCCESS
enum CUgraphInstantiateResult
|
static int |
CUDA_IPC_HANDLE_SIZE
CUDA IPC Handle Size
|
static int |
CUDA_NVSCISYNC_ATTR_SIGNAL
When \p flags of ::cuDeviceGetNvSciSyncAttributes is set to this,
it indicates that application needs signaler specific NvSciSyncAttr
to be filled by ::cuDeviceGetNvSciSyncAttributes.
|
static int |
CUDA_NVSCISYNC_ATTR_WAIT
When \p flags of ::cuDeviceGetNvSciSyncAttributes is set to this,
it indicates that application needs waiter specific NvSciSyncAttr
to be filled by ::cuDeviceGetNvSciSyncAttributes.
|
static int |
CUDA_R_16BF
enum cudaDataType
|
static int |
CUDA_R_16F
enum cudaDataType
|
static int |
CUDA_R_16I
enum cudaDataType
|
static int |
CUDA_R_16U
enum cudaDataType
|
static int |
CUDA_R_32F
enum cudaDataType
|
static int |
CUDA_R_32I
enum cudaDataType
|
static int |
CUDA_R_32U
enum cudaDataType
|
static int |
CUDA_R_4I
enum cudaDataType
|
static int |
CUDA_R_4U
enum cudaDataType
|
static int |
CUDA_R_64F
enum cudaDataType
|
static int |
CUDA_R_64I
enum cudaDataType
|
static int |
CUDA_R_64U
enum cudaDataType
|
static int |
CUDA_R_8F_E4M3
enum cudaDataType
|
static int |
CUDA_R_8F_E5M2
enum cudaDataType
|
static int |
CUDA_R_8I
enum cudaDataType
|
static int |
CUDA_R_8U
enum cudaDataType
|
static int |
CUDA_SUCCESS
enum CUresult
|
static int |
CUDA_VERSION
CUDA API version number
|
static int |
cudaAccessPropertyNormal
enum cudaAccessProperty
|
static int |
cudaAccessPropertyPersisting
enum cudaAccessProperty
|
static int |
cudaAccessPropertyStreaming
enum cudaAccessProperty
|
static int |
cudaAddressModeBorder
enum cudaTextureAddressMode
|
static int |
cudaAddressModeClamp
enum cudaTextureAddressMode
|
static int |
cudaAddressModeMirror
enum cudaTextureAddressMode
|
static int |
cudaAddressModeWrap
enum cudaTextureAddressMode
|
static int |
cudaArrayColorAttachment
Must be set in cudaExternalMemoryGetMappedMipmappedArray if the mipmapped array is used as a color target in a graphics API
|
static int |
cudaArrayCubemap
Must be set in cudaMalloc3DArray to create a cubemap CUDA array
|
static int |
cudaArrayDefault
Default CUDA array allocation flag
|
static int |
cudaArrayDeferredMapping
Must be set in cudaMallocArray, cudaMalloc3DArray or cudaMallocMipmappedArray in order to create a deferred mapping CUDA array or CUDA mipmapped array
|
static int |
cudaArrayLayered
Must be set in cudaMalloc3DArray to create a layered CUDA array
|
static int |
cudaArraySparse
Must be set in cudaMallocArray, cudaMalloc3DArray or cudaMallocMipmappedArray in order to create a sparse CUDA array or CUDA mipmapped array
|
static int |
cudaArraySparsePropertiesSingleMipTail
Indicates that the layered sparse CUDA array or CUDA mipmapped array has a single mip tail region for all layers
|
static int |
cudaArraySurfaceLoadStore
Must be set in cudaMallocArray or cudaMalloc3DArray in order to bind surfaces to the CUDA array
|
static int |
cudaArrayTextureGather
Must be set in cudaMallocArray or cudaMalloc3DArray in order to perform texture gather operations on the CUDA array
|
static int |
cudaBoundaryModeClamp
enum cudaSurfaceBoundaryMode
|
static int |
cudaBoundaryModeTrap
enum cudaSurfaceBoundaryMode
|
static int |
cudaBoundaryModeZero
enum cudaSurfaceBoundaryMode
|
static int |
cudaCGScopeGrid
enum cudaCGScope
|
static int |
cudaCGScopeInvalid
enum cudaCGScope
|
static int |
cudaCGScopeMultiGrid
enum cudaCGScope
|
static int |
cudaChannelFormatKindFloat
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindNone
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindNV12
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindSigned
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindSignedBlockCompressed4
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindSignedBlockCompressed5
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindSignedBlockCompressed6H
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindSignedNormalized16X1
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindSignedNormalized16X2
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindSignedNormalized16X4
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindSignedNormalized8X1
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindSignedNormalized8X2
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindSignedNormalized8X4
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindUnsigned
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindUnsignedBlockCompressed1
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindUnsignedBlockCompressed1SRGB
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindUnsignedBlockCompressed2
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindUnsignedBlockCompressed2SRGB
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindUnsignedBlockCompressed3
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindUnsignedBlockCompressed3SRGB
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindUnsignedBlockCompressed4
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindUnsignedBlockCompressed5
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindUnsignedBlockCompressed6H
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindUnsignedBlockCompressed7
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindUnsignedBlockCompressed7SRGB
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindUnsignedNormalized16X1
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindUnsignedNormalized16X2
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindUnsignedNormalized16X4
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindUnsignedNormalized8X1
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindUnsignedNormalized8X2
enum cudaChannelFormatKind
|
static int |
cudaChannelFormatKindUnsignedNormalized8X4
enum cudaChannelFormatKind
|
static int |
cudaClusterSchedulingPolicyDefault
enum cudaClusterSchedulingPolicy
|
static int |
cudaClusterSchedulingPolicyLoadBalancing
enum cudaClusterSchedulingPolicy
|
static int |
cudaClusterSchedulingPolicySpread
enum cudaClusterSchedulingPolicy
|
static int |
cudaComputeModeDefault
enum cudaComputeMode
|
static int |
cudaComputeModeExclusive
enum cudaComputeMode
|
static int |
cudaComputeModeExclusiveProcess
enum cudaComputeMode
|
static int |
cudaComputeModeProhibited
enum cudaComputeMode
|
static int |
cudaCooperativeLaunchMultiDeviceNoPostSync
If set, any subsequent work pushed in a stream that participated in a call to
::cudaLaunchCooperativeKernelMultiDevice will only wait for the kernel launched on
the GPU corresponding to that stream to complete before it begins execution.
|
static int |
cudaCooperativeLaunchMultiDeviceNoPreSync
If set, each kernel launched as part of ::cudaLaunchCooperativeKernelMultiDevice only
waits for prior work in the stream corresponding to that GPU to complete before the
kernel begins execution.
|
static int |
cudaCpuDeviceId
Device id that represents the CPU
|
static int |
cudaDevAttrAsyncEngineCount
enum cudaDeviceAttr
|
static int |
cudaDevAttrCanFlushRemoteWrites
enum cudaDeviceAttr
|
static int |
cudaDevAttrCanMapHostMemory
enum cudaDeviceAttr
|
static int |
cudaDevAttrCanUseHostPointerForRegisteredMem
enum cudaDeviceAttr
|
static int |
cudaDevAttrClockRate
enum cudaDeviceAttr
|
static int |
cudaDevAttrClusterLaunch
enum cudaDeviceAttr
|
static int |
cudaDevAttrComputeCapabilityMajor
enum cudaDeviceAttr
|
static int |
cudaDevAttrComputeCapabilityMinor
enum cudaDeviceAttr
|
static int |
cudaDevAttrComputeMode
enum cudaDeviceAttr
|
static int |
cudaDevAttrComputePreemptionSupported
enum cudaDeviceAttr
|
static int |
cudaDevAttrConcurrentKernels
enum cudaDeviceAttr
|
static int |
cudaDevAttrConcurrentManagedAccess
enum cudaDeviceAttr
|
static int |
cudaDevAttrCooperativeLaunch
enum cudaDeviceAttr
|
static int |
cudaDevAttrCooperativeMultiDeviceLaunch
enum cudaDeviceAttr
|
static int |
cudaDevAttrDeferredMappingCudaArraySupported
enum cudaDeviceAttr
|
static int |
cudaDevAttrDirectManagedMemAccessFromHost
enum cudaDeviceAttr
|
static int |
cudaDevAttrEccEnabled
enum cudaDeviceAttr
|
static int |
cudaDevAttrGlobalL1CacheSupported
enum cudaDeviceAttr
|
static int |
cudaDevAttrGlobalMemoryBusWidth
enum cudaDeviceAttr
|
static int |
cudaDevAttrGPUDirectRDMAFlushWritesOptions
enum cudaDeviceAttr
|
static int |
cudaDevAttrGPUDirectRDMASupported
enum cudaDeviceAttr
|
static int |
cudaDevAttrGPUDirectRDMAWritesOrdering
enum cudaDeviceAttr
|
static int |
cudaDevAttrGpuOverlap
enum cudaDeviceAttr
|
static int |
cudaDevAttrHostNativeAtomicSupported
enum cudaDeviceAttr
|
static int |
cudaDevAttrHostNumaId
enum cudaDeviceAttr
|
static int |
cudaDevAttrHostRegisterReadOnlySupported
enum cudaDeviceAttr
|
static int |
cudaDevAttrHostRegisterSupported
enum cudaDeviceAttr
|
static int |
cudaDevAttrIntegrated
enum cudaDeviceAttr
|
static int |
cudaDevAttrIpcEventSupport
enum cudaDeviceAttr
|
static int |
cudaDevAttrIsMultiGpuBoard
enum cudaDeviceAttr
|
static int |
cudaDevAttrKernelExecTimeout
enum cudaDeviceAttr
|
static int |
cudaDevAttrL2CacheSize
enum cudaDeviceAttr
|
static int |
cudaDevAttrLocalL1CacheSupported
enum cudaDeviceAttr
|
static int |
cudaDevAttrManagedMemory
enum cudaDeviceAttr
|
static int |
cudaDevAttrMax
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxAccessPolicyWindowSize
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxBlockDimX
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxBlockDimY
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxBlockDimZ
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxBlocksPerMultiprocessor
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxGridDimX
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxGridDimY
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxGridDimZ
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxPersistingL2CacheSize
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxPitch
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxRegistersPerBlock
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxRegistersPerMultiprocessor
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxSharedMemoryPerBlock
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxSharedMemoryPerBlockOptin
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxSharedMemoryPerMultiprocessor
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxSurface1DLayeredLayers
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxSurface1DLayeredWidth
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxSurface1DWidth
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxSurface2DHeight
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxSurface2DLayeredHeight
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxSurface2DLayeredLayers
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxSurface2DLayeredWidth
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxSurface2DWidth
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxSurface3DDepth
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxSurface3DHeight
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxSurface3DWidth
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxSurfaceCubemapLayeredLayers
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxSurfaceCubemapLayeredWidth
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxSurfaceCubemapWidth
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture1DLayeredLayers
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture1DLayeredWidth
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture1DLinearWidth
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture1DMipmappedWidth
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture1DWidth
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture2DGatherHeight
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture2DGatherWidth
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture2DHeight
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture2DLayeredHeight
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture2DLayeredLayers
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture2DLayeredWidth
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture2DLinearHeight
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture2DLinearPitch
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture2DLinearWidth
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture2DMipmappedHeight
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture2DMipmappedWidth
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture2DWidth
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture3DDepth
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture3DDepthAlt
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture3DHeight
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture3DHeightAlt
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture3DWidth
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTexture3DWidthAlt
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTextureCubemapLayeredLayers
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTextureCubemapLayeredWidth
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTextureCubemapWidth
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxThreadsPerBlock
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxThreadsPerMultiProcessor
enum cudaDeviceAttr
|
static int |
cudaDevAttrMaxTimelineSemaphoreInteropSupported
enum cudaDeviceAttr
|
static int |
cudaDevAttrMemoryClockRate
enum cudaDeviceAttr
|
static int |
cudaDevAttrMemoryPoolsSupported
enum cudaDeviceAttr
|
static int |
cudaDevAttrMemoryPoolSupportedHandleTypes
enum cudaDeviceAttr
|
static int |
cudaDevAttrMemSyncDomainCount
enum cudaDeviceAttr
|
static int |
cudaDevAttrMpsEnabled
enum cudaDeviceAttr
|
static int |
cudaDevAttrMultiGpuBoardGroupID
enum cudaDeviceAttr
|
static int |
cudaDevAttrMultiProcessorCount
enum cudaDeviceAttr
|
static int |
cudaDevAttrNumaConfig
enum cudaDeviceAttr
|
static int |
cudaDevAttrNumaId
enum cudaDeviceAttr
|
static int |
cudaDevAttrPageableMemoryAccess
enum cudaDeviceAttr
|
static int |
cudaDevAttrPageableMemoryAccessUsesHostPageTables
enum cudaDeviceAttr
|
static int |
cudaDevAttrPciBusId
enum cudaDeviceAttr
|
static int |
cudaDevAttrPciDeviceId
enum cudaDeviceAttr
|
static int |
cudaDevAttrPciDomainId
enum cudaDeviceAttr
|
static int |
cudaDevAttrReserved122
enum cudaDeviceAttr
|
static int |
cudaDevAttrReserved123
enum cudaDeviceAttr
|
static int |
cudaDevAttrReserved124
enum cudaDeviceAttr
|
static int |
cudaDevAttrReserved127
enum cudaDeviceAttr
|
static int |
cudaDevAttrReserved128
enum cudaDeviceAttr
|
static int |
cudaDevAttrReserved129
enum cudaDeviceAttr
|
static int |
cudaDevAttrReserved132
enum cudaDeviceAttr
|
static int |
cudaDevAttrReserved92
enum cudaDeviceAttr
|
static int |
cudaDevAttrReserved93
enum cudaDeviceAttr
|
static int |
cudaDevAttrReserved94
enum cudaDeviceAttr
|
static int |
cudaDevAttrReservedSharedMemoryPerBlock
enum cudaDeviceAttr
|
static int |
cudaDevAttrSingleToDoublePrecisionPerfRatio
enum cudaDeviceAttr
|
static int |
cudaDevAttrSparseCudaArraySupported
enum cudaDeviceAttr
|
static int |
cudaDevAttrStreamPrioritiesSupported
enum cudaDeviceAttr
|
static int |
cudaDevAttrSurfaceAlignment
enum cudaDeviceAttr
|
static int |
cudaDevAttrTccDriver
enum cudaDeviceAttr
|
static int |
cudaDevAttrTextureAlignment
enum cudaDeviceAttr
|
static int |
cudaDevAttrTexturePitchAlignment
enum cudaDeviceAttr
|
static int |
cudaDevAttrTimelineSemaphoreInteropSupported
enum cudaDeviceAttr
|
static int |
cudaDevAttrTotalConstantMemory
enum cudaDeviceAttr
|
static int |
cudaDevAttrUnifiedAddressing
enum cudaDeviceAttr
|
static int |
cudaDevAttrWarpSize
enum cudaDeviceAttr
|
static int |
cudaDeviceBlockingSync
Deprecated.
This flag was deprecated as of CUDA 4.0 and
replaced with ::cudaDeviceScheduleBlockingSync.
|
static int |
cudaDeviceLmemResizeToMax
Device flag - Keep local memory allocation after launch
|
static int |
cudaDeviceMapHost
Device flag - Support mapped pinned allocations
|
static int |
cudaDeviceMask
Device flags mask
|
static int |
cudaDeviceNumaConfigNone
enum cudaDeviceNumaConfig
|
static int |
cudaDeviceNumaConfigNumaNode
enum cudaDeviceNumaConfig
|
static int |
cudaDeviceScheduleAuto
Device flag - Automatic scheduling
|
static int |
cudaDeviceScheduleBlockingSync
Device flag - Use blocking synchronization
|
static int |
cudaDeviceScheduleMask
Device schedule flags mask
|
static int |
cudaDeviceScheduleSpin
Device flag - Spin default scheduling
|
static int |
cudaDeviceScheduleYield
Device flag - Yield default scheduling
|
static int |
cudaDeviceSyncMemops
Device flag - Ensure synchronous memory operations on this context will synchronize
|
static int |
cudaDevP2PAttrAccessSupported
enum cudaDeviceP2PAttr
|
static int |
cudaDevP2PAttrCudaArrayAccessSupported
enum cudaDeviceP2PAttr
|
static int |
cudaDevP2PAttrNativeAtomicSupported
enum cudaDeviceP2PAttr
|
static int |
cudaDevP2PAttrPerformanceRank
enum cudaDeviceP2PAttr
|
static int |
cudaDriverEntryPointSuccess
enum cudaDriverEntryPointQueryResult
|
static int |
cudaDriverEntryPointSymbolNotFound
enum cudaDriverEntryPointQueryResult
|
static int |
cudaDriverEntryPointVersionNotSufficent
enum cudaDriverEntryPointQueryResult
|
static int |
cudaEnableDefault
enum cudaGetDriverEntryPointFlags
|
static int |
cudaEnableLegacyStream
enum cudaGetDriverEntryPointFlags
|
static int |
cudaEnablePerThreadDefaultStream
enum cudaGetDriverEntryPointFlags
|
static int |
cudaErrorAddressOfConstant
enum cudaError
|
static int |
cudaErrorAlreadyAcquired
enum cudaError
|
static int |
cudaErrorAlreadyMapped
enum cudaError
|
static int |
cudaErrorApiFailureBase
enum cudaError
|
static int |
cudaErrorArrayIsMapped
enum cudaError
|
static int |
cudaErrorAssert
enum cudaError
|
static int |
cudaErrorCallRequiresNewerDriver
enum cudaError
|
static int |
cudaErrorCapturedEvent
enum cudaError
|
static int |
cudaErrorCdpNotSupported
enum cudaError
|
static int |
cudaErrorCdpVersionMismatch
enum cudaError
|
static int |
cudaErrorCompatNotSupportedOnDevice
enum cudaError
|
static int |
cudaErrorContextIsDestroyed
enum cudaError
|
static int |
cudaErrorCooperativeLaunchTooLarge
enum cudaError
|
static int |
cudaErrorCudartUnloading
enum cudaError
|
static int |
cudaErrorDeviceAlreadyInUse
enum cudaError
|
static int |
cudaErrorDeviceNotLicensed
enum cudaError
|
static int |
cudaErrorDevicesUnavailable
enum cudaError
|
static int |
cudaErrorDeviceUninitialized
enum cudaError
|
static int |
cudaErrorDuplicateSurfaceName
enum cudaError
|
static int |
cudaErrorDuplicateTextureName
enum cudaError
|
static int |
cudaErrorDuplicateVariableName
enum cudaError
|
static int |
cudaErrorECCUncorrectable
enum cudaError
|
static int |
cudaErrorExternalDevice
enum cudaError
|
static int |
cudaErrorFileNotFound
enum cudaError
|
static int |
cudaErrorGraphExecUpdateFailure
enum cudaError
|
static int |
cudaErrorHardwareStackError
enum cudaError
|
static int |
cudaErrorHostMemoryAlreadyRegistered
enum cudaError
|
static int |
cudaErrorHostMemoryNotRegistered
enum cudaError
|
static int |
cudaErrorIllegalAddress
enum cudaError
|
static int |
cudaErrorIllegalInstruction
enum cudaError
|
static int |
cudaErrorIllegalState
enum cudaError
|
static int |
cudaErrorIncompatibleDriverContext
enum cudaError
|
static int |
cudaErrorInitializationError
enum cudaError
|
static int |
cudaErrorInsufficientDriver
enum cudaError
|
static int |
cudaErrorInvalidAddressSpace
enum cudaError
|
static int |
cudaErrorInvalidChannelDescriptor
enum cudaError
|
static int |
cudaErrorInvalidClusterSize
enum cudaError
|
static int |
cudaErrorInvalidConfiguration
enum cudaError
|
static int |
cudaErrorInvalidDevice
enum cudaError
|
static int |
cudaErrorInvalidDeviceFunction
enum cudaError
|
static int |
cudaErrorInvalidDevicePointer
enum cudaError
|
static int |
cudaErrorInvalidFilterSetting
enum cudaError
|
static int |
cudaErrorInvalidGraphicsContext
enum cudaError
|
static int |
cudaErrorInvalidHostPointer
enum cudaError
|
static int |
cudaErrorInvalidKernelImage
enum cudaError
|
static int |
cudaErrorInvalidMemcpyDirection
enum cudaError
|
static int |
cudaErrorInvalidNormSetting
enum cudaError
|
static int |
cudaErrorInvalidPc
enum cudaError
|
static int |
cudaErrorInvalidPitchValue
enum cudaError
|
static int |
cudaErrorInvalidPtx
enum cudaError
|
static int |
cudaErrorInvalidResourceHandle
enum cudaError
|
static int |
cudaErrorInvalidSource
enum cudaError
|
static int |
cudaErrorInvalidSurface
enum cudaError
|
static int |
cudaErrorInvalidSymbol
enum cudaError
|
static int |
cudaErrorInvalidTexture
enum cudaError
|
static int |
cudaErrorInvalidTextureBinding
enum cudaError
|
static int |
cudaErrorInvalidValue
enum cudaError
|
static int |
cudaErrorJitCompilationDisabled
enum cudaError
|
static int |
cudaErrorJitCompilerNotFound
enum cudaError
|
static int |
cudaErrorLaunchFailure
enum cudaError
|
static int |
cudaErrorLaunchFileScopedSurf
enum cudaError
|
static int |
cudaErrorLaunchFileScopedTex
enum cudaError
|
static int |
cudaErrorLaunchIncompatibleTexturing
enum cudaError
|
static int |
cudaErrorLaunchMaxDepthExceeded
enum cudaError
|
static int |
cudaErrorLaunchOutOfResources
enum cudaError
|
static int |
cudaErrorLaunchPendingCountExceeded
enum cudaError
|
static int |
cudaErrorLaunchTimeout
enum cudaError
|
static int |
cudaErrorLossyQuery
enum cudaError
|
static int |
cudaErrorMapBufferObjectFailed
enum cudaError
|
static int |
cudaErrorMemoryAllocation
enum cudaError
|
static int |
cudaErrorMemoryValueTooLarge
enum cudaError
|
static int |
cudaErrorMisalignedAddress
enum cudaError
|
static int |
cudaErrorMissingConfiguration
enum cudaError
|
static int |
cudaErrorMixedDeviceExecution
enum cudaError
|
static int |
cudaErrorMpsClientTerminated
enum cudaError
|
static int |
cudaErrorMpsConnectionFailed
enum cudaError
|
static int |
cudaErrorMpsMaxClientsReached
enum cudaError
|
static int |
cudaErrorMpsMaxConnectionsReached
enum cudaError
|
static int |
cudaErrorMpsRpcFailure
enum cudaError
|
static int |
cudaErrorMpsServerNotReady
enum cudaError
|
static int |
cudaErrorNoDevice
enum cudaError
|
static int |
cudaErrorNoKernelImageForDevice
enum cudaError
|
static int |
cudaErrorNotMapped
enum cudaError
|
static int |
cudaErrorNotMappedAsArray
enum cudaError
|
static int |
cudaErrorNotMappedAsPointer
enum cudaError
|
static int |
cudaErrorNotPermitted
enum cudaError
|
static int |
cudaErrorNotReady
enum cudaError
|
static int |
cudaErrorNotSupported
enum cudaError
|
static int |
cudaErrorNotYetImplemented
enum cudaError
|
static int |
cudaErrorNvlinkUncorrectable
enum cudaError
|
static int |
cudaErrorOperatingSystem
enum cudaError
|
static int |
cudaErrorPeerAccessAlreadyEnabled
enum cudaError
|
static int |
cudaErrorPeerAccessNotEnabled
enum cudaError
|
static int |
cudaErrorPeerAccessUnsupported
enum cudaError
|
static int |
cudaErrorPriorLaunchFailure
enum cudaError
|
static int |
cudaErrorProfilerAlreadyStarted
enum cudaError
|
static int |
cudaErrorProfilerAlreadyStopped
enum cudaError
|
static int |
cudaErrorProfilerDisabled
enum cudaError
|
static int |
cudaErrorProfilerNotInitialized
enum cudaError
|
static int |
cudaErrorSetOnActiveProcess
enum cudaError
|
static int |
cudaErrorSharedObjectInitFailed
enum cudaError
|
static int |
cudaErrorSharedObjectSymbolNotFound
enum cudaError
|
static int |
cudaErrorSoftwareValidityNotEstablished
enum cudaError
|
static int |
cudaErrorStartupFailure
enum cudaError
|
static int |
cudaErrorStreamCaptureImplicit
enum cudaError
|
static int |
cudaErrorStreamCaptureInvalidated
enum cudaError
|
static int |
cudaErrorStreamCaptureIsolation
enum cudaError
|
static int |
cudaErrorStreamCaptureMerge
enum cudaError
|
static int |
cudaErrorStreamCaptureUnjoined
enum cudaError
|
static int |
cudaErrorStreamCaptureUnmatched
enum cudaError
|
static int |
cudaErrorStreamCaptureUnsupported
enum cudaError
|
static int |
cudaErrorStreamCaptureWrongThread
enum cudaError
|
static int |
cudaErrorStubLibrary
enum cudaError
|
static int |
cudaErrorSymbolNotFound
enum cudaError
|
static int |
cudaErrorSyncDepthExceeded
enum cudaError
|
static int |
cudaErrorSynchronizationError
enum cudaError
|
static int |
cudaErrorSystemDriverMismatch
enum cudaError
|
static int |
cudaErrorSystemNotReady
enum cudaError
|
static int |
cudaErrorTextureFetchFailed
enum cudaError
|
static int |
cudaErrorTextureNotBound
enum cudaError
|
static int |
cudaErrorTimeout
enum cudaError
|
static int |
cudaErrorTooManyPeers
enum cudaError
|
static int |
cudaErrorUnknown
enum cudaError
|
static int |
cudaErrorUnmapBufferObjectFailed
enum cudaError
|
static int |
cudaErrorUnsupportedDevSideSync
enum cudaError
|
static int |
cudaErrorUnsupportedExecAffinity
enum cudaError
|
static int |
cudaErrorUnsupportedLimit
enum cudaError
|
static int |
cudaErrorUnsupportedPtxVersion
enum cudaError
|
static int |
cudaEventBlockingSync
Event uses blocking synchronization
|
static int |
cudaEventDefault
Default event flag
|
static int |
cudaEventDisableTiming
Event will not record timing data
|
static int |
cudaEventInterprocess
Event is suitable for interprocess use.
|
static int |
cudaEventRecordDefault
Default event record flag
|
static int |
cudaEventRecordExternal
Event is captured in the graph as an external event node when performing stream capture
|
static int |
cudaEventWaitDefault
Default event wait flag
|
static int |
cudaEventWaitExternal
Event is captured in the graph as an external event node when performing stream capture
|
static int |
cudaExternalMemoryDedicated
Indicates that the external memory object is a dedicated resource
|
static int |
cudaExternalMemoryHandleTypeD3D11Resource
enum cudaExternalMemoryHandleType
|
static int |
cudaExternalMemoryHandleTypeD3D11ResourceKmt
enum cudaExternalMemoryHandleType
|
static int |
cudaExternalMemoryHandleTypeD3D12Heap
enum cudaExternalMemoryHandleType
|
static int |
cudaExternalMemoryHandleTypeD3D12Resource
enum cudaExternalMemoryHandleType
|
static int |
cudaExternalMemoryHandleTypeNvSciBuf
enum cudaExternalMemoryHandleType
|
static int |
cudaExternalMemoryHandleTypeOpaqueFd
enum cudaExternalMemoryHandleType
|
static int |
cudaExternalMemoryHandleTypeOpaqueWin32
enum cudaExternalMemoryHandleType
|
static int |
cudaExternalMemoryHandleTypeOpaqueWin32Kmt
enum cudaExternalMemoryHandleType
|
static int |
cudaExternalSemaphoreHandleTypeD3D11Fence
enum cudaExternalSemaphoreHandleType
|
static int |
cudaExternalSemaphoreHandleTypeD3D12Fence
enum cudaExternalSemaphoreHandleType
|
static int |
cudaExternalSemaphoreHandleTypeKeyedMutex
enum cudaExternalSemaphoreHandleType
|
static int |
cudaExternalSemaphoreHandleTypeKeyedMutexKmt
enum cudaExternalSemaphoreHandleType
|
static int |
cudaExternalSemaphoreHandleTypeNvSciSync
enum cudaExternalSemaphoreHandleType
|
static int |
cudaExternalSemaphoreHandleTypeOpaqueFd
enum cudaExternalSemaphoreHandleType
|
static int |
cudaExternalSemaphoreHandleTypeOpaqueWin32
enum cudaExternalSemaphoreHandleType
|
static int |
cudaExternalSemaphoreHandleTypeOpaqueWin32Kmt
enum cudaExternalSemaphoreHandleType
|
static int |
cudaExternalSemaphoreHandleTypeTimelineSemaphoreFd
enum cudaExternalSemaphoreHandleType
|
static int |
cudaExternalSemaphoreHandleTypeTimelineSemaphoreWin32
enum cudaExternalSemaphoreHandleType
|
static int |
cudaExternalSemaphoreSignalSkipNvSciBufMemSync
When the /p flags parameter of ::cudaExternalSemaphoreSignalParams
contains this flag, it indicates that signaling an external semaphore object
should skip performing appropriate memory synchronization operations over all
the external memory objects that are imported as ::cudaExternalMemoryHandleTypeNvSciBuf,
which otherwise are performed by default to ensure data coherency with other
importers of the same NvSciBuf memory objects.
|
static int |
cudaExternalSemaphoreWaitSkipNvSciBufMemSync
When the /p flags parameter of ::cudaExternalSemaphoreWaitParams
contains this flag, it indicates that waiting an external semaphore object
should skip performing appropriate memory synchronization operations over all
the external memory objects that are imported as ::cudaExternalMemoryHandleTypeNvSciBuf,
which otherwise are performed by default to ensure data coherency with other
importers of the same NvSciBuf memory objects.
|
static int |
cudaFilterModeLinear
enum cudaTextureFilterMode
|
static int |
cudaFilterModePoint
enum cudaTextureFilterMode
|
static int |
cudaFlushGPUDirectRDMAWritesOptionHost
enum cudaFlushGPUDirectRDMAWritesOptions
|
static int |
cudaFlushGPUDirectRDMAWritesOptionMemOps
enum cudaFlushGPUDirectRDMAWritesOptions
|
static int |
cudaFlushGPUDirectRDMAWritesTargetCurrentDevice
enum cudaFlushGPUDirectRDMAWritesTarget
|
static int |
cudaFlushGPUDirectRDMAWritesToAllDevices
enum cudaFlushGPUDirectRDMAWritesScope
|
static int |
cudaFlushGPUDirectRDMAWritesToOwner
enum cudaFlushGPUDirectRDMAWritesScope
|
static int |
cudaFormatModeAuto
enum cudaSurfaceFormatMode
|
static int |
cudaFormatModeForced
enum cudaSurfaceFormatMode
|
static int |
cudaFuncAttributeClusterDimMustBeSet
enum cudaFuncAttribute
|
static int |
cudaFuncAttributeClusterSchedulingPolicyPreference
enum cudaFuncAttribute
|
static int |
cudaFuncAttributeMax
enum cudaFuncAttribute
|
static int |
cudaFuncAttributeMaxDynamicSharedMemorySize
enum cudaFuncAttribute
|
static int |
cudaFuncAttributeNonPortableClusterSizeAllowed
enum cudaFuncAttribute
|
static int |
cudaFuncAttributePreferredSharedMemoryCarveout
enum cudaFuncAttribute
|
static int |
cudaFuncAttributeRequiredClusterDepth
enum cudaFuncAttribute
|
static int |
cudaFuncAttributeRequiredClusterHeight
enum cudaFuncAttribute
|
static int |
cudaFuncAttributeRequiredClusterWidth
enum cudaFuncAttribute
|
static int |
cudaFuncCachePreferEqual
enum cudaFuncCache
|
static int |
cudaFuncCachePreferL1
enum cudaFuncCache
|
static int |
cudaFuncCachePreferNone
enum cudaFuncCache
|
static int |
cudaFuncCachePreferShared
enum cudaFuncCache
|
static int |
cudaGLDeviceListAll
enum cudaGLDeviceList
|
static int |
cudaGLDeviceListCurrentFrame
enum cudaGLDeviceList
|
static int |
cudaGLDeviceListNextFrame
enum cudaGLDeviceList
|
static int |
cudaGLMapFlagsNone
enum cudaGLMapFlags
|
static int |
cudaGLMapFlagsReadOnly
enum cudaGLMapFlags
|
static int |
cudaGLMapFlagsWriteDiscard
enum cudaGLMapFlags
|
static int |
cudaGPUDirectRDMAWritesOrderingAllDevices
enum cudaGPUDirectRDMAWritesOrdering
|
static int |
cudaGPUDirectRDMAWritesOrderingNone
enum cudaGPUDirectRDMAWritesOrdering
|
static int |
cudaGPUDirectRDMAWritesOrderingOwner
enum cudaGPUDirectRDMAWritesOrdering
|
static int |
cudaGraphCondAssignDefault
enum cudaGraphConditionalHandleFlags
|
static int |
cudaGraphCondTypeIf
enum cudaGraphConditionalNodeType
|
static int |
cudaGraphCondTypeWhile
enum cudaGraphConditionalNodeType
|
static int |
cudaGraphDebugDotFlagsConditionalNodeParams
enum cudaGraphDebugDotFlags
|
static int |
cudaGraphDebugDotFlagsEventNodeParams
enum cudaGraphDebugDotFlags
|
static int |
cudaGraphDebugDotFlagsExtSemasSignalNodeParams
enum cudaGraphDebugDotFlags
|
static int |
cudaGraphDebugDotFlagsExtSemasWaitNodeParams
enum cudaGraphDebugDotFlags
|
static int |
cudaGraphDebugDotFlagsHandles
enum cudaGraphDebugDotFlags
|
static int |
cudaGraphDebugDotFlagsHostNodeParams
enum cudaGraphDebugDotFlags
|
static int |
cudaGraphDebugDotFlagsKernelNodeAttributes
enum cudaGraphDebugDotFlags
|
static int |
cudaGraphDebugDotFlagsKernelNodeParams
enum cudaGraphDebugDotFlags
|
static int |
cudaGraphDebugDotFlagsMemcpyNodeParams
enum cudaGraphDebugDotFlags
|
static int |
cudaGraphDebugDotFlagsMemsetNodeParams
enum cudaGraphDebugDotFlags
|
static int |
cudaGraphDebugDotFlagsVerbose
enum cudaGraphDebugDotFlags
|
static int |
cudaGraphDependencyTypeDefault
enum cudaGraphDependencyType
|
static int |
cudaGraphDependencyTypeProgrammatic
enum cudaGraphDependencyType
|
static int |
cudaGraphExecUpdateError
enum cudaGraphExecUpdateResult
|
static int |
cudaGraphExecUpdateErrorAttributesChanged
enum cudaGraphExecUpdateResult
|
static int |
cudaGraphExecUpdateErrorFunctionChanged
enum cudaGraphExecUpdateResult
|
static int |
cudaGraphExecUpdateErrorNodeTypeChanged
enum cudaGraphExecUpdateResult
|
static int |
cudaGraphExecUpdateErrorNotSupported
enum cudaGraphExecUpdateResult
|
static int |
cudaGraphExecUpdateErrorParametersChanged
enum cudaGraphExecUpdateResult
|
static int |
cudaGraphExecUpdateErrorTopologyChanged
enum cudaGraphExecUpdateResult
|
static int |
cudaGraphExecUpdateErrorUnsupportedFunctionChange
enum cudaGraphExecUpdateResult
|
static int |
cudaGraphExecUpdateSuccess
enum cudaGraphExecUpdateResult
|
static int |
cudaGraphicsCubeFaceNegativeX
enum cudaGraphicsCubeFace
|
static int |
cudaGraphicsCubeFaceNegativeY
enum cudaGraphicsCubeFace
|
static int |
cudaGraphicsCubeFaceNegativeZ
enum cudaGraphicsCubeFace
|
static int |
cudaGraphicsCubeFacePositiveX
enum cudaGraphicsCubeFace
|
static int |
cudaGraphicsCubeFacePositiveY
enum cudaGraphicsCubeFace
|
static int |
cudaGraphicsCubeFacePositiveZ
enum cudaGraphicsCubeFace
|
static int |
cudaGraphicsMapFlagsNone
enum cudaGraphicsMapFlags
|
static int |
cudaGraphicsMapFlagsReadOnly
enum cudaGraphicsMapFlags
|
static int |
cudaGraphicsMapFlagsWriteDiscard
enum cudaGraphicsMapFlags
|
static int |
cudaGraphicsRegisterFlagsNone
enum cudaGraphicsRegisterFlags
|
static int |
cudaGraphicsRegisterFlagsReadOnly
enum cudaGraphicsRegisterFlags
|
static int |
cudaGraphicsRegisterFlagsSurfaceLoadStore
enum cudaGraphicsRegisterFlags
|
static int |
cudaGraphicsRegisterFlagsTextureGather
enum cudaGraphicsRegisterFlags
|
static int |
cudaGraphicsRegisterFlagsWriteDiscard
enum cudaGraphicsRegisterFlags
|
static int |
cudaGraphInstantiateError
enum cudaGraphInstantiateResult
|
static int |
cudaGraphInstantiateFlagAutoFreeOnLaunch
enum cudaGraphInstantiateFlags
|
static int |
cudaGraphInstantiateFlagDeviceLaunch
enum cudaGraphInstantiateFlags
|
static int |
cudaGraphInstantiateFlagUpload
enum cudaGraphInstantiateFlags
|
static int |
cudaGraphInstantiateFlagUseNodePriority
enum cudaGraphInstantiateFlags
|
static int |
cudaGraphInstantiateInvalidStructure
enum cudaGraphInstantiateResult
|
static int |
cudaGraphInstantiateMultipleDevicesNotSupported
enum cudaGraphInstantiateResult
|
static int |
cudaGraphInstantiateNodeOperationNotSupported
enum cudaGraphInstantiateResult
|
static int |
cudaGraphInstantiateSuccess
enum cudaGraphInstantiateResult
|
static int |
cudaGraphKernelNodePortDefault
This port activates when the kernel has finished executing.
|
static int |
cudaGraphKernelNodePortLaunchCompletion
This port activates when all blocks of the kernel have begun execution.
|
static int |
cudaGraphKernelNodePortProgrammatic
This port activates when all blocks of the kernel have performed cudaTriggerProgrammaticLaunchCompletion()
or have terminated.
|
static int |
cudaGraphMemAttrReservedMemCurrent
enum cudaGraphMemAttributeType
|
static int |
cudaGraphMemAttrReservedMemHigh
enum cudaGraphMemAttributeType
|
static int |
cudaGraphMemAttrUsedMemCurrent
enum cudaGraphMemAttributeType
|
static int |
cudaGraphMemAttrUsedMemHigh
enum cudaGraphMemAttributeType
|
static int |
cudaGraphNodeTypeConditional
enum cudaGraphNodeType
|
static int |
cudaGraphNodeTypeCount
enum cudaGraphNodeType
|
static int |
cudaGraphNodeTypeEmpty
enum cudaGraphNodeType
|
static int |
cudaGraphNodeTypeEventRecord
enum cudaGraphNodeType
|
static int |
cudaGraphNodeTypeExtSemaphoreSignal
enum cudaGraphNodeType
|
static int |
cudaGraphNodeTypeExtSemaphoreWait
enum cudaGraphNodeType
|
static int |
cudaGraphNodeTypeGraph
enum cudaGraphNodeType
|
static int |
cudaGraphNodeTypeHost
enum cudaGraphNodeType
|
static int |
cudaGraphNodeTypeKernel
enum cudaGraphNodeType
|
static int |
cudaGraphNodeTypeMemAlloc
enum cudaGraphNodeType
|
static int |
cudaGraphNodeTypeMemcpy
enum cudaGraphNodeType
|
static int |
cudaGraphNodeTypeMemFree
enum cudaGraphNodeType
|
static int |
cudaGraphNodeTypeMemset
enum cudaGraphNodeType
|
static int |
cudaGraphNodeTypeWaitEvent
enum cudaGraphNodeType
|
static int |
cudaGraphUserObjectMove
enum cudaUserObjectRetainFlags
|
static int |
cudaHostAllocDefault
Default page-locked allocation flag
|
static int |
cudaHostAllocMapped
Map allocation into device space
|
static int |
cudaHostAllocPortable
Pinned memory accessible by all CUDA contexts
|
static int |
cudaHostAllocWriteCombined
Write-combined memory
|
static int |
cudaHostRegisterDefault
Default host memory registration flag
|
static int |
cudaHostRegisterIoMemory
Memory-mapped I/O space
|
static int |
cudaHostRegisterMapped
Map registered memory into device space
|
static int |
cudaHostRegisterPortable
Pinned memory accessible by all CUDA contexts
|
static int |
cudaHostRegisterReadOnly
Memory-mapped read-only
|
static int |
cudaInitDeviceFlagsAreValid
Tell the CUDA runtime that DeviceFlags is being set in cudaInitDevice call
|
static int |
cudaInvalidDeviceId
Device id that represents an invalid device
|
static int |
cudaIpcMemLazyEnablePeerAccess
Automatically enable peer access between remote devices as needed
|
static int |
cudaKernelNodeAttributeMemSyncDomain |
static int |
cudaKernelNodeAttributeMemSyncDomainMap |
static int |
cudaLaunchAttributeAccessPolicyWindow
enum cudaLaunchAttributeID
|
static int |
cudaLaunchAttributeClusterDimension
enum cudaLaunchAttributeID
|
static int |
cudaLaunchAttributeClusterSchedulingPolicyPreference
enum cudaLaunchAttributeID
|
static int |
cudaLaunchAttributeCooperative
enum cudaLaunchAttributeID
|
static int |
cudaLaunchAttributeIgnore
enum cudaLaunchAttributeID
|
static int |
cudaLaunchAttributeLaunchCompletionEvent
enum cudaLaunchAttributeID
|
static int |
cudaLaunchAttributeMemSyncDomain
enum cudaLaunchAttributeID
|
static int |
cudaLaunchAttributeMemSyncDomainMap
enum cudaLaunchAttributeID
|
static int |
cudaLaunchAttributePriority
enum cudaLaunchAttributeID
|
static int |
cudaLaunchAttributeProgrammaticEvent
enum cudaLaunchAttributeID
|
static int |
cudaLaunchAttributeProgrammaticStreamSerialization
enum cudaLaunchAttributeID
|
static int |
cudaLaunchAttributeSynchronizationPolicy
enum cudaLaunchAttributeID
|
static int |
cudaLaunchMemSyncDomainDefault
enum cudaLaunchMemSyncDomain
|
static int |
cudaLaunchMemSyncDomainRemote
enum cudaLaunchMemSyncDomain
|
static int |
cudaLimitDevRuntimePendingLaunchCount
enum cudaLimit
|
static int |
cudaLimitDevRuntimeSyncDepth
enum cudaLimit
|
static int |
cudaLimitMallocHeapSize
enum cudaLimit
|
static int |
cudaLimitMaxL2FetchGranularity
enum cudaLimit
|
static int |
cudaLimitPersistingL2CacheSize
enum cudaLimit
|
static int |
cudaLimitPrintfFifoSize
enum cudaLimit
|
static int |
cudaLimitStackSize
enum cudaLimit
|
static int |
cudaMemAccessFlagsProtNone
enum cudaMemAccessFlags
|
static int |
cudaMemAccessFlagsProtRead
enum cudaMemAccessFlags
|
static int |
cudaMemAccessFlagsProtReadWrite
enum cudaMemAccessFlags
|
static int |
cudaMemAdviseSetAccessedBy
enum cudaMemoryAdvise
|
static int |
cudaMemAdviseSetPreferredLocation
enum cudaMemoryAdvise
|
static int |
cudaMemAdviseSetReadMostly
enum cudaMemoryAdvise
|
static int |
cudaMemAdviseUnsetAccessedBy
enum cudaMemoryAdvise
|
static int |
cudaMemAdviseUnsetPreferredLocation
enum cudaMemoryAdvise
|
static int |
cudaMemAdviseUnsetReadMostly
enum cudaMemoryAdvise
|
static int |
cudaMemAllocationTypeInvalid
enum cudaMemAllocationType
|
static int |
cudaMemAllocationTypeMax
enum cudaMemAllocationType
|
static int |
cudaMemAllocationTypePinned
enum cudaMemAllocationType
|
static int |
cudaMemAttachGlobal
Memory can be accessed by any stream on any device
|
static int |
cudaMemAttachHost
Memory cannot be accessed by any stream on any device
|
static int |
cudaMemAttachSingle
Memory can only be accessed by a single stream on the associated device
|
static int |
cudaMemcpyDefault
enum cudaMemcpyKind
|
static int |
cudaMemcpyDeviceToDevice
enum cudaMemcpyKind
|
static int |
cudaMemcpyDeviceToHost
enum cudaMemcpyKind
|
static int |
cudaMemcpyHostToDevice
enum cudaMemcpyKind
|
static int |
cudaMemcpyHostToHost
enum cudaMemcpyKind
|
static int |
cudaMemHandleTypeNone
enum cudaMemAllocationHandleType
|
static int |
cudaMemHandleTypePosixFileDescriptor
enum cudaMemAllocationHandleType
|
static int |
cudaMemHandleTypeWin32
enum cudaMemAllocationHandleType
|
static int |
cudaMemHandleTypeWin32Kmt
enum cudaMemAllocationHandleType
|
static int |
cudaMemLocationTypeDevice
enum cudaMemLocationType
|
static int |
cudaMemLocationTypeHost
enum cudaMemLocationType
|
static int |
cudaMemLocationTypeHostNuma
enum cudaMemLocationType
|
static int |
cudaMemLocationTypeHostNumaCurrent
enum cudaMemLocationType
|
static int |
cudaMemLocationTypeInvalid
enum cudaMemLocationType
|
static int |
cudaMemoryTypeDevice
enum cudaMemoryType
|
static int |
cudaMemoryTypeHost
enum cudaMemoryType
|
static int |
cudaMemoryTypeManaged
enum cudaMemoryType
|
static int |
cudaMemoryTypeUnregistered
enum cudaMemoryType
|
static int |
cudaMemPoolAttrReleaseThreshold
enum cudaMemPoolAttr
|
static int |
cudaMemPoolAttrReservedMemCurrent
enum cudaMemPoolAttr
|
static int |
cudaMemPoolAttrReservedMemHigh
enum cudaMemPoolAttr
|
static int |
cudaMemPoolAttrUsedMemCurrent
enum cudaMemPoolAttr
|
static int |
cudaMemPoolAttrUsedMemHigh
enum cudaMemPoolAttr
|
static int |
cudaMemPoolReuseAllowInternalDependencies
enum cudaMemPoolAttr
|
static int |
cudaMemPoolReuseAllowOpportunistic
enum cudaMemPoolAttr
|
static int |
cudaMemPoolReuseFollowEventDependencies
enum cudaMemPoolAttr
|
static int |
cudaMemRangeAttributeAccessedBy
enum cudaMemRangeAttribute
|
static int |
cudaMemRangeAttributeLastPrefetchLocation
enum cudaMemRangeAttribute
|
static int |
cudaMemRangeAttributeLastPrefetchLocationId
enum cudaMemRangeAttribute
|
static int |
cudaMemRangeAttributeLastPrefetchLocationType
enum cudaMemRangeAttribute
|
static int |
cudaMemRangeAttributePreferredLocation
enum cudaMemRangeAttribute
|
static int |
cudaMemRangeAttributePreferredLocationId
enum cudaMemRangeAttribute
|
static int |
cudaMemRangeAttributePreferredLocationType
enum cudaMemRangeAttribute
|
static int |
cudaMemRangeAttributeReadMostly
enum cudaMemRangeAttribute
|
static int |
cudaNvSciSyncAttrSignal
When /p flags of ::cudaDeviceGetNvSciSyncAttributes is set to this,
it indicates that application need signaler specific NvSciSyncAttr
to be filled by ::cudaDeviceGetNvSciSyncAttributes.
|
static int |
cudaNvSciSyncAttrWait
When /p flags of ::cudaDeviceGetNvSciSyncAttributes is set to this,
it indicates that application need waiter specific NvSciSyncAttr
to be filled by ::cudaDeviceGetNvSciSyncAttributes.
|
static int |
cudaOccupancyDefault
Default behavior
|
static int |
cudaOccupancyDisableCachingOverride
Assume global caching is enabled and cannot be automatically turned off
|
static int |
cudaPeerAccessDefault
Default peer addressing enable flag
|
static int |
cudaReadModeElementType
enum cudaTextureReadMode
|
static int |
cudaReadModeNormalizedFloat
enum cudaTextureReadMode
|
static int |
cudaResourceTypeArray
enum cudaResourceType
|
static int |
cudaResourceTypeLinear
enum cudaResourceType
|
static int |
cudaResourceTypeMipmappedArray
enum cudaResourceType
|
static int |
cudaResourceTypePitch2D
enum cudaResourceType
|
static int |
cudaResViewFormatFloat1
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatFloat2
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatFloat4
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatHalf1
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatHalf2
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatHalf4
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatNone
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatSignedBlockCompressed4
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatSignedBlockCompressed5
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatSignedBlockCompressed6H
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatSignedChar1
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatSignedChar2
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatSignedChar4
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatSignedInt1
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatSignedInt2
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatSignedInt4
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatSignedShort1
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatSignedShort2
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatSignedShort4
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatUnsignedBlockCompressed1
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatUnsignedBlockCompressed2
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatUnsignedBlockCompressed3
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatUnsignedBlockCompressed4
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatUnsignedBlockCompressed5
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatUnsignedBlockCompressed6H
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatUnsignedBlockCompressed7
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatUnsignedChar1
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatUnsignedChar2
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatUnsignedChar4
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatUnsignedInt1
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatUnsignedInt2
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatUnsignedInt4
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatUnsignedShort1
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatUnsignedShort2
enum cudaResourceViewFormat
|
static int |
cudaResViewFormatUnsignedShort4
enum cudaResourceViewFormat
|
static int |
cudaRoundMinInf
enum cudaRoundMode
|
static int |
cudaRoundNearest
enum cudaRoundMode
|
static int |
cudaRoundPosInf
enum cudaRoundMode
|
static int |
cudaRoundZero
enum cudaRoundMode
|
static int |
CUDART_INF_BF16 |
static int |
CUDART_INF_FP16 |
static int |
CUDART_MAX_NORMAL_BF16 |
static int |
CUDART_MAX_NORMAL_FP16 |
static int |
CUDART_MIN_DENORM_BF16 |
static int |
CUDART_MIN_DENORM_FP16 |
static int |
CUDART_NAN_BF16 |
static int |
CUDART_NAN_FP16 |
static int |
CUDART_NEG_ZERO_BF16 |
static int |
CUDART_NEG_ZERO_FP16 |
static int |
CUDART_ONE_BF16 |
static int |
CUDART_ONE_FP16 |
static int |
CUDART_VERSION
CUDA Runtime API Version
|
static int |
CUDART_ZERO_BF16 |
static int |
CUDART_ZERO_FP16 |
static int |
cudaSharedMemBankSizeDefault
enum cudaSharedMemConfig
|
static int |
cudaSharedMemBankSizeEightByte
enum cudaSharedMemConfig
|
static int |
cudaSharedMemBankSizeFourByte
enum cudaSharedMemConfig
|
static int |
cudaSharedmemCarveoutDefault
enum cudaSharedCarveout
|
static int |
cudaSharedmemCarveoutMaxL1
enum cudaSharedCarveout
|
static int |
cudaSharedmemCarveoutMaxShared
enum cudaSharedCarveout
|
static int |
cudaStreamAddCaptureDependencies
enum cudaStreamUpdateCaptureDependenciesFlags
|
static int |
cudaStreamAttributeMemSyncDomain |
static int |
cudaStreamAttributeMemSyncDomainMap |
static int |
cudaStreamAttributePriority |
static int |
cudaStreamCaptureModeGlobal
enum cudaStreamCaptureMode
|
static int |
cudaStreamCaptureModeRelaxed
enum cudaStreamCaptureMode
|
static int |
cudaStreamCaptureModeThreadLocal
enum cudaStreamCaptureMode
|
static int |
cudaStreamCaptureStatusActive
enum cudaStreamCaptureStatus
|
static int |
cudaStreamCaptureStatusInvalidated
enum cudaStreamCaptureStatus
|
static int |
cudaStreamCaptureStatusNone
enum cudaStreamCaptureStatus
|
static int |
cudaStreamDefault
Default stream flag
|
static CUstream_st |
cudaStreamLegacy |
static int |
cudaStreamNonBlocking
Stream does not synchronize with stream 0 (the NULL stream)
|
static CUstream_st |
cudaStreamPerThread |
static int |
cudaStreamSetCaptureDependencies
enum cudaStreamUpdateCaptureDependenciesFlags
|
static int |
cudaSuccess
enum cudaError
|
static int |
cudaSurfaceType1D
*
*
*
|
static int |
cudaSurfaceType1DLayered |
static int |
cudaSurfaceType2D |
static int |
cudaSurfaceType2DLayered |
static int |
cudaSurfaceType3D |
static int |
cudaSurfaceTypeCubemap |
static int |
cudaSurfaceTypeCubemapLayered |
static int |
cudaSyncPolicyAuto
enum cudaSynchronizationPolicy
|
static int |
cudaSyncPolicyBlockingSync
enum cudaSynchronizationPolicy
|
static int |
cudaSyncPolicySpin
enum cudaSynchronizationPolicy
|
static int |
cudaSyncPolicyYield
enum cudaSynchronizationPolicy
|
static int |
cudaTextureType1D
*
*
*
|
static int |
cudaTextureType1DLayered |
static int |
cudaTextureType2D |
static int |
cudaTextureType2DLayered |
static int |
cudaTextureType3D |
static int |
cudaTextureTypeCubemap |
static int |
cudaTextureTypeCubemapLayered |
static int |
cudaUserObjectNoDestructorSync
enum cudaUserObjectFlags
|
static int |
MAJOR_VERSION
enum libraryPropertyType
|
static int |
MINOR_VERSION
enum libraryPropertyType
|
static int |
NVCL_CTX_SCHED_AUTO
enum cl_context_flags
|
static int |
NVCL_CTX_SCHED_BLOCKING_SYNC
enum cl_context_flags
|
static int |
NVCL_CTX_SCHED_SPIN
enum cl_context_flags
|
static int |
NVCL_CTX_SCHED_YIELD
enum cl_context_flags
|
static int |
NVCL_EVENT_SCHED_AUTO
enum cl_event_flags
|
static int |
NVCL_EVENT_SCHED_BLOCKING_SYNC
enum cl_event_flags
|
static int |
NVCL_EVENT_SCHED_SPIN
enum cl_event_flags
|
static int |
NVCL_EVENT_SCHED_YIELD
enum cl_event_flags
|
static int |
PATCH_LEVEL
enum libraryPropertyType
|
static int |
warpSize |
Constructor and Description |
---|
cudart() |
Modifier and Type | Method and Description |
---|---|
static short |
__bfloat16_as_short(__nv_bfloat16 h)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Reinterprets bits in a \p nv_bfloat16 as a signed short integer.
|
static short |
__bfloat16_as_ushort(__nv_bfloat16 h)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Reinterprets bits in a \p nv_bfloat16 as an unsigned short integer.
|
static float2 |
__bfloat1622float2(__nv_bfloat162 a)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Converts both halves of \p nv_bfloat162 to float2 and returns the result.
|
static __nv_bfloat162 |
__bfloat162bfloat162(__nv_bfloat16 a)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Returns \p nv_bfloat162 with both halves equal to the input value.
|
static byte |
__bfloat162char_rz(__nv_bfloat16 h)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Convert a nv_bfloat16 to a signed char in round-towards-zero mode.
|
static float |
__bfloat162float(__nv_bfloat16 a)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Converts \p nv_bfloat16 number to float.
|
static int |
__bfloat162int_rz(__nv_bfloat16 h)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Convert a nv_bfloat16 to a signed integer in round-towards-zero mode.
|
static long |
__bfloat162ll_rz(__nv_bfloat16 h)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Convert a nv_bfloat16 to a signed 64-bit integer in round-towards-zero mode.
|
static short |
__bfloat162short_rz(__nv_bfloat16 h)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Convert a nv_bfloat16 to a signed short integer in round-towards-zero mode.
|
static byte |
__bfloat162uchar_rz(__nv_bfloat16 h)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Convert a nv_bfloat16 to an unsigned char in round-towards-zero mode.
|
static int |
__bfloat162uint_rz(__nv_bfloat16 h)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Convert a nv_bfloat16 to an unsigned integer in round-towards-zero mode.
|
static long |
__bfloat162ull_rz(__nv_bfloat16 h)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Convert a nv_bfloat16 to an unsigned 64-bit integer in round-towards-zero
mode.
|
static short |
__bfloat162ushort_rz(__nv_bfloat16 h)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Convert a nv_bfloat16 to an unsigned short integer in round-towards-zero
mode.
|
static int |
__CUDART_API_VERSION() |
static __nv_bfloat16 |
__double2bfloat16(double a)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Converts double number to nv_bfloat16 precision in round-to-nearest-even mode
and returns \p nv_bfloat16 with converted value.
|
static __half |
__double2half(double a)
\ingroup CUDA_MATH__HALF_MISC
\brief Converts double number to half precision in round-to-nearest-even mode
and returns \p half with converted value.
|
static __nv_bfloat162 |
__float22bfloat162_rn(float2 a)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Converts both components of float2 number to nv_bfloat16 precision in
round-to-nearest-even mode and returns \p nv_bfloat162 with converted values.
|
static __half2 |
__float22half2_rn(float2 a)
\ingroup CUDA_MATH__HALF_MISC
\brief Converts both components of float2 number to half precision in
round-to-nearest-even mode and returns \p half2 with converted values.
|
static __nv_bfloat16 |
__float2bfloat16_rd(float a)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Converts float number to nv_bfloat16 precision in round-down mode
and returns \p nv_bfloat16 with converted value.
|
static __nv_bfloat16 |
__float2bfloat16_rn(float a)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Converts float number to nv_bfloat16 precision in round-to-nearest-even mode
and returns \p nv_bfloat16 with converted value.
|
static __nv_bfloat16 |
__float2bfloat16_ru(float a)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Converts float number to nv_bfloat16 precision in round-up mode
and returns \p nv_bfloat16 with converted value.
|
static __nv_bfloat16 |
__float2bfloat16_rz(float a)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Converts float number to nv_bfloat16 precision in round-towards-zero mode
and returns \p nv_bfloat16 with converted value.
|
static __nv_bfloat16 |
__float2bfloat16(float a)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Converts float number to nv_bfloat16 precision in round-to-nearest-even mode
and returns \p nv_bfloat16 with converted value.
|
static __nv_bfloat162 |
__float2bfloat162_rn(float a)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Converts input to nv_bfloat16 precision in round-to-nearest-even mode and
populates both halves of \p nv_bfloat162 with converted value.
|
static __half |
__float2half_rd(float a)
\ingroup CUDA_MATH__HALF_MISC
\brief Converts float number to half precision in round-down mode
and returns \p half with converted value.
|
static __half |
__float2half_rn(float a)
\ingroup CUDA_MATH__HALF_MISC
\brief Converts float number to half precision in round-to-nearest-even mode
and returns \p half with converted value.
|
static __half |
__float2half_ru(float a)
\ingroup CUDA_MATH__HALF_MISC
\brief Converts float number to half precision in round-up mode
and returns \p half with converted value.
|
static __half |
__float2half_rz(float a)
\ingroup CUDA_MATH__HALF_MISC
\brief Converts float number to half precision in round-towards-zero mode
and returns \p half with converted value.
|
static __half |
__float2half(float a)
\ingroup CUDA_MATH__HALF_MISC
\brief Converts float number to half precision in round-to-nearest-even mode
and returns \p half with converted value.
|
static __half2 |
__float2half2_rn(float a)
\ingroup CUDA_MATH__HALF_MISC
\brief Converts input to half precision in round-to-nearest-even mode and
populates both halves of \p half2 with converted value.
|
static __nv_bfloat162 |
__floats2bfloat162_rn(float a,
float b)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Converts both input floats to nv_bfloat16 precision in round-to-nearest-even
mode and returns \p nv_bfloat162 with converted values.
|
static __half2 |
__floats2half2_rn(float a,
float b)
\ingroup CUDA_MATH__HALF_MISC
\brief Converts both input floats to half precision in round-to-nearest-even
mode and returns \p half2 with converted values.
|
static __half2 |
__h2div(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
\brief Performs \p half2 vector division in round-to-nearest-even mode.
|
static __nv_bfloat162 |
__h2div(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
\brief Performs \p nv_bfloat162 vector division in round-to-nearest-even mode.
|
static __half |
__habs(__half a)
\ingroup CUDA_MATH__HALF_ARITHMETIC
\brief Calculates the absolute value of input \p half number and returns the result.
|
static __nv_bfloat16 |
__habs(__nv_bfloat16 a)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
\brief Calculates the absolute value of input \p nv_bfloat16 number and returns the result.
|
static __half2 |
__habs2(__half2 a)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
\brief Calculates the absolute value of both halves of the input \p half2 number and
returns the result.
|
static __nv_bfloat162 |
__habs2(__nv_bfloat162 a)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
\brief Calculates the absolute value of both halves of the input \p nv_bfloat162 number and
returns the result.
|
static __half |
__hadd_rn(__half a,
__half b)
\ingroup CUDA_MATH__HALF_ARITHMETIC
\brief Performs \p half addition in round-to-nearest-even mode.
|
static __nv_bfloat16 |
__hadd_rn(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
\brief Performs \p nv_bfloat16 addition in round-to-nearest-even mode.
|
static __half |
__hadd_sat(__half a,
__half b)
\ingroup CUDA_MATH__HALF_ARITHMETIC
\brief Performs \p half addition in round-to-nearest-even mode, with
saturation to [0.0, 1.0].
|
static __nv_bfloat16 |
__hadd_sat(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
\brief Performs \p nv_bfloat16 addition in round-to-nearest-even mode, with
saturation to [0.0, 1.0].
|
static __half |
__hadd(__half a,
__half b)
\ingroup CUDA_MATH__HALF_ARITHMETIC
\brief Performs \p half addition in round-to-nearest-even mode.
|
static __nv_bfloat16 |
__hadd(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
\brief Performs \p nv_bfloat16 addition in round-to-nearest-even mode.
|
static __half2 |
__hadd2_rn(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
\brief Performs \p half2 vector addition in round-to-nearest-even mode.
|
static __nv_bfloat162 |
__hadd2_rn(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
\brief Performs \p nv_bfloat162 vector addition in round-to-nearest-even mode.
|
static __half2 |
__hadd2_sat(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
\brief Performs \p half2 vector addition in round-to-nearest-even mode, with
saturation to [0.0, 1.0].
|
static __nv_bfloat162 |
__hadd2_sat(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
\brief Performs \p nv_bfloat162 vector addition in round-to-nearest-even mode, with
saturation to [0.0, 1.0].
|
static __half2 |
__hadd2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
\brief Performs \p half2 vector addition in round-to-nearest-even mode.
|
static __nv_bfloat162 |
__hadd2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
\brief Performs \p nv_bfloat162 vector addition in round-to-nearest-even mode.
|
static short |
__half_as_short(__half h)
\ingroup CUDA_MATH__HALF_MISC
\brief Reinterprets bits in a \p half as a signed short integer.
|
static short |
__half_as_ushort(__half h)
\ingroup CUDA_MATH__HALF_MISC
\brief Reinterprets bits in a \p half as an unsigned short integer.
|
static float2 |
__half22float2(__half2 a)
\ingroup CUDA_MATH__HALF_MISC
\brief Converts both halves of \p half2 to float2 and returns the result.
|
static byte |
__half2char_rz(__half h)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert a half to a signed char in round-towards-zero mode.
|
static float |
__half2float(__half a)
\ingroup CUDA_MATH__HALF_MISC
\brief Converts \p half number to float.
|
static __half2 |
__half2half2(__half a)
\ingroup CUDA_MATH__HALF_MISC
\brief Returns \p half2 with both halves equal to the input value.
|
static int |
__half2int_rz(__half h)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert a half to a signed integer in round-towards-zero mode.
|
static long |
__half2ll_rz(__half h)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert a half to a signed 64-bit integer in round-towards-zero mode.
|
static short |
__half2short_rz(__half h)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert a half to a signed short integer in round-towards-zero mode.
|
static byte |
__half2uchar_rz(__half h)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert a half to an unsigned char in round-towards-zero
mode.
|
static int |
__half2uint_rz(__half h)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert a half to an unsigned integer in round-towards-zero mode.
|
static long |
__half2ull_rz(__half h)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert a half to an unsigned 64-bit integer in round-towards-zero
mode.
|
static short |
__half2ushort_rz(__half h)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert a half to an unsigned short integer in round-towards-zero
mode.
|
static __nv_bfloat162 |
__halves2bfloat162(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Combines two \p nv_bfloat16 numbers into one \p nv_bfloat162 number.
|
static __half2 |
__halves2half2(__half a,
__half b)
\ingroup CUDA_MATH__HALF_MISC
\brief Combines two \p half numbers into one \p half2 number.
|
static boolean |
__hbeq2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector if-equal comparison and returns boolean true
iff both \p half results are true, boolean false otherwise.
|
static boolean |
__hbeq2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector if-equal comparison and returns boolean true
iff both \p nv_bfloat16 results are true, boolean false otherwise.
|
static boolean |
__hbequ2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector unordered if-equal comparison and returns
boolean true iff both \p half results are true, boolean false otherwise.
|
static boolean |
__hbequ2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector unordered if-equal comparison and returns
boolean true iff both \p nv_bfloat16 results are true, boolean false otherwise.
|
static boolean |
__hbge2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector greater-equal comparison and returns boolean
true iff both \p half results are true, boolean false otherwise.
|
static boolean |
__hbge2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector greater-equal comparison and returns boolean
true iff both \p nv_bfloat16 results are true, boolean false otherwise.
|
static boolean |
__hbgeu2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector unordered greater-equal comparison and
returns boolean true iff both \p half results are true, boolean false
otherwise.
|
static boolean |
__hbgeu2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector unordered greater-equal comparison and
returns boolean true iff both \p nv_bfloat16 results are true, boolean false
otherwise.
|
static boolean |
__hbgt2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector greater-than comparison and returns boolean
true iff both \p half results are true, boolean false otherwise.
|
static boolean |
__hbgt2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector greater-than comparison and returns boolean
true iff both \p nv_bfloat16 results are true, boolean false otherwise.
|
static boolean |
__hbgtu2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector unordered greater-than comparison and
returns boolean true iff both \p half results are true, boolean false
otherwise.
|
static boolean |
__hbgtu2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector unordered greater-than comparison and
returns boolean true iff both \p nv_bfloat16 results are true, boolean false
otherwise.
|
static boolean |
__hble2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector less-equal comparison and returns boolean
true iff both \p half results are true, boolean false otherwise.
|
static boolean |
__hble2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector less-equal comparison and returns boolean
true iff both \p nv_bfloat16 results are true, boolean false otherwise.
|
static boolean |
__hbleu2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector unordered less-equal comparison and returns
boolean true iff both \p half results are true, boolean false otherwise.
|
static boolean |
__hbleu2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector unordered less-equal comparison and returns
boolean true iff both \p nv_bfloat16 results are true, boolean false otherwise.
|
static boolean |
__hblt2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector less-than comparison and returns boolean
true iff both \p half results are true, boolean false otherwise.
|
static boolean |
__hblt2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector less-than comparison and returns boolean
true iff both \p nv_bfloat16 results are true, boolean false otherwise.
|
static boolean |
__hbltu2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector unordered less-than comparison and returns
boolean true iff both \p half results are true, boolean false otherwise.
|
static boolean |
__hbltu2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector unordered less-than comparison and returns
boolean true iff both \p nv_bfloat16 results are true, boolean false otherwise.
|
static boolean |
__hbne2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector not-equal comparison and returns boolean
true iff both \p half results are true, boolean false otherwise.
|
static boolean |
__hbne2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector not-equal comparison and returns boolean
true iff both \p nv_bfloat16 results are true, boolean false otherwise.
|
static boolean |
__hbneu2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector unordered not-equal comparison and returns
boolean true iff both \p half results are true, boolean false otherwise.
|
static boolean |
__hbneu2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector unordered not-equal comparison and returns
boolean true iff both \p nv_bfloat16 results are true, boolean false otherwise.
|
static __half |
__hdiv(__half a,
__half b)
\ingroup CUDA_MATH__HALF_ARITHMETIC
\brief Performs \p half division in round-to-nearest-even mode.
|
static __nv_bfloat16 |
__hdiv(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
\brief Performs \p nv_bfloat16 division in round-to-nearest-even mode.
|
static boolean |
__heq(__half a,
__half b)
\ingroup CUDA_MATH__HALF_COMPARISON
\brief Performs \p half if-equal comparison.
|
static boolean |
__heq(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
\brief Performs \p nv_bfloat16 if-equal comparison.
|
static int |
__heq2_mask(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs half2 vector if-equal comparison.
|
static int |
__heq2_mask(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs nv_bfloat162 vector if-equal comparison.
|
static __half2 |
__heq2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs half2 vector if-equal comparison.
|
static __nv_bfloat162 |
__heq2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs nv_bfloat162 vector if-equal comparison.
|
static boolean |
__hequ(__half a,
__half b)
\ingroup CUDA_MATH__HALF_COMPARISON
\brief Performs \p half unordered if-equal comparison.
|
static boolean |
__hequ(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
\brief Performs \p nv_bfloat16 unordered if-equal comparison.
|
static int |
__hequ2_mask(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector unordered if-equal comparison.
|
static int |
__hequ2_mask(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector unordered if-equal comparison.
|
static __half2 |
__hequ2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector unordered if-equal comparison.
|
static __nv_bfloat162 |
__hequ2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector unordered if-equal comparison.
|
static boolean |
__hge(__half a,
__half b)
\ingroup CUDA_MATH__HALF_COMPARISON
\brief Performs \p half greater-equal comparison.
|
static boolean |
__hge(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
\brief Performs \p nv_bfloat16 greater-equal comparison.
|
static int |
__hge2_mask(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector greater-equal comparison.
|
static int |
__hge2_mask(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector greater-equal comparison.
|
static __half2 |
__hge2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector greater-equal comparison.
|
static __nv_bfloat162 |
__hge2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector greater-equal comparison.
|
static boolean |
__hgeu(__half a,
__half b)
\ingroup CUDA_MATH__HALF_COMPARISON
\brief Performs \p half unordered greater-equal comparison.
|
static boolean |
__hgeu(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
\brief Performs \p nv_bfloat16 unordered greater-equal comparison.
|
static int |
__hgeu2_mask(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector unordered greater-equal comparison.
|
static int |
__hgeu2_mask(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector unordered greater-equal comparison.
|
static __half2 |
__hgeu2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector unordered greater-equal comparison.
|
static __nv_bfloat162 |
__hgeu2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector unordered greater-equal comparison.
|
static boolean |
__hgt(__half a,
__half b)
\ingroup CUDA_MATH__HALF_COMPARISON
\brief Performs \p half greater-than comparison.
|
static boolean |
__hgt(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
\brief Performs \p nv_bfloat16 greater-than comparison.
|
static int |
__hgt2_mask(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector greater-than comparison.
|
static int |
__hgt2_mask(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector greater-than comparison.
|
static __half2 |
__hgt2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector greater-than comparison.
|
static __nv_bfloat162 |
__hgt2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector greater-than comparison.
|
static boolean |
__hgtu(__half a,
__half b)
\ingroup CUDA_MATH__HALF_COMPARISON
\brief Performs \p half unordered greater-than comparison.
|
static boolean |
__hgtu(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
\brief Performs \p nv_bfloat16 unordered greater-than comparison.
|
static int |
__hgtu2_mask(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector unordered greater-than comparison.
|
static int |
__hgtu2_mask(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector unordered greater-than comparison.
|
static __half2 |
__hgtu2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector unordered greater-than comparison.
|
static __nv_bfloat162 |
__hgtu2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector unordered greater-than comparison.
|
static __nv_bfloat16 |
__high2bfloat16(__nv_bfloat162 a)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Returns high 16 bits of \p nv_bfloat162 input.
|
static __nv_bfloat162 |
__high2bfloat162(__nv_bfloat162 a)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Extracts high 16 bits from \p nv_bfloat162 input.
|
static float |
__high2float(__half2 a)
\ingroup CUDA_MATH__HALF_MISC
\brief Converts high 16 bits of \p half2 to float and returns the result
\details Converts high 16 bits of \p half2 input \p a to 32-bit floating-point number
and returns the result.
|
static float |
__high2float(__nv_bfloat162 a)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Converts high 16 bits of \p nv_bfloat162 to float and returns the result
\details Converts high 16 bits of \p nv_bfloat162 input \p a to 32-bit floating-point number
and returns the result.
|
static __half |
__high2half(__half2 a)
\ingroup CUDA_MATH__HALF_MISC
\brief Returns high 16 bits of \p half2 input.
|
static __half2 |
__high2half2(__half2 a)
\ingroup CUDA_MATH__HALF_MISC
\brief Extracts high 16 bits from \p half2 input.
|
static __nv_bfloat162 |
__highs2bfloat162(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Extracts high 16 bits from each of the two \p nv_bfloat162 inputs and
combines into one \p nv_bfloat162 number.
|
static __half2 |
__highs2half2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF_MISC
\brief Extracts high 16 bits from each of the two \p half2 inputs and
combines into one \p half2 number.
|
static int |
__hisinf(__half a)
\ingroup CUDA_MATH__HALF_COMPARISON
\brief Checks if the input \p half number is infinite.
|
static int |
__hisinf(__nv_bfloat16 a)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
\brief Checks if the input \p nv_bfloat16 number is infinite.
|
static boolean |
__hisnan(__half a)
\ingroup CUDA_MATH__HALF_COMPARISON
\brief Determine whether \p half argument is a NaN.
|
static boolean |
__hisnan(__nv_bfloat16 a)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
\brief Determine whether \p nv_bfloat16 argument is a NaN.
|
static __half2 |
__hisnan2(__half2 a)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Determine whether \p half2 argument is a NaN.
|
static __nv_bfloat162 |
__hisnan2(__nv_bfloat162 a)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Determine whether \p nv_bfloat162 argument is a NaN.
|
static boolean |
__hle(__half a,
__half b)
\ingroup CUDA_MATH__HALF_COMPARISON
\brief Performs \p half less-equal comparison.
|
static boolean |
__hle(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
\brief Performs \p nv_bfloat16 less-equal comparison.
|
static int |
__hle2_mask(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector less-equal comparison.
|
static int |
__hle2_mask(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector less-equal comparison.
|
static __half2 |
__hle2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector less-equal comparison.
|
static __nv_bfloat162 |
__hle2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector less-equal comparison.
|
static boolean |
__hleu(__half a,
__half b)
\ingroup CUDA_MATH__HALF_COMPARISON
\brief Performs \p half unordered less-equal comparison.
|
static boolean |
__hleu(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
\brief Performs \p nv_bfloat16 unordered less-equal comparison.
|
static int |
__hleu2_mask(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector unordered less-equal comparison.
|
static int |
__hleu2_mask(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector unordered less-equal comparison.
|
static __half2 |
__hleu2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector unordered less-equal comparison.
|
static __nv_bfloat162 |
__hleu2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector unordered less-equal comparison.
|
static boolean |
__hlt(__half a,
__half b)
\ingroup CUDA_MATH__HALF_COMPARISON
\brief Performs \p half less-than comparison.
|
static boolean |
__hlt(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
\brief Performs \p nv_bfloat16 less-than comparison.
|
static int |
__hlt2_mask(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector less-than comparison.
|
static int |
__hlt2_mask(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector less-than comparison.
|
static __half2 |
__hlt2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector less-than comparison.
|
static __nv_bfloat162 |
__hlt2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector less-than comparison.
|
static boolean |
__hltu(__half a,
__half b)
\ingroup CUDA_MATH__HALF_COMPARISON
\brief Performs \p half unordered less-than comparison.
|
static boolean |
__hltu(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
\brief Performs \p nv_bfloat16 unordered less-than comparison.
|
static int |
__hltu2_mask(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector unordered less-than comparison.
|
static int |
__hltu2_mask(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector unordered less-than comparison.
|
static __half2 |
__hltu2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector unordered less-than comparison.
|
static __nv_bfloat162 |
__hltu2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector unordered less-than comparison.
|
static __half |
__hmax_nan(__half a,
__half b)
\ingroup CUDA_MATH__HALF_COMPARISON
\brief Calculates \p half maximum of two input values, NaNs pass through.
|
static __nv_bfloat16 |
__hmax_nan(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
\brief Calculates \p nv_bfloat16 maximum of two input values, NaNs pass through.
|
static __half |
__hmax(__half a,
__half b)
\ingroup CUDA_MATH__HALF_COMPARISON
\brief Calculates \p half maximum of two input values.
|
static __nv_bfloat16 |
__hmax(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
\brief Calculates \p nv_bfloat16 maximum of two input values.
|
static __half2 |
__hmax2_nan(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Calculates \p half2 vector maximum of two inputs, NaNs pass through.
|
static __nv_bfloat162 |
__hmax2_nan(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Calculates \p nv_bfloat162 vector maximum of two inputs, NaNs pass through.
|
static __half2 |
__hmax2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Calculates \p half2 vector maximum of two inputs.
|
static __nv_bfloat162 |
__hmax2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Calculates \p nv_bfloat162 vector maximum of two inputs.
|
static __half |
__hmin_nan(__half a,
__half b)
\ingroup CUDA_MATH__HALF_COMPARISON
\brief Calculates \p half minimum of two input values, NaNs pass through.
|
static __nv_bfloat16 |
__hmin_nan(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
\brief Calculates \p nv_bfloat16 minimum of two input values, NaNs pass through.
|
static __half |
__hmin(__half a,
__half b)
\ingroup CUDA_MATH__HALF_COMPARISON
\brief Calculates \p half minimum of two input values.
|
static __nv_bfloat16 |
__hmin(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
\brief Calculates \p nv_bfloat16 minimum of two input values.
|
static __half2 |
__hmin2_nan(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Calculates \p half2 vector minimum of two inputs, NaNs pass through.
|
static __nv_bfloat162 |
__hmin2_nan(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Calculates \p nv_bfloat162 vector minimum of two inputs, NaNs pass through.
|
static __half2 |
__hmin2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Calculates \p half2 vector minimum of two inputs.
|
static __nv_bfloat162 |
__hmin2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Calculates \p nv_bfloat162 vector minimum of two inputs.
|
static __half |
__hmul_rn(__half a,
__half b)
\ingroup CUDA_MATH__HALF_ARITHMETIC
\brief Performs \p half multiplication in round-to-nearest-even mode.
|
static __nv_bfloat16 |
__hmul_rn(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
\brief Performs \p nv_bfloat16 multiplication in round-to-nearest-even mode.
|
static __half |
__hmul_sat(__half a,
__half b)
\ingroup CUDA_MATH__HALF_ARITHMETIC
\brief Performs \p half multiplication in round-to-nearest-even mode, with
saturation to [0.0, 1.0].
|
static __nv_bfloat16 |
__hmul_sat(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
\brief Performs \p nv_bfloat16 multiplication in round-to-nearest-even mode, with
saturation to [0.0, 1.0].
|
static __half |
__hmul(__half a,
__half b)
\ingroup CUDA_MATH__HALF_ARITHMETIC
\brief Performs \p half multiplication in round-to-nearest-even mode.
|
static __nv_bfloat16 |
__hmul(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
\brief Performs \p nv_bfloat16 multiplication in round-to-nearest-even mode.
|
static __half2 |
__hmul2_rn(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
\brief Performs \p half2 vector multiplication in round-to-nearest-even mode.
|
static __nv_bfloat162 |
__hmul2_rn(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
\brief Performs \p nv_bfloat162 vector multiplication in round-to-nearest-even mode.
|
static __half2 |
__hmul2_sat(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
\brief Performs \p half2 vector multiplication in round-to-nearest-even mode,
with saturation to [0.0, 1.0].
|
static __nv_bfloat162 |
__hmul2_sat(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
\brief Performs \p nv_bfloat162 vector multiplication in round-to-nearest-even mode,
with saturation to [0.0, 1.0].
|
static __half2 |
__hmul2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
\brief Performs \p half2 vector multiplication in round-to-nearest-even mode.
|
static __nv_bfloat162 |
__hmul2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
\brief Performs \p nv_bfloat162 vector multiplication in round-to-nearest-even mode.
|
static boolean |
__hne(__half a,
__half b)
\ingroup CUDA_MATH__HALF_COMPARISON
\brief Performs \p half not-equal comparison.
|
static boolean |
__hne(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
\brief Performs \p nv_bfloat16 not-equal comparison.
|
static int |
__hne2_mask(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector not-equal comparison.
|
static int |
__hne2_mask(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector not-equal comparison.
|
static __half2 |
__hne2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector not-equal comparison.
|
static __nv_bfloat162 |
__hne2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector not-equal comparison.
|
static __half |
__hneg(__half a)
\ingroup CUDA_MATH__HALF_ARITHMETIC
\brief Negates input \p half number and returns the result.
|
static __nv_bfloat16 |
__hneg(__nv_bfloat16 a)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
\brief Negates input \p nv_bfloat16 number and returns the result.
|
static __half2 |
__hneg2(__half2 a)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
\brief Negates both halves of the input \p half2 number and returns the
result.
|
static __nv_bfloat162 |
__hneg2(__nv_bfloat162 a)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
\brief Negates both halves of the input \p nv_bfloat162 number and returns the
result.
|
static boolean |
__hneu(__half a,
__half b)
\ingroup CUDA_MATH__HALF_COMPARISON
\brief Performs \p half unordered not-equal comparison.
|
static boolean |
__hneu(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
\brief Performs \p nv_bfloat16 unordered not-equal comparison.
|
static int |
__hneu2_mask(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector unordered not-equal comparison.
|
static int |
__hneu2_mask(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector unordered not-equal comparison.
|
static __half2 |
__hneu2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_COMPARISON
\brief Performs \p half2 vector unordered not-equal comparison.
|
static __nv_bfloat162 |
__hneu2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
\brief Performs \p nv_bfloat162 vector unordered not-equal comparison.
|
static __half |
__hsub_rn(__half a,
__half b)
\ingroup CUDA_MATH__HALF_ARITHMETIC
\brief Performs \p half subtraction in round-to-nearest-even mode.
|
static __nv_bfloat16 |
__hsub_rn(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
\brief Performs \p nv_bfloat16 subtraction in round-to-nearest-even mode.
|
static __half |
__hsub_sat(__half a,
__half b)
\ingroup CUDA_MATH__HALF_ARITHMETIC
\brief Performs \p half subtraction in round-to-nearest-even mode, with
saturation to [0.0, 1.0].
|
static __nv_bfloat16 |
__hsub_sat(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
\brief Performs \p nv_bfloat16 subtraction in round-to-nearest-even mode, with
saturation to [0.0, 1.0].
|
static __half |
__hsub(__half a,
__half b)
\ingroup CUDA_MATH__HALF_ARITHMETIC
\brief Performs \p half subtraction in round-to-nearest-even mode.
|
static __nv_bfloat16 |
__hsub(__nv_bfloat16 a,
__nv_bfloat16 b)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
\brief Performs \p nv_bfloat16 subtraction in round-to-nearest-even mode.
|
static __half2 |
__hsub2_rn(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
\brief Performs \p half2 vector subtraction in round-to-nearest-even mode.
|
static __nv_bfloat162 |
__hsub2_rn(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
\brief Performs \p nv_bfloat162 vector subtraction in round-to-nearest-even mode.
|
static __half2 |
__hsub2_sat(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
\brief Performs \p half2 vector subtraction in round-to-nearest-even mode,
with saturation to [0.0, 1.0].
|
static __nv_bfloat162 |
__hsub2_sat(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
\brief Performs \p nv_bfloat162 vector subtraction in round-to-nearest-even mode,
with saturation to [0.0, 1.0].
|
static __half2 |
__hsub2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
\brief Performs \p half2 vector subtraction in round-to-nearest-even mode.
|
static __nv_bfloat162 |
__hsub2(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
\brief Performs \p nv_bfloat162 vector subtraction in round-to-nearest-even mode.
|
static __nv_bfloat16 |
__int2bfloat16_rn(int i)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Convert a signed integer to a nv_bfloat16 in round-to-nearest-even mode.
|
static __half |
__int2half_rd(int i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert a signed integer to a half in round-down mode.
|
static __half |
__int2half_rn(int i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert a signed integer to a half in round-to-nearest-even mode.
|
static __half |
__int2half_ru(int i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert a signed integer to a half in round-up mode.
|
static __half |
__int2half_rz(int i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert a signed integer to a half in round-towards-zero mode.
|
static float |
__internal_bfloat162float(short h) |
static int |
__internal_bfloat162int_rz(__nv_bfloat16 h) |
static int |
__internal_bfloat162uint_rz(__nv_bfloat16 h) |
static float |
__internal_double2float_rn(double x) |
static int |
__internal_float_as_uint(float f) |
static short |
__internal_float2bfloat16(float f,
int[] sign,
int[] remainder) |
static short |
__internal_float2bfloat16(float f,
IntBuffer sign,
IntBuffer remainder) |
static short |
__internal_float2bfloat16(float f,
IntPointer sign,
IntPointer remainder) |
static double |
__internal_float2double(float x) |
static short |
__internal_float2half(float f,
int[] sign,
int[] remainder) |
static short |
__internal_float2half(float f,
IntBuffer sign,
IntBuffer remainder) |
static short |
__internal_float2half(float f,
IntPointer sign,
IntPointer remainder) |
static float |
__internal_half2float(short h) |
static float |
__internal_uint_as_float(int u) |
static __nv_bfloat16 |
__ll2bfloat16_rn(long i)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Convert a signed 64-bit integer to a nv_bfloat16 in round-to-nearest-even
mode.
|
static __half |
__ll2half_rd(long i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert a signed 64-bit integer to a half in round-down mode.
|
static __half |
__ll2half_rn(long i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert a signed 64-bit integer to a half in round-to-nearest-even
mode.
|
static __half |
__ll2half_ru(long i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert a signed 64-bit integer to a half in round-up mode.
|
static __half |
__ll2half_rz(long i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert a signed 64-bit integer to a half in round-towards-zero mode.
|
static __nv_bfloat16 |
__low2bfloat16(__nv_bfloat162 a)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Returns low 16 bits of \p nv_bfloat162 input.
|
static __nv_bfloat162 |
__low2bfloat162(__nv_bfloat162 a)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Extracts low 16 bits from \p nv_bfloat162 input.
|
static float |
__low2float(__half2 a)
\ingroup CUDA_MATH__HALF_MISC
\brief Converts low 16 bits of \p half2 to float and returns the result
\details Converts low 16 bits of \p half2 input \p a to 32-bit floating-point number
and returns the result.
|
static float |
__low2float(__nv_bfloat162 a)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Converts low 16 bits of \p nv_bfloat162 to float and returns the result
\details Converts low 16 bits of \p nv_bfloat162 input \p a to 32-bit floating-point number
and returns the result.
|
static __half |
__low2half(__half2 a)
\ingroup CUDA_MATH__HALF_MISC
\brief Returns low 16 bits of \p half2 input.
|
static __half2 |
__low2half2(__half2 a)
\ingroup CUDA_MATH__HALF_MISC
\brief Extracts low 16 bits from \p half2 input.
|
static __half2 |
__lowhigh2highlow(__half2 a)
\ingroup CUDA_MATH__HALF_MISC
\brief Swaps both halves of the \p half2 input.
|
static __nv_bfloat162 |
__lowhigh2highlow(__nv_bfloat162 a)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Swaps both halves of the \p nv_bfloat162 input.
|
static __nv_bfloat162 |
__lows2bfloat162(__nv_bfloat162 a,
__nv_bfloat162 b)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Extracts low 16 bits from each of the two \p nv_bfloat162 inputs and combines
into one \p nv_bfloat162 number.
|
static __half2 |
__lows2half2(__half2 a,
__half2 b)
\ingroup CUDA_MATH__HALF_MISC
\brief Extracts low 16 bits from each of the two \p half2 inputs and combines
into one \p half2 number.
|
static __nv_bfloat16 |
__short_as_bfloat16(short i)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Reinterprets bits in a signed short integer as a \p nv_bfloat16.
|
static __half |
__short_as_half(short i)
\ingroup CUDA_MATH__HALF_MISC
\brief Reinterprets bits in a signed short integer as a \p half.
|
static __nv_bfloat16 |
__short2bfloat16_rn(short i)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Convert a signed short integer to a nv_bfloat16 in round-to-nearest-even
mode.
|
static __half |
__short2half_rd(short i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert a signed short integer to a half in round-down mode.
|
static __half |
__short2half_rn(short i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert a signed short integer to a half in round-to-nearest-even
mode.
|
static __half |
__short2half_ru(short i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert a signed short integer to a half in round-up mode.
|
static __half |
__short2half_rz(short i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert a signed short integer to a half in round-towards-zero mode.
|
static __nv_bfloat16 |
__uint2bfloat16_rn(int i)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Convert an unsigned integer to a nv_bfloat16 in round-to-nearest-even mode.
|
static __half |
__uint2half_rd(int i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert an unsigned integer to a half in round-down mode.
|
static __half |
__uint2half_rn(int i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert an unsigned integer to a half in round-to-nearest-even mode.
|
static __half |
__uint2half_ru(int i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert an unsigned integer to a half in round-up mode.
|
static __half |
__uint2half_rz(int i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert an unsigned integer to a half in round-towards-zero mode.
|
static __nv_bfloat16 |
__ull2bfloat16_rn(long i)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Convert an unsigned 64-bit integer to a nv_bfloat16 in round-to-nearest-even
mode.
|
static __half |
__ull2half_rd(long i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert an unsigned 64-bit integer to a half in round-down mode.
|
static __half |
__ull2half_rn(long i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert an unsigned 64-bit integer to a half in round-to-nearest-even
mode.
|
static __half |
__ull2half_ru(long i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert an unsigned 64-bit integer to a half in round-up mode.
|
static __half |
__ull2half_rz(long i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert an unsigned 64-bit integer to a half in round-towards-zero
mode.
|
static __nv_bfloat16 |
__ushort_as_bfloat16(short i)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Reinterprets bits in an unsigned short integer as a \p nv_bfloat16.
|
static __half |
__ushort_as_half(short i)
\ingroup CUDA_MATH__HALF_MISC
\brief Reinterprets bits in an unsigned short integer as a \p half.
|
static __nv_bfloat16 |
__ushort2bfloat16_rn(short i)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Convert an unsigned short integer to a nv_bfloat16 in round-to-nearest-even
mode.
|
static __half |
__ushort2half_rd(short i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert an unsigned short integer to a half in round-down mode.
|
static __half |
__ushort2half_rn(short i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert an unsigned short integer to a half in round-to-nearest-even
mode.
|
static __half |
__ushort2half_ru(short i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert an unsigned short integer to a half in round-up mode.
|
static __half |
__ushort2half_rz(short i)
\ingroup CUDA_MATH__HALF_MISC
\brief Convert an unsigned short integer to a half in round-towards-zero
mode.
|
static __half |
add(__half lh,
__half rh)
\ingroup CUDA_MATH__HALF_ARITHMETIC
Performs \p half addition operation.
|
static __half |
add(__half h)
\ingroup CUDA_MATH__HALF_ARITHMETIC
Implements \p half unary plus operator, returns input value.
|
static __half2 |
add(__half2 lh,
__half2 rh)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
Performs packed \p half addition operation.
|
static __half2 |
add(__half2 h)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
Implements packed \p half unary plus operator, returns input value.
|
static __nv_bfloat16 |
add(__nv_bfloat16 lh,
__nv_bfloat16 rh)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
Performs \p nv_bfloat16 addition operation.
|
static __nv_bfloat16 |
add(__nv_bfloat16 h)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
Implements \p nv_bfloat16 unary plus operator, returns input value.
|
static __nv_bfloat162 |
add(__nv_bfloat162 lh,
__nv_bfloat162 rh)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
Performs packed \p nv_bfloat16 addition operation.
|
static __nv_bfloat162 |
add(__nv_bfloat162 h)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
Implements packed \p nv_bfloat16 unary plus operator, returns input value.
|
static __half |
addPut(__half lh,
__half rh)
\ingroup CUDA_MATH__HALF_ARITHMETIC
Performs \p half compound assignment with addition operation.
|
static __half2 |
addPut(__half2 lh,
__half2 rh)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
Performs packed \p half compound assignment with addition operation.
|
static __nv_bfloat16 |
addPut(__nv_bfloat16 lh,
__nv_bfloat16 rh)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
Performs \p nv_bfloat16 compound assignment with addition operation.
|
static __nv_bfloat162 |
addPut(__nv_bfloat162 lh,
__nv_bfloat162 rh)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
Performs packed \p nv_bfloat16 compound assignment with addition operation.
|
static int |
CU_DEVICE_CPU()
Device that represents the CPU
|
static int |
CU_DEVICE_INVALID()
Device that represents an invalid device
|
static Pointer |
CU_LAUNCH_PARAM_BUFFER_POINTER()
Indicator that the next value in the \p extra parameter to
::cuLaunchKernel will be a pointer to a buffer containing all kernel
parameters used for launching kernel \p f.
|
static Pointer |
CU_LAUNCH_PARAM_BUFFER_SIZE()
Indicator that the next value in the \p extra parameter to
::cuLaunchKernel will be a pointer to a size_t which contains the
size of the buffer specified with ::CU_LAUNCH_PARAM_BUFFER_POINTER.
|
static Pointer |
CU_LAUNCH_PARAM_END()
End of array terminator for the \p extra parameter to
::cuLaunchKernel
|
static CUstream_st |
CU_STREAM_LEGACY()
Legacy stream handle
Stream handle that can be passed as a CUstream to use an implicit stream
with legacy synchronization behavior.
|
static CUstream_st |
CU_STREAM_PER_THREAD()
Per-thread stream handle
Stream handle that can be passed as a CUstream to use an implicit stream
with per-thread synchronization behavior.
|
static int |
cuArray3DCreate(CUarray_st pHandle,
CUDA_ARRAY3D_DESCRIPTOR_v2 pAllocateArray)
\brief Creates a 3D CUDA array
Creates a CUDA array according to the ::CUDA_ARRAY3D_DESCRIPTOR structure
\p pAllocateArray and returns a handle to the new CUDA array in \p *pHandle.
|
static int |
cuArray3DGetDescriptor(CUDA_ARRAY3D_DESCRIPTOR_v2 pArrayDescriptor,
CUarray_st hArray)
\brief Get a 3D CUDA array descriptor
Returns in \p *pArrayDescriptor a descriptor containing information on the
format and dimensions of the CUDA array \p hArray.
|
static int |
cuArrayCreate(CUarray_st pHandle,
CUDA_ARRAY_DESCRIPTOR_v2 pAllocateArray)
\brief Creates a 1D or 2D CUDA array
Creates a CUDA array according to the ::CUDA_ARRAY_DESCRIPTOR structure
\p pAllocateArray and returns a handle to the new CUDA array in \p *pHandle.
|
static int |
cuArrayDestroy(CUarray_st hArray)
\brief Destroys a CUDA array
Destroys the CUDA array \p hArray.
|
static int |
cuArrayGetDescriptor(CUDA_ARRAY_DESCRIPTOR_v2 pArrayDescriptor,
CUarray_st hArray)
\brief Get a 1D or 2D CUDA array descriptor
Returns in \p *pArrayDescriptor a descriptor containing information on the
format and dimensions of the CUDA array \p hArray.
|
static int |
cuArrayGetMemoryRequirements(CUDA_ARRAY_MEMORY_REQUIREMENTS_v1 memoryRequirements,
CUarray_st array,
int device)
\brief Returns the memory requirements of a CUDA array
Returns the memory requirements of a CUDA array in \p memoryRequirements
If the CUDA array is not allocated with flag ::CUDA_ARRAY3D_DEFERRED_MAPPING
::CUDA_ERROR_INVALID_VALUE will be returned.
|
static int |
cuArrayGetPlane(CUarray_st pPlaneArray,
CUarray_st hArray,
int planeIdx)
\brief Gets a CUDA array plane from a CUDA array
Returns in \p pPlaneArray a CUDA array that represents a single format plane
of the CUDA array \p hArray.
|
static int |
cuArrayGetSparseProperties(CUDA_ARRAY_SPARSE_PROPERTIES_v1 sparseProperties,
CUarray_st array)
\brief Returns the layout properties of a sparse CUDA array
Returns the layout properties of a sparse CUDA array in \p sparseProperties
If the CUDA array is not allocated with flag ::CUDA_ARRAY3D_SPARSE
::CUDA_ERROR_INVALID_VALUE will be returned.
|
static double |
cuCabs(double2 x) |
static float |
cuCabsf(float2 x) |
static double2 |
cuCadd(double2 x,
double2 y) |
static float2 |
cuCaddf(float2 x,
float2 y) |
static double2 |
cuCdiv(double2 x,
double2 y) |
static float2 |
cuCdivf(float2 x,
float2 y) |
static double2 |
cuCfma(double2 x,
double2 y,
double2 d) |
static float2 |
cuCfmaf(float2 x,
float2 y,
float2 d) |
static double |
cuCimag(double2 x) |
static float |
cuCimagf(float2 x) |
static double2 |
cuCmul(double2 x,
double2 y) |
static float2 |
cuCmulf(float2 x,
float2 y) |
static float2 |
cuComplexDoubleToFloat(double2 c) |
static double2 |
cuComplexFloatToDouble(float2 c) |
static double2 |
cuConj(double2 x) |
static float2 |
cuConjf(float2 x) |
static int |
cuCoredumpGetAttribute(int attrib,
Pointer value,
SizeTPointer size)
\brief Allows caller to fetch a coredump attribute value for the current context
Returns in \p *value the requested value specified by \p attrib.
|
static int |
cuCoredumpGetAttributeGlobal(int attrib,
Pointer value,
SizeTPointer size)
\brief Allows caller to fetch a coredump attribute value for the entire application
Returns in \p *value the requested value specified by \p attrib.
|
static int |
cuCoredumpSetAttribute(int attrib,
Pointer value,
SizeTPointer size)
\brief Allows caller to set a coredump attribute value for the current context
This function should be considered an alternate interface to the CUDA-GDB environment
variables defined in this document: https://docs.nvidia.com/cuda/cuda-gdb/index.html#gpu-coredump
An important design decision to note is that any coredump environment variable values
set before CUDA initializes will take permanent precedence over any values set with this
this function.
|
static int |
cuCoredumpSetAttributeGlobal(int attrib,
Pointer value,
SizeTPointer size)
\brief Allows caller to set a coredump attribute value globally
This function should be considered an alternate interface to the CUDA-GDB environment
variables defined in this document: https://docs.nvidia.com/cuda/cuda-gdb/index.html#gpu-coredump
An important design decision to note is that any coredump environment variable values
set before CUDA initializes will take permanent precedence over any values set with this
this function.
|
static double |
cuCreal(double2 x) |
static float |
cuCrealf(float2 x) |
static double2 |
cuCsub(double2 x,
double2 y) |
static float2 |
cuCsubf(float2 x,
float2 y) |
static int |
cuCtxAttach(CUctx_st pctx,
int flags)
Deprecated.
Note that this function is deprecated and should not be used.
Increments the usage count of the context and passes back a context handle
in \p *pctx that must be passed to ::cuCtxDetach() when the application is
done with the context. ::cuCtxAttach() fails if there is no context current
to the thread.
Currently, the \p flags parameter must be 0.
|
static int |
cuCtxCreate_v3(CUctx_st pctx,
CUexecAffinityParam_v1 paramsArray,
int numParams,
int flags,
int dev)
\brief Create a CUDA context with execution affinity
Creates a new CUDA context with execution affinity and associates it with
the calling thread.
|
static int |
cuCtxCreate(CUctx_st pctx,
int flags,
int dev)
\brief Create a CUDA context
\note In most cases it is recommended to use ::cuDevicePrimaryCtxRetain.
|
static int |
cuCtxDestroy(CUctx_st ctx)
\brief Destroy a CUDA context
Destroys the CUDA context specified by \p ctx.
|
static int |
cuCtxDetach(CUctx_st ctx)
Deprecated.
Note that this function is deprecated and should not be used.
Decrements the usage count of the context \p ctx, and destroys the context
if the usage count goes to 0. The context must be a handle that was passed
back by ::cuCtxCreate() or ::cuCtxAttach(), and must be current to the
calling thread.
|
static int |
cuCtxDisablePeerAccess(CUctx_st peerContext)
\brief Disables direct access to memory allocations in a peer context and
unregisters any registered allocations.
|
static int |
cuCtxEnablePeerAccess(CUctx_st peerContext,
int Flags)
\brief Enables direct access to memory allocations in a peer context.
|
static int |
cuCtxGetApiVersion(CUctx_st ctx,
int[] version) |
static int |
cuCtxGetApiVersion(CUctx_st ctx,
IntBuffer version) |
static int |
cuCtxGetApiVersion(CUctx_st ctx,
IntPointer version)
\brief Gets the context's API version.
|
static int |
cuCtxGetCacheConfig(int[] pconfig) |
static int |
cuCtxGetCacheConfig(IntBuffer pconfig) |
static int |
cuCtxGetCacheConfig(IntPointer pconfig)
\brief Returns the preferred cache configuration for the current context.
|
static int |
cuCtxGetCurrent(CUctx_st pctx)
\brief Returns the CUDA context bound to the calling CPU thread.
|
static int |
cuCtxGetDevice(int[] device) |
static int |
cuCtxGetDevice(IntBuffer device) |
static int |
cuCtxGetDevice(IntPointer device)
\brief Returns the device ID for the current context
Returns in \p *device the ordinal of the current context's device.
|
static int |
cuCtxGetExecAffinity(CUexecAffinityParam_v1 pExecAffinity,
int type)
\brief Returns the execution affinity setting for the current context.
|
static int |
cuCtxGetFlags(int[] flags) |
static int |
cuCtxGetFlags(IntBuffer flags) |
static int |
cuCtxGetFlags(IntPointer flags)
\brief Returns the flags for the current context
Returns in \p *flags the flags of the current context.
|
static int |
cuCtxGetId(CUctx_st ctx,
long[] ctxId) |
static int |
cuCtxGetId(CUctx_st ctx,
LongBuffer ctxId) |
static int |
cuCtxGetId(CUctx_st ctx,
LongPointer ctxId)
\brief Returns the unique Id associated with the context supplied
Returns in \p ctxId the unique Id which is associated with a given context.
|
static int |
cuCtxGetLimit(SizeTPointer pvalue,
int _limit)
\brief Returns resource limits
Returns in \p *pvalue the current size of \p limit.
|
static int |
cuCtxGetSharedMemConfig(int[] pConfig) |
static int |
cuCtxGetSharedMemConfig(IntBuffer pConfig) |
static int |
cuCtxGetSharedMemConfig(IntPointer pConfig)
\brief Returns the current shared memory configuration for the current context.
|
static int |
cuCtxGetStreamPriorityRange(int[] leastPriority,
int[] greatestPriority) |
static int |
cuCtxGetStreamPriorityRange(IntBuffer leastPriority,
IntBuffer greatestPriority) |
static int |
cuCtxGetStreamPriorityRange(IntPointer leastPriority,
IntPointer greatestPriority)
\brief Returns numerical values that correspond to the least and
greatest stream priorities.
|
static int |
cuCtxPopCurrent(CUctx_st pctx)
\brief Pops the current CUDA context from the current CPU thread.
|
static int |
cuCtxPushCurrent(CUctx_st ctx)
\brief Pushes a context on the current CPU thread
Pushes the given context \p ctx onto the CPU thread's stack of current
contexts.
|
static int |
cuCtxResetPersistingL2Cache()
\brief Resets all persisting lines in cache to normal status.
|
static int |
cuCtxSetCacheConfig(int config)
\brief Sets the preferred cache configuration for the current context.
|
static int |
cuCtxSetCurrent(CUctx_st ctx)
\brief Binds the specified CUDA context to the calling CPU thread
Binds the specified CUDA context to the calling CPU thread.
|
static int |
cuCtxSetFlags(int flags)
\brief Sets the flags for the current context
Sets the flags for the current context overwriting previously set ones.
|
static int |
cuCtxSetLimit(int _limit,
long value)
\brief Set resource limits
Setting \p limit to \p value is a request by the application to update
the current limit maintained by the context.
|
static int |
cuCtxSetSharedMemConfig(int config)
\brief Sets the shared memory configuration for the current context.
|
static int |
cuCtxSynchronize()
\brief Block for a context's tasks to complete
Blocks until the device has completed all preceding requested tasks.
|
static int |
cudaArrayGetInfo(cudaChannelFormatDesc desc,
cudaExtent extent,
int[] flags,
cudaArray array) |
static int |
cudaArrayGetInfo(cudaChannelFormatDesc desc,
cudaExtent extent,
IntBuffer flags,
cudaArray array) |
static int |
cudaArrayGetInfo(cudaChannelFormatDesc desc,
cudaExtent extent,
IntPointer flags,
cudaArray array)
\brief Gets info about the specified cudaArray
Returns in \p *desc, \p *extent and \p *flags respectively, the type, shape
and flags of \p array.
|
static int |
cudaArrayGetMemoryRequirements(cudaArrayMemoryRequirements memoryRequirements,
cudaArray array,
int device)
\brief Returns the memory requirements of a CUDA array
Returns the memory requirements of a CUDA array in \p memoryRequirements
If the CUDA array is not allocated with flag ::cudaArrayDeferredMapping
::cudaErrorInvalidValue will be returned.
|
static int |
cudaArrayGetPlane(cudaArray pPlaneArray,
cudaArray hArray,
int planeIdx)
\brief Gets a CUDA array plane from a CUDA array
Returns in \p pPlaneArray a CUDA array that represents a single format plane
of the CUDA array \p hArray.
|
static int |
cudaArrayGetSparseProperties(cudaArraySparseProperties sparseProperties,
cudaArray array)
\brief Returns the layout properties of a sparse CUDA array
Returns the layout properties of a sparse CUDA array in \p sparseProperties.
|
static int |
cudaChooseDevice(int[] device,
cudaDeviceProp prop) |
static int |
cudaChooseDevice(IntBuffer device,
cudaDeviceProp prop) |
static int |
cudaChooseDevice(IntPointer device,
cudaDeviceProp prop)
\brief Select compute-device which best matches criteria
Returns in \p *device the device which has properties that best match
\p *prop.
|
static cudaChannelFormatDesc |
cudaCreateChannelDesc(int x,
int y,
int z,
int w,
int f)
\brief Returns a channel descriptor using the specified format
Returns a channel descriptor with format \p f and number of bits of each
component \p x, \p y, \p z, and \p w.
|
static int |
cudaCreateSurfaceObject(long[] pSurfObject,
cudaResourceDesc pResDesc) |
static int |
cudaCreateSurfaceObject(LongBuffer pSurfObject,
cudaResourceDesc pResDesc) |
static int |
cudaCreateSurfaceObject(LongPointer pSurfObject,
cudaResourceDesc pResDesc)
\brief Creates a surface object
Creates a surface object and returns it in \p pSurfObject.
|
static int |
cudaCreateTextureObject(long[] pTexObject,
cudaResourceDesc pResDesc,
cudaTextureDesc pTexDesc,
cudaResourceViewDesc pResViewDesc) |
static int |
cudaCreateTextureObject(LongBuffer pTexObject,
cudaResourceDesc pResDesc,
cudaTextureDesc pTexDesc,
cudaResourceViewDesc pResViewDesc) |
static int |
cudaCreateTextureObject(LongPointer pTexObject,
cudaResourceDesc pResDesc,
cudaTextureDesc pTexDesc,
cudaResourceViewDesc pResViewDesc)
\brief Creates a texture object
Creates a texture object and returns it in \p pTexObject.
|
static int |
cudaCtxResetPersistingL2Cache()
\brief Resets all persisting lines in cache to normal status.
|
static int |
cudaDestroyExternalMemory(CUexternalMemory_st extMem)
\brief Destroys an external memory object.
|
static int |
cudaDestroyExternalSemaphore(CUexternalSemaphore_st extSem)
\brief Destroys an external semaphore
Destroys an external semaphore object and releases any references
to the underlying resource.
|
static int |
cudaDestroySurfaceObject(long surfObject)
\brief Destroys a surface object
Destroys the surface object specified by \p surfObject.
|
static int |
cudaDestroyTextureObject(long texObject)
\brief Destroys a texture object
Destroys the texture object specified by \p texObject.
|
static int |
cudaDeviceCanAccessPeer(int[] canAccessPeer,
int device,
int peerDevice) |
static int |
cudaDeviceCanAccessPeer(IntBuffer canAccessPeer,
int device,
int peerDevice) |
static int |
cudaDeviceCanAccessPeer(IntPointer canAccessPeer,
int device,
int peerDevice)
\brief Queries if a device may directly access a peer device's memory.
|
static int |
cudaDeviceDisablePeerAccess(int peerDevice)
\brief Disables direct access to memory allocations on a peer device.
|
static int |
cudaDeviceEnablePeerAccess(int peerDevice,
int flags)
\brief Enables direct access to memory allocations on a peer device.
|
static int |
cudaDeviceFlushGPUDirectRDMAWrites(int target,
int scope)
\brief Blocks until remote writes are visible to the specified scope
Blocks until remote writes to the target context via mappings created
through GPUDirect RDMA APIs, like nvidia_p2p_get_pages (see
https://docs.nvidia.com/cuda/gpudirect-rdma for more information), are
visible to the specified scope.
|
static int |
cudaDeviceGetAttribute(int[] value,
int attr,
int device) |
static int |
cudaDeviceGetAttribute(IntBuffer value,
int attr,
int device) |
static int |
cudaDeviceGetAttribute(IntPointer value,
int attr,
int device)
\brief Returns information about the device
Returns in \p *value the integer value of the attribute \p attr on device
\p device.
|
static int |
cudaDeviceGetByPCIBusId(int[] device,
BytePointer pciBusId) |
static int |
cudaDeviceGetByPCIBusId(int[] device,
String pciBusId) |
static int |
cudaDeviceGetByPCIBusId(IntBuffer device,
BytePointer pciBusId) |
static int |
cudaDeviceGetByPCIBusId(IntBuffer device,
String pciBusId) |
static int |
cudaDeviceGetByPCIBusId(IntPointer device,
BytePointer pciBusId)
\brief Returns a handle to a compute device
Returns in \p *device a device ordinal given a PCI bus ID string.
|
static int |
cudaDeviceGetByPCIBusId(IntPointer device,
String pciBusId) |
static int |
cudaDeviceGetCacheConfig(int[] pCacheConfig) |
static int |
cudaDeviceGetCacheConfig(IntBuffer pCacheConfig) |
static int |
cudaDeviceGetCacheConfig(IntPointer pCacheConfig)
\brief Returns the preferred cache configuration for the current device.
|
static int |
cudaDeviceGetDefaultMemPool(CUmemPoolHandle_st memPool,
int device)
\brief Returns the default mempool of a device
The default mempool of a device contains device memory from that device.
|
static int |
cudaDeviceGetGraphMemAttribute(int device,
int attr,
Pointer value)
\brief Query asynchronous allocation attributes related to graphs
Valid attributes are:
- ::cudaGraphMemAttrUsedMemCurrent: Amount of memory, in bytes, currently associated with graphs
- ::cudaGraphMemAttrUsedMemHigh: High watermark of memory, in bytes, associated with graphs since the
last time it was reset.
|
static int |
cudaDeviceGetLimit(SizeTPointer pValue,
int _limit)
\brief Return resource limits
Returns in \p *pValue the current size of \p limit.
|
static int |
cudaDeviceGetMemPool(CUmemPoolHandle_st memPool,
int device)
\brief Gets the current mempool for a device
Returns the last pool provided to ::cudaDeviceSetMemPool for this device
or the device's default memory pool if ::cudaDeviceSetMemPool has never been called.
|
static int |
cudaDeviceGetP2PAttribute(int[] value,
int attr,
int srcDevice,
int dstDevice) |
static int |
cudaDeviceGetP2PAttribute(IntBuffer value,
int attr,
int srcDevice,
int dstDevice) |
static int |
cudaDeviceGetP2PAttribute(IntPointer value,
int attr,
int srcDevice,
int dstDevice)
\brief Queries attributes of the link between two devices.
|
static int |
cudaDeviceGetPCIBusId(byte[] pciBusId,
int len,
int device) |
static int |
cudaDeviceGetPCIBusId(ByteBuffer pciBusId,
int len,
int device) |
static int |
cudaDeviceGetPCIBusId(BytePointer pciBusId,
int len,
int device)
\brief Returns a PCI Bus Id string for the device
Returns an ASCII string identifying the device \p dev in the NULL-terminated
string pointed to by \p pciBusId.
|
static int |
cudaDeviceGetSharedMemConfig(int[] pConfig) |
static int |
cudaDeviceGetSharedMemConfig(IntBuffer pConfig) |
static int |
cudaDeviceGetSharedMemConfig(IntPointer pConfig)
\brief Returns the shared memory configuration for the current device.
|
static int |
cudaDeviceGetStreamPriorityRange(int[] leastPriority,
int[] greatestPriority) |
static int |
cudaDeviceGetStreamPriorityRange(IntBuffer leastPriority,
IntBuffer greatestPriority) |
static int |
cudaDeviceGetStreamPriorityRange(IntPointer leastPriority,
IntPointer greatestPriority)
\brief Returns numerical values that correspond to the least and
greatest stream priorities.
|
static int |
cudaDeviceGetTexture1DLinearMaxWidth(SizeTPointer maxWidthInElements,
cudaChannelFormatDesc fmtDesc,
int device)
\brief Returns the maximum number of elements allocatable in a 1D linear texture for a given element size.
|
static int |
cudaDeviceGraphMemTrim(int device)
\brief Free unused memory that was cached on the specified device for use with graphs back to the OS.
|
static int |
cudaDeviceReset()
\brief Destroy all allocations and reset all state on the current device
in the current process.
|
static int |
cudaDeviceSetCacheConfig(int cacheConfig)
\brief Sets the preferred cache configuration for the current device.
|
static int |
cudaDeviceSetGraphMemAttribute(int device,
int attr,
Pointer value)
\brief Set asynchronous allocation attributes related to graphs
Valid attributes are:
- ::cudaGraphMemAttrUsedMemHigh: High watermark of memory, in bytes, associated with graphs since the
last time it was reset.
|
static int |
cudaDeviceSetLimit(int _limit,
long value)
\brief Set resource limits
Setting \p limit to \p value is a request by the application to update
the current limit maintained by the device.
|
static int |
cudaDeviceSetMemPool(int device,
CUmemPoolHandle_st memPool)
\brief Sets the current memory pool of a device
The memory pool must be local to the specified device.
|
static int |
cudaDeviceSetSharedMemConfig(int config)
\brief Sets the shared memory configuration for the current device.
|
static int |
cudaDeviceSynchronize()
\brief Wait for compute device to finish
Blocks until the device has completed all preceding requested tasks.
|
static int |
cudaDriverGetVersion(int[] driverVersion) |
static int |
cudaDriverGetVersion(IntBuffer driverVersion) |
static int |
cudaDriverGetVersion(IntPointer driverVersion)
\brief Returns the latest version of CUDA supported by the driver
Returns in \p *driverVersion the latest version of CUDA supported by
the driver.
|
static int |
cudaEventCreate(CUevent_st event)
\brief Creates an event object
Creates an event object for the current device using ::cudaEventDefault.
|
static int |
cudaEventCreateWithFlags(CUevent_st event,
int flags)
\brief Creates an event object with the specified flags
Creates an event object for the current device with the specified flags.
|
static int |
cudaEventDestroy(CUevent_st event)
\brief Destroys an event object
Destroys the event specified by \p event.
|
static int |
cudaEventElapsedTime(float[] ms,
CUevent_st start,
CUevent_st end) |
static int |
cudaEventElapsedTime(FloatBuffer ms,
CUevent_st start,
CUevent_st end) |
static int |
cudaEventElapsedTime(FloatPointer ms,
CUevent_st start,
CUevent_st end)
\brief Computes the elapsed time between events
Computes the elapsed time between two events (in milliseconds with a
resolution of around 0.5 microseconds).
|
static int |
cudaEventQuery(CUevent_st event)
\brief Queries an event's status
Queries the status of all work currently captured by \p event.
|
static int |
cudaEventRecord(CUevent_st event) |
static int |
cudaEventRecord(CUevent_st event,
CUstream_st stream)
\brief Records an event
Captures in \p event the contents of \p stream at the time of this call.
|
static int |
cudaEventRecordWithFlags(CUevent_st event) |
static int |
cudaEventRecordWithFlags(CUevent_st event,
CUstream_st stream,
int flags)
\brief Records an event
Captures in \p event the contents of \p stream at the time of this call.
|
static int |
cudaEventSynchronize(CUevent_st event)
\brief Waits for an event to complete
Waits until the completion of all work currently captured in \p event.
|
static int |
cudaExternalMemoryGetMappedBuffer(Pointer devPtr,
CUexternalMemory_st extMem,
cudaExternalMemoryBufferDesc bufferDesc) |
static int |
cudaExternalMemoryGetMappedBuffer(PointerPointer devPtr,
CUexternalMemory_st extMem,
cudaExternalMemoryBufferDesc bufferDesc)
\brief Maps a buffer onto an imported memory object
Maps a buffer onto an imported memory object and returns a device
pointer in \p devPtr.
|
static int |
cudaExternalMemoryGetMappedMipmappedArray(cudaMipmappedArray mipmap,
CUexternalMemory_st extMem,
cudaExternalMemoryMipmappedArrayDesc mipmapDesc)
\brief Maps a CUDA mipmapped array onto an external memory object
Maps a CUDA mipmapped array onto an external object and returns a
handle to it in \p mipmap.
|
static int |
cudaFree(Pointer devPtr)
\brief Frees memory on the device
Frees the memory space pointed to by \p devPtr, which must have been
returned by a previous call to one of the following memory allocation APIs -
::cudaMalloc(), ::cudaMallocPitch(), ::cudaMallocManaged(), ::cudaMallocAsync(),
::cudaMallocFromPoolAsync().
|
static int |
cudaFreeArray(cudaArray array)
\brief Frees an array on the device
Frees the CUDA array \p array, which must have been returned by a
previous call to ::cudaMallocArray().
|
static int |
cudaFreeAsync(Pointer devPtr,
CUstream_st hStream)
\brief Frees memory with stream ordered semantics
Inserts a free operation into \p hStream.
|
static int |
cudaFreeHost(Pointer ptr)
\brief Frees page-locked memory
Frees the memory space pointed to by \p hostPtr, which must have been
returned by a previous call to ::cudaMallocHost() or ::cudaHostAlloc().
|
static int |
cudaFreeMipmappedArray(cudaMipmappedArray mipmappedArray)
\brief Frees a mipmapped array on the device
Frees the CUDA mipmapped array \p mipmappedArray, which must have been
returned by a previous call to ::cudaMallocMipmappedArray().
|
static int |
cudaFuncGetAttributes(cudaFuncAttributes attr,
Pointer func)
\brief Find out attributes for a given function
This function obtains the attributes of a function specified via \p func.
|
static int |
cudaFuncGetName(byte[] name,
Pointer func) |
static int |
cudaFuncGetName(ByteBuffer name,
Pointer func) |
static int |
cudaFuncGetName(BytePointer name,
Pointer func) |
static int |
cudaFuncGetName(PointerPointer name,
Pointer func)
\brief Returns the function name for a device entry function pointer.
|
static int |
cudaFuncSetAttribute(Pointer func,
int attr,
int value)
\brief Set attributes for a given function
This function sets the attributes of a function specified via \p func.
|
static int |
cudaFuncSetCacheConfig(Pointer func,
int cacheConfig)
\brief Sets the preferred cache configuration for a device function
On devices where the L1 cache and shared memory use the same hardware
resources, this sets through \p cacheConfig the preferred cache configuration
for the function specified via \p func.
|
static int |
cudaFuncSetSharedMemConfig(Pointer func,
int config)
\brief Sets the shared memory configuration for a device function
On devices with configurable shared memory banks, this function will
force all subsequent launches of the specified device function to have
the given shared memory bank size configuration.
|
static int |
cudaGetChannelDesc(cudaChannelFormatDesc desc,
cudaArray array)
\brief Get the channel descriptor of an array
Returns in \p *desc the channel descriptor of the CUDA array \p array.
|
static int |
cudaGetDevice(int[] device) |
static int |
cudaGetDevice(IntBuffer device) |
static int |
cudaGetDevice(IntPointer device)
\brief Returns which device is currently being used
Returns in \p *device the current device for the calling host thread.
|
static int |
cudaGetDeviceCount(int[] count) |
static int |
cudaGetDeviceCount(IntBuffer count) |
static int |
cudaGetDeviceCount(IntPointer count)
\brief Returns the number of compute-capable devices
Returns in \p *count the number of devices with compute capability greater
or equal to 2.0 that are available for execution.
|
static int |
cudaGetDeviceFlags(int[] flags) |
static int |
cudaGetDeviceFlags(IntBuffer flags) |
static int |
cudaGetDeviceFlags(IntPointer flags)
\brief Gets the flags for the current device
Returns in \p flags the flags for the current device.
|
static int |
cudaGetDeviceProperties(cudaDeviceProp prop,
int device)
\brief Returns information about the compute-device
Returns in \p *prop the properties of device \p dev.
|
static int |
cudaGetDriverEntryPoint(BytePointer symbol,
Pointer funcPtr,
long flags) |
static int |
cudaGetDriverEntryPoint(BytePointer symbol,
Pointer funcPtr,
long flags,
int[] driverStatus) |
static int |
cudaGetDriverEntryPoint(BytePointer symbol,
Pointer funcPtr,
long flags,
IntBuffer driverStatus) |
static int |
cudaGetDriverEntryPoint(BytePointer symbol,
Pointer funcPtr,
long flags,
IntPointer driverStatus) |
static int |
cudaGetDriverEntryPoint(BytePointer symbol,
PointerPointer funcPtr,
long flags,
IntPointer driverStatus)
\brief Returns the requested driver API function pointer
Returns in \p **funcPtr the address of the CUDA driver function for the requested flags.
|
static int |
cudaGetDriverEntryPoint(String symbol,
Pointer funcPtr,
long flags) |
static int |
cudaGetDriverEntryPoint(String symbol,
Pointer funcPtr,
long flags,
int[] driverStatus) |
static int |
cudaGetDriverEntryPoint(String symbol,
Pointer funcPtr,
long flags,
IntBuffer driverStatus) |
static int |
cudaGetDriverEntryPoint(String symbol,
Pointer funcPtr,
long flags,
IntPointer driverStatus) |
static BytePointer |
cudaGetErrorName(int error)
\brief Returns the string representation of an error code enum name
Returns a string containing the name of an error code in the enum.
|
static BytePointer |
cudaGetErrorString(int error)
\brief Returns the description string for an error code
Returns the description string for an error code.
|
static int |
cudaGetExportTable(Pointer ppExportTable,
cudaUUID_t pExportTableId) |
static int |
cudaGetExportTable(PointerPointer ppExportTable,
cudaUUID_t pExportTableId)
\cond impl_private
|
static int |
cudaGetFuncBySymbol(CUfunc_st functionPtr,
Pointer symbolPtr)
\brief Get pointer to device entry function that matches entry function \p symbolPtr
Returns in \p functionPtr the device entry function corresponding to the symbol \p symbolPtr.
|
static int |
cudaGetKernel(CUkern_st kernelPtr,
Pointer entryFuncAddr)
\brief Get pointer to device kernel that matches entry function \p entryFuncAddr
Returns in \p kernelPtr the device kernel corresponding to the entry function \p entryFuncAddr.
|
static int |
cudaGetLastError()
\brief Returns the last error from a runtime call
Returns the last error that has been produced by any of the runtime calls
in the same instance of the CUDA Runtime library in the host thread and
resets it to ::cudaSuccess.
|
static int |
cudaGetMipmappedArrayLevel(cudaArray levelArray,
cudaMipmappedArray mipmappedArray,
int level)
\brief Gets a mipmap level of a CUDA mipmapped array
Returns in \p *levelArray a CUDA array that represents a single mipmap level
of the CUDA mipmapped array \p mipmappedArray.
|
static int |
cudaGetSurfaceObjectResourceDesc(cudaResourceDesc pResDesc,
long surfObject)
\brief Returns a surface object's resource descriptor
Returns the resource descriptor for the surface object specified by \p surfObject.
|
static int |
cudaGetSymbolAddress(Pointer devPtr,
Pointer symbol) |
static int |
cudaGetSymbolAddress(PointerPointer devPtr,
Pointer symbol)
\brief Finds the address associated with a CUDA symbol
Returns in \p *devPtr the address of symbol \p symbol on the device.
|
static int |
cudaGetSymbolSize(SizeTPointer size,
Pointer symbol)
\brief Finds the size of the object associated with a CUDA symbol
Returns in \p *size the size of symbol \p symbol.
|
static int |
cudaGetTextureObjectResourceDesc(cudaResourceDesc pResDesc,
long texObject)
\brief Returns a texture object's resource descriptor
Returns the resource descriptor for the texture object specified by \p texObject.
|
static int |
cudaGetTextureObjectResourceViewDesc(cudaResourceViewDesc pResViewDesc,
long texObject)
\brief Returns a texture object's resource view descriptor
Returns the resource view descriptor for the texture object specified by \p texObject.
|
static int |
cudaGetTextureObjectTextureDesc(cudaTextureDesc pTexDesc,
long texObject)
\brief Returns a texture object's texture descriptor
Returns the texture descriptor for the texture object specified by \p texObject.
|
static int |
cudaGLGetDevices(int[] pCudaDeviceCount,
int[] pCudaDevices,
int cudaDeviceCount,
int deviceList) |
static int |
cudaGLGetDevices(IntBuffer pCudaDeviceCount,
IntBuffer pCudaDevices,
int cudaDeviceCount,
int deviceList) |
static int |
cudaGLGetDevices(IntPointer pCudaDeviceCount,
IntPointer pCudaDevices,
int cudaDeviceCount,
int deviceList)
\brief Gets the CUDA devices associated with the current OpenGL context
Returns in \p *pCudaDeviceCount the number of CUDA-compatible devices
corresponding to the current OpenGL context.
|
static int |
cudaGLMapBufferObject(Pointer devPtr,
int bufObj)
Deprecated.
|
static int |
cudaGLMapBufferObject(PointerPointer devPtr,
int bufObj)
Deprecated.
This function is deprecated as of CUDA 3.0.
Maps the buffer object of ID \p bufObj into the address space of
CUDA and returns in \p *devPtr the base pointer of the resulting
mapping. The buffer must have previously been registered by
calling ::cudaGLRegisterBufferObject(). While a buffer is mapped
by CUDA, any OpenGL operation which references the buffer will
result in undefined behavior. The OpenGL context used to create
the buffer, or another context from the same share group, must be
bound to the current thread when this is called.
All streams in the current thread are synchronized with the current
GL context.
|
static int |
cudaGLMapBufferObjectAsync(Pointer devPtr,
int bufObj,
CUstream_st stream)
Deprecated.
|
static int |
cudaGLMapBufferObjectAsync(PointerPointer devPtr,
int bufObj,
CUstream_st stream)
Deprecated.
This function is deprecated as of CUDA 3.0.
Maps the buffer object of ID \p bufObj into the address space of
CUDA and returns in \p *devPtr the base pointer of the resulting
mapping. The buffer must have previously been registered by
calling ::cudaGLRegisterBufferObject(). While a buffer is mapped
by CUDA, any OpenGL operation which references the buffer will
result in undefined behavior. The OpenGL context used to create
the buffer, or another context from the same share group, must be
bound to the current thread when this is called.
Stream /p stream is synchronized with the current GL context.
|
static int |
cudaGLRegisterBufferObject(int bufObj)
Deprecated.
This function is deprecated as of CUDA 3.0.
Registers the buffer object of ID \p bufObj for access by
CUDA. This function must be called before CUDA can map the buffer
object. The OpenGL context used to create the buffer, or another
context from the same share group, must be bound to the current
thread when this is called.
|
static int |
cudaGLSetBufferObjectMapFlags(int bufObj,
int flags)
Deprecated.
This function is deprecated as of CUDA 3.0.
Set flags for mapping the OpenGL buffer \p bufObj
Changes to flags will take effect the next time \p bufObj is mapped.
The \p flags argument may be any of the following:
- ::cudaGLMapFlagsNone: Specifies no hints about how this buffer will
be used. It is therefore assumed that this buffer will be read from and
written to by CUDA kernels. This is the default value.
- ::cudaGLMapFlagsReadOnly: Specifies that CUDA kernels which access this
buffer will not write to the buffer.
- ::cudaGLMapFlagsWriteDiscard: Specifies that CUDA kernels which access
this buffer will not read from the buffer and will write over the
entire contents of the buffer, so none of the data previously stored in
the buffer will be preserved.
If \p bufObj has not been registered for use with CUDA, then
::cudaErrorInvalidResourceHandle is returned. If \p bufObj is presently
mapped for access by CUDA, then ::cudaErrorUnknown is returned.
|
static int |
cudaGLSetGLDevice(int device)
Deprecated.
This function is deprecated as of CUDA 5.0.
This function is deprecated and should no longer be used. It is
no longer necessary to associate a CUDA device with an OpenGL
context in order to achieve maximum interoperability performance.
This function will immediately initialize the primary context on
\p device if needed.
|
static int |
cudaGLUnmapBufferObject(int bufObj)
Deprecated.
This function is deprecated as of CUDA 3.0.
Unmaps the buffer object of ID \p bufObj for access by CUDA. When
a buffer is unmapped, the base address returned by
::cudaGLMapBufferObject() is invalid and subsequent references to
the address result in undefined behavior. The OpenGL context used
to create the buffer, or another context from the same share group,
must be bound to the current thread when this is called.
All streams in the current thread are synchronized with the current
GL context.
|
static int |
cudaGLUnmapBufferObjectAsync(int bufObj,
CUstream_st stream)
Deprecated.
This function is deprecated as of CUDA 3.0.
Unmaps the buffer object of ID \p bufObj for access by CUDA. When
a buffer is unmapped, the base address returned by
::cudaGLMapBufferObject() is invalid and subsequent references to
the address result in undefined behavior. The OpenGL context used
to create the buffer, or another context from the same share group,
must be bound to the current thread when this is called.
Stream /p stream is synchronized with the current GL context.
|
static int |
cudaGLUnregisterBufferObject(int bufObj)
Deprecated.
This function is deprecated as of CUDA 3.0.
Unregisters the buffer object of ID \p bufObj for access by CUDA
and releases any CUDA resources associated with the buffer. Once a
buffer is unregistered, it may no longer be mapped by CUDA. The GL
context used to create the buffer, or another context from the
same share group, must be bound to the current thread when this is
called.
|
static int |
cudaGraphAddChildGraphNode(CUgraphNode_st pGraphNode,
CUgraph_st graph,
CUgraphNode_st pDependencies,
long numDependencies,
CUgraph_st childGraph)
\brief Creates a child graph node and adds it to a graph
Creates a new node which executes an embedded graph, and adds it to \p graph with
\p numDependencies dependencies specified via \p pDependencies.
|
static int |
cudaGraphAddDependencies_v2(CUgraph_st graph,
CUgraphNode_st from,
CUgraphNode_st to,
cudaGraphEdgeData edgeData,
long numDependencies)
\brief Adds dependency edges to a graph.
|
static int |
cudaGraphAddDependencies(CUgraph_st graph,
CUgraphNode_st from,
CUgraphNode_st to,
long numDependencies)
\brief Adds dependency edges to a graph.
|
static int |
cudaGraphAddEmptyNode(CUgraphNode_st pGraphNode,
CUgraph_st graph,
CUgraphNode_st pDependencies,
long numDependencies)
\brief Creates an empty node and adds it to a graph
Creates a new node which performs no operation, and adds it to \p graph with
\p numDependencies dependencies specified via \p pDependencies.
|
static int |
cudaGraphAddEventRecordNode(CUgraphNode_st pGraphNode,
CUgraph_st graph,
CUgraphNode_st pDependencies,
long numDependencies,
CUevent_st event)
\brief Creates an event record node and adds it to a graph
Creates a new event record node and adds it to \p hGraph with \p numDependencies
dependencies specified via \p dependencies and event specified in \p event.
|
static int |
cudaGraphAddEventWaitNode(CUgraphNode_st pGraphNode,
CUgraph_st graph,
CUgraphNode_st pDependencies,
long numDependencies,
CUevent_st event)
\brief Creates an event wait node and adds it to a graph
Creates a new event wait node and adds it to \p hGraph with \p numDependencies
dependencies specified via \p dependencies and event specified in \p event.
|
static int |
cudaGraphAddExternalSemaphoresSignalNode(CUgraphNode_st pGraphNode,
CUgraph_st graph,
CUgraphNode_st pDependencies,
long numDependencies,
cudaExternalSemaphoreSignalNodeParams nodeParams)
\brief Creates an external semaphore signal node and adds it to a graph
Creates a new external semaphore signal node and adds it to \p graph with \p
numDependencies dependencies specified via \p dependencies and arguments specified
in \p nodeParams.
|
static int |
cudaGraphAddExternalSemaphoresWaitNode(CUgraphNode_st pGraphNode,
CUgraph_st graph,
CUgraphNode_st pDependencies,
long numDependencies,
cudaExternalSemaphoreWaitNodeParams nodeParams)
\brief Creates an external semaphore wait node and adds it to a graph
Creates a new external semaphore wait node and adds it to \p graph with \p numDependencies
dependencies specified via \p dependencies and arguments specified in \p nodeParams.
|
static int |
cudaGraphAddHostNode(CUgraphNode_st pGraphNode,
CUgraph_st graph,
CUgraphNode_st pDependencies,
long numDependencies,
cudaHostNodeParams pNodeParams)
\brief Creates a host execution node and adds it to a graph
Creates a new CPU execution node and adds it to \p graph with \p numDependencies
dependencies specified via \p pDependencies and arguments specified in \p pNodeParams.
|
static int |
cudaGraphAddKernelNode(CUgraphNode_st pGraphNode,
CUgraph_st graph,
CUgraphNode_st pDependencies,
long numDependencies,
cudaKernelNodeParams pNodeParams)
\brief Creates a kernel execution node and adds it to a graph
Creates a new kernel execution node and adds it to \p graph with \p numDependencies
dependencies specified via \p pDependencies and arguments specified in \p pNodeParams.
|
static int |
cudaGraphAddMemAllocNode(CUgraphNode_st pGraphNode,
CUgraph_st graph,
CUgraphNode_st pDependencies,
long numDependencies,
cudaMemAllocNodeParams nodeParams)
\brief Creates an allocation node and adds it to a graph
Creates a new allocation node and adds it to \p graph with \p numDependencies
dependencies specified via \p pDependencies and arguments specified in \p nodeParams.
|
static int |
cudaGraphAddMemcpyNode(CUgraphNode_st pGraphNode,
CUgraph_st graph,
CUgraphNode_st pDependencies,
long numDependencies,
cudaMemcpy3DParms pCopyParams)
\brief Creates a memcpy node and adds it to a graph
Creates a new memcpy node and adds it to \p graph with \p numDependencies
dependencies specified via \p pDependencies.
|
static int |
cudaGraphAddMemcpyNode1D(CUgraphNode_st pGraphNode,
CUgraph_st graph,
CUgraphNode_st pDependencies,
long numDependencies,
Pointer dst,
Pointer src,
long count,
int kind)
\brief Creates a 1D memcpy node and adds it to a graph
Creates a new 1D memcpy node and adds it to \p graph with \p numDependencies
dependencies specified via \p pDependencies.
|
static int |
cudaGraphAddMemcpyNodeFromSymbol(CUgraphNode_st pGraphNode,
CUgraph_st graph,
CUgraphNode_st pDependencies,
long numDependencies,
Pointer dst,
Pointer symbol,
long count,
long offset,
int kind)
\brief Creates a memcpy node to copy from a symbol on the device and adds it to a graph
Creates a new memcpy node to copy from \p symbol and adds it to \p graph with
\p numDependencies dependencies specified via \p pDependencies.
|
static int |
cudaGraphAddMemcpyNodeToSymbol(CUgraphNode_st pGraphNode,
CUgraph_st graph,
CUgraphNode_st pDependencies,
long numDependencies,
Pointer symbol,
Pointer src,
long count,
long offset,
int kind)
\brief Creates a memcpy node to copy to a symbol on the device and adds it to a graph
Creates a new memcpy node to copy to \p symbol and adds it to \p graph with
\p numDependencies dependencies specified via \p pDependencies.
|
static int |
cudaGraphAddMemFreeNode(CUgraphNode_st pGraphNode,
CUgraph_st graph,
CUgraphNode_st pDependencies,
long numDependencies,
Pointer dptr)
\brief Creates a memory free node and adds it to a graph
Creates a new memory free node and adds it to \p graph with \p numDependencies
dependencies specified via \p pDependencies and address specified in \p dptr.
|
static int |
cudaGraphAddMemsetNode(CUgraphNode_st pGraphNode,
CUgraph_st graph,
CUgraphNode_st pDependencies,
long numDependencies,
cudaMemsetParams pMemsetParams)
\brief Creates a memset node and adds it to a graph
Creates a new memset node and adds it to \p graph with \p numDependencies
dependencies specified via \p pDependencies.
|
static int |
cudaGraphAddNode_v2(CUgraphNode_st pGraphNode,
CUgraph_st graph,
CUgraphNode_st pDependencies,
cudaGraphEdgeData dependencyData,
long numDependencies,
cudaGraphNodeParams nodeParams)
\brief Adds a node of arbitrary type to a graph (12.3+)
Creates a new node in \p graph described by \p nodeParams with \p numDependencies
dependencies specified via \p pDependencies.
|
static int |
cudaGraphAddNode(CUgraphNode_st pGraphNode,
CUgraph_st graph,
CUgraphNode_st pDependencies,
long numDependencies,
cudaGraphNodeParams nodeParams)
\brief Adds a node of arbitrary type to a graph
Creates a new node in \p graph described by \p nodeParams with \p numDependencies
dependencies specified via \p pDependencies.
|
static int |
cudaGraphChildGraphNodeGetGraph(CUgraphNode_st node,
CUgraph_st pGraph)
\brief Gets a handle to the embedded graph of a child graph node
Gets a handle to the embedded graph in a child graph node.
|
static int |
cudaGraphClone(CUgraph_st pGraphClone,
CUgraph_st originalGraph)
\brief Clones a graph
This function creates a copy of \p originalGraph and returns it in \p pGraphClone.
|
static int |
cudaGraphConditionalHandleCreate(long[] pHandle_out,
CUgraph_st graph) |
static int |
cudaGraphConditionalHandleCreate(long[] pHandle_out,
CUgraph_st graph,
int defaultLaunchValue,
int flags) |
static int |
cudaGraphConditionalHandleCreate(LongBuffer pHandle_out,
CUgraph_st graph) |
static int |
cudaGraphConditionalHandleCreate(LongBuffer pHandle_out,
CUgraph_st graph,
int defaultLaunchValue,
int flags) |
static int |
cudaGraphConditionalHandleCreate(LongPointer pHandle_out,
CUgraph_st graph) |
static int |
cudaGraphConditionalHandleCreate(LongPointer pHandle_out,
CUgraph_st graph,
int defaultLaunchValue,
int flags)
\brief Create a conditional handle
Creates a conditional handle associated with \p hGraph.
|
static int |
cudaGraphCreate(CUgraph_st pGraph,
int flags)
\brief Creates a graph
Creates an empty graph, which is returned via \p pGraph.
|
static int |
cudaGraphDebugDotPrint(CUgraph_st graph,
BytePointer path,
int flags)
\brief Write a DOT file describing graph structure
Using the provided \p graph, write to \p path a DOT formatted description of the graph.
|
static int |
cudaGraphDebugDotPrint(CUgraph_st graph,
String path,
int flags) |
static int |
cudaGraphDestroy(CUgraph_st graph)
\brief Destroys a graph
Destroys the graph specified by \p graph, as well as all of its nodes.
|
static int |
cudaGraphDestroyNode(CUgraphNode_st node)
\brief Remove a node from the graph
Removes \p node from its graph.
|
static int |
cudaGraphEventRecordNodeGetEvent(CUgraphNode_st node,
CUevent_st event_out)
\brief Returns the event associated with an event record node
Returns the event of event record node \p hNode in \p event_out.
|
static int |
cudaGraphEventRecordNodeSetEvent(CUgraphNode_st node,
CUevent_st event)
\brief Sets an event record node's event
Sets the event of event record node \p hNode to \p event.
|
static int |
cudaGraphEventWaitNodeGetEvent(CUgraphNode_st node,
CUevent_st event_out)
\brief Returns the event associated with an event wait node
Returns the event of event wait node \p hNode in \p event_out.
|
static int |
cudaGraphEventWaitNodeSetEvent(CUgraphNode_st node,
CUevent_st event)
\brief Sets an event wait node's event
Sets the event of event wait node \p hNode to \p event.
|
static int |
cudaGraphExecChildGraphNodeSetParams(CUgraphExec_st hGraphExec,
CUgraphNode_st node,
CUgraph_st childGraph)
\brief Updates node parameters in the child graph node in the given graphExec.
|
static int |
cudaGraphExecDestroy(CUgraphExec_st graphExec)
\brief Destroys an executable graph
Destroys the executable graph specified by \p graphExec.
|
static int |
cudaGraphExecEventRecordNodeSetEvent(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
CUevent_st event)
\brief Sets the event for an event record node in the given graphExec
Sets the event of an event record node in an executable graph \p hGraphExec.
|
static int |
cudaGraphExecEventWaitNodeSetEvent(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
CUevent_st event)
\brief Sets the event for an event wait node in the given graphExec
Sets the event of an event wait node in an executable graph \p hGraphExec.
|
static int |
cudaGraphExecExternalSemaphoresSignalNodeSetParams(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
cudaExternalSemaphoreSignalNodeParams nodeParams)
\brief Sets the parameters for an external semaphore signal node in the given graphExec
Sets the parameters of an external semaphore signal node in an executable graph \p hGraphExec.
|
static int |
cudaGraphExecExternalSemaphoresWaitNodeSetParams(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
cudaExternalSemaphoreWaitNodeParams nodeParams)
\brief Sets the parameters for an external semaphore wait node in the given graphExec
Sets the parameters of an external semaphore wait node in an executable graph \p hGraphExec.
|
static int |
cudaGraphExecGetFlags(CUgraphExec_st graphExec,
long[] flags) |
static int |
cudaGraphExecGetFlags(CUgraphExec_st graphExec,
LongBuffer flags) |
static int |
cudaGraphExecGetFlags(CUgraphExec_st graphExec,
LongPointer flags)
\brief Query the instantiation flags of an executable graph
Returns the flags that were passed to instantiation for the given executable graph.
|
static int |
cudaGraphExecHostNodeSetParams(CUgraphExec_st hGraphExec,
CUgraphNode_st node,
cudaHostNodeParams pNodeParams)
\brief Sets the parameters for a host node in the given graphExec.
|
static int |
cudaGraphExecKernelNodeSetParams(CUgraphExec_st hGraphExec,
CUgraphNode_st node,
cudaKernelNodeParams pNodeParams)
\brief Sets the parameters for a kernel node in the given graphExec
Sets the parameters of a kernel node in an executable graph \p hGraphExec.
|
static int |
cudaGraphExecMemcpyNodeSetParams(CUgraphExec_st hGraphExec,
CUgraphNode_st node,
cudaMemcpy3DParms pNodeParams)
\brief Sets the parameters for a memcpy node in the given graphExec.
|
static int |
cudaGraphExecMemcpyNodeSetParams1D(CUgraphExec_st hGraphExec,
CUgraphNode_st node,
Pointer dst,
Pointer src,
long count,
int kind)
\brief Sets the parameters for a memcpy node in the given graphExec to perform a 1-dimensional copy
Updates the work represented by \p node in \p hGraphExec as though \p node had
contained the given params at instantiation.
|
static int |
cudaGraphExecMemcpyNodeSetParamsFromSymbol(CUgraphExec_st hGraphExec,
CUgraphNode_st node,
Pointer dst,
Pointer symbol,
long count,
long offset,
int kind)
\brief Sets the parameters for a memcpy node in the given graphExec to copy from a symbol on the device
Updates the work represented by \p node in \p hGraphExec as though \p node had
contained the given params at instantiation.
|
static int |
cudaGraphExecMemcpyNodeSetParamsToSymbol(CUgraphExec_st hGraphExec,
CUgraphNode_st node,
Pointer symbol,
Pointer src,
long count,
long offset,
int kind)
\brief Sets the parameters for a memcpy node in the given graphExec to copy to a symbol on the device
Updates the work represented by \p node in \p hGraphExec as though \p node had
contained the given params at instantiation.
|
static int |
cudaGraphExecMemsetNodeSetParams(CUgraphExec_st hGraphExec,
CUgraphNode_st node,
cudaMemsetParams pNodeParams)
\brief Sets the parameters for a memset node in the given graphExec.
|
static int |
cudaGraphExecNodeSetParams(CUgraphExec_st graphExec,
CUgraphNode_st node,
cudaGraphNodeParams nodeParams)
\brief Update's a graph node's parameters in an instantiated graph
Sets the parameters of a node in an executable graph \p graphExec.
|
static int |
cudaGraphExecUpdate(CUgraphExec_st hGraphExec,
CUgraph_st hGraph,
cudaGraphExecUpdateResultInfo resultInfo)
\brief Check whether an executable graph can be updated with a graph and perform the update if possible
Updates the node parameters in the instantiated graph specified by \p hGraphExec with the
node parameters in a topologically identical graph specified by \p hGraph.
|
static int |
cudaGraphExternalSemaphoresSignalNodeGetParams(CUgraphNode_st hNode,
cudaExternalSemaphoreSignalNodeParams params_out)
\brief Returns an external semaphore signal node's parameters
Returns the parameters of an external semaphore signal node \p hNode in \p params_out.
|
static int |
cudaGraphExternalSemaphoresSignalNodeSetParams(CUgraphNode_st hNode,
cudaExternalSemaphoreSignalNodeParams nodeParams)
\brief Sets an external semaphore signal node's parameters
Sets the parameters of an external semaphore signal node \p hNode to \p nodeParams.
|
static int |
cudaGraphExternalSemaphoresWaitNodeGetParams(CUgraphNode_st hNode,
cudaExternalSemaphoreWaitNodeParams params_out)
\brief Returns an external semaphore wait node's parameters
Returns the parameters of an external semaphore wait node \p hNode in \p params_out.
|
static int |
cudaGraphExternalSemaphoresWaitNodeSetParams(CUgraphNode_st hNode,
cudaExternalSemaphoreWaitNodeParams nodeParams)
\brief Sets an external semaphore wait node's parameters
Sets the parameters of an external semaphore wait node \p hNode to \p nodeParams.
|
static int |
cudaGraphGetEdges_v2(CUgraph_st graph,
CUgraphNode_st from,
CUgraphNode_st to,
cudaGraphEdgeData edgeData,
SizeTPointer numEdges)
\brief Returns a graph's dependency edges (12.3+)
Returns a list of \p graph's dependency edges.
|
static int |
cudaGraphGetEdges(CUgraph_st graph,
CUgraphNode_st from,
CUgraphNode_st to,
SizeTPointer numEdges)
\brief Returns a graph's dependency edges
Returns a list of \p graph's dependency edges.
|
static int |
cudaGraphGetNodes(CUgraph_st graph,
CUgraphNode_st nodes,
SizeTPointer numNodes)
\brief Returns a graph's nodes
Returns a list of \p graph's nodes.
|
static int |
cudaGraphGetRootNodes(CUgraph_st graph,
CUgraphNode_st pRootNodes,
SizeTPointer pNumRootNodes)
\brief Returns a graph's root nodes
Returns a list of \p graph's root nodes.
|
static int |
cudaGraphHostNodeGetParams(CUgraphNode_st node,
cudaHostNodeParams pNodeParams)
\brief Returns a host node's parameters
Returns the parameters of host node \p node in \p pNodeParams.
|
static int |
cudaGraphHostNodeSetParams(CUgraphNode_st node,
cudaHostNodeParams pNodeParams)
\brief Sets a host node's parameters
Sets the parameters of host node \p node to \p nodeParams.
|
static int |
cudaGraphicsGLRegisterBuffer(cudaGraphicsResource resource,
int buffer,
int flags) |
static int |
cudaGraphicsGLRegisterBuffer(PointerPointer resource,
int buffer,
int flags)
\brief Registers an OpenGL buffer object
Registers the buffer object specified by \p buffer for access by
CUDA.
|
static int |
cudaGraphicsGLRegisterImage(cudaGraphicsResource resource,
int image,
int target,
int flags) |
static int |
cudaGraphicsGLRegisterImage(PointerPointer resource,
int image,
int target,
int flags)
\brief Register an OpenGL texture or renderbuffer object
Registers the texture or renderbuffer object specified by \p image for access by CUDA.
|
static int |
cudaGraphicsMapResources(int count,
cudaGraphicsResource resources) |
static int |
cudaGraphicsMapResources(int count,
cudaGraphicsResource resources,
CUstream_st stream)
\brief Map graphics resources for access by CUDA
Maps the \p count graphics resources in \p resources for access by CUDA.
|
static int |
cudaGraphicsResourceGetMappedMipmappedArray(cudaMipmappedArray mipmappedArray,
cudaGraphicsResource resource)
\brief Get a mipmapped array through which to access a mapped graphics resource.
|
static int |
cudaGraphicsResourceGetMappedPointer(PointerPointer devPtr,
SizeTPointer size,
cudaGraphicsResource resource)
\brief Get an device pointer through which to access a mapped graphics resource.
|
static int |
cudaGraphicsResourceGetMappedPointer(Pointer devPtr,
SizeTPointer size,
cudaGraphicsResource resource) |
static int |
cudaGraphicsResourceSetMapFlags(cudaGraphicsResource resource,
int flags)
\brief Set usage flags for mapping a graphics resource
Set \p flags for mapping the graphics resource \p resource.
|
static int |
cudaGraphicsSubResourceGetMappedArray(cudaArray array,
cudaGraphicsResource resource,
int arrayIndex,
int mipLevel)
\brief Get an array through which to access a subresource of a mapped graphics resource.
|
static int |
cudaGraphicsUnmapResources(int count,
cudaGraphicsResource resources) |
static int |
cudaGraphicsUnmapResources(int count,
cudaGraphicsResource resources,
CUstream_st stream)
\brief Unmap graphics resources.
|
static int |
cudaGraphicsUnregisterResource(cudaGraphicsResource resource)
\brief Unregisters a graphics resource for access by CUDA
Unregisters the graphics resource \p resource so it is not accessible by
CUDA unless registered again.
|
static int |
cudaGraphInstantiate(CUgraphExec_st pGraphExec,
CUgraph_st graph) |
static int |
cudaGraphInstantiate(CUgraphExec_st pGraphExec,
CUgraph_st graph,
long flags)
\brief Creates an executable graph from a graph
Instantiates \p graph as an executable graph.
|
static int |
cudaGraphInstantiateWithFlags(CUgraphExec_st pGraphExec,
CUgraph_st graph) |
static int |
cudaGraphInstantiateWithFlags(CUgraphExec_st pGraphExec,
CUgraph_st graph,
long flags)
\brief Creates an executable graph from a graph
Instantiates \p graph as an executable graph.
|
static int |
cudaGraphInstantiateWithParams(CUgraphExec_st pGraphExec,
CUgraph_st graph,
cudaGraphInstantiateParams instantiateParams)
\brief Creates an executable graph from a graph
Instantiates \p graph as an executable graph according to the \p instantiateParams structure.
|
static int |
cudaGraphKernelNodeCopyAttributes(CUgraphNode_st hSrc,
CUgraphNode_st hDst)
\brief Copies attributes from source node to destination node.
|
static int |
cudaGraphKernelNodeGetAttribute(CUgraphNode_st hNode,
int attr,
cudaLaunchAttributeValue value_out)
\brief Queries node attribute.
|
static int |
cudaGraphKernelNodeGetParams(CUgraphNode_st node,
cudaKernelNodeParams pNodeParams)
\brief Returns a kernel node's parameters
Returns the parameters of kernel node \p node in \p pNodeParams.
|
static int |
cudaGraphKernelNodeSetAttribute(CUgraphNode_st hNode,
int attr,
cudaLaunchAttributeValue value)
\brief Sets node attribute.
|
static int |
cudaGraphKernelNodeSetParams(CUgraphNode_st node,
cudaKernelNodeParams pNodeParams)
\brief Sets a kernel node's parameters
Sets the parameters of kernel node \p node to \p pNodeParams.
|
static int |
cudaGraphLaunch(CUgraphExec_st graphExec,
CUstream_st stream)
\brief Launches an executable graph in a stream
Executes \p graphExec in \p stream.
|
static int |
cudaGraphMemAllocNodeGetParams(CUgraphNode_st node,
cudaMemAllocNodeParams params_out)
\brief Returns a memory alloc node's parameters
Returns the parameters of a memory alloc node \p hNode in \p params_out.
|
static int |
cudaGraphMemcpyNodeGetParams(CUgraphNode_st node,
cudaMemcpy3DParms pNodeParams)
\brief Returns a memcpy node's parameters
Returns the parameters of memcpy node \p node in \p pNodeParams.
|
static int |
cudaGraphMemcpyNodeSetParams(CUgraphNode_st node,
cudaMemcpy3DParms pNodeParams)
\brief Sets a memcpy node's parameters
Sets the parameters of memcpy node \p node to \p pNodeParams.
|
static int |
cudaGraphMemcpyNodeSetParams1D(CUgraphNode_st node,
Pointer dst,
Pointer src,
long count,
int kind)
\brief Sets a memcpy node's parameters to perform a 1-dimensional copy
Sets the parameters of memcpy node \p node to the copy described by the provided parameters.
|
static int |
cudaGraphMemcpyNodeSetParamsFromSymbol(CUgraphNode_st node,
Pointer dst,
Pointer symbol,
long count,
long offset,
int kind)
\brief Sets a memcpy node's parameters to copy from a symbol on the device
Sets the parameters of memcpy node \p node to the copy described by the provided parameters.
|
static int |
cudaGraphMemcpyNodeSetParamsToSymbol(CUgraphNode_st node,
Pointer symbol,
Pointer src,
long count,
long offset,
int kind)
\brief Sets a memcpy node's parameters to copy to a symbol on the device
Sets the parameters of memcpy node \p node to the copy described by the provided parameters.
|
static int |
cudaGraphMemFreeNodeGetParams(CUgraphNode_st node,
Pointer dptr_out)
\brief Returns a memory free node's parameters
Returns the address of a memory free node \p hNode in \p dptr_out.
|
static int |
cudaGraphMemsetNodeGetParams(CUgraphNode_st node,
cudaMemsetParams pNodeParams)
\brief Returns a memset node's parameters
Returns the parameters of memset node \p node in \p pNodeParams.
|
static int |
cudaGraphMemsetNodeSetParams(CUgraphNode_st node,
cudaMemsetParams pNodeParams)
\brief Sets a memset node's parameters
Sets the parameters of memset node \p node to \p pNodeParams.
|
static int |
cudaGraphNodeFindInClone(CUgraphNode_st pNode,
CUgraphNode_st originalNode,
CUgraph_st clonedGraph)
\brief Finds a cloned version of a node
This function returns the node in \p clonedGraph corresponding to \p originalNode
in the original graph.
|
static int |
cudaGraphNodeGetDependencies_v2(CUgraphNode_st node,
CUgraphNode_st pDependencies,
cudaGraphEdgeData edgeData,
SizeTPointer pNumDependencies)
\brief Returns a node's dependencies (12.3+)
Returns a list of \p node's dependencies.
|
static int |
cudaGraphNodeGetDependencies(CUgraphNode_st node,
CUgraphNode_st pDependencies,
SizeTPointer pNumDependencies)
\brief Returns a node's dependencies
Returns a list of \p node's dependencies.
|
static int |
cudaGraphNodeGetDependentNodes_v2(CUgraphNode_st node,
CUgraphNode_st pDependentNodes,
cudaGraphEdgeData edgeData,
SizeTPointer pNumDependentNodes)
\brief Returns a node's dependent nodes (12.3+)
Returns a list of \p node's dependent nodes.
|
static int |
cudaGraphNodeGetDependentNodes(CUgraphNode_st node,
CUgraphNode_st pDependentNodes,
SizeTPointer pNumDependentNodes)
\brief Returns a node's dependent nodes
Returns a list of \p node's dependent nodes.
|
static int |
cudaGraphNodeGetEnabled(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
int[] isEnabled) |
static int |
cudaGraphNodeGetEnabled(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
IntBuffer isEnabled) |
static int |
cudaGraphNodeGetEnabled(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
IntPointer isEnabled)
\brief Query whether a node in the given graphExec is enabled
Sets isEnabled to 1 if \p hNode is enabled, or 0 if \p hNode is disabled.
|
static int |
cudaGraphNodeGetType(CUgraphNode_st node,
int[] pType) |
static int |
cudaGraphNodeGetType(CUgraphNode_st node,
IntBuffer pType) |
static int |
cudaGraphNodeGetType(CUgraphNode_st node,
IntPointer pType)
\brief Returns a node's type
Returns the node type of \p node in \p pType.
|
static int |
cudaGraphNodeSetEnabled(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
int isEnabled)
\brief Enables or disables the specified node in the given graphExec
Sets \p hNode to be either enabled or disabled.
|
static int |
cudaGraphNodeSetParams(CUgraphNode_st node,
cudaGraphNodeParams nodeParams)
\brief Update's a graph node's parameters
Sets the parameters of graph node \p node to \p nodeParams.
|
static int |
cudaGraphReleaseUserObject(CUgraph_st graph,
CUuserObject_st object) |
static int |
cudaGraphReleaseUserObject(CUgraph_st graph,
CUuserObject_st object,
int count)
\brief Release a user object reference from a graph
Releases user object references owned by a graph.
|
static int |
cudaGraphRemoveDependencies_v2(CUgraph_st graph,
CUgraphNode_st from,
CUgraphNode_st to,
cudaGraphEdgeData edgeData,
long numDependencies)
\brief Removes dependency edges from a graph.
|
static int |
cudaGraphRemoveDependencies(CUgraph_st graph,
CUgraphNode_st from,
CUgraphNode_st to,
long numDependencies)
\brief Removes dependency edges from a graph.
|
static int |
cudaGraphRetainUserObject(CUgraph_st graph,
CUuserObject_st object) |
static int |
cudaGraphRetainUserObject(CUgraph_st graph,
CUuserObject_st object,
int count,
int flags)
\brief Retain a reference to a user object from a graph
Creates or moves user object references that will be owned by a CUDA graph.
|
static int |
cudaGraphUpload(CUgraphExec_st graphExec,
CUstream_st stream)
\brief Uploads an executable graph in a stream
Uploads \p hGraphExec to the device in \p hStream without executing it.
|
static int |
cudaHostAlloc(Pointer pHost,
long size,
int flags) |
static int |
cudaHostAlloc(PointerPointer pHost,
long size,
int flags)
\brief Allocates page-locked memory on the host
Allocates \p size bytes of host memory that is page-locked and accessible
to the device.
|
static int |
cudaHostGetDevicePointer(Pointer pDevice,
Pointer pHost,
int flags) |
static int |
cudaHostGetDevicePointer(PointerPointer pDevice,
Pointer pHost,
int flags)
\brief Passes back device pointer of mapped host memory allocated by
cudaHostAlloc or registered by cudaHostRegister
Passes back the device pointer corresponding to the mapped, pinned host
buffer allocated by ::cudaHostAlloc() or registered by ::cudaHostRegister().
|
static int |
cudaHostGetFlags(int[] pFlags,
Pointer pHost) |
static int |
cudaHostGetFlags(IntBuffer pFlags,
Pointer pHost) |
static int |
cudaHostGetFlags(IntPointer pFlags,
Pointer pHost)
\brief Passes back flags used to allocate pinned host memory allocated by
cudaHostAlloc
::cudaHostGetFlags() will fail if the input pointer does not
reside in an address range allocated by ::cudaHostAlloc().
|
static int |
cudaHostRegister(Pointer ptr,
long size,
int flags)
\brief Registers an existing host memory range for use by CUDA
Page-locks the memory range specified by \p ptr and \p size and maps it
for the device(s) as specified by \p flags.
|
static int |
cudaHostUnregister(Pointer ptr)
\brief Unregisters a memory range that was registered with cudaHostRegister
Unmaps the memory range whose base address is specified by \p ptr, and makes
it pageable again.
|
static int |
cudaImportExternalMemory(CUexternalMemory_st extMem_out,
cudaExternalMemoryHandleDesc memHandleDesc)
\brief Imports an external memory object
Imports an externally allocated memory object and returns
a handle to that in \p extMem_out.
|
static int |
cudaImportExternalSemaphore(CUexternalSemaphore_st extSem_out,
cudaExternalSemaphoreHandleDesc semHandleDesc)
\brief Imports an external semaphore
Imports an externally allocated synchronization object and returns
a handle to that in \p extSem_out.
|
static int |
cudaInitDevice(int device,
int deviceFlags,
int flags)
\brief Initialize device to be used for GPU executions
This function will initialize the CUDA Runtime structures and primary context on \p device when called,
but the context will not be made current to \p device.
|
static int |
cudaIpcCloseMemHandle(Pointer devPtr)
\brief Attempts to close memory mapped with cudaIpcOpenMemHandle
Decrements the reference count of the memory returnd by ::cudaIpcOpenMemHandle by 1.
|
static int |
cudaIpcGetEventHandle(cudaIpcEventHandle_t handle,
CUevent_st event)
\brief Gets an interprocess handle for a previously allocated event
Takes as input a previously allocated event.
|
static int |
cudaIpcGetMemHandle(cudaIpcMemHandle_t handle,
Pointer devPtr)
\brief Gets an interprocess memory handle for an existing device memory
allocation
Takes a pointer to the base of an existing device memory allocation created
with ::cudaMalloc and exports it for use in another process.
|
static int |
cudaIpcOpenEventHandle(CUevent_st event,
cudaIpcEventHandle_t handle)
\brief Opens an interprocess event handle for use in the current process
Opens an interprocess event handle exported from another process with
::cudaIpcGetEventHandle.
|
static int |
cudaIpcOpenMemHandle(Pointer devPtr,
cudaIpcMemHandle_t handle,
int flags) |
static int |
cudaIpcOpenMemHandle(PointerPointer devPtr,
cudaIpcMemHandle_t handle,
int flags)
\brief Opens an interprocess memory handle exported from another process
and returns a device pointer usable in the local process.
|
static int |
cudaLaunchCooperativeKernel(Pointer func,
dim3 gridDim,
dim3 blockDim,
Pointer args,
long sharedMem,
CUstream_st stream) |
static int |
cudaLaunchCooperativeKernel(Pointer func,
dim3 gridDim,
dim3 blockDim,
PointerPointer args,
long sharedMem,
CUstream_st stream)
\brief Launches a device function where thread blocks can cooperate and synchronize as they execute
The function invokes kernel \p func on \p gridDim (\p gridDim.x × \p gridDim.y
× \p gridDim.z) grid of blocks.
|
static int |
cudaLaunchCooperativeKernelMultiDevice(cudaLaunchParams launchParamsList,
int numDevices)
Deprecated.
|
static int |
cudaLaunchCooperativeKernelMultiDevice(cudaLaunchParams launchParamsList,
int numDevices,
int flags)
Deprecated.
This function is deprecated as of CUDA 11.3.
Invokes kernels as specified in the \p launchParamsList array where each element
of the array specifies all the parameters required to perform a single kernel launch.
These kernels can cooperate and synchronize as they execute. The size of the array is
specified by \p numDevices.
No two kernels can be launched on the same device. All the devices targeted by this
multi-device launch must be identical. All devices must have a non-zero value for the
device attribute ::cudaDevAttrCooperativeMultiDeviceLaunch.
The same kernel must be launched on all devices. Note that any __device__ or __constant__
variables are independently instantiated on every device. It is the application's
responsiblity to ensure these variables are initialized and used appropriately.
The size of the grids as specified in blocks, the size of the blocks themselves and the
amount of shared memory used by each thread block must also match across all launched kernels.
The streams used to launch these kernels must have been created via either ::cudaStreamCreate
or ::cudaStreamCreateWithPriority or ::cudaStreamCreateWithPriority. The NULL stream or
::cudaStreamLegacy or ::cudaStreamPerThread cannot be used.
The total number of blocks launched per kernel cannot exceed the maximum number of blocks
per multiprocessor as returned by ::cudaOccupancyMaxActiveBlocksPerMultiprocessor (or
::cudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlags) times the number of multiprocessors
as specified by the device attribute ::cudaDevAttrMultiProcessorCount. Since the
total number of blocks launched per device has to match across all devices, the maximum
number of blocks that can be launched per device will be limited by the device with the
least number of multiprocessors.
The kernel cannot make use of CUDA dynamic parallelism.
The ::cudaLaunchParams structure is defined as:
where:
- ::cudaLaunchParams::func specifies the kernel to be launched. This same functions must
be launched on all devices. For templated functions, pass the function symbol as follows:
func_name |
static int |
cudaLaunchHostFunc(CUstream_st stream,
cudaHostFn_t fn,
Pointer userData)
\brief Enqueues a host function call in a stream
Enqueues a host function to run in a stream.
|
static int |
cudaLaunchKernel(Pointer func,
dim3 gridDim,
dim3 blockDim,
Pointer args,
long sharedMem,
CUstream_st stream) |
static int |
cudaLaunchKernel(Pointer func,
dim3 gridDim,
dim3 blockDim,
PointerPointer args,
long sharedMem,
CUstream_st stream)
\brief Launches a device function
The function invokes kernel \p func on \p gridDim (\p gridDim.x × \p gridDim.y
× \p gridDim.z) grid of blocks.
|
static int |
cudaLaunchKernelExC(cudaLaunchConfig_t config,
Pointer func,
Pointer args) |
static int |
cudaLaunchKernelExC(cudaLaunchConfig_t config,
Pointer func,
PointerPointer args)
\brief Launches a CUDA function with launch-time configuration
Note that the functionally equivalent variadic template ::cudaLaunchKernelEx
is available for C++11 and newer.
|
static int |
cudaMalloc(Pointer devPtr,
long size) |
static int |
cudaMalloc(PointerPointer devPtr,
long size)
\brief Allocate memory on the device
Allocates \p size bytes of linear memory on the device and returns in
\p *devPtr a pointer to the allocated memory.
|
static int |
cudaMalloc3D(cudaPitchedPtr pitchedDevPtr,
cudaExtent extent)
\brief Allocates logical 1D, 2D, or 3D memory objects on the device
Allocates at least \p width * \p height * \p depth bytes of linear memory
on the device and returns a ::cudaPitchedPtr in which \p ptr is a pointer
to the allocated memory.
|
static int |
cudaMalloc3DArray(cudaArray array,
cudaChannelFormatDesc desc,
cudaExtent extent) |
static int |
cudaMalloc3DArray(cudaArray array,
cudaChannelFormatDesc desc,
cudaExtent extent,
int flags)
\brief Allocate an array on the device
Allocates a CUDA array according to the ::cudaChannelFormatDesc structure
\p desc and returns a handle to the new CUDA array in \p *array.
|
static int |
cudaMallocArray(cudaArray array,
cudaChannelFormatDesc desc,
long width) |
static int |
cudaMallocArray(cudaArray array,
cudaChannelFormatDesc desc,
long width,
long height,
int flags)
\brief Allocate an array on the device
Allocates a CUDA array according to the ::cudaChannelFormatDesc structure
\p desc and returns a handle to the new CUDA array in \p *array.
|
static int |
cudaMallocAsync(Pointer devPtr,
long size,
CUstream_st hStream) |
static int |
cudaMallocAsync(PointerPointer devPtr,
long size,
CUstream_st hStream)
\brief Allocates memory with stream ordered semantics
Inserts an allocation operation into \p hStream.
|
static int |
cudaMallocFromPoolAsync(Pointer ptr,
long size,
CUmemPoolHandle_st memPool,
CUstream_st stream) |
static int |
cudaMallocFromPoolAsync(PointerPointer ptr,
long size,
CUmemPoolHandle_st memPool,
CUstream_st stream)
\brief Allocates memory from a specified pool with stream ordered semantics.
|
static int |
cudaMallocHost(Pointer ptr,
long size) |
static int |
cudaMallocHost(PointerPointer ptr,
long size)
\brief Allocates page-locked memory on the host
Allocates \p size bytes of host memory that is page-locked and accessible
to the device.
|
static int |
cudaMallocManaged(Pointer devPtr,
long size) |
static int |
cudaMallocManaged(Pointer devPtr,
long size,
int flags) |
static int |
cudaMallocManaged(PointerPointer devPtr,
long size,
int flags)
\brief Allocates memory that will be automatically managed by the Unified Memory system
Allocates \p size bytes of managed memory on the device and returns in
\p *devPtr a pointer to the allocated memory.
|
static int |
cudaMallocMipmappedArray(cudaMipmappedArray mipmappedArray,
cudaChannelFormatDesc desc,
cudaExtent extent,
int numLevels) |
static int |
cudaMallocMipmappedArray(cudaMipmappedArray mipmappedArray,
cudaChannelFormatDesc desc,
cudaExtent extent,
int numLevels,
int flags)
\brief Allocate a mipmapped array on the device
Allocates a CUDA mipmapped array according to the ::cudaChannelFormatDesc structure
\p desc and returns a handle to the new CUDA mipmapped array in \p *mipmappedArray.
|
static int |
cudaMallocPitch(PointerPointer devPtr,
SizeTPointer pitch,
long width,
long height)
\brief Allocates pitched memory on the device
Allocates at least \p width (in bytes) * \p height bytes of linear memory
on the device and returns in \p *devPtr a pointer to the allocated memory.
|
static int |
cudaMallocPitch(Pointer devPtr,
SizeTPointer pitch,
long width,
long height) |
static int |
cudaMemAdvise_v2(Pointer devPtr,
long count,
int advice,
cudaMemLocation location)
\brief Advise about the usage of a given memory range
Advise the Unified Memory subsystem about the usage pattern for the memory range
starting at \p devPtr with a size of \p count bytes.
|
static int |
cudaMemAdvise(Pointer devPtr,
long count,
int advice,
int device)
\brief Advise about the usage of a given memory range
Advise the Unified Memory subsystem about the usage pattern for the memory range
starting at \p devPtr with a size of \p count bytes.
|
static int |
cudaMemcpy(Pointer dst,
Pointer src,
long count,
int kind)
\brief Copies data between host and device
Copies \p count bytes from the memory area pointed to by \p src to the
memory area pointed to by \p dst, where \p kind specifies the direction
of the copy, and must be one of ::cudaMemcpyHostToHost,
::cudaMemcpyHostToDevice, ::cudaMemcpyDeviceToHost,
::cudaMemcpyDeviceToDevice, or ::cudaMemcpyDefault.
|
static int |
cudaMemcpy2D(Pointer dst,
long dpitch,
Pointer src,
long spitch,
long width,
long height,
int kind)
\brief Copies data between host and device
Copies a matrix (\p height rows of \p width bytes each) from the memory
area pointed to by \p src to the memory area pointed to by \p dst, where
\p kind specifies the direction of the copy, and must be one of
::cudaMemcpyHostToHost, ::cudaMemcpyHostToDevice, ::cudaMemcpyDeviceToHost,
::cudaMemcpyDeviceToDevice, or ::cudaMemcpyDefault.
|
static int |
cudaMemcpy2DArrayToArray(cudaArray dst,
long wOffsetDst,
long hOffsetDst,
cudaArray src,
long wOffsetSrc,
long hOffsetSrc,
long width,
long height) |
static int |
cudaMemcpy2DArrayToArray(cudaArray dst,
long wOffsetDst,
long hOffsetDst,
cudaArray src,
long wOffsetSrc,
long hOffsetSrc,
long width,
long height,
int kind)
\brief Copies data between host and device
Copies a matrix (\p height rows of \p width bytes each) from the CUDA
array \p src starting at \p hOffsetSrc rows and \p wOffsetSrc bytes from the
upper left corner to the CUDA array \p dst starting at \p hOffsetDst rows
and \p wOffsetDst bytes from the upper left corner, where \p kind
specifies the direction of the copy, and must be one of
::cudaMemcpyHostToHost, ::cudaMemcpyHostToDevice, ::cudaMemcpyDeviceToHost,
::cudaMemcpyDeviceToDevice, or ::cudaMemcpyDefault.
|
static int |
cudaMemcpy2DAsync(Pointer dst,
long dpitch,
Pointer src,
long spitch,
long width,
long height,
int kind) |
static int |
cudaMemcpy2DAsync(Pointer dst,
long dpitch,
Pointer src,
long spitch,
long width,
long height,
int kind,
CUstream_st stream)
\brief Copies data between host and device
Copies a matrix (\p height rows of \p width bytes each) from the memory
area pointed to by \p src to the memory area pointed to by \p dst, where
\p kind specifies the direction of the copy, and must be one of
::cudaMemcpyHostToHost, ::cudaMemcpyHostToDevice, ::cudaMemcpyDeviceToHost,
::cudaMemcpyDeviceToDevice, or ::cudaMemcpyDefault.
|
static int |
cudaMemcpy2DFromArray(Pointer dst,
long dpitch,
cudaArray src,
long wOffset,
long hOffset,
long width,
long height,
int kind)
\brief Copies data between host and device
Copies a matrix (\p height rows of \p width bytes each) from the CUDA
array \p src starting at \p hOffset rows and \p wOffset bytes from the
upper left corner to the memory area pointed to by \p dst, where
\p kind specifies the direction of the copy, and must be one of
::cudaMemcpyHostToHost, ::cudaMemcpyHostToDevice, ::cudaMemcpyDeviceToHost,
::cudaMemcpyDeviceToDevice, or ::cudaMemcpyDefault.
|
static int |
cudaMemcpy2DFromArrayAsync(Pointer dst,
long dpitch,
cudaArray src,
long wOffset,
long hOffset,
long width,
long height,
int kind) |
static int |
cudaMemcpy2DFromArrayAsync(Pointer dst,
long dpitch,
cudaArray src,
long wOffset,
long hOffset,
long width,
long height,
int kind,
CUstream_st stream)
\brief Copies data between host and device
Copies a matrix (\p height rows of \p width bytes each) from the CUDA
array \p src starting at \p hOffset rows and \p wOffset bytes from the
upper left corner to the memory area pointed to by \p dst,
where \p kind specifies the direction of the copy, and must be one of
::cudaMemcpyHostToHost, ::cudaMemcpyHostToDevice, ::cudaMemcpyDeviceToHost,
::cudaMemcpyDeviceToDevice, or ::cudaMemcpyDefault.
|
static int |
cudaMemcpy2DToArray(cudaArray dst,
long wOffset,
long hOffset,
Pointer src,
long spitch,
long width,
long height,
int kind)
\brief Copies data between host and device
Copies a matrix (\p height rows of \p width bytes each) from the memory
area pointed to by \p src to the CUDA array \p dst starting at
\p hOffset rows and \p wOffset bytes from the upper left corner,
where \p kind specifies the direction of the copy, and must be one
of ::cudaMemcpyHostToHost, ::cudaMemcpyHostToDevice, ::cudaMemcpyDeviceToHost,
::cudaMemcpyDeviceToDevice, or ::cudaMemcpyDefault.
|
static int |
cudaMemcpy2DToArrayAsync(cudaArray dst,
long wOffset,
long hOffset,
Pointer src,
long spitch,
long width,
long height,
int kind) |
static int |
cudaMemcpy2DToArrayAsync(cudaArray dst,
long wOffset,
long hOffset,
Pointer src,
long spitch,
long width,
long height,
int kind,
CUstream_st stream)
\brief Copies data between host and device
Copies a matrix (\p height rows of \p width bytes each) from the memory
area pointed to by \p src to the CUDA array \p dst starting at \p hOffset
rows and \p wOffset bytes from the upper left corner, where \p kind specifies
the direction of the copy, and must be one of ::cudaMemcpyHostToHost,
::cudaMemcpyHostToDevice, ::cudaMemcpyDeviceToHost,
::cudaMemcpyDeviceToDevice, or ::cudaMemcpyDefault.
|
static int |
cudaMemcpy3D(cudaMemcpy3DParms p)
\brief Copies data between 3D objects
|
static int |
cudaMemcpy3DAsync(cudaMemcpy3DParms p) |
static int |
cudaMemcpy3DAsync(cudaMemcpy3DParms p,
CUstream_st stream)
\brief Copies data between 3D objects
|
static int |
cudaMemcpy3DPeer(cudaMemcpy3DPeerParms p)
\brief Copies memory between devices
Perform a 3D memory copy according to the parameters specified in
\p p.
|
static int |
cudaMemcpy3DPeerAsync(cudaMemcpy3DPeerParms p) |
static int |
cudaMemcpy3DPeerAsync(cudaMemcpy3DPeerParms p,
CUstream_st stream)
\brief Copies memory between devices asynchronously.
|
static int |
cudaMemcpyArrayToArray(cudaArray dst,
long wOffsetDst,
long hOffsetDst,
cudaArray src,
long wOffsetSrc,
long hOffsetSrc,
long count)
Deprecated.
|
static int |
cudaMemcpyArrayToArray(cudaArray dst,
long wOffsetDst,
long hOffsetDst,
cudaArray src,
long wOffsetSrc,
long hOffsetSrc,
long count,
int kind)
Deprecated.
Copies \p count bytes from the CUDA array \p src starting at \p hOffsetSrc
rows and \p wOffsetSrc bytes from the upper left corner to the CUDA array
\p dst starting at \p hOffsetDst rows and \p wOffsetDst bytes from the upper
left corner, where \p kind specifies the direction of the copy, and must be one of
::cudaMemcpyHostToHost, ::cudaMemcpyHostToDevice, ::cudaMemcpyDeviceToHost,
::cudaMemcpyDeviceToDevice, or ::cudaMemcpyDefault. Passing
::cudaMemcpyDefault is recommended, in which case the type of transfer is
inferred from the pointer values. However, ::cudaMemcpyDefault is only
allowed on systems that support unified virtual addressing.
|
static int |
cudaMemcpyAsync(Pointer dst,
Pointer src,
long count,
int kind) |
static int |
cudaMemcpyAsync(Pointer dst,
Pointer src,
long count,
int kind,
CUstream_st stream)
\brief Copies data between host and device
Copies \p count bytes from the memory area pointed to by \p src to the
memory area pointed to by \p dst, where \p kind specifies the
direction of the copy, and must be one of ::cudaMemcpyHostToHost,
::cudaMemcpyHostToDevice, ::cudaMemcpyDeviceToHost,
::cudaMemcpyDeviceToDevice, or ::cudaMemcpyDefault.
|
static int |
cudaMemcpyFromArray(Pointer dst,
cudaArray src,
long wOffset,
long hOffset,
long count,
int kind)
Deprecated.
Copies \p count bytes from the CUDA array \p src starting at \p hOffset rows
and \p wOffset bytes from the upper left corner to the memory area pointed to
by \p dst, where \p kind specifies the direction of the copy, and must be one of
::cudaMemcpyHostToHost, ::cudaMemcpyHostToDevice, ::cudaMemcpyDeviceToHost,
::cudaMemcpyDeviceToDevice, or ::cudaMemcpyDefault. Passing
::cudaMemcpyDefault is recommended, in which case the type of transfer is
inferred from the pointer values. However, ::cudaMemcpyDefault is only
allowed on systems that support unified virtual addressing.
|
static int |
cudaMemcpyFromArrayAsync(Pointer dst,
cudaArray src,
long wOffset,
long hOffset,
long count,
int kind)
Deprecated.
|
static int |
cudaMemcpyFromArrayAsync(Pointer dst,
cudaArray src,
long wOffset,
long hOffset,
long count,
int kind,
CUstream_st stream)
Deprecated.
Copies \p count bytes from the CUDA array \p src starting at \p hOffset rows
and \p wOffset bytes from the upper left corner to the memory area pointed to
by \p dst, where \p kind specifies the direction of the copy, and must be one of
::cudaMemcpyHostToHost, ::cudaMemcpyHostToDevice, ::cudaMemcpyDeviceToHost,
::cudaMemcpyDeviceToDevice, or ::cudaMemcpyDefault. Passing
::cudaMemcpyDefault is recommended, in which case the type of transfer is
inferred from the pointer values. However, ::cudaMemcpyDefault is only
allowed on systems that support unified virtual addressing.
::cudaMemcpyFromArrayAsync() is asynchronous with respect to the host, so
the call may return before the copy is complete. The copy can optionally
be associated to a stream by passing a non-zero \p stream argument. If \p
kind is ::cudaMemcpyHostToDevice or ::cudaMemcpyDeviceToHost and \p stream
is non-zero, the copy may overlap with operations in other streams.
|
static int |
cudaMemcpyFromSymbol(Pointer dst,
Pointer symbol,
long count) |
static int |
cudaMemcpyFromSymbol(Pointer dst,
Pointer symbol,
long count,
long offset,
int kind)
\brief Copies data from the given symbol on the device
Copies \p count bytes from the memory area pointed to by \p offset bytes
from the start of symbol \p symbol to the memory area pointed to by \p dst.
|
static int |
cudaMemcpyFromSymbolAsync(Pointer dst,
Pointer symbol,
long count,
long offset,
int kind) |
static int |
cudaMemcpyFromSymbolAsync(Pointer dst,
Pointer symbol,
long count,
long offset,
int kind,
CUstream_st stream)
\brief Copies data from the given symbol on the device
Copies \p count bytes from the memory area pointed to by \p offset bytes
from the start of symbol \p symbol to the memory area pointed to by \p dst.
|
static int |
cudaMemcpyPeer(Pointer dst,
int dstDevice,
Pointer src,
int srcDevice,
long count)
\brief Copies memory between two devices
Copies memory from one device to memory on another device.
|
static int |
cudaMemcpyPeerAsync(Pointer dst,
int dstDevice,
Pointer src,
int srcDevice,
long count) |
static int |
cudaMemcpyPeerAsync(Pointer dst,
int dstDevice,
Pointer src,
int srcDevice,
long count,
CUstream_st stream)
\brief Copies memory between two devices asynchronously.
|
static int |
cudaMemcpyToArray(cudaArray dst,
long wOffset,
long hOffset,
Pointer src,
long count,
int kind)
Deprecated.
Copies \p count bytes from the memory area pointed to by \p src to the
CUDA array \p dst starting at \p hOffset rows and \p wOffset bytes from
the upper left corner, where \p kind specifies the direction
of the copy, and must be one of ::cudaMemcpyHostToHost,
::cudaMemcpyHostToDevice, ::cudaMemcpyDeviceToHost,
::cudaMemcpyDeviceToDevice, or ::cudaMemcpyDefault. Passing
::cudaMemcpyDefault is recommended, in which case the type of transfer is
inferred from the pointer values. However, ::cudaMemcpyDefault is only
allowed on systems that support unified virtual addressing.
|
static int |
cudaMemcpyToArrayAsync(cudaArray dst,
long wOffset,
long hOffset,
Pointer src,
long count,
int kind)
Deprecated.
|
static int |
cudaMemcpyToArrayAsync(cudaArray dst,
long wOffset,
long hOffset,
Pointer src,
long count,
int kind,
CUstream_st stream)
Deprecated.
Copies \p count bytes from the memory area pointed to by \p src to the
CUDA array \p dst starting at \p hOffset rows and \p wOffset bytes from
the upper left corner, where \p kind specifies the
direction of the copy, and must be one of ::cudaMemcpyHostToHost,
::cudaMemcpyHostToDevice, ::cudaMemcpyDeviceToHost,
::cudaMemcpyDeviceToDevice, or ::cudaMemcpyDefault. Passing
::cudaMemcpyDefault is recommended, in which case the type of transfer is
inferred from the pointer values. However, ::cudaMemcpyDefault is only
allowed on systems that support unified virtual addressing.
::cudaMemcpyToArrayAsync() is asynchronous with respect to the host, so
the call may return before the copy is complete. The copy can optionally
be associated to a stream by passing a non-zero \p stream argument. If \p
kind is ::cudaMemcpyHostToDevice or ::cudaMemcpyDeviceToHost and \p stream
is non-zero, the copy may overlap with operations in other streams.
|
static int |
cudaMemcpyToSymbol(Pointer symbol,
Pointer src,
long count) |
static int |
cudaMemcpyToSymbol(Pointer symbol,
Pointer src,
long count,
long offset,
int kind)
\brief Copies data to the given symbol on the device
Copies \p count bytes from the memory area pointed to by \p src
to the memory area pointed to by \p offset bytes from the start of symbol
\p symbol.
|
static int |
cudaMemcpyToSymbolAsync(Pointer symbol,
Pointer src,
long count,
long offset,
int kind) |
static int |
cudaMemcpyToSymbolAsync(Pointer symbol,
Pointer src,
long count,
long offset,
int kind,
CUstream_st stream)
\brief Copies data to the given symbol on the device
Copies \p count bytes from the memory area pointed to by \p src
to the memory area pointed to by \p offset bytes from the start of symbol
\p symbol.
|
static int |
cudaMemGetInfo(SizeTPointer _free,
SizeTPointer total)
\brief Gets free and total device memory
Returns in \p *total the total amount of memory available to the the current context.
|
static int |
cudaMemPoolCreate(CUmemPoolHandle_st memPool,
cudaMemPoolProps poolProps)
\brief Creates a memory pool
Creates a CUDA memory pool and returns the handle in \p pool.
|
static int |
cudaMemPoolDestroy(CUmemPoolHandle_st memPool)
\brief Destroys the specified memory pool
If any pointers obtained from this pool haven't been freed or
the pool has free operations that haven't completed
when ::cudaMemPoolDestroy is invoked, the function will return immediately and the
resources associated with the pool will be released automatically
once there are no more outstanding allocations.
|
static int |
cudaMemPoolExportPointer(cudaMemPoolPtrExportData exportData,
Pointer ptr)
\brief Export data to share a memory pool allocation between processes.
|
static int |
cudaMemPoolExportToShareableHandle(Pointer shareableHandle,
CUmemPoolHandle_st memPool,
int handleType,
int flags)
\brief Exports a memory pool to the requested handle type.
|
static int |
cudaMemPoolGetAccess(int[] flags,
CUmemPoolHandle_st memPool,
cudaMemLocation location) |
static int |
cudaMemPoolGetAccess(IntBuffer flags,
CUmemPoolHandle_st memPool,
cudaMemLocation location) |
static int |
cudaMemPoolGetAccess(IntPointer flags,
CUmemPoolHandle_st memPool,
cudaMemLocation location)
\brief Returns the accessibility of a pool from a device
Returns the accessibility of the pool's memory from the specified location.
|
static int |
cudaMemPoolGetAttribute(CUmemPoolHandle_st memPool,
int attr,
Pointer value)
\brief Gets attributes of a memory pool
Supported attributes are:
- ::cudaMemPoolAttrReleaseThreshold: (value type = cuuint64_t)
Amount of reserved memory in bytes to hold onto before trying
to release memory back to the OS.
|
static int |
cudaMemPoolImportFromShareableHandle(CUmemPoolHandle_st memPool,
Pointer shareableHandle,
int handleType,
int flags)
\brief imports a memory pool from a shared handle.
|
static int |
cudaMemPoolImportPointer(Pointer ptr,
CUmemPoolHandle_st memPool,
cudaMemPoolPtrExportData exportData) |
static int |
cudaMemPoolImportPointer(PointerPointer ptr,
CUmemPoolHandle_st memPool,
cudaMemPoolPtrExportData exportData)
\brief Import a memory pool allocation from another process.
|
static int |
cudaMemPoolSetAccess(CUmemPoolHandle_st memPool,
cudaMemAccessDesc descList,
long count)
\brief Controls visibility of pools between devices
|
static int |
cudaMemPoolSetAttribute(CUmemPoolHandle_st memPool,
int attr,
Pointer value)
\brief Sets attributes of a memory pool
Supported attributes are:
- ::cudaMemPoolAttrReleaseThreshold: (value type = cuuint64_t)
Amount of reserved memory in bytes to hold onto before trying
to release memory back to the OS.
|
static int |
cudaMemPoolTrimTo(CUmemPoolHandle_st memPool,
long minBytesToKeep)
\brief Tries to release memory back to the OS
Releases memory back to the OS until the pool contains fewer than minBytesToKeep
reserved bytes, or there is no more memory that the allocator can safely release.
|
static int |
cudaMemPrefetchAsync_v2(Pointer devPtr,
long count,
cudaMemLocation location,
int flags) |
static int |
cudaMemPrefetchAsync_v2(Pointer devPtr,
long count,
cudaMemLocation location,
int flags,
CUstream_st stream) |
static int |
cudaMemPrefetchAsync(Pointer devPtr,
long count,
int dstDevice) |
static int |
cudaMemPrefetchAsync(Pointer devPtr,
long count,
int dstDevice,
CUstream_st stream)
\brief Prefetches memory to the specified destination device
Prefetches memory to the specified destination device.
|
static int |
cudaMemRangeGetAttribute(Pointer data,
long dataSize,
int attribute,
Pointer devPtr,
long count)
\brief Query an attribute of a given memory range
Query an attribute about the memory range starting at \p devPtr with a size of \p count bytes.
|
static int |
cudaMemRangeGetAttributes(PointerPointer data,
SizeTPointer dataSizes,
IntPointer attributes,
long numAttributes,
Pointer devPtr,
long count)
\brief Query attributes of a given memory range.
|
static int |
cudaMemRangeGetAttributes(Pointer data,
SizeTPointer dataSizes,
int[] attributes,
long numAttributes,
Pointer devPtr,
long count) |
static int |
cudaMemRangeGetAttributes(Pointer data,
SizeTPointer dataSizes,
IntBuffer attributes,
long numAttributes,
Pointer devPtr,
long count) |
static int |
cudaMemRangeGetAttributes(Pointer data,
SizeTPointer dataSizes,
IntPointer attributes,
long numAttributes,
Pointer devPtr,
long count) |
static int |
cudaMemset(Pointer devPtr,
int value,
long count)
\brief Initializes or sets device memory to a value
Fills the first \p count bytes of the memory area pointed to by \p devPtr
with the constant byte value \p value.
|
static int |
cudaMemset2D(Pointer devPtr,
long pitch,
int value,
long width,
long height)
\brief Initializes or sets device memory to a value
Sets to the specified value \p value a matrix (\p height rows of \p width
bytes each) pointed to by \p dstPtr.
|
static int |
cudaMemset2DAsync(Pointer devPtr,
long pitch,
int value,
long width,
long height) |
static int |
cudaMemset2DAsync(Pointer devPtr,
long pitch,
int value,
long width,
long height,
CUstream_st stream)
\brief Initializes or sets device memory to a value
Sets to the specified value \p value a matrix (\p height rows of \p width
bytes each) pointed to by \p dstPtr.
|
static int |
cudaMemset3D(cudaPitchedPtr pitchedDevPtr,
int value,
cudaExtent extent)
\brief Initializes or sets device memory to a value
Initializes each element of a 3D array to the specified value \p value.
|
static int |
cudaMemset3DAsync(cudaPitchedPtr pitchedDevPtr,
int value,
cudaExtent extent) |
static int |
cudaMemset3DAsync(cudaPitchedPtr pitchedDevPtr,
int value,
cudaExtent extent,
CUstream_st stream)
\brief Initializes or sets device memory to a value
Initializes each element of a 3D array to the specified value \p value.
|
static int |
cudaMemsetAsync(Pointer devPtr,
int value,
long count) |
static int |
cudaMemsetAsync(Pointer devPtr,
int value,
long count,
CUstream_st stream)
\brief Initializes or sets device memory to a value
Fills the first \p count bytes of the memory area pointed to by \p devPtr
with the constant byte value \p value.
|
static int |
cudaMipmappedArrayGetMemoryRequirements(cudaArrayMemoryRequirements memoryRequirements,
cudaMipmappedArray mipmap,
int device)
\brief Returns the memory requirements of a CUDA mipmapped array
Returns the memory requirements of a CUDA mipmapped array in \p memoryRequirements
If the CUDA mipmapped array is not allocated with flag ::cudaArrayDeferredMapping
::cudaErrorInvalidValue will be returned.
|
static int |
cudaMipmappedArrayGetSparseProperties(cudaArraySparseProperties sparseProperties,
cudaMipmappedArray mipmap)
\brief Returns the layout properties of a sparse CUDA mipmapped array
Returns the sparse array layout properties in \p sparseProperties.
|
static int |
cudaOccupancyAvailableDynamicSMemPerBlock(SizeTPointer dynamicSmemSize,
Pointer func,
int numBlocks,
int blockSize)
\brief Returns dynamic shared memory available per block when launching \p numBlocks blocks on SM.
|
static int |
cudaOccupancyMaxActiveBlocksPerMultiprocessor(int[] numBlocks,
Pointer func,
int blockSize,
long dynamicSMemSize) |
static int |
cudaOccupancyMaxActiveBlocksPerMultiprocessor(IntBuffer numBlocks,
Pointer func,
int blockSize,
long dynamicSMemSize) |
static int |
cudaOccupancyMaxActiveBlocksPerMultiprocessor(IntPointer numBlocks,
Pointer func,
int blockSize,
long dynamicSMemSize)
\brief Returns occupancy for a device function
Returns in \p *numBlocks the maximum number of active blocks per
streaming multiprocessor for the device function.
|
static int |
cudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlags(int[] numBlocks,
Pointer func,
int blockSize,
long dynamicSMemSize,
int flags) |
static int |
cudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlags(IntBuffer numBlocks,
Pointer func,
int blockSize,
long dynamicSMemSize,
int flags) |
static int |
cudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlags(IntPointer numBlocks,
Pointer func,
int blockSize,
long dynamicSMemSize,
int flags)
\brief Returns occupancy for a device function with the specified flags
Returns in \p *numBlocks the maximum number of active blocks per
streaming multiprocessor for the device function.
|
static int |
cudaOccupancyMaxActiveClusters(int[] numClusters,
Pointer func,
cudaLaunchConfig_t launchConfig) |
static int |
cudaOccupancyMaxActiveClusters(IntBuffer numClusters,
Pointer func,
cudaLaunchConfig_t launchConfig) |
static int |
cudaOccupancyMaxActiveClusters(IntPointer numClusters,
Pointer func,
cudaLaunchConfig_t launchConfig)
\brief Given the kernel function (\p func) and launch configuration
(\p config), return the maximum number of clusters that could co-exist
on the target device in \p *numClusters.
|
static int |
cudaOccupancyMaxPotentialClusterSize(int[] clusterSize,
Pointer func,
cudaLaunchConfig_t launchConfig) |
static int |
cudaOccupancyMaxPotentialClusterSize(IntBuffer clusterSize,
Pointer func,
cudaLaunchConfig_t launchConfig) |
static int |
cudaOccupancyMaxPotentialClusterSize(IntPointer clusterSize,
Pointer func,
cudaLaunchConfig_t launchConfig)
\brief Given the kernel function (\p func) and launch configuration
(\p config), return the maximum cluster size in \p *clusterSize.
|
static int |
cudaPeekAtLastError()
\brief Returns the last error from a runtime call
Returns the last error that has been produced by any of the runtime calls
in the same instance of the CUDA Runtime library in the host thread.
|
static int |
cudaPointerGetAttributes(cudaPointerAttributes attributes,
Pointer ptr)
\brief Returns attributes about a specified pointer
Returns in \p *attributes the attributes of the pointer \p ptr.
|
static int |
CUDART_INF_BF16()
\ingroup CUDA_MATH_INTRINSIC_BFLOAT16_CONSTANTS
\brief Defines floating-point positive infinity value for the \p nv_bfloat16 data type
|
static int |
CUDART_INF_FP16()
\ingroup CUDA_MATH_INTRINSIC_HALF_CONSTANTS
\brief Defines floating-point positive infinity value for the \p half data type
|
static int |
CUDART_MAX_NORMAL_BF16()
\ingroup CUDA_MATH_INTRINSIC_BFLOAT16_CONSTANTS
\brief Defines a maximum representable value for the \p nv_bfloat16 data type
|
static int |
CUDART_MAX_NORMAL_FP16()
\ingroup CUDA_MATH_INTRINSIC_HALF_CONSTANTS
\brief Defines a maximum representable value for the \p half data type
|
static int |
CUDART_MIN_DENORM_BF16()
\ingroup CUDA_MATH_INTRINSIC_BFLOAT16_CONSTANTS
\brief Defines a minimum representable (denormalized) value for the \p nv_bfloat16 data type
|
static int |
CUDART_MIN_DENORM_FP16()
\ingroup CUDA_MATH_INTRINSIC_HALF_CONSTANTS
\brief Defines a minimum representable (denormalized) value for the \p half data type
|
static int |
CUDART_NAN_BF16()
\ingroup CUDA_MATH_INTRINSIC_BFLOAT16_CONSTANTS
\brief Defines canonical NaN value for the \p nv_bfloat16 data type
|
static int |
CUDART_NAN_FP16()
\ingroup CUDA_MATH_INTRINSIC_HALF_CONSTANTS
\brief Defines canonical NaN value for the \p half data type
|
static int |
CUDART_NEG_ZERO_BF16()
\ingroup CUDA_MATH_INTRINSIC_BFLOAT16_CONSTANTS
\brief Defines a negative zero value for the \p nv_bfloat16 data type
|
static int |
CUDART_NEG_ZERO_FP16()
\ingroup CUDA_MATH_INTRINSIC_HALF_CONSTANTS
\brief Defines a negative zero value for the \p half data type
|
static int |
CUDART_ONE_BF16()
\ingroup CUDA_MATH_INTRINSIC_BFLOAT16_CONSTANTS
\brief Defines a value of 1.0 for the \p nv_bfloat16 data type
|
static int |
CUDART_ONE_FP16()
\ingroup CUDA_MATH_INTRINSIC_HALF_CONSTANTS
\brief Defines a value of 1.0 for the \p half data type
|
static int |
CUDART_ZERO_BF16()
\ingroup CUDA_MATH_INTRINSIC_BFLOAT16_CONSTANTS
\brief Defines a positive zero value for the \p nv_bfloat16 data type
|
static int |
CUDART_ZERO_FP16()
\ingroup CUDA_MATH_INTRINSIC_HALF_CONSTANTS
\brief Defines a positive zero value for the \p half data type
|
static int |
cudaRuntimeGetVersion(int[] runtimeVersion) |
static int |
cudaRuntimeGetVersion(IntBuffer runtimeVersion) |
static int |
cudaRuntimeGetVersion(IntPointer runtimeVersion)
\brief Returns the CUDA Runtime version
Returns in \p *runtimeVersion the version number of the current CUDA
Runtime instance.
|
static int |
cudaSetDevice(int device)
\brief Set device to be used for GPU executions
Sets \p device as the current device for the calling host thread.
|
static int |
cudaSetDeviceFlags(int flags)
\brief Sets flags to be used for device executions
Records \p flags as the flags for the current device.
|
static int |
cudaSetDoubleForDevice(double[] d)
Deprecated.
|
static int |
cudaSetDoubleForDevice(DoubleBuffer d)
Deprecated.
|
static int |
cudaSetDoubleForDevice(DoublePointer d)
Deprecated.
This function is deprecated as of CUDA 7.5
Converts the double value of \p d to an internal float representation if
the device does not support double arithmetic. If the device does natively
support doubles, then this function does nothing.
|
static int |
cudaSetDoubleForHost(double[] d)
Deprecated.
|
static int |
cudaSetDoubleForHost(DoubleBuffer d)
Deprecated.
|
static int |
cudaSetDoubleForHost(DoublePointer d)
Deprecated.
This function is deprecated as of CUDA 7.5
Converts the double value of \p d from a potentially internal float
representation if the device does not support double arithmetic. If the
device does natively support doubles, then this function does nothing.
|
static int |
cudaSetValidDevices(int[] device_arr,
int len) |
static int |
cudaSetValidDevices(IntBuffer device_arr,
int len) |
static int |
cudaSetValidDevices(IntPointer device_arr,
int len)
\brief Set a list of devices that can be used for CUDA
Sets a list of devices for CUDA execution in priority order using
\p device_arr.
|
static int |
cudaSignalExternalSemaphoresAsync(CUexternalSemaphore_st extSemArray,
cudaExternalSemaphoreSignalParams paramsArray,
int numExtSems) |
static int |
cudaSignalExternalSemaphoresAsync(CUexternalSemaphore_st extSemArray,
cudaExternalSemaphoreSignalParams paramsArray,
int numExtSems,
CUstream_st stream)
\brief Signals a set of external semaphore objects
Enqueues a signal operation on a set of externally allocated
semaphore object in the specified stream.
|
static int |
cudaStreamAddCallback(CUstream_st stream,
cudaStreamCallback_t callback,
Pointer userData,
int flags)
\brief Add a callback to a compute stream
\note This function is slated for eventual deprecation and removal.
|
static int |
cudaStreamAttachMemAsync(CUstream_st stream,
Pointer devPtr) |
static int |
cudaStreamAttachMemAsync(CUstream_st stream,
Pointer devPtr,
int flags) |
static int |
cudaStreamAttachMemAsync(CUstream_st stream,
Pointer devPtr,
long length,
int flags)
\brief Attach memory to a stream asynchronously
Enqueues an operation in \p stream to specify stream association of
\p length bytes of memory starting from \p devPtr.
|
static int |
cudaStreamBeginCapture(CUstream_st stream,
int mode)
\brief Begins graph capture on a stream
Begin graph capture on \p stream.
|
static int |
cudaStreamBeginCaptureToGraph(CUstream_st stream,
CUgraph_st graph,
CUgraphNode_st dependencies,
cudaGraphEdgeData dependencyData,
long numDependencies,
int mode)
\brief Begins graph capture on a stream to an existing graph
Begin graph capture on \p stream.
|
static int |
cudaStreamCopyAttributes(CUstream_st dst,
CUstream_st src)
\brief Copies attributes from source stream to destination stream.
|
static int |
cudaStreamCreate(CUstream_st pStream)
\brief Create an asynchronous stream
Creates a new asynchronous stream.
|
static int |
cudaStreamCreateWithFlags(CUstream_st pStream,
int flags)
\brief Create an asynchronous stream
Creates a new asynchronous stream.
|
static int |
cudaStreamCreateWithPriority(CUstream_st pStream,
int flags,
int priority)
\brief Create an asynchronous stream with the specified priority
Creates a stream with the specified priority and returns a handle in \p pStream.
|
static int |
cudaStreamDestroy(CUstream_st stream)
\brief Destroys and cleans up an asynchronous stream
Destroys and cleans up the asynchronous stream specified by \p stream.
|
static int |
cudaStreamEndCapture(CUstream_st stream,
CUgraph_st pGraph)
\brief Ends capture on a stream, returning the captured graph
End capture on \p stream, returning the captured graph via \p pGraph.
|
static int |
cudaStreamGetAttribute(CUstream_st hStream,
int attr,
cudaLaunchAttributeValue value_out)
\brief Queries stream attribute.
|
static int |
cudaStreamGetCaptureInfo_v3(CUstream_st stream,
int[] captureStatus_out) |
static int |
cudaStreamGetCaptureInfo_v3(CUstream_st stream,
int[] captureStatus_out,
long[] id_out,
CUgraph_st graph_out,
PointerPointer dependencies_out,
cudaGraphEdgeData edgeData_out,
SizeTPointer numDependencies_out) |
static int |
cudaStreamGetCaptureInfo_v3(CUstream_st stream,
IntBuffer captureStatus_out) |
static int |
cudaStreamGetCaptureInfo_v3(CUstream_st stream,
IntBuffer captureStatus_out,
LongBuffer id_out,
CUgraph_st graph_out,
PointerPointer dependencies_out,
cudaGraphEdgeData edgeData_out,
SizeTPointer numDependencies_out) |
static int |
cudaStreamGetCaptureInfo_v3(CUstream_st stream,
IntPointer captureStatus_out) |
static int |
cudaStreamGetCaptureInfo_v3(CUstream_st stream,
IntPointer captureStatus_out,
LongPointer id_out,
CUgraph_st graph_out,
PointerPointer dependencies_out,
cudaGraphEdgeData edgeData_out,
SizeTPointer numDependencies_out) |
static int |
cudaStreamGetCaptureInfo_v3(CUstream_st stream,
IntPointer captureStatus_out,
LongPointer id_out,
CUgraph_st graph_out,
PointerPointer dependencies_out,
PointerPointer edgeData_out,
SizeTPointer numDependencies_out)
\brief Query a stream's capture state (12.3+)
Query stream state related to stream capture.
|
static int |
cudaStreamGetCaptureInfo(CUstream_st stream,
int[] captureStatus_out) |
static int |
cudaStreamGetCaptureInfo(CUstream_st stream,
int[] captureStatus_out,
long[] id_out,
CUgraph_st graph_out,
PointerPointer dependencies_out,
SizeTPointer numDependencies_out) |
static int |
cudaStreamGetCaptureInfo(CUstream_st stream,
IntBuffer captureStatus_out) |
static int |
cudaStreamGetCaptureInfo(CUstream_st stream,
IntBuffer captureStatus_out,
LongBuffer id_out,
CUgraph_st graph_out,
PointerPointer dependencies_out,
SizeTPointer numDependencies_out) |
static int |
cudaStreamGetCaptureInfo(CUstream_st stream,
IntPointer captureStatus_out) |
static int |
cudaStreamGetCaptureInfo(CUstream_st stream,
IntPointer captureStatus_out,
LongPointer id_out,
CUgraph_st graph_out,
PointerPointer dependencies_out,
SizeTPointer numDependencies_out)
\brief Query a stream's capture state
Query stream state related to stream capture.
|
static int |
cudaStreamGetFlags(CUstream_st hStream,
int[] flags) |
static int |
cudaStreamGetFlags(CUstream_st hStream,
IntBuffer flags) |
static int |
cudaStreamGetFlags(CUstream_st hStream,
IntPointer flags)
\brief Query the flags of a stream
Query the flags of a stream.
|
static int |
cudaStreamGetId(CUstream_st hStream,
long[] streamId) |
static int |
cudaStreamGetId(CUstream_st hStream,
LongBuffer streamId) |
static int |
cudaStreamGetId(CUstream_st hStream,
LongPointer streamId)
\brief Query the Id of a stream
Query the Id of a stream.
|
static int |
cudaStreamGetPriority(CUstream_st hStream,
int[] priority) |
static int |
cudaStreamGetPriority(CUstream_st hStream,
IntBuffer priority) |
static int |
cudaStreamGetPriority(CUstream_st hStream,
IntPointer priority)
\brief Query the priority of a stream
Query the priority of a stream.
|
static int |
cudaStreamIsCapturing(CUstream_st stream,
int[] pCaptureStatus) |
static int |
cudaStreamIsCapturing(CUstream_st stream,
IntBuffer pCaptureStatus) |
static int |
cudaStreamIsCapturing(CUstream_st stream,
IntPointer pCaptureStatus)
\brief Returns a stream's capture status
Return the capture status of \p stream via \p pCaptureStatus.
|
static CUstream_st |
cudaStreamLegacy()
Legacy stream handle
Stream handle that can be passed as a cudaStream_t to use an implicit stream
with legacy synchronization behavior.
|
static CUstream_st |
cudaStreamPerThread()
Per-thread stream handle
Stream handle that can be passed as a cudaStream_t to use an implicit stream
with per-thread synchronization behavior.
|
static int |
cudaStreamQuery(CUstream_st stream)
\brief Queries an asynchronous stream for completion status
Returns ::cudaSuccess if all operations in \p stream have
completed, or ::cudaErrorNotReady if not.
|
static int |
cudaStreamSetAttribute(CUstream_st hStream,
int attr,
cudaLaunchAttributeValue value)
\brief Sets stream attribute.
|
static int |
cudaStreamSynchronize(CUstream_st stream)
\brief Waits for stream tasks to complete
Blocks until \p stream has completed all operations.
|
static int |
cudaStreamUpdateCaptureDependencies_v2(CUstream_st stream,
CUgraphNode_st dependencies,
cudaGraphEdgeData dependencyData,
long numDependencies) |
static int |
cudaStreamUpdateCaptureDependencies_v2(CUstream_st stream,
CUgraphNode_st dependencies,
cudaGraphEdgeData dependencyData,
long numDependencies,
int flags)
\brief Update the set of dependencies in a capturing stream (12.3+)
Modifies the dependency set of a capturing stream.
|
static int |
cudaStreamUpdateCaptureDependencies(CUstream_st stream,
CUgraphNode_st dependencies,
long numDependencies) |
static int |
cudaStreamUpdateCaptureDependencies(CUstream_st stream,
CUgraphNode_st dependencies,
long numDependencies,
int flags)
\brief Update the set of dependencies in a capturing stream (11.3+)
Modifies the dependency set of a capturing stream.
|
static int |
cudaStreamWaitEvent(CUstream_st stream,
CUevent_st event) |
static int |
cudaStreamWaitEvent(CUstream_st stream,
CUevent_st event,
int flags)
\brief Make a compute stream wait on an event
Makes all future work submitted to \p stream wait for all work captured in
\p event.
|
static int |
cudaThreadExchangeStreamCaptureMode(int[] mode) |
static int |
cudaThreadExchangeStreamCaptureMode(IntBuffer mode) |
static int |
cudaThreadExchangeStreamCaptureMode(IntPointer mode)
\brief Swaps the stream capture interaction mode for a thread
Sets the calling thread's stream capture interaction mode to the value contained
in \p *mode, and overwrites \p *mode with the previous mode for the thread.
|
static int |
cudaThreadExit()
Deprecated.
Note that this function is deprecated because its name does not
reflect its behavior. Its functionality is identical to the
non-deprecated function ::cudaDeviceReset(), which should be used
instead.
Explicitly destroys all cleans up all resources associated with the current
device in the current process. Any subsequent API call to this device will
reinitialize the device.
Note that this function will reset the device immediately. It is the caller's
responsibility to ensure that the device is not being accessed by any
other host threads from the process when this function is called.
|
static int |
cudaThreadGetCacheConfig(int[] pCacheConfig)
Deprecated.
|
static int |
cudaThreadGetCacheConfig(IntBuffer pCacheConfig)
Deprecated.
|
static int |
cudaThreadGetCacheConfig(IntPointer pCacheConfig)
Deprecated.
Note that this function is deprecated because its name does not
reflect its behavior. Its functionality is identical to the
non-deprecated function ::cudaDeviceGetCacheConfig(), which should be
used instead.
On devices where the L1 cache and shared memory use the same hardware
resources, this returns through \p pCacheConfig the preferred cache
configuration for the current device. This is only a preference. The
runtime will use the requested configuration if possible, but it is free to
choose a different configuration if required to execute functions.
This will return a \p pCacheConfig of ::cudaFuncCachePreferNone on devices
where the size of the L1 cache and shared memory are fixed.
The supported cache configurations are:
- ::cudaFuncCachePreferNone: no preference for shared memory or L1 (default)
- ::cudaFuncCachePreferShared: prefer larger shared memory and smaller L1 cache
- ::cudaFuncCachePreferL1: prefer larger L1 cache and smaller shared memory
|
static int |
cudaThreadGetLimit(SizeTPointer pValue,
int _limit)
Deprecated.
Note that this function is deprecated because its name does not
reflect its behavior. Its functionality is identical to the
non-deprecated function ::cudaDeviceGetLimit(), which should be used
instead.
Returns in \p *pValue the current size of \p limit. The supported
::cudaLimit values are:
- ::cudaLimitStackSize: stack size of each GPU thread;
- ::cudaLimitPrintfFifoSize: size of the shared FIFO used by the
::printf() device system call.
- ::cudaLimitMallocHeapSize: size of the heap used by the
::malloc() and ::free() device system calls;
|
static int |
cudaThreadSetCacheConfig(int cacheConfig)
Deprecated.
Note that this function is deprecated because its name does not
reflect its behavior. Its functionality is identical to the
non-deprecated function ::cudaDeviceSetCacheConfig(), which should be
used instead.
On devices where the L1 cache and shared memory use the same hardware
resources, this sets through \p cacheConfig the preferred cache
configuration for the current device. This is only a preference. The
runtime will use the requested configuration if possible, but it is free to
choose a different configuration if required to execute the function. Any
function preference set via
\ref ::cudaFuncSetCacheConfig(const void*, enum cudaFuncCache) "cudaFuncSetCacheConfig (C API)"
or
\ref ::cudaFuncSetCacheConfig(T*, enum cudaFuncCache) "cudaFuncSetCacheConfig (C++ API)"
will be preferred over this device-wide setting. Setting the device-wide
cache configuration to ::cudaFuncCachePreferNone will cause subsequent
kernel launches to prefer to not change the cache configuration unless
required to launch the kernel.
This setting does nothing on devices where the size of the L1 cache and
shared memory are fixed.
Launching a kernel with a different preference than the most recent
preference setting may insert a device-side synchronization point.
The supported cache configurations are:
- ::cudaFuncCachePreferNone: no preference for shared memory or L1 (default)
- ::cudaFuncCachePreferShared: prefer larger shared memory and smaller L1 cache
- ::cudaFuncCachePreferL1: prefer larger L1 cache and smaller shared memory
|
static int |
cudaThreadSetLimit(int _limit,
long value)
Deprecated.
Note that this function is deprecated because its name does not
reflect its behavior. Its functionality is identical to the
non-deprecated function ::cudaDeviceSetLimit(), which should be used
instead.
Setting \p limit to \p value is a request by the application to update
the current limit maintained by the device. The driver is free to
modify the requested value to meet h/w requirements (this could be
clamping to minimum or maximum values, rounding up to nearest element
size, etc). The application can use ::cudaThreadGetLimit() to find out
exactly what the limit has been set to.
Setting each ::cudaLimit has its own specific restrictions, so each is
discussed here.
- ::cudaLimitStackSize controls the stack size of each GPU thread.
- ::cudaLimitPrintfFifoSize controls the size of the shared FIFO
used by the ::printf() device system call.
Setting ::cudaLimitPrintfFifoSize must be performed before
launching any kernel that uses the ::printf() device
system call, otherwise ::cudaErrorInvalidValue will be returned.
- ::cudaLimitMallocHeapSize controls the size of the heap used
by the ::malloc() and ::free() device system calls. Setting
::cudaLimitMallocHeapSize must be performed before launching
any kernel that uses the ::malloc() or ::free() device system calls,
otherwise ::cudaErrorInvalidValue will be returned.
|
static int |
cudaThreadSynchronize()
Deprecated.
Note that this function is deprecated because its name does not
reflect its behavior. Its functionality is similar to the
non-deprecated function ::cudaDeviceSynchronize(), which should be used
instead.
Blocks until the device has completed all preceding requested tasks.
::cudaThreadSynchronize() returns an error if one of the preceding tasks
has failed. If the ::cudaDeviceScheduleBlockingSync flag was set for
this device, the host thread will block until the device has finished
its work.
|
static int |
cudaUserObjectCreate(CUuserObject_st object_out,
Pointer ptr,
cudaHostFn_t destroy,
int initialRefcount,
int flags)
\brief Create a user object
Create a user object with the specified destructor callback and initial reference count.
|
static int |
cudaUserObjectRelease(CUuserObject_st object) |
static int |
cudaUserObjectRelease(CUuserObject_st object,
int count)
\brief Release a reference to a user object
Releases user object references owned by the caller.
|
static int |
cudaUserObjectRetain(CUuserObject_st object) |
static int |
cudaUserObjectRetain(CUuserObject_st object,
int count)
\brief Retain a reference to a user object
Retains new references to a user object.
|
static int |
cudaWaitExternalSemaphoresAsync(CUexternalSemaphore_st extSemArray,
cudaExternalSemaphoreWaitParams paramsArray,
int numExtSems) |
static int |
cudaWaitExternalSemaphoresAsync(CUexternalSemaphore_st extSemArray,
cudaExternalSemaphoreWaitParams paramsArray,
int numExtSems,
CUstream_st stream)
\brief Waits on a set of external semaphore objects
Enqueues a wait operation on a set of externally allocated
semaphore object in the specified stream.
|
static int |
cuDestroyExternalMemory(CUextMemory_st extMem)
\brief Destroys an external memory object.
|
static int |
cuDestroyExternalSemaphore(CUextSemaphore_st extSem)
\brief Destroys an external semaphore
Destroys an external semaphore object and releases any references
to the underlying resource.
|
static int |
cuDeviceCanAccessPeer(int[] canAccessPeer,
int dev,
int peerDev) |
static int |
cuDeviceCanAccessPeer(IntBuffer canAccessPeer,
int dev,
int peerDev) |
static int |
cuDeviceCanAccessPeer(IntPointer canAccessPeer,
int dev,
int peerDev)
\brief Queries if a device may directly access a peer device's memory.
|
static int |
cuDeviceComputeCapability(int[] major,
int[] minor,
int dev)
Deprecated.
|
static int |
cuDeviceComputeCapability(IntBuffer major,
IntBuffer minor,
int dev)
Deprecated.
|
static int |
cuDeviceComputeCapability(IntPointer major,
IntPointer minor,
int dev)
Deprecated.
This function was deprecated as of CUDA 5.0 and its functionality superseded
by ::cuDeviceGetAttribute().
Returns in \p *major and \p *minor the major and minor revision numbers that
define the compute capability of the device \p dev.
|
static int |
cuDeviceGet(int[] device,
int ordinal) |
static int |
cuDeviceGet(IntBuffer device,
int ordinal) |
static int |
cuDeviceGet(IntPointer device,
int ordinal)
\brief Returns a handle to a compute device
Returns in \p *device a device handle given an ordinal in the range [0,
::cuDeviceGetCount()-1].
|
static int |
cuDeviceGetAttribute(int[] pi,
int attrib,
int dev) |
static int |
cuDeviceGetAttribute(IntBuffer pi,
int attrib,
int dev) |
static int |
cuDeviceGetAttribute(IntPointer pi,
int attrib,
int dev)
\brief Returns information about the device
Returns in \p *pi the integer value of the attribute \p attrib on device
\p dev.
|
static int |
cuDeviceGetByPCIBusId(int[] dev,
BytePointer pciBusId) |
static int |
cuDeviceGetByPCIBusId(int[] dev,
String pciBusId) |
static int |
cuDeviceGetByPCIBusId(IntBuffer dev,
BytePointer pciBusId) |
static int |
cuDeviceGetByPCIBusId(IntBuffer dev,
String pciBusId) |
static int |
cuDeviceGetByPCIBusId(IntPointer dev,
BytePointer pciBusId)
\brief Returns a handle to a compute device
Returns in \p *device a device handle given a PCI bus ID string.
|
static int |
cuDeviceGetByPCIBusId(IntPointer dev,
String pciBusId) |
static int |
cuDeviceGetCount(int[] count) |
static int |
cuDeviceGetCount(IntBuffer count) |
static int |
cuDeviceGetCount(IntPointer count)
\brief Returns the number of compute-capable devices
Returns in \p *count the number of devices with compute capability greater
than or equal to 2.0 that are available for execution.
|
static int |
cuDeviceGetDefaultMemPool(CUmemPoolHandle_st pool_out,
int dev)
\brief Returns the default mempool of a device
The default mempool of a device contains device memory from that device.
|
static int |
cuDeviceGetExecAffinitySupport(int[] pi,
int type,
int dev) |
static int |
cuDeviceGetExecAffinitySupport(IntBuffer pi,
int type,
int dev) |
static int |
cuDeviceGetExecAffinitySupport(IntPointer pi,
int type,
int dev)
\brief Returns information about the execution affinity support of the device.
|
static int |
cuDeviceGetGraphMemAttribute(int device,
int attr,
Pointer value)
\brief Query asynchronous allocation attributes related to graphs
Valid attributes are:
- ::CU_GRAPH_MEM_ATTR_USED_MEM_CURRENT: Amount of memory, in bytes, currently associated with graphs
- ::CU_GRAPH_MEM_ATTR_USED_MEM_HIGH: High watermark of memory, in bytes, associated with graphs since the
last time it was reset.
|
static int |
cuDeviceGetMemPool(CUmemPoolHandle_st pool,
int dev)
\brief Gets the current mempool for a device
Returns the last pool provided to ::cuDeviceSetMemPool for this device
or the device's default memory pool if ::cuDeviceSetMemPool has never been called.
|
static int |
cuDeviceGetName(byte[] name,
int len,
int dev) |
static int |
cuDeviceGetName(ByteBuffer name,
int len,
int dev) |
static int |
cuDeviceGetName(BytePointer name,
int len,
int dev)
\brief Returns an identifier string for the device
Returns an ASCII string identifying the device \p dev in the NULL-terminated
string pointed to by \p name.
|
static int |
cuDeviceGetPCIBusId(byte[] pciBusId,
int len,
int dev) |
static int |
cuDeviceGetPCIBusId(ByteBuffer pciBusId,
int len,
int dev) |
static int |
cuDeviceGetPCIBusId(BytePointer pciBusId,
int len,
int dev)
\brief Returns a PCI Bus Id string for the device
Returns an ASCII string identifying the device \p dev in the NULL-terminated
string pointed to by \p pciBusId.
|
static int |
cuDeviceGetProperties(CUdevprop_v1 prop,
int dev)
Deprecated.
This function was deprecated as of CUDA 5.0 and replaced by ::cuDeviceGetAttribute().
Returns in \p *prop the properties of device \p dev. The ::CUdevprop
structure is defined as:
where:
- ::maxThreadsPerBlock is the maximum number of threads per block;
- ::maxThreadsDim[3] is the maximum sizes of each dimension of a block;
- ::maxGridSize[3] is the maximum sizes of each dimension of a grid;
- ::sharedMemPerBlock is the total amount of shared memory available per
block in bytes;
- ::totalConstantMemory is the total amount of constant memory available on
the device in bytes;
- ::SIMDWidth is the warp size;
- ::memPitch is the maximum pitch allowed by the memory copy functions that
involve memory regions allocated through ::cuMemAllocPitch();
- ::regsPerBlock is the total number of registers available per block;
- ::clockRate is the clock frequency in kilohertz;
- ::textureAlign is the alignment requirement; texture base addresses that
are aligned to ::textureAlign bytes do not need an offset applied to
texture fetches. |
static int |
cuDeviceGetTexture1DLinearMaxWidth(SizeTPointer maxWidthInElements,
int format,
int numChannels,
int dev)
\brief Returns the maximum number of elements allocatable in a 1D linear texture for a given texture element size.
|
static int |
cuDeviceGetUuid_v2(CUuuid uuid,
int dev)
\brief Return an UUID for the device (11.4+)
Returns 16-octets identifying the device \p dev in the structure
pointed by the \p uuid.
|
static int |
cuDeviceGetUuid(CUuuid uuid,
int dev)
\brief Return an UUID for the device
Note there is a later version of this API, ::cuDeviceGetUuid_v2.
|
static int |
cuDeviceGraphMemTrim(int device)
\brief Free unused memory that was cached on the specified device for use with graphs back to the OS.
|
static int |
cuDevicePrimaryCtxGetState(int dev,
int[] flags,
int[] active) |
static int |
cuDevicePrimaryCtxGetState(int dev,
IntBuffer flags,
IntBuffer active) |
static int |
cuDevicePrimaryCtxGetState(int dev,
IntPointer flags,
IntPointer active)
\brief Get the state of the primary context
Returns in \p *flags the flags for the primary context of \p dev, and in
\p *active whether it is active.
|
static int |
cuDevicePrimaryCtxRelease(int dev)
\brief Release the primary context on the GPU
Releases the primary context interop on the device.
|
static int |
cuDevicePrimaryCtxReset(int dev)
\brief Destroy all allocations and reset all state on the primary context
Explicitly destroys and cleans up all resources associated with the current
device in the current process.
|
static int |
cuDevicePrimaryCtxRetain(CUctx_st pctx,
int dev)
\brief Retain the primary context on the GPU
Retains the primary context on the device.
|
static int |
cuDevicePrimaryCtxSetFlags(int dev,
int flags)
\brief Set flags for the primary context
Sets the flags for the primary context on the device overwriting perviously
set ones.
|
static int |
cuDeviceSetGraphMemAttribute(int device,
int attr,
Pointer value)
\brief Set asynchronous allocation attributes related to graphs
Valid attributes are:
- ::CU_GRAPH_MEM_ATTR_USED_MEM_HIGH: High watermark of memory, in bytes, associated with graphs since the
last time it was reset.
|
static int |
cuDeviceSetMemPool(int dev,
CUmemPoolHandle_st pool)
\brief Sets the current memory pool of a device
The memory pool must be local to the specified device.
|
static int |
cuDeviceTotalMem(SizeTPointer bytes,
int dev)
\brief Returns the total amount of memory on the device
Returns in \p *bytes the total amount of memory available on the device
\p dev in bytes.
|
static int |
cuDriverGetVersion(int[] driverVersion) |
static int |
cuDriverGetVersion(IntBuffer driverVersion) |
static int |
cuDriverGetVersion(IntPointer driverVersion)
\brief Returns the latest CUDA version supported by driver
Returns in \p *driverVersion the version of CUDA supported by
the driver.
|
static int |
cuEventCreate(CUevent_st phEvent,
int Flags)
\brief Creates an event
Creates an event *phEvent for the current context with the flags specified via
\p Flags.
|
static int |
cuEventDestroy(CUevent_st hEvent)
\brief Destroys an event
Destroys the event specified by \p hEvent.
|
static int |
cuEventElapsedTime(float[] pMilliseconds,
CUevent_st hStart,
CUevent_st hEnd) |
static int |
cuEventElapsedTime(FloatBuffer pMilliseconds,
CUevent_st hStart,
CUevent_st hEnd) |
static int |
cuEventElapsedTime(FloatPointer pMilliseconds,
CUevent_st hStart,
CUevent_st hEnd)
\brief Computes the elapsed time between two events
Computes the elapsed time between two events (in milliseconds with a
resolution of around 0.5 microseconds).
|
static int |
cuEventQuery(CUevent_st hEvent)
\brief Queries an event's status
Queries the status of all work currently captured by \p hEvent.
|
static int |
cuEventRecord(CUevent_st hEvent,
CUstream_st hStream)
\brief Records an event
Captures in \p hEvent the contents of \p hStream at the time of this call.
|
static int |
cuEventRecordWithFlags(CUevent_st hEvent,
CUstream_st hStream,
int flags)
\brief Records an event
Captures in \p hEvent the contents of \p hStream at the time of this call.
|
static int |
cuEventSynchronize(CUevent_st hEvent)
\brief Waits for an event to complete
Waits until the completion of all work currently captured in \p hEvent.
|
static int |
cuExternalMemoryGetMappedBuffer(long[] devPtr,
CUextMemory_st extMem,
CUDA_EXTERNAL_MEMORY_BUFFER_DESC_v1 bufferDesc) |
static int |
cuExternalMemoryGetMappedBuffer(LongBuffer devPtr,
CUextMemory_st extMem,
CUDA_EXTERNAL_MEMORY_BUFFER_DESC_v1 bufferDesc) |
static int |
cuExternalMemoryGetMappedBuffer(LongPointer devPtr,
CUextMemory_st extMem,
CUDA_EXTERNAL_MEMORY_BUFFER_DESC_v1 bufferDesc)
\brief Maps a buffer onto an imported memory object
Maps a buffer onto an imported memory object and returns a device
pointer in \p devPtr.
|
static int |
cuExternalMemoryGetMappedMipmappedArray(CUmipmappedArray_st mipmap,
CUextMemory_st extMem,
CUDA_EXTERNAL_MEMORY_MIPMAPPED_ARRAY_DESC_v1 mipmapDesc)
\brief Maps a CUDA mipmapped array onto an external memory object
Maps a CUDA mipmapped array onto an external object and returns a
handle to it in \p mipmap.
|
static int |
cuFlushGPUDirectRDMAWrites(int target,
int scope)
\brief Blocks until remote writes are visible to the specified scope
Blocks until GPUDirect RDMA writes to the target context via mappings
created through APIs like nvidia_p2p_get_pages (see
https://docs.nvidia.com/cuda/gpudirect-rdma for more information), are
visible to the specified scope.
|
static int |
cuFuncGetAttribute(int[] pi,
int attrib,
CUfunc_st hfunc) |
static int |
cuFuncGetAttribute(IntBuffer pi,
int attrib,
CUfunc_st hfunc) |
static int |
cuFuncGetAttribute(IntPointer pi,
int attrib,
CUfunc_st hfunc)
\brief Returns information about a function
Returns in \p *pi the integer value of the attribute \p attrib on the kernel
given by \p hfunc.
|
static int |
cuFuncGetModule(CUmod_st hmod,
CUfunc_st hfunc)
\brief Returns a module handle
Returns in \p *hmod the handle of the module that function \p hfunc
is located in.
|
static int |
cuFuncGetName(byte[] name,
CUfunc_st hfunc) |
static int |
cuFuncGetName(ByteBuffer name,
CUfunc_st hfunc) |
static int |
cuFuncGetName(BytePointer name,
CUfunc_st hfunc) |
static int |
cuFuncGetName(PointerPointer name,
CUfunc_st hfunc)
\brief Returns the function name for a ::CUfunction handle
Returns in \p **name the function name associated with the function handle \p hfunc .
|
static int |
cuFuncSetAttribute(CUfunc_st hfunc,
int attrib,
int value)
\brief Sets information about a function
This call sets the value of a specified attribute \p attrib on the kernel given
by \p hfunc to an integer value specified by \p val
This function returns CUDA_SUCCESS if the new value of the attribute could be
successfully set.
|
static int |
cuFuncSetBlockShape(CUfunc_st hfunc,
int x,
int y,
int z)
Deprecated.
Specifies the \p x, \p y, and \p z dimensions of the thread blocks that are
created when the kernel given by \p hfunc is launched.
|
static int |
cuFuncSetCacheConfig(CUfunc_st hfunc,
int config)
\brief Sets the preferred cache configuration for a device function
On devices where the L1 cache and shared memory use the same hardware
resources, this sets through \p config the preferred cache configuration for
the device function \p hfunc.
|
static int |
cuFuncSetSharedMemConfig(CUfunc_st hfunc,
int config)
\brief Sets the shared memory configuration for a device function.
|
static int |
cuFuncSetSharedSize(CUfunc_st hfunc,
int bytes)
Deprecated.
Sets through \p bytes the amount of dynamic shared memory that will be
available to each thread block when the kernel given by \p hfunc is launched.
|
static int |
cuGetErrorName(int error,
byte[] pStr) |
static int |
cuGetErrorName(int error,
ByteBuffer pStr) |
static int |
cuGetErrorName(int error,
BytePointer pStr) |
static int |
cuGetErrorName(int error,
PointerPointer pStr)
\brief Gets the string representation of an error code enum name
Sets \p *pStr to the address of a NULL-terminated string representation
of the name of the enum error code \p error.
|
static int |
cuGetErrorString(int error,
byte[] pStr) |
static int |
cuGetErrorString(int error,
ByteBuffer pStr) |
static int |
cuGetErrorString(int error,
BytePointer pStr) |
static int |
cuGetErrorString(int error,
PointerPointer pStr)
\brief Gets the string description of an error code
Sets \p *pStr to the address of a NULL-terminated string description
of the error code \p error.
|
static int |
cuGetExportTable(Pointer ppExportTable,
CUuuid pExportTableId) |
static int |
cuGetExportTable(PointerPointer ppExportTable,
CUuuid pExportTableId)
\}
|
static int |
cuGetProcAddress(BytePointer symbol,
Pointer pfn,
int cudaVersion,
long flags,
int[] symbolStatus) |
static int |
cuGetProcAddress(BytePointer symbol,
Pointer pfn,
int cudaVersion,
long flags,
IntBuffer symbolStatus) |
static int |
cuGetProcAddress(BytePointer symbol,
Pointer pfn,
int cudaVersion,
long flags,
IntPointer symbolStatus) |
static int |
cuGetProcAddress(BytePointer symbol,
PointerPointer pfn,
int cudaVersion,
long flags,
IntPointer symbolStatus)
\brief Returns the requested driver API function pointer
Returns in \p **pfn the address of the CUDA driver function for the requested
CUDA version and flags.
|
static int |
cuGetProcAddress(String symbol,
Pointer pfn,
int cudaVersion,
long flags,
int[] symbolStatus) |
static int |
cuGetProcAddress(String symbol,
Pointer pfn,
int cudaVersion,
long flags,
IntBuffer symbolStatus) |
static int |
cuGetProcAddress(String symbol,
Pointer pfn,
int cudaVersion,
long flags,
IntPointer symbolStatus) |
static int |
cuGLCtxCreate(CUctx_st pCtx,
int Flags,
int device)
Deprecated.
This function is deprecated as of Cuda 5.0.
This function is deprecated and should no longer be used. It is
no longer necessary to associate a CUDA context with an OpenGL
context in order to achieve maximum interoperability performance.
|
static int |
cuGLGetDevices(int[] pCudaDeviceCount,
int[] pCudaDevices,
int cudaDeviceCount,
int deviceList) |
static int |
cuGLGetDevices(IntBuffer pCudaDeviceCount,
IntBuffer pCudaDevices,
int cudaDeviceCount,
int deviceList) |
static int |
cuGLGetDevices(IntPointer pCudaDeviceCount,
IntPointer pCudaDevices,
int cudaDeviceCount,
int deviceList)
\brief Gets the CUDA devices associated with the current OpenGL context
Returns in \p *pCudaDeviceCount the number of CUDA-compatible devices
corresponding to the current OpenGL context.
|
static int |
cuGLInit()
Deprecated.
This function is deprecated as of Cuda 3.0.
Initializes OpenGL interoperability. This function is deprecated
and calling it is no longer required. It may fail if the needed
OpenGL driver facilities are not available.
|
static int |
cuGLMapBufferObject(long[] dptr,
SizeTPointer size,
int buffer)
Deprecated.
|
static int |
cuGLMapBufferObject(LongBuffer dptr,
SizeTPointer size,
int buffer)
Deprecated.
|
static int |
cuGLMapBufferObject(LongPointer dptr,
SizeTPointer size,
int buffer)
Deprecated.
This function is deprecated as of Cuda 3.0.
Maps the buffer object specified by \p buffer into the address space of the
current CUDA context and returns in \p *dptr and \p *size the base pointer
and size of the resulting mapping.
There must be a valid OpenGL context bound to the current thread
when this function is called. This must be the same context, or a
member of the same shareGroup, as the context that was bound when
the buffer was registered.
All streams in the current CUDA context are synchronized with the
current GL context.
|
static int |
cuGLMapBufferObjectAsync(long[] dptr,
SizeTPointer size,
int buffer,
CUstream_st hStream)
Deprecated.
|
static int |
cuGLMapBufferObjectAsync(LongBuffer dptr,
SizeTPointer size,
int buffer,
CUstream_st hStream)
Deprecated.
|
static int |
cuGLMapBufferObjectAsync(LongPointer dptr,
SizeTPointer size,
int buffer,
CUstream_st hStream)
Deprecated.
This function is deprecated as of Cuda 3.0.
Maps the buffer object specified by \p buffer into the address space of the
current CUDA context and returns in \p *dptr and \p *size the base pointer
and size of the resulting mapping.
There must be a valid OpenGL context bound to the current thread
when this function is called. This must be the same context, or a
member of the same shareGroup, as the context that was bound when
the buffer was registered.
Stream \p hStream in the current CUDA context is synchronized with
the current GL context.
|
static int |
cuGLRegisterBufferObject(int buffer)
Deprecated.
This function is deprecated as of Cuda 3.0.
Registers the buffer object specified by \p buffer for access by
CUDA. This function must be called before CUDA can map the buffer
object. There must be a valid OpenGL context bound to the current
thread when this function is called, and the buffer name is
resolved by that context.
|
static int |
cuGLSetBufferObjectMapFlags(int buffer,
int Flags)
Deprecated.
This function is deprecated as of Cuda 3.0.
Sets the map flags for the buffer object specified by \p buffer.
Changes to \p Flags will take effect the next time \p buffer is mapped.
The \p Flags argument may be any of the following:
- ::CU_GL_MAP_RESOURCE_FLAGS_NONE: Specifies no hints about how this
resource will be used. It is therefore assumed that this resource will be
read from and written to by CUDA kernels. This is the default value.
- ::CU_GL_MAP_RESOURCE_FLAGS_READ_ONLY: Specifies that CUDA kernels which
access this resource will not write to this resource.
- ::CU_GL_MAP_RESOURCE_FLAGS_WRITE_DISCARD: Specifies that CUDA kernels
which access this resource will not read from this resource and will
write over the entire contents of the resource, so none of the data
previously stored in the resource will be preserved.
If \p buffer has not been registered for use with CUDA, then
::CUDA_ERROR_INVALID_HANDLE is returned. If \p buffer is presently
mapped for access by CUDA, then ::CUDA_ERROR_ALREADY_MAPPED is returned.
There must be a valid OpenGL context bound to the current thread
when this function is called. This must be the same context, or a
member of the same shareGroup, as the context that was bound when
the buffer was registered.
|
static int |
cuGLUnmapBufferObject(int buffer)
Deprecated.
This function is deprecated as of Cuda 3.0.
Unmaps the buffer object specified by \p buffer for access by CUDA.
There must be a valid OpenGL context bound to the current thread
when this function is called. This must be the same context, or a
member of the same shareGroup, as the context that was bound when
the buffer was registered.
All streams in the current CUDA context are synchronized with the
current GL context.
|
static int |
cuGLUnmapBufferObjectAsync(int buffer,
CUstream_st hStream)
Deprecated.
This function is deprecated as of Cuda 3.0.
Unmaps the buffer object specified by \p buffer for access by CUDA.
There must be a valid OpenGL context bound to the current thread
when this function is called. This must be the same context, or a
member of the same shareGroup, as the context that was bound when
the buffer was registered.
Stream \p hStream in the current CUDA context is synchronized with
the current GL context.
|
static int |
cuGLUnregisterBufferObject(int buffer)
Deprecated.
This function is deprecated as of Cuda 3.0.
Unregisters the buffer object specified by \p buffer. This
releases any resources associated with the registered buffer.
After this call, the buffer may no longer be mapped for access by
CUDA.
There must be a valid OpenGL context bound to the current thread
when this function is called. This must be the same context, or a
member of the same shareGroup, as the context that was bound when
the buffer was registered.
|
static int |
cuGraphAddBatchMemOpNode(CUgraphNode_st phGraphNode,
CUgraph_st hGraph,
CUgraphNode_st dependencies,
long numDependencies,
CUDA_BATCH_MEM_OP_NODE_PARAMS_v1 nodeParams)
\brief Creates a batch memory operation node and adds it to a graph
Creates a new batch memory operation node and adds it to \p hGraph with \p
numDependencies dependencies specified via \p dependencies and arguments specified in \p nodeParams.
|
static int |
cuGraphAddChildGraphNode(CUgraphNode_st phGraphNode,
CUgraph_st hGraph,
CUgraphNode_st dependencies,
long numDependencies,
CUgraph_st childGraph)
\brief Creates a child graph node and adds it to a graph
Creates a new node which executes an embedded graph, and adds it to \p hGraph with
\p numDependencies dependencies specified via \p dependencies.
|
static int |
cuGraphAddDependencies_v2(CUgraph_st hGraph,
CUgraphNode_st from,
CUgraphNode_st to,
CUgraphEdgeData edgeData,
long numDependencies)
\brief Adds dependency edges to a graph (12.3+)
The number of dependencies to be added is defined by \p numDependencies
Elements in \p from and \p to at corresponding indices define a dependency.
|
static int |
cuGraphAddDependencies(CUgraph_st hGraph,
CUgraphNode_st from,
CUgraphNode_st to,
long numDependencies)
\brief Adds dependency edges to a graph
The number of dependencies to be added is defined by \p numDependencies
Elements in \p from and \p to at corresponding indices define a dependency.
|
static int |
cuGraphAddEmptyNode(CUgraphNode_st phGraphNode,
CUgraph_st hGraph,
CUgraphNode_st dependencies,
long numDependencies)
\brief Creates an empty node and adds it to a graph
Creates a new node which performs no operation, and adds it to \p hGraph with
\p numDependencies dependencies specified via \p dependencies.
|
static int |
cuGraphAddEventRecordNode(CUgraphNode_st phGraphNode,
CUgraph_st hGraph,
CUgraphNode_st dependencies,
long numDependencies,
CUevent_st event)
\brief Creates an event record node and adds it to a graph
Creates a new event record node and adds it to \p hGraph with \p numDependencies
dependencies specified via \p dependencies and event specified in \p event.
|
static int |
cuGraphAddEventWaitNode(CUgraphNode_st phGraphNode,
CUgraph_st hGraph,
CUgraphNode_st dependencies,
long numDependencies,
CUevent_st event)
\brief Creates an event wait node and adds it to a graph
Creates a new event wait node and adds it to \p hGraph with \p numDependencies
dependencies specified via \p dependencies and event specified in \p event.
|
static int |
cuGraphAddExternalSemaphoresSignalNode(CUgraphNode_st phGraphNode,
CUgraph_st hGraph,
CUgraphNode_st dependencies,
long numDependencies,
CUDA_EXT_SEM_SIGNAL_NODE_PARAMS_v1 nodeParams)
\brief Creates an external semaphore signal node and adds it to a graph
Creates a new external semaphore signal node and adds it to \p hGraph with \p
numDependencies dependencies specified via \p dependencies and arguments specified
in \p nodeParams.
|
static int |
cuGraphAddExternalSemaphoresWaitNode(CUgraphNode_st phGraphNode,
CUgraph_st hGraph,
CUgraphNode_st dependencies,
long numDependencies,
CUDA_EXT_SEM_WAIT_NODE_PARAMS_v1 nodeParams)
\brief Creates an external semaphore wait node and adds it to a graph
Creates a new external semaphore wait node and adds it to \p hGraph with \p numDependencies
dependencies specified via \p dependencies and arguments specified in \p nodeParams.
|
static int |
cuGraphAddHostNode(CUgraphNode_st phGraphNode,
CUgraph_st hGraph,
CUgraphNode_st dependencies,
long numDependencies,
CUDA_HOST_NODE_PARAMS_v1 nodeParams)
\brief Creates a host execution node and adds it to a graph
Creates a new CPU execution node and adds it to \p hGraph with \p numDependencies
dependencies specified via \p dependencies and arguments specified in \p nodeParams.
|
static int |
cuGraphAddKernelNode(CUgraphNode_st phGraphNode,
CUgraph_st hGraph,
CUgraphNode_st dependencies,
long numDependencies,
CUDA_KERNEL_NODE_PARAMS_v2 nodeParams)
\brief Creates a kernel execution node and adds it to a graph
Creates a new kernel execution node and adds it to \p hGraph with \p numDependencies
dependencies specified via \p dependencies and arguments specified in \p nodeParams.
|
static int |
cuGraphAddMemAllocNode(CUgraphNode_st phGraphNode,
CUgraph_st hGraph,
CUgraphNode_st dependencies,
long numDependencies,
CUDA_MEM_ALLOC_NODE_PARAMS_v1 nodeParams)
\brief Creates an allocation node and adds it to a graph
Creates a new allocation node and adds it to \p hGraph with \p numDependencies
dependencies specified via \p dependencies and arguments specified in \p nodeParams.
|
static int |
cuGraphAddMemcpyNode(CUgraphNode_st phGraphNode,
CUgraph_st hGraph,
CUgraphNode_st dependencies,
long numDependencies,
CUDA_MEMCPY3D_v2 copyParams,
CUctx_st ctx)
\brief Creates a memcpy node and adds it to a graph
Creates a new memcpy node and adds it to \p hGraph with \p numDependencies
dependencies specified via \p dependencies.
|
static int |
cuGraphAddMemFreeNode(CUgraphNode_st phGraphNode,
CUgraph_st hGraph,
CUgraphNode_st dependencies,
long numDependencies,
long dptr)
\brief Creates a memory free node and adds it to a graph
Creates a new memory free node and adds it to \p hGraph with \p numDependencies
dependencies specified via \p dependencies and arguments specified in \p nodeParams.
|
static int |
cuGraphAddMemsetNode(CUgraphNode_st phGraphNode,
CUgraph_st hGraph,
CUgraphNode_st dependencies,
long numDependencies,
CUDA_MEMSET_NODE_PARAMS_v1 memsetParams,
CUctx_st ctx)
\brief Creates a memset node and adds it to a graph
Creates a new memset node and adds it to \p hGraph with \p numDependencies
dependencies specified via \p dependencies.
|
static int |
cuGraphAddNode_v2(CUgraphNode_st phGraphNode,
CUgraph_st hGraph,
CUgraphNode_st dependencies,
CUgraphEdgeData dependencyData,
long numDependencies,
CUgraphNodeParams nodeParams)
\brief Adds a node of arbitrary type to a graph (12.3+)
Creates a new node in \p hGraph described by \p nodeParams with \p numDependencies
dependencies specified via \p dependencies.
|
static int |
cuGraphAddNode(CUgraphNode_st phGraphNode,
CUgraph_st hGraph,
CUgraphNode_st dependencies,
long numDependencies,
CUgraphNodeParams nodeParams)
\brief Adds a node of arbitrary type to a graph
Creates a new node in \p hGraph described by \p nodeParams with \p numDependencies
dependencies specified via \p dependencies.
|
static int |
cuGraphBatchMemOpNodeGetParams(CUgraphNode_st hNode,
CUDA_BATCH_MEM_OP_NODE_PARAMS_v1 nodeParams_out)
\brief Returns a batch mem op node's parameters
Returns the parameters of batch mem op node \p hNode in \p nodeParams_out.
|
static int |
cuGraphBatchMemOpNodeSetParams(CUgraphNode_st hNode,
CUDA_BATCH_MEM_OP_NODE_PARAMS_v1 nodeParams)
\brief Sets a batch mem op node's parameters
Sets the parameters of batch mem op node \p hNode to \p nodeParams.
|
static int |
cuGraphChildGraphNodeGetGraph(CUgraphNode_st hNode,
CUgraph_st phGraph)
\brief Gets a handle to the embedded graph of a child graph node
Gets a handle to the embedded graph in a child graph node.
|
static int |
cuGraphClone(CUgraph_st phGraphClone,
CUgraph_st originalGraph)
\brief Clones a graph
This function creates a copy of \p originalGraph and returns it in \p phGraphClone.
|
static int |
cuGraphConditionalHandleCreate(long[] pHandle_out,
CUgraph_st hGraph,
CUctx_st ctx,
int defaultLaunchValue,
int flags) |
static int |
cuGraphConditionalHandleCreate(LongBuffer pHandle_out,
CUgraph_st hGraph,
CUctx_st ctx,
int defaultLaunchValue,
int flags) |
static int |
cuGraphConditionalHandleCreate(LongPointer pHandle_out,
CUgraph_st hGraph,
CUctx_st ctx,
int defaultLaunchValue,
int flags)
\brief Create a conditional handle
Creates a conditional handle associated with \p hGraph.
|
static int |
cuGraphCreate(CUgraph_st phGraph,
int flags)
\brief Creates a graph
Creates an empty graph, which is returned via \p phGraph.
|
static int |
cuGraphDebugDotPrint(CUgraph_st hGraph,
BytePointer path,
int flags)
\brief Write a DOT file describing graph structure
Using the provided \p hGraph, write to \p path a DOT formatted description of the graph.
|
static int |
cuGraphDebugDotPrint(CUgraph_st hGraph,
String path,
int flags) |
static int |
cuGraphDestroy(CUgraph_st hGraph)
\brief Destroys a graph
Destroys the graph specified by \p hGraph, as well as all of its nodes.
|
static int |
cuGraphDestroyNode(CUgraphNode_st hNode)
\brief Remove a node from the graph
Removes \p hNode from its graph.
|
static int |
cuGraphEventRecordNodeGetEvent(CUgraphNode_st hNode,
CUevent_st event_out)
\brief Returns the event associated with an event record node
Returns the event of event record node \p hNode in \p event_out.
|
static int |
cuGraphEventRecordNodeSetEvent(CUgraphNode_st hNode,
CUevent_st event)
\brief Sets an event record node's event
Sets the event of event record node \p hNode to \p event.
|
static int |
cuGraphEventWaitNodeGetEvent(CUgraphNode_st hNode,
CUevent_st event_out)
\brief Returns the event associated with an event wait node
Returns the event of event wait node \p hNode in \p event_out.
|
static int |
cuGraphEventWaitNodeSetEvent(CUgraphNode_st hNode,
CUevent_st event)
\brief Sets an event wait node's event
Sets the event of event wait node \p hNode to \p event.
|
static int |
cuGraphExecBatchMemOpNodeSetParams(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
CUDA_BATCH_MEM_OP_NODE_PARAMS_v1 nodeParams)
\brief Sets the parameters for a batch mem op node in the given graphExec
Sets the parameters of a batch mem op node in an executable graph \p hGraphExec.
|
static int |
cuGraphExecChildGraphNodeSetParams(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
CUgraph_st childGraph)
\brief Updates node parameters in the child graph node in the given graphExec.
|
static int |
cuGraphExecDestroy(CUgraphExec_st hGraphExec)
\brief Destroys an executable graph
Destroys the executable graph specified by \p hGraphExec, as well
as all of its executable nodes.
|
static int |
cuGraphExecEventRecordNodeSetEvent(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
CUevent_st event)
\brief Sets the event for an event record node in the given graphExec
Sets the event of an event record node in an executable graph \p hGraphExec.
|
static int |
cuGraphExecEventWaitNodeSetEvent(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
CUevent_st event)
\brief Sets the event for an event wait node in the given graphExec
Sets the event of an event wait node in an executable graph \p hGraphExec.
|
static int |
cuGraphExecExternalSemaphoresSignalNodeSetParams(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
CUDA_EXT_SEM_SIGNAL_NODE_PARAMS_v1 nodeParams)
\brief Sets the parameters for an external semaphore signal node in the given graphExec
Sets the parameters of an external semaphore signal node in an executable graph \p hGraphExec.
|
static int |
cuGraphExecExternalSemaphoresWaitNodeSetParams(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
CUDA_EXT_SEM_WAIT_NODE_PARAMS_v1 nodeParams)
\brief Sets the parameters for an external semaphore wait node in the given graphExec
Sets the parameters of an external semaphore wait node in an executable graph \p hGraphExec.
|
static int |
cuGraphExecGetFlags(CUgraphExec_st hGraphExec,
long[] flags) |
static int |
cuGraphExecGetFlags(CUgraphExec_st hGraphExec,
LongBuffer flags) |
static int |
cuGraphExecGetFlags(CUgraphExec_st hGraphExec,
LongPointer flags)
\brief Query the instantiation flags of an executable graph
Returns the flags that were passed to instantiation for the given executable graph.
|
static int |
cuGraphExecHostNodeSetParams(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
CUDA_HOST_NODE_PARAMS_v1 nodeParams)
\brief Sets the parameters for a host node in the given graphExec.
|
static int |
cuGraphExecKernelNodeSetParams(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
CUDA_KERNEL_NODE_PARAMS_v2 nodeParams)
\brief Sets the parameters for a kernel node in the given graphExec
Sets the parameters of a kernel node in an executable graph \p hGraphExec.
|
static int |
cuGraphExecMemcpyNodeSetParams(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
CUDA_MEMCPY3D_v2 copyParams,
CUctx_st ctx)
\brief Sets the parameters for a memcpy node in the given graphExec.
|
static int |
cuGraphExecMemsetNodeSetParams(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
CUDA_MEMSET_NODE_PARAMS_v1 memsetParams,
CUctx_st ctx)
\brief Sets the parameters for a memset node in the given graphExec.
|
static int |
cuGraphExecNodeSetParams(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
CUgraphNodeParams nodeParams)
\brief Update's a graph node's parameters in an instantiated graph
Sets the parameters of a node in an executable graph \p hGraphExec.
|
static int |
cuGraphExecUpdate(CUgraphExec_st hGraphExec,
CUgraph_st hGraph,
CUgraphExecUpdateResultInfo_v1 resultInfo)
\brief Check whether an executable graph can be updated with a graph and perform the update if possible
Updates the node parameters in the instantiated graph specified by \p hGraphExec with the
node parameters in a topologically identical graph specified by \p hGraph.
|
static int |
cuGraphExternalSemaphoresSignalNodeGetParams(CUgraphNode_st hNode,
CUDA_EXT_SEM_SIGNAL_NODE_PARAMS_v1 params_out)
\brief Returns an external semaphore signal node's parameters
Returns the parameters of an external semaphore signal node \p hNode in \p params_out.
|
static int |
cuGraphExternalSemaphoresSignalNodeSetParams(CUgraphNode_st hNode,
CUDA_EXT_SEM_SIGNAL_NODE_PARAMS_v1 nodeParams)
\brief Sets an external semaphore signal node's parameters
Sets the parameters of an external semaphore signal node \p hNode to \p nodeParams.
|
static int |
cuGraphExternalSemaphoresWaitNodeGetParams(CUgraphNode_st hNode,
CUDA_EXT_SEM_WAIT_NODE_PARAMS_v1 params_out)
\brief Returns an external semaphore wait node's parameters
Returns the parameters of an external semaphore wait node \p hNode in \p params_out.
|
static int |
cuGraphExternalSemaphoresWaitNodeSetParams(CUgraphNode_st hNode,
CUDA_EXT_SEM_WAIT_NODE_PARAMS_v1 nodeParams)
\brief Sets an external semaphore wait node's parameters
Sets the parameters of an external semaphore wait node \p hNode to \p nodeParams.
|
static int |
cuGraphGetEdges_v2(CUgraph_st hGraph,
CUgraphNode_st from,
CUgraphNode_st to,
CUgraphEdgeData edgeData,
SizeTPointer numEdges)
\brief Returns a graph's dependency edges (12.3+)
Returns a list of \p hGraph's dependency edges.
|
static int |
cuGraphGetEdges(CUgraph_st hGraph,
CUgraphNode_st from,
CUgraphNode_st to,
SizeTPointer numEdges)
\brief Returns a graph's dependency edges
Returns a list of \p hGraph's dependency edges.
|
static int |
cuGraphGetNodes(CUgraph_st hGraph,
CUgraphNode_st nodes,
SizeTPointer numNodes)
\brief Returns a graph's nodes
Returns a list of \p hGraph's nodes.
|
static int |
cuGraphGetRootNodes(CUgraph_st hGraph,
CUgraphNode_st rootNodes,
SizeTPointer numRootNodes)
\brief Returns a graph's root nodes
Returns a list of \p hGraph's root nodes.
|
static int |
cuGraphHostNodeGetParams(CUgraphNode_st hNode,
CUDA_HOST_NODE_PARAMS_v1 nodeParams)
\brief Returns a host node's parameters
Returns the parameters of host node \p hNode in \p nodeParams.
|
static int |
cuGraphHostNodeSetParams(CUgraphNode_st hNode,
CUDA_HOST_NODE_PARAMS_v1 nodeParams)
\brief Sets a host node's parameters
Sets the parameters of host node \p hNode to \p nodeParams.
|
static int |
cuGraphicsGLRegisterBuffer(CUgraphicsResource_st pCudaResource,
int buffer,
int Flags)
\brief Registers an OpenGL buffer object
Registers the buffer object specified by \p buffer for access by
CUDA.
|
static int |
cuGraphicsGLRegisterImage(CUgraphicsResource_st pCudaResource,
int image,
int target,
int Flags)
\brief Register an OpenGL texture or renderbuffer object
Registers the texture or renderbuffer object specified by \p image for access by CUDA.
|
static int |
cuGraphicsMapResources(int count,
CUgraphicsResource_st resources,
CUstream_st hStream)
\brief Map graphics resources for access by CUDA
Maps the \p count graphics resources in \p resources for access by CUDA.
|
static int |
cuGraphicsResourceGetMappedMipmappedArray(CUmipmappedArray_st pMipmappedArray,
CUgraphicsResource_st resource)
\brief Get a mipmapped array through which to access a mapped graphics resource.
|
static int |
cuGraphicsResourceGetMappedPointer(long[] pDevPtr,
SizeTPointer pSize,
CUgraphicsResource_st resource) |
static int |
cuGraphicsResourceGetMappedPointer(LongBuffer pDevPtr,
SizeTPointer pSize,
CUgraphicsResource_st resource) |
static int |
cuGraphicsResourceGetMappedPointer(LongPointer pDevPtr,
SizeTPointer pSize,
CUgraphicsResource_st resource)
\brief Get a device pointer through which to access a mapped graphics resource.
|
static int |
cuGraphicsResourceSetMapFlags(CUgraphicsResource_st resource,
int flags)
\brief Set usage flags for mapping a graphics resource
Set \p flags for mapping the graphics resource \p resource.
|
static int |
cuGraphicsSubResourceGetMappedArray(CUarray_st pArray,
CUgraphicsResource_st resource,
int arrayIndex,
int mipLevel)
\brief Get an array through which to access a subresource of a mapped graphics resource.
|
static int |
cuGraphicsUnmapResources(int count,
CUgraphicsResource_st resources,
CUstream_st hStream)
\brief Unmap graphics resources.
|
static int |
cuGraphicsUnregisterResource(CUgraphicsResource_st resource)
\brief Unregisters a graphics resource for access by CUDA
Unregisters the graphics resource \p resource so it is not accessible by
CUDA unless registered again.
|
static int |
cuGraphInstantiate(CUgraphExec_st phGraphExec,
CUgraph_st hGraph,
long flags)
\brief Creates an executable graph from a graph
Instantiates \p hGraph as an executable graph.
|
static int |
cuGraphInstantiateWithParams(CUgraphExec_st phGraphExec,
CUgraph_st hGraph,
CUDA_GRAPH_INSTANTIATE_PARAMS instantiateParams)
\brief Creates an executable graph from a graph
Instantiates \p hGraph as an executable graph according to the \p instantiateParams structure.
|
static int |
cuGraphKernelNodeCopyAttributes(CUgraphNode_st dst,
CUgraphNode_st src)
\brief Copies attributes from source node to destination node.
|
static int |
cuGraphKernelNodeGetAttribute(CUgraphNode_st hNode,
int attr,
CUlaunchAttributeValue value_out)
\brief Queries node attribute.
|
static int |
cuGraphKernelNodeGetParams(CUgraphNode_st hNode,
CUDA_KERNEL_NODE_PARAMS_v2 nodeParams)
\brief Returns a kernel node's parameters
Returns the parameters of kernel node \p hNode in \p nodeParams.
|
static int |
cuGraphKernelNodeSetAttribute(CUgraphNode_st hNode,
int attr,
CUlaunchAttributeValue value)
\brief Sets node attribute.
|
static int |
cuGraphKernelNodeSetParams(CUgraphNode_st hNode,
CUDA_KERNEL_NODE_PARAMS_v2 nodeParams)
\brief Sets a kernel node's parameters
Sets the parameters of kernel node \p hNode to \p nodeParams.
|
static int |
cuGraphLaunch(CUgraphExec_st hGraphExec,
CUstream_st hStream)
\brief Launches an executable graph in a stream
Executes \p hGraphExec in \p hStream.
|
static int |
cuGraphMemAllocNodeGetParams(CUgraphNode_st hNode,
CUDA_MEM_ALLOC_NODE_PARAMS_v1 params_out)
\brief Returns a memory alloc node's parameters
Returns the parameters of a memory alloc node \p hNode in \p params_out.
|
static int |
cuGraphMemcpyNodeGetParams(CUgraphNode_st hNode,
CUDA_MEMCPY3D_v2 nodeParams)
\brief Returns a memcpy node's parameters
Returns the parameters of memcpy node \p hNode in \p nodeParams.
|
static int |
cuGraphMemcpyNodeSetParams(CUgraphNode_st hNode,
CUDA_MEMCPY3D_v2 nodeParams)
\brief Sets a memcpy node's parameters
Sets the parameters of memcpy node \p hNode to \p nodeParams.
|
static int |
cuGraphMemFreeNodeGetParams(CUgraphNode_st hNode,
long[] dptr_out) |
static int |
cuGraphMemFreeNodeGetParams(CUgraphNode_st hNode,
LongBuffer dptr_out) |
static int |
cuGraphMemFreeNodeGetParams(CUgraphNode_st hNode,
LongPointer dptr_out)
\brief Returns a memory free node's parameters
Returns the address of a memory free node \p hNode in \p dptr_out.
|
static int |
cuGraphMemsetNodeGetParams(CUgraphNode_st hNode,
CUDA_MEMSET_NODE_PARAMS_v1 nodeParams)
\brief Returns a memset node's parameters
Returns the parameters of memset node \p hNode in \p nodeParams.
|
static int |
cuGraphMemsetNodeSetParams(CUgraphNode_st hNode,
CUDA_MEMSET_NODE_PARAMS_v1 nodeParams)
\brief Sets a memset node's parameters
Sets the parameters of memset node \p hNode to \p nodeParams.
|
static int |
cuGraphNodeFindInClone(CUgraphNode_st phNode,
CUgraphNode_st hOriginalNode,
CUgraph_st hClonedGraph)
\brief Finds a cloned version of a node
This function returns the node in \p hClonedGraph corresponding to \p hOriginalNode
in the original graph.
|
static int |
cuGraphNodeGetDependencies_v2(CUgraphNode_st hNode,
CUgraphNode_st dependencies,
CUgraphEdgeData edgeData,
SizeTPointer numDependencies)
\brief Returns a node's dependencies (12.3+)
Returns a list of \p node's dependencies.
|
static int |
cuGraphNodeGetDependencies(CUgraphNode_st hNode,
CUgraphNode_st dependencies,
SizeTPointer numDependencies)
\brief Returns a node's dependencies
Returns a list of \p node's dependencies.
|
static int |
cuGraphNodeGetDependentNodes_v2(CUgraphNode_st hNode,
CUgraphNode_st dependentNodes,
CUgraphEdgeData edgeData,
SizeTPointer numDependentNodes)
\brief Returns a node's dependent nodes (12.3+)
Returns a list of \p node's dependent nodes.
|
static int |
cuGraphNodeGetDependentNodes(CUgraphNode_st hNode,
CUgraphNode_st dependentNodes,
SizeTPointer numDependentNodes)
\brief Returns a node's dependent nodes
Returns a list of \p node's dependent nodes.
|
static int |
cuGraphNodeGetEnabled(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
int[] isEnabled) |
static int |
cuGraphNodeGetEnabled(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
IntBuffer isEnabled) |
static int |
cuGraphNodeGetEnabled(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
IntPointer isEnabled)
\brief Query whether a node in the given graphExec is enabled
Sets isEnabled to 1 if \p hNode is enabled, or 0 if \p hNode is disabled.
|
static int |
cuGraphNodeGetType(CUgraphNode_st hNode,
int[] type) |
static int |
cuGraphNodeGetType(CUgraphNode_st hNode,
IntBuffer type) |
static int |
cuGraphNodeGetType(CUgraphNode_st hNode,
IntPointer type)
\brief Returns a node's type
Returns the node type of \p hNode in \p type.
|
static int |
cuGraphNodeSetEnabled(CUgraphExec_st hGraphExec,
CUgraphNode_st hNode,
int isEnabled)
\brief Enables or disables the specified node in the given graphExec
Sets \p hNode to be either enabled or disabled.
|
static int |
cuGraphNodeSetParams(CUgraphNode_st hNode,
CUgraphNodeParams nodeParams)
\brief Update's a graph node's parameters
Sets the parameters of graph node \p hNode to \p nodeParams.
|
static int |
cuGraphReleaseUserObject(CUgraph_st graph,
CUuserObject_st object,
int count)
\brief Release a user object reference from a graph
Releases user object references owned by a graph.
|
static int |
cuGraphRemoveDependencies_v2(CUgraph_st hGraph,
CUgraphNode_st from,
CUgraphNode_st to,
CUgraphEdgeData edgeData,
long numDependencies)
\brief Removes dependency edges from a graph (12.3+)
The number of \p dependencies to be removed is defined by \p numDependencies.
|
static int |
cuGraphRemoveDependencies(CUgraph_st hGraph,
CUgraphNode_st from,
CUgraphNode_st to,
long numDependencies)
\brief Removes dependency edges from a graph
The number of \p dependencies to be removed is defined by \p numDependencies.
|
static int |
cuGraphRetainUserObject(CUgraph_st graph,
CUuserObject_st object,
int count,
int flags)
\brief Retain a reference to a user object from a graph
Creates or moves user object references that will be owned by a CUDA graph.
|
static int |
cuGraphUpload(CUgraphExec_st hGraphExec,
CUstream_st hStream)
\brief Uploads an executable graph in a stream
Uploads \p hGraphExec to the device in \p hStream without executing it.
|
static int |
cuImportExternalMemory(CUextMemory_st extMem_out,
CUDA_EXTERNAL_MEMORY_HANDLE_DESC_v1 memHandleDesc)
\brief Imports an external memory object
Imports an externally allocated memory object and returns
a handle to that in \p extMem_out.
|
static int |
cuImportExternalSemaphore(CUextSemaphore_st extSem_out,
CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC_v1 semHandleDesc)
\brief Imports an external semaphore
Imports an externally allocated synchronization object and returns
a handle to that in \p extSem_out.
|
static int |
cuInit(int Flags)
\brief Initialize the CUDA driver API
Initializes the driver API and must be called before any other function from
the driver API in the current process.
|
static int |
cuIpcCloseMemHandle(long dptr)
\brief Attempts to close memory mapped with ::cuIpcOpenMemHandle
Decrements the reference count of the memory returned by ::cuIpcOpenMemHandle by 1.
|
static int |
cuIpcGetEventHandle(CUipcEventHandle_v1 pHandle,
CUevent_st event)
\brief Gets an interprocess handle for a previously allocated event
Takes as input a previously allocated event.
|
static int |
cuIpcGetMemHandle(CUipcMemHandle_v1 pHandle,
long dptr)
\brief Gets an interprocess memory handle for an existing device memory
allocation
Takes a pointer to the base of an existing device memory allocation created
with ::cuMemAlloc and exports it for use in another process.
|
static int |
cuIpcOpenEventHandle(CUevent_st phEvent,
CUipcEventHandle_v1 handle)
\brief Opens an interprocess event handle for use in the current process
Opens an interprocess event handle exported from another process with
::cuIpcGetEventHandle.
|
static int |
cuIpcOpenMemHandle(long[] pdptr,
CUipcMemHandle_v1 handle,
int Flags) |
static int |
cuIpcOpenMemHandle(LongBuffer pdptr,
CUipcMemHandle_v1 handle,
int Flags) |
static int |
cuIpcOpenMemHandle(LongPointer pdptr,
CUipcMemHandle_v1 handle,
int Flags)
\brief Opens an interprocess memory handle exported from another process
and returns a device pointer usable in the local process.
|
static int |
cuKernelGetAttribute(int[] pi,
int attrib,
CUkern_st kernel,
int dev) |
static int |
cuKernelGetAttribute(IntBuffer pi,
int attrib,
CUkern_st kernel,
int dev) |
static int |
cuKernelGetAttribute(IntPointer pi,
int attrib,
CUkern_st kernel,
int dev)
\brief Returns information about a kernel
Returns in \p *pi the integer value of the attribute \p attrib for the kernel
\p kernel for the requested device \p dev.
|
static int |
cuKernelGetFunction(CUfunc_st pFunc,
CUkern_st kernel)
\brief Returns a function handle
Returns in \p pFunc the handle of the function for the requested kernel \p kernel and
the current context.
|
static int |
cuKernelGetName(byte[] name,
CUkern_st hfunc) |
static int |
cuKernelGetName(ByteBuffer name,
CUkern_st hfunc) |
static int |
cuKernelGetName(BytePointer name,
CUkern_st hfunc) |
static int |
cuKernelGetName(PointerPointer name,
CUkern_st hfunc)
\brief Returns the function name for a ::CUkernel handle
Returns in \p **name the function name associated with the kernel handle \p hfunc .
|
static int |
cuKernelSetAttribute(int attrib,
int val,
CUkern_st kernel,
int dev)
\brief Sets information about a kernel
This call sets the value of a specified attribute \p attrib on the kernel \p kernel
for the requested device \p dev to an integer value specified by \p val.
|
static int |
cuKernelSetCacheConfig(CUkern_st kernel,
int config,
int dev)
\brief Sets the preferred cache configuration for a device kernel.
|
static int |
cuLaunch(CUfunc_st f)
Deprecated.
Invokes the kernel \p f on a 1 x 1 x 1 grid of blocks. The block
contains the number of threads specified by a previous call to
::cuFuncSetBlockShape().
The block shape, dynamic shared memory size, and parameter information
must be set using
::cuFuncSetBlockShape(),
::cuFuncSetSharedSize(),
::cuParamSetSize(),
::cuParamSeti(),
::cuParamSetf(), and
::cuParamSetv()
prior to calling this function.
Launching a function via ::cuLaunchKernel() invalidates the function's
block shape, dynamic shared memory size, and parameter information. After
launching via cuLaunchKernel, this state must be re-initialized prior to
calling this function. Failure to do so results in undefined behavior.
|
static int |
cuLaunchCooperativeKernel(CUfunc_st f,
int gridDimX,
int gridDimY,
int gridDimZ,
int blockDimX,
int blockDimY,
int blockDimZ,
int sharedMemBytes,
CUstream_st hStream,
Pointer kernelParams) |
static int |
cuLaunchCooperativeKernel(CUfunc_st f,
int gridDimX,
int gridDimY,
int gridDimZ,
int blockDimX,
int blockDimY,
int blockDimZ,
int sharedMemBytes,
CUstream_st hStream,
PointerPointer kernelParams)
\brief Launches a CUDA function ::CUfunction or a CUDA kernel ::CUkernel where thread blocks
can cooperate and synchronize as they execute
Invokes the function ::CUfunction or the kernel ::CUkernel \p f on a \p gridDimX x \p gridDimY x \p gridDimZ
grid of blocks.
|
static int |
cuLaunchCooperativeKernelMultiDevice(CUDA_LAUNCH_PARAMS_v1 launchParamsList,
int numDevices,
int flags)
Deprecated.
This function is deprecated as of CUDA 11.3.
Invokes kernels as specified in the \p launchParamsList array where each element
of the array specifies all the parameters required to perform a single kernel launch.
These kernels can cooperate and synchronize as they execute. The size of the array is
specified by \p numDevices.
No two kernels can be launched on the same device. All the devices targeted by this
multi-device launch must be identical. All devices must have a non-zero value for the
device attribute ::CU_DEVICE_ATTRIBUTE_COOPERATIVE_MULTI_DEVICE_LAUNCH.
All kernels launched must be identical with respect to the compiled code. Note that
any __device__, __constant__ or __managed__ variables present in the module that owns
the kernel launched on each device, are independently instantiated on every device.
It is the application's responsibility to ensure these variables are initialized and
used appropriately.
The size of the grids as specified in blocks, the size of the blocks themselves
and the amount of shared memory used by each thread block must also match across
all launched kernels.
The streams used to launch these kernels must have been created via either ::cuStreamCreate
or ::cuStreamCreateWithPriority. The NULL stream or ::CU_STREAM_LEGACY or ::CU_STREAM_PER_THREAD
cannot be used.
The total number of blocks launched per kernel cannot exceed the maximum number of blocks
per multiprocessor as returned by ::cuOccupancyMaxActiveBlocksPerMultiprocessor (or
::cuOccupancyMaxActiveBlocksPerMultiprocessorWithFlags) times the number of multiprocessors
as specified by the device attribute ::CU_DEVICE_ATTRIBUTE_MULTIPROCESSOR_COUNT. Since the
total number of blocks launched per device has to match across all devices, the maximum
number of blocks that can be launched per device will be limited by the device with the
least number of multiprocessors.
The kernels cannot make use of CUDA dynamic parallelism.
The ::CUDA_LAUNCH_PARAMS structure is defined as:
where:
- ::CUDA_LAUNCH_PARAMS::function specifies the kernel to be launched. All functions must
be identical with respect to the compiled code.
Note that you can also specify context-less kernel ::CUkernel by querying the handle
using ::cuLibraryGetKernel() and then casting to ::CUfunction. In this case, the context to
launch the kernel on be taken from the specified stream ::CUDA_LAUNCH_PARAMS::hStream.
- ::CUDA_LAUNCH_PARAMS::gridDimX is the width of the grid in blocks. This must match across
all kernels launched.
- ::CUDA_LAUNCH_PARAMS::gridDimY is the height of the grid in blocks. This must match across
all kernels launched.
- ::CUDA_LAUNCH_PARAMS::gridDimZ is the depth of the grid in blocks. This must match across
all kernels launched.
- ::CUDA_LAUNCH_PARAMS::blockDimX is the X dimension of each thread block. This must match across
all kernels launched.
- ::CUDA_LAUNCH_PARAMS::blockDimX is the Y dimension of each thread block. This must match across
all kernels launched.
- ::CUDA_LAUNCH_PARAMS::blockDimZ is the Z dimension of each thread block. This must match across
all kernels launched.
- ::CUDA_LAUNCH_PARAMS::sharedMemBytes is the dynamic shared-memory size per thread block in bytes.
This must match across all kernels launched.
- ::CUDA_LAUNCH_PARAMS::hStream is the handle to the stream to perform the launch in. This cannot
be the NULL stream or ::CU_STREAM_LEGACY or ::CU_STREAM_PER_THREAD. The CUDA context associated
with this stream must match that associated with ::CUDA_LAUNCH_PARAMS::function.
- ::CUDA_LAUNCH_PARAMS::kernelParams is an array of pointers to kernel parameters. If
::CUDA_LAUNCH_PARAMS::function has N parameters, then ::CUDA_LAUNCH_PARAMS::kernelParams
needs to be an array of N pointers. Each of ::CUDA_LAUNCH_PARAMS::kernelParams[0] through
::CUDA_LAUNCH_PARAMS::kernelParams[N-1] must point to a region of memory from which the actual
kernel parameter will be copied. The number of kernel parameters and their offsets and sizes
do not need to be specified as that information is retrieved directly from the kernel's image.
By default, the kernel won't begin execution on any GPU until all prior work in all the specified
streams has completed. This behavior can be overridden by specifying the flag
::CUDA_COOPERATIVE_LAUNCH_MULTI_DEVICE_NO_PRE_LAUNCH_SYNC. When this flag is specified, each kernel
will only wait for prior work in the stream corresponding to that GPU to complete before it begins
execution.
Similarly, by default, any subsequent work pushed in any of the specified streams will not begin
execution until the kernels on all GPUs have completed. This behavior can be overridden by specifying
the flag ::CUDA_COOPERATIVE_LAUNCH_MULTI_DEVICE_NO_POST_LAUNCH_SYNC. When this flag is specified,
any subsequent work pushed in any of the specified streams will only wait for the kernel launched
on the GPU corresponding to that stream to complete before it begins execution.
Calling ::cuLaunchCooperativeKernelMultiDevice() sets persistent function state that is
the same as function state set through ::cuLaunchKernel API when called individually for each
element in \p launchParamsList.
When kernels are launched via ::cuLaunchCooperativeKernelMultiDevice(), the previous
block shape, shared size and parameter info associated with each ::CUDA_LAUNCH_PARAMS::function
in \p launchParamsList is overwritten.
Note that to use ::cuLaunchCooperativeKernelMultiDevice(), the kernels must either have
been compiled with toolchain version 3.2 or later so that it will
contain kernel parameter information, or have no kernel parameters.
If either of these conditions is not met, then ::cuLaunchCooperativeKernelMultiDevice() will
return ::CUDA_ERROR_INVALID_IMAGE. |
static int |
cuLaunchGrid(CUfunc_st f,
int grid_width,
int grid_height)
Deprecated.
Invokes the kernel \p f on a \p grid_width x \p grid_height grid of
blocks. Each block contains the number of threads specified by a previous
call to ::cuFuncSetBlockShape().
The block shape, dynamic shared memory size, and parameter information
must be set using
::cuFuncSetBlockShape(),
::cuFuncSetSharedSize(),
::cuParamSetSize(),
::cuParamSeti(),
::cuParamSetf(), and
::cuParamSetv()
prior to calling this function.
Launching a function via ::cuLaunchKernel() invalidates the function's
block shape, dynamic shared memory size, and parameter information. After
launching via cuLaunchKernel, this state must be re-initialized prior to
calling this function. Failure to do so results in undefined behavior.
|
static int |
cuLaunchGridAsync(CUfunc_st f,
int grid_width,
int grid_height,
CUstream_st hStream)
Deprecated.
Invokes the kernel \p f on a \p grid_width x \p grid_height grid of
blocks. Each block contains the number of threads specified by a previous
call to ::cuFuncSetBlockShape().
The block shape, dynamic shared memory size, and parameter information
must be set using
::cuFuncSetBlockShape(),
::cuFuncSetSharedSize(),
::cuParamSetSize(),
::cuParamSeti(),
::cuParamSetf(), and
::cuParamSetv()
prior to calling this function.
Launching a function via ::cuLaunchKernel() invalidates the function's
block shape, dynamic shared memory size, and parameter information. After
launching via cuLaunchKernel, this state must be re-initialized prior to
calling this function. Failure to do so results in undefined behavior.
|
static int |
cuLaunchHostFunc(CUstream_st hStream,
CUhostFn fn,
Pointer userData)
\brief Enqueues a host function call in a stream
Enqueues a host function to run in a stream.
|
static int |
cuLaunchKernel(CUfunc_st f,
int gridDimX,
int gridDimY,
int gridDimZ,
int blockDimX,
int blockDimY,
int blockDimZ,
int sharedMemBytes,
CUstream_st hStream,
Pointer kernelParams,
Pointer extra) |
static int |
cuLaunchKernel(CUfunc_st f,
int gridDimX,
int gridDimY,
int gridDimZ,
int blockDimX,
int blockDimY,
int blockDimZ,
int sharedMemBytes,
CUstream_st hStream,
PointerPointer kernelParams,
PointerPointer extra)
\brief Launches a CUDA function ::CUfunction or a CUDA kernel ::CUkernel
Invokes the function ::CUfunction or the kernel ::CUkernel \p f
on a \p gridDimX x \p gridDimY x \p gridDimZ grid of blocks.
|
static int |
cuLaunchKernelEx(CUlaunchConfig config,
CUfunc_st f,
Pointer kernelParams,
Pointer extra) |
static int |
cuLaunchKernelEx(CUlaunchConfig config,
CUfunc_st f,
PointerPointer kernelParams,
PointerPointer extra)
\brief Launches a CUDA function ::CUfunction or a CUDA kernel ::CUkernel with launch-time configuration
Invokes the function ::CUfunction or the kernel ::CUkernel \p f with the specified launch-time configuration
\p config.
|
static int |
cuLibraryGetGlobal(long[] dptr,
SizeTPointer bytes,
CUlib_st library,
BytePointer name) |
static int |
cuLibraryGetGlobal(long[] dptr,
SizeTPointer bytes,
CUlib_st library,
String name) |
static int |
cuLibraryGetGlobal(LongBuffer dptr,
SizeTPointer bytes,
CUlib_st library,
BytePointer name) |
static int |
cuLibraryGetGlobal(LongBuffer dptr,
SizeTPointer bytes,
CUlib_st library,
String name) |
static int |
cuLibraryGetGlobal(LongPointer dptr,
SizeTPointer bytes,
CUlib_st library,
BytePointer name)
\brief Returns a global device pointer
Returns in \p *dptr and \p *bytes the base pointer and size of the global with
name \p name for the requested library \p library and the current context.
|
static int |
cuLibraryGetGlobal(LongPointer dptr,
SizeTPointer bytes,
CUlib_st library,
String name) |
static int |
cuLibraryGetKernel(CUkern_st pKernel,
CUlib_st library,
BytePointer name)
\brief Returns a kernel handle
Returns in \p pKernel the handle of the kernel with name \p name located in library \p library.
|
static int |
cuLibraryGetKernel(CUkern_st pKernel,
CUlib_st library,
String name) |
static int |
cuLibraryGetManaged(long[] dptr,
SizeTPointer bytes,
CUlib_st library,
BytePointer name) |
static int |
cuLibraryGetManaged(long[] dptr,
SizeTPointer bytes,
CUlib_st library,
String name) |
static int |
cuLibraryGetManaged(LongBuffer dptr,
SizeTPointer bytes,
CUlib_st library,
BytePointer name) |
static int |
cuLibraryGetManaged(LongBuffer dptr,
SizeTPointer bytes,
CUlib_st library,
String name) |
static int |
cuLibraryGetManaged(LongPointer dptr,
SizeTPointer bytes,
CUlib_st library,
BytePointer name)
\brief Returns a pointer to managed memory
Returns in \p *dptr and \p *bytes the base pointer and size of the managed memory with
name \p name for the requested library \p library.
|
static int |
cuLibraryGetManaged(LongPointer dptr,
SizeTPointer bytes,
CUlib_st library,
String name) |
static int |
cuLibraryGetModule(CUmod_st pMod,
CUlib_st library)
\brief Returns a module handle
Returns in \p pMod the module handle associated with the current context located in
library \p library.
|
static int |
cuLibraryGetUnifiedFunction(Pointer fptr,
CUlib_st library,
BytePointer symbol) |
static int |
cuLibraryGetUnifiedFunction(Pointer fptr,
CUlib_st library,
String symbol) |
static int |
cuLibraryGetUnifiedFunction(PointerPointer fptr,
CUlib_st library,
BytePointer symbol)
\brief Returns a pointer to a unified function
Returns in \p *fptr the function pointer to a unified function denoted by \p symbol.
|
static int |
cuLibraryLoadData(CUlib_st library,
Pointer code,
int[] jitOptions,
Pointer jitOptionsValues,
int numJitOptions,
int[] libraryOptions,
Pointer libraryOptionValues,
int numLibraryOptions) |
static int |
cuLibraryLoadData(CUlib_st library,
Pointer code,
IntBuffer jitOptions,
Pointer jitOptionsValues,
int numJitOptions,
IntBuffer libraryOptions,
Pointer libraryOptionValues,
int numLibraryOptions) |
static int |
cuLibraryLoadData(CUlib_st library,
Pointer code,
IntPointer jitOptions,
Pointer jitOptionsValues,
int numJitOptions,
IntPointer libraryOptions,
Pointer libraryOptionValues,
int numLibraryOptions) |
static int |
cuLibraryLoadData(CUlib_st library,
Pointer code,
IntPointer jitOptions,
PointerPointer jitOptionsValues,
int numJitOptions,
IntPointer libraryOptions,
PointerPointer libraryOptionValues,
int numLibraryOptions)
\brief Load a library with specified code and options
Takes a pointer \p code and loads the corresponding library \p library into
all contexts existent at the time of the call and future contexts at the time
of creation until the library is unloaded with ::cuLibraryUnload().
|
static int |
cuLibraryLoadFromFile(CUlib_st library,
BytePointer fileName,
int[] jitOptions,
Pointer jitOptionsValues,
int numJitOptions,
int[] libraryOptions,
Pointer libraryOptionValues,
int numLibraryOptions) |
static int |
cuLibraryLoadFromFile(CUlib_st library,
BytePointer fileName,
IntBuffer jitOptions,
Pointer jitOptionsValues,
int numJitOptions,
IntBuffer libraryOptions,
Pointer libraryOptionValues,
int numLibraryOptions) |
static int |
cuLibraryLoadFromFile(CUlib_st library,
BytePointer fileName,
IntPointer jitOptions,
Pointer jitOptionsValues,
int numJitOptions,
IntPointer libraryOptions,
Pointer libraryOptionValues,
int numLibraryOptions) |
static int |
cuLibraryLoadFromFile(CUlib_st library,
BytePointer fileName,
IntPointer jitOptions,
PointerPointer jitOptionsValues,
int numJitOptions,
IntPointer libraryOptions,
PointerPointer libraryOptionValues,
int numLibraryOptions)
\brief Load a library with specified file and options
Takes a filename \p fileName and loads the corresponding library \p library into
all contexts existent at the time of the call and future contexts at the time of
creation until the library is unloaded with ::cuLibraryUnload().
|
static int |
cuLibraryLoadFromFile(CUlib_st library,
String fileName,
int[] jitOptions,
Pointer jitOptionsValues,
int numJitOptions,
int[] libraryOptions,
Pointer libraryOptionValues,
int numLibraryOptions) |
static int |
cuLibraryLoadFromFile(CUlib_st library,
String fileName,
IntBuffer jitOptions,
Pointer jitOptionsValues,
int numJitOptions,
IntBuffer libraryOptions,
Pointer libraryOptionValues,
int numLibraryOptions) |
static int |
cuLibraryLoadFromFile(CUlib_st library,
String fileName,
IntPointer jitOptions,
Pointer jitOptionsValues,
int numJitOptions,
IntPointer libraryOptions,
Pointer libraryOptionValues,
int numLibraryOptions) |
static int |
cuLibraryUnload(CUlib_st library)
\brief Unloads a library
Unloads the library specified with \p library
|
static int |
cuLinkAddData(CUlinkState_st state,
int type,
Pointer data,
long size,
BytePointer name,
int numOptions,
int[] options,
Pointer optionValues) |
static int |
cuLinkAddData(CUlinkState_st state,
int type,
Pointer data,
long size,
BytePointer name,
int numOptions,
IntBuffer options,
Pointer optionValues) |
static int |
cuLinkAddData(CUlinkState_st state,
int type,
Pointer data,
long size,
BytePointer name,
int numOptions,
IntPointer options,
Pointer optionValues) |
static int |
cuLinkAddData(CUlinkState_st state,
int type,
Pointer data,
long size,
BytePointer name,
int numOptions,
IntPointer options,
PointerPointer optionValues)
\brief Add an input to a pending linker invocation
Ownership of \p data is retained by the caller.
|
static int |
cuLinkAddData(CUlinkState_st state,
int type,
Pointer data,
long size,
String name,
int numOptions,
int[] options,
Pointer optionValues) |
static int |
cuLinkAddData(CUlinkState_st state,
int type,
Pointer data,
long size,
String name,
int numOptions,
IntBuffer options,
Pointer optionValues) |
static int |
cuLinkAddData(CUlinkState_st state,
int type,
Pointer data,
long size,
String name,
int numOptions,
IntPointer options,
Pointer optionValues) |
static int |
cuLinkAddFile(CUlinkState_st state,
int type,
BytePointer path,
int numOptions,
int[] options,
Pointer optionValues) |
static int |
cuLinkAddFile(CUlinkState_st state,
int type,
BytePointer path,
int numOptions,
IntBuffer options,
Pointer optionValues) |
static int |
cuLinkAddFile(CUlinkState_st state,
int type,
BytePointer path,
int numOptions,
IntPointer options,
Pointer optionValues) |
static int |
cuLinkAddFile(CUlinkState_st state,
int type,
BytePointer path,
int numOptions,
IntPointer options,
PointerPointer optionValues)
\brief Add a file input to a pending linker invocation
No reference is retained to any inputs after this call returns.
|
static int |
cuLinkAddFile(CUlinkState_st state,
int type,
String path,
int numOptions,
int[] options,
Pointer optionValues) |
static int |
cuLinkAddFile(CUlinkState_st state,
int type,
String path,
int numOptions,
IntBuffer options,
Pointer optionValues) |
static int |
cuLinkAddFile(CUlinkState_st state,
int type,
String path,
int numOptions,
IntPointer options,
Pointer optionValues) |
static int |
cuLinkComplete(CUlinkState_st state,
PointerPointer cubinOut,
SizeTPointer sizeOut)
\brief Complete a pending linker invocation
Completes the pending linker action and returns the cubin image for the linked
device code, which can be used with ::cuModuleLoadData.
|
static int |
cuLinkComplete(CUlinkState_st state,
Pointer cubinOut,
SizeTPointer sizeOut) |
static int |
cuLinkCreate(int numOptions,
int[] options,
Pointer optionValues,
CUlinkState_st stateOut) |
static int |
cuLinkCreate(int numOptions,
IntBuffer options,
Pointer optionValues,
CUlinkState_st stateOut) |
static int |
cuLinkCreate(int numOptions,
IntPointer options,
Pointer optionValues,
CUlinkState_st stateOut) |
static int |
cuLinkCreate(int numOptions,
IntPointer options,
PointerPointer optionValues,
CUlinkState_st stateOut)
\brief Creates a pending JIT linker invocation.
|
static int |
cuLinkDestroy(CUlinkState_st state)
\brief Destroys state for a JIT linker invocation.
|
static int |
cuMemAddressFree(long ptr,
long size)
\brief Free an address range reservation.
|
static int |
cuMemAddressReserve(long[] ptr,
long size,
long alignment,
long addr,
long flags) |
static int |
cuMemAddressReserve(LongBuffer ptr,
long size,
long alignment,
long addr,
long flags) |
static int |
cuMemAddressReserve(LongPointer ptr,
long size,
long alignment,
long addr,
long flags)
\brief Allocate an address range reservation.
|
static int |
cuMemAdvise_v2(long devPtr,
long count,
int advice,
CUmemLocation_v1 location)
\brief Advise about the usage of a given memory range
Advise the Unified Memory subsystem about the usage pattern for the memory range
starting at \p devPtr with a size of \p count bytes.
|
static int |
cuMemAdvise(long devPtr,
long count,
int advice,
int device)
\brief Advise about the usage of a given memory range
Note there is a later version of this API, ::cuMemAdvise_v2.
|
static int |
cuMemAlloc(long[] dptr,
long bytesize) |
static int |
cuMemAlloc(LongBuffer dptr,
long bytesize) |
static int |
cuMemAlloc(LongPointer dptr,
long bytesize)
\brief Allocates device memory
Allocates \p bytesize bytes of linear memory on the device and returns in
\p *dptr a pointer to the allocated memory.
|
static int |
cuMemAllocAsync(long[] dptr,
long bytesize,
CUstream_st hStream) |
static int |
cuMemAllocAsync(LongBuffer dptr,
long bytesize,
CUstream_st hStream) |
static int |
cuMemAllocAsync(LongPointer dptr,
long bytesize,
CUstream_st hStream)
\brief Allocates memory with stream ordered semantics
Inserts an allocation operation into \p hStream.
|
static int |
cuMemAllocFromPoolAsync(long[] dptr,
long bytesize,
CUmemPoolHandle_st pool,
CUstream_st hStream) |
static int |
cuMemAllocFromPoolAsync(LongBuffer dptr,
long bytesize,
CUmemPoolHandle_st pool,
CUstream_st hStream) |
static int |
cuMemAllocFromPoolAsync(LongPointer dptr,
long bytesize,
CUmemPoolHandle_st pool,
CUstream_st hStream)
\brief Allocates memory from a specified pool with stream ordered semantics.
|
static int |
cuMemAllocHost(Pointer pp,
long bytesize) |
static int |
cuMemAllocHost(PointerPointer pp,
long bytesize)
\brief Allocates page-locked host memory
Allocates \p bytesize bytes of host memory that is page-locked and
accessible to the device.
|
static int |
cuMemAllocManaged(long[] dptr,
long bytesize,
int flags) |
static int |
cuMemAllocManaged(LongBuffer dptr,
long bytesize,
int flags) |
static int |
cuMemAllocManaged(LongPointer dptr,
long bytesize,
int flags)
\brief Allocates memory that will be automatically managed by the Unified Memory system
Allocates \p bytesize bytes of managed memory on the device and returns in
\p *dptr a pointer to the allocated memory.
|
static int |
cuMemAllocPitch(long[] dptr,
SizeTPointer pPitch,
long WidthInBytes,
long Height,
int ElementSizeBytes) |
static int |
cuMemAllocPitch(LongBuffer dptr,
SizeTPointer pPitch,
long WidthInBytes,
long Height,
int ElementSizeBytes) |
static int |
cuMemAllocPitch(LongPointer dptr,
SizeTPointer pPitch,
long WidthInBytes,
long Height,
int ElementSizeBytes)
\brief Allocates pitched device memory
Allocates at least \p WidthInBytes * \p Height bytes of linear memory on
the device and returns in \p *dptr a pointer to the allocated memory.
|
static int |
cuMemcpy(long dst,
long src,
long ByteCount)
\brief Copies memory
Copies data between two pointers.
|
static int |
cuMemcpy2D(CUDA_MEMCPY2D_v2 pCopy)
\brief Copies memory for 2D arrays
Perform a 2D memory copy according to the parameters specified in \p pCopy.
|
static int |
cuMemcpy2DAsync(CUDA_MEMCPY2D_v2 pCopy,
CUstream_st hStream)
\brief Copies memory for 2D arrays
Perform a 2D memory copy according to the parameters specified in \p pCopy.
|
static int |
cuMemcpy2DUnaligned(CUDA_MEMCPY2D_v2 pCopy)
\brief Copies memory for 2D arrays
Perform a 2D memory copy according to the parameters specified in \p pCopy.
|
static int |
cuMemcpy3D(CUDA_MEMCPY3D_v2 pCopy)
\brief Copies memory for 3D arrays
Perform a 3D memory copy according to the parameters specified in
\p pCopy.
|
static int |
cuMemcpy3DAsync(CUDA_MEMCPY3D_v2 pCopy,
CUstream_st hStream)
\brief Copies memory for 3D arrays
Perform a 3D memory copy according to the parameters specified in
\p pCopy.
|
static int |
cuMemcpy3DPeer(CUDA_MEMCPY3D_PEER_v1 pCopy)
\brief Copies memory between contexts
Perform a 3D memory copy according to the parameters specified in
\p pCopy.
|
static int |
cuMemcpy3DPeerAsync(CUDA_MEMCPY3D_PEER_v1 pCopy,
CUstream_st hStream)
\brief Copies memory between contexts asynchronously.
|
static int |
cuMemcpyAsync(long dst,
long src,
long ByteCount,
CUstream_st hStream)
\brief Copies memory asynchronously
Copies data between two pointers.
|
static int |
cuMemcpyAtoA(CUarray_st dstArray,
long dstOffset,
CUarray_st srcArray,
long srcOffset,
long ByteCount)
\brief Copies memory from Array to Array
Copies from one 1D CUDA array to another.
|
static int |
cuMemcpyAtoD(long dstDevice,
CUarray_st srcArray,
long srcOffset,
long ByteCount)
\brief Copies memory from Array to Device
Copies from one 1D CUDA array to device memory.
|
static int |
cuMemcpyAtoH(Pointer dstHost,
CUarray_st srcArray,
long srcOffset,
long ByteCount)
\brief Copies memory from Array to Host
Copies from one 1D CUDA array to host memory.
|
static int |
cuMemcpyAtoHAsync(Pointer dstHost,
CUarray_st srcArray,
long srcOffset,
long ByteCount,
CUstream_st hStream)
\brief Copies memory from Array to Host
Copies from one 1D CUDA array to host memory.
|
static int |
cuMemcpyDtoA(CUarray_st dstArray,
long dstOffset,
long srcDevice,
long ByteCount)
\brief Copies memory from Device to Array
Copies from device memory to a 1D CUDA array.
|
static int |
cuMemcpyDtoD(long dstDevice,
long srcDevice,
long ByteCount)
\brief Copies memory from Device to Device
Copies from device memory to device memory.
|
static int |
cuMemcpyDtoDAsync(long dstDevice,
long srcDevice,
long ByteCount,
CUstream_st hStream)
\brief Copies memory from Device to Device
Copies from device memory to device memory.
|
static int |
cuMemcpyDtoH(Pointer dstHost,
long srcDevice,
long ByteCount)
\brief Copies memory from Device to Host
Copies from device to host memory.
|
static int |
cuMemcpyDtoHAsync(Pointer dstHost,
long srcDevice,
long ByteCount,
CUstream_st hStream)
\brief Copies memory from Device to Host
Copies from device to host memory.
|
static int |
cuMemcpyHtoA(CUarray_st dstArray,
long dstOffset,
Pointer srcHost,
long ByteCount)
\brief Copies memory from Host to Array
Copies from host memory to a 1D CUDA array.
|
static int |
cuMemcpyHtoAAsync(CUarray_st dstArray,
long dstOffset,
Pointer srcHost,
long ByteCount,
CUstream_st hStream)
\brief Copies memory from Host to Array
Copies from host memory to a 1D CUDA array.
|
static int |
cuMemcpyHtoD(long dstDevice,
Pointer srcHost,
long ByteCount)
\brief Copies memory from Host to Device
Copies from host memory to device memory.
|
static int |
cuMemcpyHtoDAsync(long dstDevice,
Pointer srcHost,
long ByteCount,
CUstream_st hStream)
\brief Copies memory from Host to Device
Copies from host memory to device memory.
|
static int |
cuMemcpyPeer(long dstDevice,
CUctx_st dstContext,
long srcDevice,
CUctx_st srcContext,
long ByteCount)
\brief Copies device memory between two contexts
Copies from device memory in one context to device memory in another
context.
|
static int |
cuMemcpyPeerAsync(long dstDevice,
CUctx_st dstContext,
long srcDevice,
CUctx_st srcContext,
long ByteCount,
CUstream_st hStream)
\brief Copies device memory between two contexts asynchronously.
|
static int |
cuMemCreate(long[] handle,
long size,
CUmemAllocationProp_v1 prop,
long flags) |
static int |
cuMemCreate(LongBuffer handle,
long size,
CUmemAllocationProp_v1 prop,
long flags) |
static int |
cuMemCreate(LongPointer handle,
long size,
CUmemAllocationProp_v1 prop,
long flags)
\brief Create a CUDA memory handle representing a memory allocation of a given size described by the given properties
This creates a memory allocation on the target device specified through the
\p prop structure.
|
static int |
cuMemExportToShareableHandle(Pointer shareableHandle,
long handle,
int handleType,
long flags)
\brief Exports an allocation to a requested shareable handle type
Given a CUDA memory handle, create a shareable memory
allocation handle that can be used to share the memory with other
processes.
|
static int |
cuMemFree(long dptr)
\brief Frees device memory
Frees the memory space pointed to by \p dptr, which must have been returned
by a previous call to one of the following memory allocation APIs - ::cuMemAlloc(),
::cuMemAllocPitch(), ::cuMemAllocManaged(), ::cuMemAllocAsync(), ::cuMemAllocFromPoolAsync()
Note - This API will not perform any implict synchronization when the pointer was allocated with
::cuMemAllocAsync or ::cuMemAllocFromPoolAsync.
|
static int |
cuMemFreeAsync(long dptr,
CUstream_st hStream)
\brief Frees memory with stream ordered semantics
Inserts a free operation into \p hStream.
|
static int |
cuMemFreeHost(Pointer p)
\brief Frees page-locked host memory
Frees the memory space pointed to by \p p, which must have been returned by
a previous call to ::cuMemAllocHost().
|
static int |
cuMemGetAccess(long[] flags,
CUmemLocation_v1 location,
long ptr) |
static int |
cuMemGetAccess(LongBuffer flags,
CUmemLocation_v1 location,
long ptr) |
static int |
cuMemGetAccess(LongPointer flags,
CUmemLocation_v1 location,
long ptr)
\brief Get the access \p flags set for the given \p location and \p ptr
|
static int |
cuMemGetAddressRange(long[] pbase,
SizeTPointer psize,
long dptr) |
static int |
cuMemGetAddressRange(LongBuffer pbase,
SizeTPointer psize,
long dptr) |
static int |
cuMemGetAddressRange(LongPointer pbase,
SizeTPointer psize,
long dptr)
\brief Get information on memory allocations
Returns the base address in \p *pbase and size in \p *psize of the
allocation by ::cuMemAlloc() or ::cuMemAllocPitch() that contains the input
pointer \p dptr.
|
static int |
cuMemGetAllocationGranularity(SizeTPointer granularity,
CUmemAllocationProp_v1 prop,
int option)
\brief Calculates either the minimal or recommended granularity
Calculates either the minimal or recommended granularity
for a given allocation specification and returns it in granularity.
|
static int |
cuMemGetAllocationPropertiesFromHandle(CUmemAllocationProp_v1 prop,
long handle)
\brief Retrieve the contents of the property structure defining properties for this handle
|
static int |
cuMemGetHandleForAddressRange(Pointer handle,
long dptr,
long size,
int handleType,
long flags)
\brief Retrieve handle for an address range
Get a handle of the specified type to an address range.
|
static int |
cuMemGetInfo(SizeTPointer _free,
SizeTPointer total)
\brief Gets free and total memory
Returns in \p *total the total amount of memory available to the the current context.
|
static int |
cuMemHostAlloc(Pointer pp,
long bytesize,
int Flags) |
static int |
cuMemHostAlloc(PointerPointer pp,
long bytesize,
int Flags)
\brief Allocates page-locked host memory
Allocates \p bytesize bytes of host memory that is page-locked and accessible
to the device.
|
static int |
cuMemHostGetDevicePointer(long[] pdptr,
Pointer p,
int Flags) |
static int |
cuMemHostGetDevicePointer(LongBuffer pdptr,
Pointer p,
int Flags) |
static int |
cuMemHostGetDevicePointer(LongPointer pdptr,
Pointer p,
int Flags)
\brief Passes back device pointer of mapped pinned memory
Passes back the device pointer \p pdptr corresponding to the mapped, pinned
host buffer \p p allocated by ::cuMemHostAlloc.
|
static int |
cuMemHostGetFlags(int[] pFlags,
Pointer p) |
static int |
cuMemHostGetFlags(IntBuffer pFlags,
Pointer p) |
static int |
cuMemHostGetFlags(IntPointer pFlags,
Pointer p)
\brief Passes back flags that were used for a pinned allocation
Passes back the flags \p pFlags that were specified when allocating
the pinned host buffer \p p allocated by ::cuMemHostAlloc.
|
static int |
cuMemHostRegister(Pointer p,
long bytesize,
int Flags)
\brief Registers an existing host memory range for use by CUDA
Page-locks the memory range specified by \p p and \p bytesize and maps it
for the device(s) as specified by \p Flags.
|
static int |
cuMemHostUnregister(Pointer p)
\brief Unregisters a memory range that was registered with cuMemHostRegister.
|
static int |
cuMemImportFromShareableHandle(long[] handle,
Pointer osHandle,
int shHandleType) |
static int |
cuMemImportFromShareableHandle(LongBuffer handle,
Pointer osHandle,
int shHandleType) |
static int |
cuMemImportFromShareableHandle(LongPointer handle,
Pointer osHandle,
int shHandleType)
\brief Imports an allocation from a requested shareable handle type.
|
static int |
cuMemMap(long ptr,
long size,
long offset,
long handle,
long flags)
\brief Maps an allocation handle to a reserved virtual address range.
|
static int |
cuMemMapArrayAsync(CUarrayMapInfo_v1 mapInfoList,
int count,
CUstream_st hStream)
\brief Maps or unmaps subregions of sparse CUDA arrays and sparse CUDA mipmapped arrays
Performs map or unmap operations on subregions of sparse CUDA arrays and sparse CUDA mipmapped arrays.
|
static int |
cuMemPoolCreate(CUmemPoolHandle_st pool,
CUmemPoolProps_v1 poolProps)
\brief Creates a memory pool
Creates a CUDA memory pool and returns the handle in \p pool.
|
static int |
cuMemPoolDestroy(CUmemPoolHandle_st pool)
\brief Destroys the specified memory pool
If any pointers obtained from this pool haven't been freed or
the pool has free operations that haven't completed
when ::cuMemPoolDestroy is invoked, the function will return immediately and the
resources associated with the pool will be released automatically
once there are no more outstanding allocations.
|
static int |
cuMemPoolExportPointer(CUmemPoolPtrExportData_v1 shareData_out,
long ptr)
\brief Export data to share a memory pool allocation between processes.
|
static int |
cuMemPoolExportToShareableHandle(Pointer handle_out,
CUmemPoolHandle_st pool,
int handleType,
long flags)
\brief Exports a memory pool to the requested handle type.
|
static int |
cuMemPoolGetAccess(int[] flags,
CUmemPoolHandle_st memPool,
CUmemLocation_v1 location) |
static int |
cuMemPoolGetAccess(IntBuffer flags,
CUmemPoolHandle_st memPool,
CUmemLocation_v1 location) |
static int |
cuMemPoolGetAccess(IntPointer flags,
CUmemPoolHandle_st memPool,
CUmemLocation_v1 location)
\brief Returns the accessibility of a pool from a device
Returns the accessibility of the pool's memory from the specified location.
|
static int |
cuMemPoolGetAttribute(CUmemPoolHandle_st pool,
int attr,
Pointer value)
\brief Gets attributes of a memory pool
Supported attributes are:
- ::CU_MEMPOOL_ATTR_RELEASE_THRESHOLD: (value type = cuuint64_t)
Amount of reserved memory in bytes to hold onto before trying
to release memory back to the OS.
|
static int |
cuMemPoolImportFromShareableHandle(CUmemPoolHandle_st pool_out,
Pointer handle,
int handleType,
long flags)
\brief imports a memory pool from a shared handle.
|
static int |
cuMemPoolImportPointer(long[] ptr_out,
CUmemPoolHandle_st pool,
CUmemPoolPtrExportData_v1 shareData) |
static int |
cuMemPoolImportPointer(LongBuffer ptr_out,
CUmemPoolHandle_st pool,
CUmemPoolPtrExportData_v1 shareData) |
static int |
cuMemPoolImportPointer(LongPointer ptr_out,
CUmemPoolHandle_st pool,
CUmemPoolPtrExportData_v1 shareData)
\brief Import a memory pool allocation from another process.
|
static int |
cuMemPoolSetAccess(CUmemPoolHandle_st pool,
CUmemAccessDesc_v1 map,
long count)
\brief Controls visibility of pools between devices
|
static int |
cuMemPoolSetAttribute(CUmemPoolHandle_st pool,
int attr,
Pointer value)
\brief Sets attributes of a memory pool
Supported attributes are:
- ::CU_MEMPOOL_ATTR_RELEASE_THRESHOLD: (value type = cuuint64_t)
Amount of reserved memory in bytes to hold onto before trying
to release memory back to the OS.
|
static int |
cuMemPoolTrimTo(CUmemPoolHandle_st pool,
long minBytesToKeep)
\brief Tries to release memory back to the OS
Releases memory back to the OS until the pool contains fewer than minBytesToKeep
reserved bytes, or there is no more memory that the allocator can safely release.
|
static int |
cuMemPrefetchAsync_v2(long devPtr,
long count,
CUmemLocation_v1 location,
int flags,
CUstream_st hStream)
\brief Prefetches memory to the specified destination location
Prefetches memory to the specified destination location.
|
static int |
cuMemPrefetchAsync(long devPtr,
long count,
int dstDevice,
CUstream_st hStream)
\brief Prefetches memory to the specified destination device
Note there is a later version of this API, ::cuMemPrefetchAsync_v2.
|
static int |
cuMemRelease(long handle)
\brief Release a memory handle representing a memory allocation which was previously allocated through cuMemCreate.
|
static int |
cuMemSetAccess(long ptr,
long size,
CUmemAccessDesc_v1 desc,
long count)
\brief Set the access flags for each location specified in \p desc for the given virtual address range
Given the virtual address range via \p ptr and \p size, and the locations
in the array given by \p desc and \p count, set the access flags for the
target locations.
|
static int |
cuMemsetD16(long dstDevice,
short us,
long N)
\brief Initializes device memory
Sets the memory range of \p N 16-bit values to the specified value
\p us.
|
static int |
cuMemsetD16Async(long dstDevice,
short us,
long N,
CUstream_st hStream)
\brief Sets device memory
Sets the memory range of \p N 16-bit values to the specified value
\p us.
|
static int |
cuMemsetD2D16(long dstDevice,
long dstPitch,
short us,
long Width,
long Height)
\brief Initializes device memory
Sets the 2D memory range of \p Width 16-bit values to the specified value
\p us.
|
static int |
cuMemsetD2D16Async(long dstDevice,
long dstPitch,
short us,
long Width,
long Height,
CUstream_st hStream)
\brief Sets device memory
Sets the 2D memory range of \p Width 16-bit values to the specified value
\p us.
|
static int |
cuMemsetD2D32(long dstDevice,
long dstPitch,
int ui,
long Width,
long Height)
\brief Initializes device memory
Sets the 2D memory range of \p Width 32-bit values to the specified value
\p ui.
|
static int |
cuMemsetD2D32Async(long dstDevice,
long dstPitch,
int ui,
long Width,
long Height,
CUstream_st hStream)
\brief Sets device memory
Sets the 2D memory range of \p Width 32-bit values to the specified value
\p ui.
|
static int |
cuMemsetD2D8(long dstDevice,
long dstPitch,
byte uc,
long Width,
long Height)
\brief Initializes device memory
Sets the 2D memory range of \p Width 8-bit values to the specified value
\p uc.
|
static int |
cuMemsetD2D8Async(long dstDevice,
long dstPitch,
byte uc,
long Width,
long Height,
CUstream_st hStream)
\brief Sets device memory
Sets the 2D memory range of \p Width 8-bit values to the specified value
\p uc.
|
static int |
cuMemsetD32(long dstDevice,
int ui,
long N)
\brief Initializes device memory
Sets the memory range of \p N 32-bit values to the specified value
\p ui.
|
static int |
cuMemsetD32Async(long dstDevice,
int ui,
long N,
CUstream_st hStream)
\brief Sets device memory
Sets the memory range of \p N 32-bit values to the specified value
\p ui.
|
static int |
cuMemsetD8(long dstDevice,
byte uc,
long N)
\brief Initializes device memory
Sets the memory range of \p N 8-bit values to the specified value
\p uc.
|
static int |
cuMemsetD8Async(long dstDevice,
byte uc,
long N,
CUstream_st hStream)
\brief Sets device memory
Sets the memory range of \p N 8-bit values to the specified value
\p uc.
|
static int |
cuMemUnmap(long ptr,
long size)
\brief Unmap the backing memory of a given address range.
|
static int |
cuMipmappedArrayCreate(CUmipmappedArray_st pHandle,
CUDA_ARRAY3D_DESCRIPTOR_v2 pMipmappedArrayDesc,
int numMipmapLevels)
\brief Creates a CUDA mipmapped array
Creates a CUDA mipmapped array according to the ::CUDA_ARRAY3D_DESCRIPTOR structure
\p pMipmappedArrayDesc and returns a handle to the new CUDA mipmapped array in \p *pHandle.
|
static int |
cuMipmappedArrayDestroy(CUmipmappedArray_st hMipmappedArray)
\brief Destroys a CUDA mipmapped array
Destroys the CUDA mipmapped array \p hMipmappedArray.
|
static int |
cuMipmappedArrayGetLevel(CUarray_st pLevelArray,
CUmipmappedArray_st hMipmappedArray,
int level)
\brief Gets a mipmap level of a CUDA mipmapped array
Returns in \p *pLevelArray a CUDA array that represents a single mipmap level
of the CUDA mipmapped array \p hMipmappedArray.
|
static int |
cuMipmappedArrayGetMemoryRequirements(CUDA_ARRAY_MEMORY_REQUIREMENTS_v1 memoryRequirements,
CUmipmappedArray_st mipmap,
int device)
\brief Returns the memory requirements of a CUDA mipmapped array
Returns the memory requirements of a CUDA mipmapped array in \p memoryRequirements
If the CUDA mipmapped array is not allocated with flag ::CUDA_ARRAY3D_DEFERRED_MAPPING
::CUDA_ERROR_INVALID_VALUE will be returned.
|
static int |
cuMipmappedArrayGetSparseProperties(CUDA_ARRAY_SPARSE_PROPERTIES_v1 sparseProperties,
CUmipmappedArray_st mipmap)
\brief Returns the layout properties of a sparse CUDA mipmapped array
Returns the sparse array layout properties in \p sparseProperties
If the CUDA mipmapped array is not allocated with flag ::CUDA_ARRAY3D_SPARSE
::CUDA_ERROR_INVALID_VALUE will be returned.
|
static int |
cuModuleGetFunction(CUfunc_st hfunc,
CUmod_st hmod,
BytePointer name)
\brief Returns a function handle
Returns in \p *hfunc the handle of the function of name \p name located in
module \p hmod.
|
static int |
cuModuleGetFunction(CUfunc_st hfunc,
CUmod_st hmod,
String name) |
static int |
cuModuleGetGlobal(long[] dptr,
SizeTPointer bytes,
CUmod_st hmod,
BytePointer name) |
static int |
cuModuleGetGlobal(long[] dptr,
SizeTPointer bytes,
CUmod_st hmod,
String name) |
static int |
cuModuleGetGlobal(LongBuffer dptr,
SizeTPointer bytes,
CUmod_st hmod,
BytePointer name) |
static int |
cuModuleGetGlobal(LongBuffer dptr,
SizeTPointer bytes,
CUmod_st hmod,
String name) |
static int |
cuModuleGetGlobal(LongPointer dptr,
SizeTPointer bytes,
CUmod_st hmod,
BytePointer name)
\brief Returns a global pointer from a module
Returns in \p *dptr and \p *bytes the base pointer and size of the
global of name \p name located in module \p hmod.
|
static int |
cuModuleGetGlobal(LongPointer dptr,
SizeTPointer bytes,
CUmod_st hmod,
String name) |
static int |
cuModuleGetLoadingMode(int[] mode) |
static int |
cuModuleGetLoadingMode(IntBuffer mode) |
static int |
cuModuleGetLoadingMode(IntPointer mode)
\brief Query lazy loading mode
Returns lazy loading mode
Module loading mode is controlled by CUDA_MODULE_LOADING env variable
|
static int |
cuModuleGetSurfRef(CUsurfref_st pSurfRef,
CUmod_st hmod,
BytePointer name)
Deprecated.
Returns in \p *pSurfRef the handle of the surface reference of name \p name
in the module \p hmod. If no surface reference of that name exists,
::cuModuleGetSurfRef() returns ::CUDA_ERROR_NOT_FOUND.
|
static int |
cuModuleGetSurfRef(CUsurfref_st pSurfRef,
CUmod_st hmod,
String name)
Deprecated.
|
static int |
cuModuleGetTexRef(CUtexref_st pTexRef,
CUmod_st hmod,
BytePointer name)
Deprecated.
Returns in \p *pTexRef the handle of the texture reference of name \p name
in the module \p hmod. If no texture reference of that name exists,
::cuModuleGetTexRef() returns ::CUDA_ERROR_NOT_FOUND. This texture reference
handle should not be destroyed, since it will be destroyed when the module
is unloaded.
|
static int |
cuModuleGetTexRef(CUtexref_st pTexRef,
CUmod_st hmod,
String name)
Deprecated.
|
static int |
cuModuleLoad(CUmod_st module,
BytePointer fname)
\brief Loads a compute module
Takes a filename \p fname and loads the corresponding module \p module into
the current context.
|
static int |
cuModuleLoad(CUmod_st module,
String fname) |
static int |
cuModuleLoadData(CUmod_st module,
Pointer image)
\brief Load a module's data
Takes a pointer \p image and loads the corresponding module \p module into
the current context.
|
static int |
cuModuleLoadDataEx(CUmod_st module,
Pointer image,
int numOptions,
int[] options,
Pointer optionValues) |
static int |
cuModuleLoadDataEx(CUmod_st module,
Pointer image,
int numOptions,
IntBuffer options,
Pointer optionValues) |
static int |
cuModuleLoadDataEx(CUmod_st module,
Pointer image,
int numOptions,
IntPointer options,
Pointer optionValues) |
static int |
cuModuleLoadDataEx(CUmod_st module,
Pointer image,
int numOptions,
IntPointer options,
PointerPointer optionValues)
\brief Load a module's data with options
Takes a pointer \p image and loads the corresponding module \p module into
the current context.
|
static int |
cuModuleLoadFatBinary(CUmod_st module,
Pointer fatCubin)
\brief Load a module's data
Takes a pointer \p fatCubin and loads the corresponding module \p module
into the current context.
|
static int |
cuModuleUnload(CUmod_st hmod)
\brief Unloads a module
Unloads a module \p hmod from the current context.
|
static int |
cuMulticastAddDevice(long mcHandle,
int dev)
\brief Associate a device to a multicast object.
|
static int |
cuMulticastBindAddr(long mcHandle,
long mcOffset,
long memptr,
long size,
long flags)
\brief Bind a memory allocation represented by a virtual address to a multicast object.
|
static int |
cuMulticastBindMem(long mcHandle,
long mcOffset,
long memHandle,
long memOffset,
long size,
long flags)
\brief Bind a memory allocation represented by a handle to a multicast object.
|
static int |
cuMulticastCreate(long[] mcHandle,
CUmulticastObjectProp_v1 prop) |
static int |
cuMulticastCreate(LongBuffer mcHandle,
CUmulticastObjectProp_v1 prop) |
static int |
cuMulticastCreate(LongPointer mcHandle,
CUmulticastObjectProp_v1 prop)
\brief Create a generic allocation handle representing a multicast object described by the given properties.
|
static int |
cuMulticastGetGranularity(SizeTPointer granularity,
CUmulticastObjectProp_v1 prop,
int option)
\brief Calculates either the minimal or recommended granularity for multicast object
Calculates either the minimal or recommended granularity for a given set of
multicast object properties and returns it in granularity.
|
static int |
cuMulticastUnbind(long mcHandle,
int dev,
long mcOffset,
long size)
\brief Unbind any memory allocations bound to a multicast object at a given offset and upto a given size.
|
static int |
cuOccupancyAvailableDynamicSMemPerBlock(SizeTPointer dynamicSmemSize,
CUfunc_st func,
int numBlocks,
int blockSize)
\brief Returns dynamic shared memory available per block when launching \p numBlocks blocks on SM
Returns in \p *dynamicSmemSize the maximum size of dynamic shared memory to allow \p numBlocks blocks per SM.
|
static int |
cuOccupancyMaxActiveBlocksPerMultiprocessor(int[] numBlocks,
CUfunc_st func,
int blockSize,
long dynamicSMemSize) |
static int |
cuOccupancyMaxActiveBlocksPerMultiprocessor(IntBuffer numBlocks,
CUfunc_st func,
int blockSize,
long dynamicSMemSize) |
static int |
cuOccupancyMaxActiveBlocksPerMultiprocessor(IntPointer numBlocks,
CUfunc_st func,
int blockSize,
long dynamicSMemSize)
\brief Returns occupancy of a function
Returns in \p *numBlocks the number of the maximum active blocks per
streaming multiprocessor.
|
static int |
cuOccupancyMaxActiveBlocksPerMultiprocessorWithFlags(int[] numBlocks,
CUfunc_st func,
int blockSize,
long dynamicSMemSize,
int flags) |
static int |
cuOccupancyMaxActiveBlocksPerMultiprocessorWithFlags(IntBuffer numBlocks,
CUfunc_st func,
int blockSize,
long dynamicSMemSize,
int flags) |
static int |
cuOccupancyMaxActiveBlocksPerMultiprocessorWithFlags(IntPointer numBlocks,
CUfunc_st func,
int blockSize,
long dynamicSMemSize,
int flags)
\brief Returns occupancy of a function
Returns in \p *numBlocks the number of the maximum active blocks per
streaming multiprocessor.
|
static int |
cuOccupancyMaxActiveClusters(int[] numClusters,
CUfunc_st func,
CUlaunchConfig config) |
static int |
cuOccupancyMaxActiveClusters(IntBuffer numClusters,
CUfunc_st func,
CUlaunchConfig config) |
static int |
cuOccupancyMaxActiveClusters(IntPointer numClusters,
CUfunc_st func,
CUlaunchConfig config)
\brief Given the kernel function (\p func) and launch configuration
(\p config), return the maximum number of clusters that could co-exist
on the target device in \p *numClusters.
|
static int |
cuOccupancyMaxPotentialBlockSize(int[] minGridSize,
int[] blockSize,
CUfunc_st func,
CUoccupancyB2DSize blockSizeToDynamicSMemSize,
long dynamicSMemSize,
int blockSizeLimit) |
static int |
cuOccupancyMaxPotentialBlockSize(IntBuffer minGridSize,
IntBuffer blockSize,
CUfunc_st func,
CUoccupancyB2DSize blockSizeToDynamicSMemSize,
long dynamicSMemSize,
int blockSizeLimit) |
static int |
cuOccupancyMaxPotentialBlockSize(IntPointer minGridSize,
IntPointer blockSize,
CUfunc_st func,
CUoccupancyB2DSize blockSizeToDynamicSMemSize,
long dynamicSMemSize,
int blockSizeLimit)
\brief Suggest a launch configuration with reasonable occupancy
Returns in \p *blockSize a reasonable block size that can achieve
the maximum occupancy (or, the maximum number of active warps with
the fewest blocks per multiprocessor), and in \p *minGridSize the
minimum grid size to achieve the maximum occupancy.
|
static int |
cuOccupancyMaxPotentialBlockSizeWithFlags(int[] minGridSize,
int[] blockSize,
CUfunc_st func,
CUoccupancyB2DSize blockSizeToDynamicSMemSize,
long dynamicSMemSize,
int blockSizeLimit,
int flags) |
static int |
cuOccupancyMaxPotentialBlockSizeWithFlags(IntBuffer minGridSize,
IntBuffer blockSize,
CUfunc_st func,
CUoccupancyB2DSize blockSizeToDynamicSMemSize,
long dynamicSMemSize,
int blockSizeLimit,
int flags) |
static int |
cuOccupancyMaxPotentialBlockSizeWithFlags(IntPointer minGridSize,
IntPointer blockSize,
CUfunc_st func,
CUoccupancyB2DSize blockSizeToDynamicSMemSize,
long dynamicSMemSize,
int blockSizeLimit,
int flags)
\brief Suggest a launch configuration with reasonable occupancy
An extended version of ::cuOccupancyMaxPotentialBlockSize.
|
static int |
cuOccupancyMaxPotentialClusterSize(int[] clusterSize,
CUfunc_st func,
CUlaunchConfig config) |
static int |
cuOccupancyMaxPotentialClusterSize(IntBuffer clusterSize,
CUfunc_st func,
CUlaunchConfig config) |
static int |
cuOccupancyMaxPotentialClusterSize(IntPointer clusterSize,
CUfunc_st func,
CUlaunchConfig config)
\brief Given the kernel function (\p func) and launch configuration
(\p config), return the maximum cluster size in \p *clusterSize.
|
static int |
cuParamSetf(CUfunc_st hfunc,
int offset,
float value)
Deprecated.
Sets a floating-point parameter that will be specified the next time the
kernel corresponding to \p hfunc will be invoked. \p offset is a byte offset.
|
static int |
cuParamSeti(CUfunc_st hfunc,
int offset,
int value)
Deprecated.
Sets an integer parameter that will be specified the next time the
kernel corresponding to \p hfunc will be invoked. \p offset is a byte offset.
|
static int |
cuParamSetSize(CUfunc_st hfunc,
int numbytes)
Deprecated.
Sets through \p numbytes the total size in bytes needed by the function
parameters of the kernel corresponding to \p hfunc.
|
static int |
cuParamSetTexRef(CUfunc_st hfunc,
int texunit,
CUtexref_st hTexRef)
Deprecated.
Makes the CUDA array or linear memory bound to the texture reference
\p hTexRef available to a device program as a texture. In this version of
CUDA, the texture-reference must be obtained via ::cuModuleGetTexRef() and
the \p texunit parameter must be set to ::CU_PARAM_TR_DEFAULT.
|
static int |
cuParamSetv(CUfunc_st hfunc,
int offset,
Pointer ptr,
int numbytes)
Deprecated.
Copies an arbitrary amount of data (specified in \p numbytes) from \p ptr
into the parameter space of the kernel corresponding to \p hfunc. \p offset
is a byte offset.
|
static int |
cuPointerGetAttribute(Pointer data,
int attribute,
long ptr)
\brief Returns information about a pointer
The supported attributes are:
- ::CU_POINTER_ATTRIBUTE_CONTEXT:
Returns in \p *data the ::CUcontext in which \p ptr was allocated or
registered.
|
static int |
cuPointerGetAttributes(int numAttributes,
int[] attributes,
Pointer data,
long ptr) |
static int |
cuPointerGetAttributes(int numAttributes,
IntBuffer attributes,
Pointer data,
long ptr) |
static int |
cuPointerGetAttributes(int numAttributes,
IntPointer attributes,
Pointer data,
long ptr) |
static int |
cuPointerGetAttributes(int numAttributes,
IntPointer attributes,
PointerPointer data,
long ptr)
\brief Returns information about a pointer.
|
static int |
cuPointerSetAttribute(Pointer value,
int attribute,
long ptr)
\brief Set attributes on a previously allocated memory region
The supported attributes are:
- ::CU_POINTER_ATTRIBUTE_SYNC_MEMOPS:
A boolean attribute that can either be set (1) or unset (0).
|
static int |
cuSignalExternalSemaphoresAsync(CUextSemaphore_st extSemArray,
CUDA_EXTERNAL_SEMAPHORE_SIGNAL_PARAMS_v1 paramsArray,
int numExtSems,
CUstream_st stream)
\brief Signals a set of external semaphore objects
Enqueues a signal operation on a set of externally allocated
semaphore object in the specified stream.
|
static int |
cuStreamAddCallback(CUstream_st hStream,
CUstreamCallback callback,
Pointer userData,
int flags)
\brief Add a callback to a compute stream
\note This function is slated for eventual deprecation and removal.
|
static int |
cuStreamAttachMemAsync(CUstream_st hStream,
long dptr,
long length,
int flags)
\brief Attach memory to a stream asynchronously
Enqueues an operation in \p hStream to specify stream association of
\p length bytes of memory starting from \p dptr.
|
static int |
cuStreamBatchMemOp(CUstream_st stream,
int count,
CUstreamBatchMemOpParams_v1 paramArray,
int flags)
\brief Batch operations to synchronize the stream via memory operations
This is a batch version of ::cuStreamWaitValue32() and ::cuStreamWriteValue32().
|
static int |
cuStreamBeginCapture(CUstream_st hStream,
int mode)
\brief Begins graph capture on a stream
Begin graph capture on \p hStream.
|
static int |
cuStreamBeginCaptureToGraph(CUstream_st hStream,
CUgraph_st hGraph,
CUgraphNode_st dependencies,
CUgraphEdgeData dependencyData,
long numDependencies,
int mode)
\brief Begins graph capture on a stream to an existing graph
Begin graph capture on \p hStream, placing new nodes into an existing graph.
|
static int |
cuStreamCopyAttributes(CUstream_st dst,
CUstream_st src)
\brief Copies attributes from source stream to destination stream.
|
static int |
cuStreamCreate(CUstream_st phStream,
int Flags)
\brief Create a stream
Creates a stream and returns a handle in \p phStream.
|
static int |
cuStreamCreateWithPriority(CUstream_st phStream,
int flags,
int priority)
\brief Create a stream with the given priority
Creates a stream with the specified priority and returns a handle in \p phStream.
|
static int |
cuStreamDestroy(CUstream_st hStream)
\brief Destroys a stream
Destroys the stream specified by \p hStream.
|
static int |
cuStreamEndCapture(CUstream_st hStream,
CUgraph_st phGraph)
\brief Ends capture on a stream, returning the captured graph
End capture on \p hStream, returning the captured graph via \p phGraph.
|
static int |
cuStreamGetAttribute(CUstream_st hStream,
int attr,
CUlaunchAttributeValue value_out)
\brief Queries stream attribute.
|
static int |
cuStreamGetCaptureInfo_v3(CUstream_st hStream,
int[] captureStatus_out,
long[] id_out,
CUgraph_st graph_out,
PointerPointer dependencies_out,
CUgraphEdgeData edgeData_out,
SizeTPointer numDependencies_out) |
static int |
cuStreamGetCaptureInfo_v3(CUstream_st hStream,
IntBuffer captureStatus_out,
LongBuffer id_out,
CUgraph_st graph_out,
PointerPointer dependencies_out,
CUgraphEdgeData edgeData_out,
SizeTPointer numDependencies_out) |
static int |
cuStreamGetCaptureInfo_v3(CUstream_st hStream,
IntPointer captureStatus_out,
LongPointer id_out,
CUgraph_st graph_out,
PointerPointer dependencies_out,
CUgraphEdgeData edgeData_out,
SizeTPointer numDependencies_out) |
static int |
cuStreamGetCaptureInfo_v3(CUstream_st hStream,
IntPointer captureStatus_out,
LongPointer id_out,
CUgraph_st graph_out,
PointerPointer dependencies_out,
PointerPointer edgeData_out,
SizeTPointer numDependencies_out)
\brief Query a stream's capture state (12.3+)
Query stream state related to stream capture.
|
static int |
cuStreamGetCaptureInfo(CUstream_st hStream,
int[] captureStatus_out,
long[] id_out,
CUgraph_st graph_out,
PointerPointer dependencies_out,
SizeTPointer numDependencies_out) |
static int |
cuStreamGetCaptureInfo(CUstream_st hStream,
IntBuffer captureStatus_out,
LongBuffer id_out,
CUgraph_st graph_out,
PointerPointer dependencies_out,
SizeTPointer numDependencies_out) |
static int |
cuStreamGetCaptureInfo(CUstream_st hStream,
IntPointer captureStatus_out,
LongPointer id_out,
CUgraph_st graph_out,
PointerPointer dependencies_out,
SizeTPointer numDependencies_out)
\brief Query a stream's capture state
Query stream state related to stream capture.
|
static int |
cuStreamGetCtx(CUstream_st hStream,
CUctx_st pctx)
\brief Query the context associated with a stream
Returns the CUDA context that the stream is associated with.
|
static int |
cuStreamGetFlags(CUstream_st hStream,
int[] flags) |
static int |
cuStreamGetFlags(CUstream_st hStream,
IntBuffer flags) |
static int |
cuStreamGetFlags(CUstream_st hStream,
IntPointer flags)
\brief Query the flags of a given stream
Query the flags of a stream created using ::cuStreamCreate or ::cuStreamCreateWithPriority
and return the flags in \p flags.
|
static int |
cuStreamGetId(CUstream_st hStream,
long[] streamId) |
static int |
cuStreamGetId(CUstream_st hStream,
LongBuffer streamId) |
static int |
cuStreamGetId(CUstream_st hStream,
LongPointer streamId)
\brief Returns the unique Id associated with the stream handle supplied
Returns in \p streamId the unique Id which is associated with the given stream handle.
|
static int |
cuStreamGetPriority(CUstream_st hStream,
int[] priority) |
static int |
cuStreamGetPriority(CUstream_st hStream,
IntBuffer priority) |
static int |
cuStreamGetPriority(CUstream_st hStream,
IntPointer priority)
\brief Query the priority of a given stream
Query the priority of a stream created using ::cuStreamCreate or ::cuStreamCreateWithPriority
and return the priority in \p priority.
|
static int |
cuStreamIsCapturing(CUstream_st hStream,
int[] captureStatus) |
static int |
cuStreamIsCapturing(CUstream_st hStream,
IntBuffer captureStatus) |
static int |
cuStreamIsCapturing(CUstream_st hStream,
IntPointer captureStatus)
\brief Returns a stream's capture status
Return the capture status of \p hStream via \p captureStatus.
|
static int |
cuStreamQuery(CUstream_st hStream)
\brief Determine status of a compute stream
Returns ::CUDA_SUCCESS if all operations in the stream specified by
\p hStream have completed, or ::CUDA_ERROR_NOT_READY if not.
|
static int |
cuStreamSetAttribute(CUstream_st hStream,
int attr,
CUlaunchAttributeValue value)
\brief Sets stream attribute.
|
static int |
cuStreamSynchronize(CUstream_st hStream)
\brief Wait until a stream's tasks are completed
Waits until the device has completed all operations in the stream specified
by \p hStream.
|
static int |
cuStreamUpdateCaptureDependencies_v2(CUstream_st hStream,
CUgraphNode_st dependencies,
CUgraphEdgeData dependencyData,
long numDependencies,
int flags)
\brief Update the set of dependencies in a capturing stream (12.3+)
Modifies the dependency set of a capturing stream.
|
static int |
cuStreamUpdateCaptureDependencies(CUstream_st hStream,
CUgraphNode_st dependencies,
long numDependencies,
int flags)
\brief Update the set of dependencies in a capturing stream (11.3+)
Modifies the dependency set of a capturing stream.
|
static int |
cuStreamWaitEvent(CUstream_st hStream,
CUevent_st hEvent,
int Flags)
\brief Make a compute stream wait on an event
Makes all future work submitted to \p hStream wait for all work captured in
\p hEvent.
|
static int |
cuStreamWaitValue32(CUstream_st stream,
long addr,
int value,
int flags)
\brief Wait on a memory location
Enqueues a synchronization of the stream on the given memory location.
|
static int |
cuStreamWaitValue64(CUstream_st stream,
long addr,
long value,
int flags)
\brief Wait on a memory location
Enqueues a synchronization of the stream on the given memory location.
|
static int |
cuStreamWriteValue32(CUstream_st stream,
long addr,
int value,
int flags)
\brief Write a value to memory
Write a value to memory.
|
static int |
cuStreamWriteValue64(CUstream_st stream,
long addr,
long value,
int flags)
\brief Write a value to memory
Write a value to memory.
|
static int |
cuSurfObjectCreate(long[] pSurfObject,
CUDA_RESOURCE_DESC_v1 pResDesc) |
static int |
cuSurfObjectCreate(LongBuffer pSurfObject,
CUDA_RESOURCE_DESC_v1 pResDesc) |
static int |
cuSurfObjectCreate(LongPointer pSurfObject,
CUDA_RESOURCE_DESC_v1 pResDesc)
\brief Creates a surface object
Creates a surface object and returns it in \p pSurfObject.
|
static int |
cuSurfObjectDestroy(long surfObject)
\brief Destroys a surface object
Destroys the surface object specified by \p surfObject.
|
static int |
cuSurfObjectGetResourceDesc(CUDA_RESOURCE_DESC_v1 pResDesc,
long surfObject)
\brief Returns a surface object's resource descriptor
Returns the resource descriptor for the surface object specified by \p surfObject.
|
static int |
cuSurfRefGetArray(CUarray_st phArray,
CUsurfref_st hSurfRef)
Deprecated.
Returns in \p *phArray the CUDA array bound to the surface reference
\p hSurfRef, or returns ::CUDA_ERROR_INVALID_VALUE if the surface reference
is not bound to any CUDA array.
|
static int |
cuSurfRefSetArray(CUsurfref_st hSurfRef,
CUarray_st hArray,
int Flags)
Deprecated.
Sets the CUDA array \p hArray to be read and written by the surface reference
\p hSurfRef. Any previous CUDA array state associated with the surface
reference is superseded by this function. \p Flags must be set to 0.
The ::CUDA_ARRAY3D_SURFACE_LDST flag must have been set for the CUDA array.
Any CUDA array previously bound to \p hSurfRef is unbound.
|
static int |
cuTensorMapEncodeIm2col(CUtensorMap tensorMap,
int tensorDataType,
int tensorRank,
Pointer globalAddress,
long[] globalDim,
long[] globalStrides,
int[] pixelBoxLowerCorner,
int[] pixelBoxUpperCorner,
int channelsPerPixel,
int pixelsPerColumn,
int[] elementStrides,
int interleave,
int swizzle,
int l2Promotion,
int oobFill) |
static int |
cuTensorMapEncodeIm2col(CUtensorMap tensorMap,
int tensorDataType,
int tensorRank,
Pointer globalAddress,
LongBuffer globalDim,
LongBuffer globalStrides,
IntBuffer pixelBoxLowerCorner,
IntBuffer pixelBoxUpperCorner,
int channelsPerPixel,
int pixelsPerColumn,
IntBuffer elementStrides,
int interleave,
int swizzle,
int l2Promotion,
int oobFill) |
static int |
cuTensorMapEncodeIm2col(CUtensorMap tensorMap,
int tensorDataType,
int tensorRank,
Pointer globalAddress,
LongPointer globalDim,
LongPointer globalStrides,
IntPointer pixelBoxLowerCorner,
IntPointer pixelBoxUpperCorner,
int channelsPerPixel,
int pixelsPerColumn,
IntPointer elementStrides,
int interleave,
int swizzle,
int l2Promotion,
int oobFill)
\brief Create a tensor map descriptor object representing im2col memory region
Creates a descriptor for Tensor Memory Access (TMA) object specified
by the parameters describing a im2col memory layout and returns it in \p tensorMap.
|
static int |
cuTensorMapEncodeTiled(CUtensorMap tensorMap,
int tensorDataType,
int tensorRank,
Pointer globalAddress,
long[] globalDim,
long[] globalStrides,
int[] boxDim,
int[] elementStrides,
int interleave,
int swizzle,
int l2Promotion,
int oobFill) |
static int |
cuTensorMapEncodeTiled(CUtensorMap tensorMap,
int tensorDataType,
int tensorRank,
Pointer globalAddress,
LongBuffer globalDim,
LongBuffer globalStrides,
IntBuffer boxDim,
IntBuffer elementStrides,
int interleave,
int swizzle,
int l2Promotion,
int oobFill) |
static int |
cuTensorMapEncodeTiled(CUtensorMap tensorMap,
int tensorDataType,
int tensorRank,
Pointer globalAddress,
LongPointer globalDim,
LongPointer globalStrides,
IntPointer boxDim,
IntPointer elementStrides,
int interleave,
int swizzle,
int l2Promotion,
int oobFill)
\brief Create a tensor map descriptor object representing tiled memory region
Creates a descriptor for Tensor Memory Access (TMA) object specified
by the parameters describing a tiled region and returns it in \p tensorMap.
|
static int |
cuTensorMapReplaceAddress(CUtensorMap tensorMap,
Pointer globalAddress)
\brief Modify an existing tensor map descriptor with an updated global address
Modifies the descriptor for Tensor Memory Access (TMA) object passed in \p tensorMap with
an updated \p globalAddress.
|
static int |
cuTexObjectCreate(long[] pTexObject,
CUDA_RESOURCE_DESC_v1 pResDesc,
CUDA_TEXTURE_DESC_v1 pTexDesc,
CUDA_RESOURCE_VIEW_DESC_v1 pResViewDesc) |
static int |
cuTexObjectCreate(LongBuffer pTexObject,
CUDA_RESOURCE_DESC_v1 pResDesc,
CUDA_TEXTURE_DESC_v1 pTexDesc,
CUDA_RESOURCE_VIEW_DESC_v1 pResViewDesc) |
static int |
cuTexObjectCreate(LongPointer pTexObject,
CUDA_RESOURCE_DESC_v1 pResDesc,
CUDA_TEXTURE_DESC_v1 pTexDesc,
CUDA_RESOURCE_VIEW_DESC_v1 pResViewDesc)
\brief Creates a texture object
Creates a texture object and returns it in \p pTexObject.
|
static int |
cuTexObjectDestroy(long texObject)
\brief Destroys a texture object
Destroys the texture object specified by \p texObject.
|
static int |
cuTexObjectGetResourceDesc(CUDA_RESOURCE_DESC_v1 pResDesc,
long texObject)
\brief Returns a texture object's resource descriptor
Returns the resource descriptor for the texture object specified by \p texObject.
|
static int |
cuTexObjectGetResourceViewDesc(CUDA_RESOURCE_VIEW_DESC_v1 pResViewDesc,
long texObject)
\brief Returns a texture object's resource view descriptor
Returns the resource view descriptor for the texture object specified by \p texObject.
|
static int |
cuTexObjectGetTextureDesc(CUDA_TEXTURE_DESC_v1 pTexDesc,
long texObject)
\brief Returns a texture object's texture descriptor
Returns the texture descriptor for the texture object specified by \p texObject.
|
static int |
cuTexRefCreate(CUtexref_st pTexRef)
Deprecated.
Creates a texture reference and returns its handle in \p *pTexRef. Once
created, the application must call ::cuTexRefSetArray() or
::cuTexRefSetAddress() to associate the reference with allocated memory.
Other texture reference functions are used to specify the format and
interpretation (addressing, filtering, etc.) to be used when the memory is
read through this texture reference.
|
static int |
cuTexRefDestroy(CUtexref_st hTexRef)
Deprecated.
Destroys the texture reference specified by \p hTexRef.
|
static int |
cuTexRefGetAddress(long[] pdptr,
CUtexref_st hTexRef)
Deprecated.
|
static int |
cuTexRefGetAddress(LongBuffer pdptr,
CUtexref_st hTexRef)
Deprecated.
|
static int |
cuTexRefGetAddress(LongPointer pdptr,
CUtexref_st hTexRef)
Deprecated.
Returns in \p *pdptr the base address bound to the texture reference
\p hTexRef, or returns ::CUDA_ERROR_INVALID_VALUE if the texture reference
is not bound to any device memory range.
|
static int |
cuTexRefGetAddressMode(int[] pam,
CUtexref_st hTexRef,
int dim)
Deprecated.
|
static int |
cuTexRefGetAddressMode(IntBuffer pam,
CUtexref_st hTexRef,
int dim)
Deprecated.
|
static int |
cuTexRefGetAddressMode(IntPointer pam,
CUtexref_st hTexRef,
int dim)
Deprecated.
Returns in \p *pam the addressing mode corresponding to the
dimension \p dim of the texture reference \p hTexRef. Currently, the only
valid value for \p dim are 0 and 1.
|
static int |
cuTexRefGetArray(CUarray_st phArray,
CUtexref_st hTexRef)
Deprecated.
Returns in \p *phArray the CUDA array bound to the texture reference
\p hTexRef, or returns ::CUDA_ERROR_INVALID_VALUE if the texture reference
is not bound to any CUDA array.
|
static int |
cuTexRefGetBorderColor(float[] pBorderColor,
CUtexref_st hTexRef)
Deprecated.
|
static int |
cuTexRefGetBorderColor(FloatBuffer pBorderColor,
CUtexref_st hTexRef)
Deprecated.
|
static int |
cuTexRefGetBorderColor(FloatPointer pBorderColor,
CUtexref_st hTexRef)
Deprecated.
Returns in \p pBorderColor, values of the RGBA color used by
the texture reference \p hTexRef.
The color value is of type float and holds color components in
the following sequence:
pBorderColor[0] holds 'R' component
pBorderColor[1] holds 'G' component
pBorderColor[2] holds 'B' component
pBorderColor[3] holds 'A' component
|
static int |
cuTexRefGetFilterMode(int[] pfm,
CUtexref_st hTexRef)
Deprecated.
|
static int |
cuTexRefGetFilterMode(IntBuffer pfm,
CUtexref_st hTexRef)
Deprecated.
|
static int |
cuTexRefGetFilterMode(IntPointer pfm,
CUtexref_st hTexRef)
Deprecated.
Returns in \p *pfm the filtering mode of the texture reference
\p hTexRef.
|
static int |
cuTexRefGetFlags(int[] pFlags,
CUtexref_st hTexRef)
Deprecated.
|
static int |
cuTexRefGetFlags(IntBuffer pFlags,
CUtexref_st hTexRef)
Deprecated.
|
static int |
cuTexRefGetFlags(IntPointer pFlags,
CUtexref_st hTexRef)
Deprecated.
Returns in \p *pFlags the flags of the texture reference \p hTexRef.
|
static int |
cuTexRefGetFormat(int[] pFormat,
int[] pNumChannels,
CUtexref_st hTexRef)
Deprecated.
|
static int |
cuTexRefGetFormat(IntBuffer pFormat,
IntBuffer pNumChannels,
CUtexref_st hTexRef)
Deprecated.
|
static int |
cuTexRefGetFormat(IntPointer pFormat,
IntPointer pNumChannels,
CUtexref_st hTexRef)
Deprecated.
Returns in \p *pFormat and \p *pNumChannels the format and number
of components of the CUDA array bound to the texture reference \p hTexRef.
If \p pFormat or \p pNumChannels is NULL, it will be ignored.
|
static int |
cuTexRefGetMaxAnisotropy(int[] pmaxAniso,
CUtexref_st hTexRef)
Deprecated.
|
static int |
cuTexRefGetMaxAnisotropy(IntBuffer pmaxAniso,
CUtexref_st hTexRef)
Deprecated.
|
static int |
cuTexRefGetMaxAnisotropy(IntPointer pmaxAniso,
CUtexref_st hTexRef)
Deprecated.
Returns the maximum anisotropy in \p pmaxAniso that's used when reading memory through
the texture reference \p hTexRef.
|
static int |
cuTexRefGetMipmapFilterMode(int[] pfm,
CUtexref_st hTexRef)
Deprecated.
|
static int |
cuTexRefGetMipmapFilterMode(IntBuffer pfm,
CUtexref_st hTexRef)
Deprecated.
|
static int |
cuTexRefGetMipmapFilterMode(IntPointer pfm,
CUtexref_st hTexRef)
Deprecated.
Returns the mipmap filtering mode in \p pfm that's used when reading memory through
the texture reference \p hTexRef.
|
static int |
cuTexRefGetMipmapLevelBias(float[] pbias,
CUtexref_st hTexRef)
Deprecated.
|
static int |
cuTexRefGetMipmapLevelBias(FloatBuffer pbias,
CUtexref_st hTexRef)
Deprecated.
|
static int |
cuTexRefGetMipmapLevelBias(FloatPointer pbias,
CUtexref_st hTexRef)
Deprecated.
Returns the mipmap level bias in \p pBias that's added to the specified mipmap
level when reading memory through the texture reference \p hTexRef.
|
static int |
cuTexRefGetMipmapLevelClamp(float[] pminMipmapLevelClamp,
float[] pmaxMipmapLevelClamp,
CUtexref_st hTexRef)
Deprecated.
|
static int |
cuTexRefGetMipmapLevelClamp(FloatBuffer pminMipmapLevelClamp,
FloatBuffer pmaxMipmapLevelClamp,
CUtexref_st hTexRef)
Deprecated.
|
static int |
cuTexRefGetMipmapLevelClamp(FloatPointer pminMipmapLevelClamp,
FloatPointer pmaxMipmapLevelClamp,
CUtexref_st hTexRef)
Deprecated.
Returns the min/max mipmap level clamps in \p pminMipmapLevelClamp and \p pmaxMipmapLevelClamp
that's used when reading memory through the texture reference \p hTexRef.
|
static int |
cuTexRefGetMipmappedArray(CUmipmappedArray_st phMipmappedArray,
CUtexref_st hTexRef)
Deprecated.
Returns in \p *phMipmappedArray the CUDA mipmapped array bound to the texture
reference \p hTexRef, or returns ::CUDA_ERROR_INVALID_VALUE if the texture reference
is not bound to any CUDA mipmapped array.
|
static int |
cuTexRefSetAddress(SizeTPointer ByteOffset,
CUtexref_st hTexRef,
long dptr,
long bytes)
Deprecated.
Binds a linear address range to the texture reference \p hTexRef. Any
previous address or CUDA array state associated with the texture reference
is superseded by this function. Any memory previously bound to \p hTexRef
is unbound.
Since the hardware enforces an alignment requirement on texture base
addresses, ::cuTexRefSetAddress() passes back a byte offset in
\p *ByteOffset that must be applied to texture fetches in order to read from
the desired memory. This offset must be divided by the texel size and
passed to kernels that read from the texture so they can be applied to the
::tex1Dfetch() function.
If the device memory pointer was returned from ::cuMemAlloc(), the offset
is guaranteed to be 0 and NULL may be passed as the \p ByteOffset parameter.
The total number of elements (or texels) in the linear address range
cannot exceed ::CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE1D_LINEAR_WIDTH.
The number of elements is computed as (\p bytes / bytesPerElement),
where bytesPerElement is determined from the data format and number of
components set using ::cuTexRefSetFormat().
|
static int |
cuTexRefSetAddress2D(CUtexref_st hTexRef,
CUDA_ARRAY_DESCRIPTOR_v2 desc,
long dptr,
long Pitch)
Deprecated.
Binds a linear address range to the texture reference \p hTexRef. Any
previous address or CUDA array state associated with the texture reference
is superseded by this function. Any memory previously bound to \p hTexRef
is unbound.
Using a ::tex2D() function inside a kernel requires a call to either
::cuTexRefSetArray() to bind the corresponding texture reference to an
array, or ::cuTexRefSetAddress2D() to bind the texture reference to linear
memory.
Function calls to ::cuTexRefSetFormat() cannot follow calls to
::cuTexRefSetAddress2D() for the same texture reference.
It is required that \p dptr be aligned to the appropriate hardware-specific
texture alignment. You can query this value using the device attribute
::CU_DEVICE_ATTRIBUTE_TEXTURE_ALIGNMENT. If an unaligned \p dptr is
supplied, ::CUDA_ERROR_INVALID_VALUE is returned.
\p Pitch has to be aligned to the hardware-specific texture pitch alignment.
This value can be queried using the device attribute
::CU_DEVICE_ATTRIBUTE_TEXTURE_PITCH_ALIGNMENT. If an unaligned \p Pitch is
supplied, ::CUDA_ERROR_INVALID_VALUE is returned.
Width and Height, which are specified in elements (or texels), cannot exceed
::CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_LINEAR_WIDTH and
::CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_LINEAR_HEIGHT respectively.
\p Pitch, which is specified in bytes, cannot exceed
::CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_LINEAR_PITCH.
|
static int |
cuTexRefSetAddressMode(CUtexref_st hTexRef,
int dim,
int am)
Deprecated.
Specifies the addressing mode \p am for the given dimension \p dim of the
texture reference \p hTexRef. If \p dim is zero, the addressing mode is
applied to the first parameter of the functions used to fetch from the
texture; if \p dim is 1, the second, and so on. ::CUaddress_mode is defined
as:
Note that this call has no effect if \p hTexRef is bound to linear memory.
Also, if the flag, ::CU_TRSF_NORMALIZED_COORDINATES, is not set, the only
supported address mode is ::CU_TR_ADDRESS_MODE_CLAMP. |
static int |
cuTexRefSetArray(CUtexref_st hTexRef,
CUarray_st hArray,
int Flags)
Deprecated.
Binds the CUDA array \p hArray to the texture reference \p hTexRef. Any
previous address or CUDA array state associated with the texture reference
is superseded by this function. \p Flags must be set to
::CU_TRSA_OVERRIDE_FORMAT. Any CUDA array previously bound to \p hTexRef is
unbound.
|
static int |
cuTexRefSetBorderColor(CUtexref_st hTexRef,
float[] pBorderColor)
Deprecated.
|
static int |
cuTexRefSetBorderColor(CUtexref_st hTexRef,
FloatBuffer pBorderColor)
Deprecated.
|
static int |
cuTexRefSetBorderColor(CUtexref_st hTexRef,
FloatPointer pBorderColor)
Deprecated.
Specifies the value of the RGBA color via the \p pBorderColor to the texture reference
\p hTexRef. The color value supports only float type and holds color components in
the following sequence:
pBorderColor[0] holds 'R' component
pBorderColor[1] holds 'G' component
pBorderColor[2] holds 'B' component
pBorderColor[3] holds 'A' component
Note that the color values can be set only when the Address mode is set to
CU_TR_ADDRESS_MODE_BORDER using ::cuTexRefSetAddressMode.
Applications using integer border color values have to "reinterpret_cast" their values to float.
|
static int |
cuTexRefSetFilterMode(CUtexref_st hTexRef,
int fm)
Deprecated.
Specifies the filtering mode \p fm to be used when reading memory through
the texture reference \p hTexRef. ::CUfilter_mode_enum is defined as:
Note that this call has no effect if \p hTexRef is bound to linear memory. |
static int |
cuTexRefSetFlags(CUtexref_st hTexRef,
int Flags)
Deprecated.
Specifies optional flags via \p Flags to specify the behavior of data
returned through the texture reference \p hTexRef. The valid flags are:
- ::CU_TRSF_READ_AS_INTEGER, which suppresses the default behavior of
having the texture promote integer data to floating point data in the
range [0, 1]. Note that texture with 32-bit integer format
would not be promoted, regardless of whether or not this
flag is specified;
- ::CU_TRSF_NORMALIZED_COORDINATES, which suppresses the
default behavior of having the texture coordinates range
from [0, Dim) where Dim is the width or height of the CUDA
array. Instead, the texture coordinates [0, 1.0) reference
the entire breadth of the array dimension;
- ::CU_TRSF_DISABLE_TRILINEAR_OPTIMIZATION, which disables any trilinear
filtering optimizations. Trilinear optimizations improve texture filtering
performance by allowing bilinear filtering on textures in scenarios where
it can closely approximate the expected results.
|
static int |
cuTexRefSetFormat(CUtexref_st hTexRef,
int fmt,
int NumPackedComponents)
Deprecated.
Specifies the format of the data to be read by the texture reference
\p hTexRef. \p fmt and \p NumPackedComponents are exactly analogous to the
::Format and ::NumChannels members of the ::CUDA_ARRAY_DESCRIPTOR structure:
They specify the format of each component and the number of components per
array element.
|
static int |
cuTexRefSetMaxAnisotropy(CUtexref_st hTexRef,
int maxAniso)
Deprecated.
Specifies the maximum anisotropy \p maxAniso to be used when reading memory through
the texture reference \p hTexRef.
Note that this call has no effect if \p hTexRef is bound to linear memory.
|
static int |
cuTexRefSetMipmapFilterMode(CUtexref_st hTexRef,
int fm)
Deprecated.
Specifies the mipmap filtering mode \p fm to be used when reading memory through
the texture reference \p hTexRef. ::CUfilter_mode_enum is defined as:
Note that this call has no effect if \p hTexRef is not bound to a mipmapped array. |
static int |
cuTexRefSetMipmapLevelBias(CUtexref_st hTexRef,
float bias)
Deprecated.
Specifies the mipmap level bias \p bias to be added to the specified mipmap level when
reading memory through the texture reference \p hTexRef.
Note that this call has no effect if \p hTexRef is not bound to a mipmapped array.
|
static int |
cuTexRefSetMipmapLevelClamp(CUtexref_st hTexRef,
float minMipmapLevelClamp,
float maxMipmapLevelClamp)
Deprecated.
Specifies the min/max mipmap level clamps, \p minMipmapLevelClamp and \p maxMipmapLevelClamp
respectively, to be used when reading memory through the texture reference
\p hTexRef.
Note that this call has no effect if \p hTexRef is not bound to a mipmapped array.
|
static int |
cuTexRefSetMipmappedArray(CUtexref_st hTexRef,
CUmipmappedArray_st hMipmappedArray,
int Flags)
Deprecated.
Binds the CUDA mipmapped array \p hMipmappedArray to the texture reference \p hTexRef.
Any previous address or CUDA array state associated with the texture reference
is superseded by this function. \p Flags must be set to ::CU_TRSA_OVERRIDE_FORMAT.
Any CUDA array previously bound to \p hTexRef is unbound.
|
static int |
cuThreadExchangeStreamCaptureMode(int[] mode) |
static int |
cuThreadExchangeStreamCaptureMode(IntBuffer mode) |
static int |
cuThreadExchangeStreamCaptureMode(IntPointer mode)
\brief Swaps the stream capture interaction mode for a thread
Sets the calling thread's stream capture interaction mode to the value contained
in \p *mode, and overwrites \p *mode with the previous mode for the thread.
|
static int |
cuUserObjectCreate(CUuserObject_st object_out,
Pointer ptr,
CUhostFn destroy,
int initialRefcount,
int flags)
\brief Create a user object
Create a user object with the specified destructor callback and initial reference count.
|
static int |
cuUserObjectRelease(CUuserObject_st object,
int count)
\brief Release a reference to a user object
Releases user object references owned by the caller.
|
static int |
cuUserObjectRetain(CUuserObject_st object,
int count)
\brief Retain a reference to a user object
Retains new references to a user object.
|
static int |
cuWaitExternalSemaphoresAsync(CUextSemaphore_st extSemArray,
CUDA_EXTERNAL_SEMAPHORE_WAIT_PARAMS_v1 paramsArray,
int numExtSems,
CUstream_st stream)
\brief Waits on a set of external semaphore objects
Enqueues a wait operation on a set of externally allocated
semaphore object in the specified stream.
|
static __half |
decrement(__half h)
\ingroup CUDA_MATH__HALF_ARITHMETIC
Performs \p half prefix decrement operation.
|
static __half2 |
decrement(__half2 h)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
Performs packed \p half prefix decrement operation.
|
static __half2 |
decrement(__half2 h,
int ignored)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
Performs packed \p half postfix decrement operation.
|
static __half |
decrement(__half h,
int ignored)
\ingroup CUDA_MATH__HALF_ARITHMETIC
Performs \p half postfix decrement operation.
|
static __nv_bfloat16 |
decrement(__nv_bfloat16 h)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
Performs \p nv_bfloat16 prefix decrement operation.
|
static __nv_bfloat162 |
decrement(__nv_bfloat162 h)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
Performs packed \p nv_bfloat16 prefix decrement operation.
|
static __nv_bfloat162 |
decrement(__nv_bfloat162 h,
int ignored)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
Performs packed \p nv_bfloat16 postfix decrement operation.
|
static __nv_bfloat16 |
decrement(__nv_bfloat16 h,
int ignored)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
Performs \p nv_bfloat16 postfix decrement operation.
|
static __half |
divide(__half lh,
__half rh)
\ingroup CUDA_MATH__HALF_ARITHMETIC
Performs \p half division operation.
|
static __half2 |
divide(__half2 lh,
__half2 rh)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
Performs packed \p half division operation.
|
static __nv_bfloat16 |
divide(__nv_bfloat16 lh,
__nv_bfloat16 rh)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
Performs \p nv_bfloat16 division operation.
|
static __nv_bfloat162 |
divide(__nv_bfloat162 lh,
__nv_bfloat162 rh)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
Performs packed \p nv_bfloat16 division operation.
|
static __half |
dividePut(__half lh,
__half rh)
\ingroup CUDA_MATH__HALF_ARITHMETIC
Performs \p half compound assignment with division operation.
|
static __half2 |
dividePut(__half2 lh,
__half2 rh)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
Performs packed \p half compound assignment with division operation.
|
static __nv_bfloat16 |
dividePut(__nv_bfloat16 lh,
__nv_bfloat16 rh)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
Performs \p nv_bfloat16 compound assignment with division operation.
|
static __nv_bfloat162 |
dividePut(__nv_bfloat162 lh,
__nv_bfloat162 rh)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
Performs packed \p nv_bfloat16 compound assignment with division operation.
|
static boolean |
equals(__half lh,
__half rh)
\ingroup CUDA_MATH__HALF_COMPARISON
Performs \p half ordered compare equal operation.
|
static boolean |
equals(__half2 lh,
__half2 rh)
\ingroup CUDA_MATH__HALF2_COMPARISON
Performs packed \p half ordered compare equal operation.
|
static boolean |
equals(__nv_bfloat16 lh,
__nv_bfloat16 rh)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
Performs \p nv_bfloat16 ordered compare equal operation.
|
static boolean |
equals(__nv_bfloat162 lh,
__nv_bfloat162 rh)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
Performs packed \p nv_bfloat16 ordered compare equal operation.
|
static boolean |
greaterThan(__half lh,
__half rh)
\ingroup CUDA_MATH__HALF_COMPARISON
Performs \p half ordered greater-than compare operation.
|
static boolean |
greaterThan(__half2 lh,
__half2 rh)
\ingroup CUDA_MATH__HALF2_COMPARISON
Performs packed \p half ordered greater-than compare operation.
|
static boolean |
greaterThan(__nv_bfloat16 lh,
__nv_bfloat16 rh)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
Performs \p nv_bfloat16 ordered greater-than compare operation.
|
static boolean |
greaterThan(__nv_bfloat162 lh,
__nv_bfloat162 rh)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
Performs packed \p nv_bfloat16 ordered greater-than compare operation.
|
static boolean |
greaterThanEquals(__half lh,
__half rh)
\ingroup CUDA_MATH__HALF_COMPARISON
Performs \p half ordered greater-or-equal compare operation.
|
static boolean |
greaterThanEquals(__half2 lh,
__half2 rh)
\ingroup CUDA_MATH__HALF2_COMPARISON
Performs packed \p half ordered greater-or-equal compare operation.
|
static boolean |
greaterThanEquals(__nv_bfloat16 lh,
__nv_bfloat16 rh)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
Performs \p nv_bfloat16 ordered greater-or-equal compare operation.
|
static boolean |
greaterThanEquals(__nv_bfloat162 lh,
__nv_bfloat162 rh)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
Performs packed \p nv_bfloat16 ordered greater-or-equal compare operation.
|
static __half |
increment(__half h)
\ingroup CUDA_MATH__HALF_ARITHMETIC
Performs \p half prefix increment operation.
|
static __half2 |
increment(__half2 h)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
Performs packed \p half prefix increment operation.
|
static __half2 |
increment(__half2 h,
int ignored)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
Performs packed \p half postfix increment operation.
|
static __half |
increment(__half h,
int ignored)
\ingroup CUDA_MATH__HALF_ARITHMETIC
Performs \p half postfix increment operation.
|
static __nv_bfloat16 |
increment(__nv_bfloat16 h)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
Performs \p nv_bfloat16 prefix increment operation.
|
static __nv_bfloat162 |
increment(__nv_bfloat162 h)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
Performs packed \p nv_bfloat16 prefix increment operation.
|
static __nv_bfloat162 |
increment(__nv_bfloat162 h,
int ignored)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
Performs packed \p nv_bfloat16 postfix increment operation.
|
static __nv_bfloat16 |
increment(__nv_bfloat16 h,
int ignored)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
Performs \p nv_bfloat16 postfix increment operation.
|
static boolean |
lessThan(__half lh,
__half rh)
\ingroup CUDA_MATH__HALF_COMPARISON
Performs \p half ordered less-than compare operation.
|
static boolean |
lessThan(__half2 lh,
__half2 rh)
\ingroup CUDA_MATH__HALF2_COMPARISON
Performs packed \p half ordered less-than compare operation.
|
static boolean |
lessThan(__nv_bfloat16 lh,
__nv_bfloat16 rh)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
Performs \p nv_bfloat16 ordered less-than compare operation.
|
static boolean |
lessThan(__nv_bfloat162 lh,
__nv_bfloat162 rh)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
Performs packed \p nv_bfloat16 ordered less-than compare operation.
|
static boolean |
lessThanEquals(__half lh,
__half rh)
\ingroup CUDA_MATH__HALF_COMPARISON
Performs \p half ordered less-or-equal compare operation.
|
static boolean |
lessThanEquals(__half2 lh,
__half2 rh)
\ingroup CUDA_MATH__HALF2_COMPARISON
Performs packed \p half ordered less-or-equal compare operation.
|
static boolean |
lessThanEquals(__nv_bfloat16 lh,
__nv_bfloat16 rh)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
Performs \p nv_bfloat16 ordered less-or-equal compare operation.
|
static boolean |
lessThanEquals(__nv_bfloat162 lh,
__nv_bfloat162 rh)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
Performs packed \p nv_bfloat16 ordered less-or-equal compare operation.
|
static __nv_bfloat162 |
make_bfloat162(__nv_bfloat16 x,
__nv_bfloat16 y)
\ingroup CUDA_MATH__BFLOAT16_MISC
\brief Vector function, combines two \p nv_bfloat16 numbers into one \p nv_bfloat162 number.
|
static char1 |
make_char1(byte x)
*
*
*
|
static char2 |
make_char2(byte x,
byte y) |
static char3 |
make_char3(byte x,
byte y,
byte z) |
static char4 |
make_char4(byte x,
byte y,
byte z,
byte w) |
static float2 |
make_cuComplex(float x,
float y) |
static cudaExtent |
make_cudaExtent(long w,
long h,
long d)
\brief Returns a cudaExtent based on input parameters
Returns a ::cudaExtent based on the specified input parameters \p w,
\p h, and \p d.
|
static cudaPitchedPtr |
make_cudaPitchedPtr(Pointer d,
long p,
long xsz,
long ysz)
\brief Returns a cudaPitchedPtr based on input parameters
Returns a ::cudaPitchedPtr based on the specified input parameters \p d,
\p p, \p xsz, and \p ysz.
|
static cudaPos |
make_cudaPos(long x,
long y,
long z)
\brief Returns a cudaPos based on input parameters
Returns a ::cudaPos based on the specified input parameters \p x,
\p y, and \p z.
|
static double2 |
make_cuDoubleComplex(double r,
double i) |
static float2 |
make_cuFloatComplex(float r,
float i) |
static double1 |
make_double1(double x) |
static double2 |
make_double2(double x,
double y) |
static double3 |
make_double3(double x,
double y,
double z) |
static double4 |
make_double4(double x,
double y,
double z,
double w) |
static float1 |
make_float1(float x) |
static float2 |
make_float2(float x,
float y) |
static float3 |
make_float3(float x,
float y,
float z) |
static float4 |
make_float4(float x,
float y,
float z,
float w) |
static __half2 |
make_half2(__half x,
__half y)
\ingroup CUDA_MATH__HALF_MISC
\brief Vector function, combines two \p __half numbers into one \p __half2 number.
|
static int1 |
make_int1(int x) |
static int2 |
make_int2(int x,
int y) |
static int3 |
make_int3(int x,
int y,
int z) |
static int4 |
make_int4(int x,
int y,
int z,
int w) |
static long1 |
make_long1(long x) |
static long2 |
make_long2(long x,
long y) |
static long3 |
make_long3(long x,
long y,
long z) |
static long4 |
make_long4(long x,
long y,
long z,
long w) |
static longlong1 |
make_longlong1(long x) |
static longlong2 |
make_longlong2(long x,
long y) |
static longlong3 |
make_longlong3(long x,
long y,
long z) |
static longlong4 |
make_longlong4(long x,
long y,
long z,
long w) |
static short1 |
make_short1(short x) |
static short2 |
make_short2(short x,
short y) |
static short3 |
make_short3(short x,
short y,
short z) |
static short4 |
make_short4(short x,
short y,
short z,
short w) |
static uchar1 |
make_uchar1(byte x) |
static uchar2 |
make_uchar2(byte x,
byte y) |
static uchar3 |
make_uchar3(byte x,
byte y,
byte z) |
static uchar4 |
make_uchar4(byte x,
byte y,
byte z,
byte w) |
static uint1 |
make_uint1(int x) |
static uint2 |
make_uint2(int x,
int y) |
static uint3 |
make_uint3(int x,
int y,
int z) |
static uint4 |
make_uint4(int x,
int y,
int z,
int w) |
static ulong1 |
make_ulong1(long x) |
static ulong2 |
make_ulong2(long x,
long y) |
static ulong3 |
make_ulong3(long x,
long y,
long z) |
static ulong4 |
make_ulong4(long x,
long y,
long z,
long w) |
static ulonglong1 |
make_ulonglong1(long x) |
static ulonglong2 |
make_ulonglong2(long x,
long y) |
static ulonglong3 |
make_ulonglong3(long x,
long y,
long z) |
static ulonglong4 |
make_ulonglong4(long x,
long y,
long z,
long w) |
static ushort1 |
make_ushort1(short x) |
static ushort2 |
make_ushort2(short x,
short y) |
static ushort3 |
make_ushort3(short x,
short y,
short z) |
static ushort4 |
make_ushort4(short x,
short y,
short z,
short w) |
static __half |
multiply(__half lh,
__half rh)
\ingroup CUDA_MATH__HALF_ARITHMETIC
Performs \p half multiplication operation.
|
static __half2 |
multiply(__half2 lh,
__half2 rh)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
Performs packed \p half multiplication operation.
|
static __nv_bfloat16 |
multiply(__nv_bfloat16 lh,
__nv_bfloat16 rh)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
Performs \p nv_bfloat16 multiplication operation.
|
static __nv_bfloat162 |
multiply(__nv_bfloat162 lh,
__nv_bfloat162 rh)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
Performs packed \p nv_bfloat16 multiplication operation.
|
static __half |
multiplyPut(__half lh,
__half rh)
\ingroup CUDA_MATH__HALF_ARITHMETIC
Performs \p half compound assignment with multiplication operation.
|
static __half2 |
multiplyPut(__half2 lh,
__half2 rh)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
Performs packed \p half compound assignment with multiplication operation.
|
static __nv_bfloat16 |
multiplyPut(__nv_bfloat16 lh,
__nv_bfloat16 rh)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
Performs \p nv_bfloat16 compound assignment with multiplication operation.
|
static __nv_bfloat162 |
multiplyPut(__nv_bfloat162 lh,
__nv_bfloat162 rh)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
Performs packed \p nv_bfloat16 compound assignment with multiplication operation.
|
static boolean |
notEquals(__half lh,
__half rh)
\ingroup CUDA_MATH__HALF_COMPARISON
Performs \p half unordered compare not-equal operation.
|
static boolean |
notEquals(__half2 lh,
__half2 rh)
\ingroup CUDA_MATH__HALF2_COMPARISON
Performs packed \p half unordered compare not-equal operation.
|
static boolean |
notEquals(__nv_bfloat16 lh,
__nv_bfloat16 rh)
\ingroup CUDA_MATH__BFLOAT16_COMPARISON
Performs \p nv_bfloat16 unordered compare not-equal operation.
|
static boolean |
notEquals(__nv_bfloat162 lh,
__nv_bfloat162 rh)
\ingroup CUDA_MATH__BFLOAT162_COMPARISON
Performs packed \p nv_bfloat16 unordered compare not-equal operation.
|
static __half |
subtract(__half lh,
__half rh)
\ingroup CUDA_MATH__HALF_ARITHMETIC
Performs \p half subtraction operation.
|
static __half |
subtract(__half h)
\ingroup CUDA_MATH__HALF_ARITHMETIC
Implements \p half unary minus operator.
|
static __half2 |
subtract(__half2 lh,
__half2 rh)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
Performs packed \p half subtraction operation.
|
static __half2 |
subtract(__half2 h)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
Implements packed \p half unary minus operator.
|
static __nv_bfloat16 |
subtract(__nv_bfloat16 lh,
__nv_bfloat16 rh)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
Performs \p nv_bfloat16 subtraction operation.
|
static __nv_bfloat16 |
subtract(__nv_bfloat16 h)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
Implements \p nv_bfloat16 unary minus operator.
|
static __nv_bfloat162 |
subtract(__nv_bfloat162 lh,
__nv_bfloat162 rh)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
Performs packed \p nv_bfloat16 subtraction operation.
|
static __nv_bfloat162 |
subtract(__nv_bfloat162 h)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
Implements packed \p nv_bfloat16 unary minus operator.
|
static __half |
subtractPut(__half lh,
__half rh)
\ingroup CUDA_MATH__HALF_ARITHMETIC
Performs \p half compound assignment with subtraction operation.
|
static __half2 |
subtractPut(__half2 lh,
__half2 rh)
\ingroup CUDA_MATH__HALF2_ARITHMETIC
Performs packed \p half compound assignment with subtraction operation.
|
static __nv_bfloat16 |
subtractPut(__nv_bfloat16 lh,
__nv_bfloat16 rh)
\ingroup CUDA_MATH__BFLOAT16_ARITHMETIC
Performs \p nv_bfloat16 compound assignment with subtraction operation.
|
static __nv_bfloat162 |
subtractPut(__nv_bfloat162 lh,
__nv_bfloat162 rh)
\ingroup CUDA_MATH__BFLOAT162_ARITHMETIC
Performs packed \p nv_bfloat16 compound assignment with subtraction operation.
|
public static final int CUDA_VERSION
public static final int CU_IPC_HANDLE_SIZE
public static final int CU_IPC_MEM_LAZY_ENABLE_PEER_ACCESS
public static final int CU_MEM_ATTACH_GLOBAL
public static final int CU_MEM_ATTACH_HOST
public static final int CU_MEM_ATTACH_SINGLE
public static final int CU_CTX_SCHED_AUTO
public static final int CU_CTX_SCHED_SPIN
public static final int CU_CTX_SCHED_YIELD
public static final int CU_CTX_SCHED_BLOCKING_SYNC
public static final int CU_CTX_BLOCKING_SYNC
public static final int CU_CTX_SCHED_MASK
public static final int CU_CTX_MAP_HOST
public static final int CU_CTX_LMEM_RESIZE_TO_MAX
public static final int CU_CTX_COREDUMP_ENABLE
public static final int CU_CTX_USER_COREDUMP_ENABLE
public static final int CU_CTX_SYNC_MEMOPS
public static final int CU_CTX_FLAGS_MASK
public static final int CU_EVENT_SCHED_AUTO
public static final int CU_EVENT_SCHED_SPIN
public static final int CU_EVENT_SCHED_YIELD
public static final int CU_EVENT_SCHED_BLOCKING_SYNC
public static final int NVCL_EVENT_SCHED_AUTO
public static final int NVCL_EVENT_SCHED_SPIN
public static final int NVCL_EVENT_SCHED_YIELD
public static final int NVCL_EVENT_SCHED_BLOCKING_SYNC
public static final int NVCL_CTX_SCHED_AUTO
public static final int NVCL_CTX_SCHED_SPIN
public static final int NVCL_CTX_SCHED_YIELD
public static final int NVCL_CTX_SCHED_BLOCKING_SYNC
public static final int CU_STREAM_DEFAULT
public static final int CU_STREAM_NON_BLOCKING
public static final CUstream_st CU_STREAM_LEGACY
public static final CUstream_st CU_STREAM_PER_THREAD
public static final int CU_EVENT_DEFAULT
public static final int CU_EVENT_BLOCKING_SYNC
public static final int CU_EVENT_DISABLE_TIMING
public static final int CU_EVENT_INTERPROCESS
public static final int CU_EVENT_RECORD_DEFAULT
public static final int CU_EVENT_RECORD_EXTERNAL
public static final int CU_EVENT_WAIT_DEFAULT
public static final int CU_EVENT_WAIT_EXTERNAL
public static final int CU_STREAM_WAIT_VALUE_GEQ
public static final int CU_STREAM_WAIT_VALUE_EQ
public static final int CU_STREAM_WAIT_VALUE_AND
public static final int CU_STREAM_WAIT_VALUE_NOR
public static final int CU_STREAM_WAIT_VALUE_FLUSH
public static final int CU_STREAM_WRITE_VALUE_DEFAULT
public static final int CU_STREAM_WRITE_VALUE_NO_MEMORY_BARRIER
public static final int CU_STREAM_MEM_OP_WAIT_VALUE_32
public static final int CU_STREAM_MEM_OP_WRITE_VALUE_32
public static final int CU_STREAM_MEM_OP_WAIT_VALUE_64
public static final int CU_STREAM_MEM_OP_WRITE_VALUE_64
public static final int CU_STREAM_MEM_OP_BARRIER
public static final int CU_STREAM_MEM_OP_FLUSH_REMOTE_WRITES
public static final int CU_STREAM_MEMORY_BARRIER_TYPE_SYS
public static final int CU_STREAM_MEMORY_BARRIER_TYPE_GPU
public static final int CU_OCCUPANCY_DEFAULT
public static final int CU_OCCUPANCY_DISABLE_CACHING_OVERRIDE
public static final int CU_STREAM_ADD_CAPTURE_DEPENDENCIES
public static final int CU_STREAM_SET_CAPTURE_DEPENDENCIES
public static final int CU_AD_FORMAT_UNSIGNED_INT8
public static final int CU_AD_FORMAT_UNSIGNED_INT16
public static final int CU_AD_FORMAT_UNSIGNED_INT32
public static final int CU_AD_FORMAT_SIGNED_INT8
public static final int CU_AD_FORMAT_SIGNED_INT16
public static final int CU_AD_FORMAT_SIGNED_INT32
public static final int CU_AD_FORMAT_HALF
public static final int CU_AD_FORMAT_FLOAT
public static final int CU_AD_FORMAT_NV12
public static final int CU_AD_FORMAT_UNORM_INT8X1
public static final int CU_AD_FORMAT_UNORM_INT8X2
public static final int CU_AD_FORMAT_UNORM_INT8X4
public static final int CU_AD_FORMAT_UNORM_INT16X1
public static final int CU_AD_FORMAT_UNORM_INT16X2
public static final int CU_AD_FORMAT_UNORM_INT16X4
public static final int CU_AD_FORMAT_SNORM_INT8X1
public static final int CU_AD_FORMAT_SNORM_INT8X2
public static final int CU_AD_FORMAT_SNORM_INT8X4
public static final int CU_AD_FORMAT_SNORM_INT16X1
public static final int CU_AD_FORMAT_SNORM_INT16X2
public static final int CU_AD_FORMAT_SNORM_INT16X4
public static final int CU_AD_FORMAT_BC1_UNORM
public static final int CU_AD_FORMAT_BC1_UNORM_SRGB
public static final int CU_AD_FORMAT_BC2_UNORM
public static final int CU_AD_FORMAT_BC2_UNORM_SRGB
public static final int CU_AD_FORMAT_BC3_UNORM
public static final int CU_AD_FORMAT_BC3_UNORM_SRGB
public static final int CU_AD_FORMAT_BC4_UNORM
public static final int CU_AD_FORMAT_BC4_SNORM
public static final int CU_AD_FORMAT_BC5_UNORM
public static final int CU_AD_FORMAT_BC5_SNORM
public static final int CU_AD_FORMAT_BC6H_UF16
public static final int CU_AD_FORMAT_BC6H_SF16
public static final int CU_AD_FORMAT_BC7_UNORM
public static final int CU_AD_FORMAT_BC7_UNORM_SRGB
public static final int CU_TR_ADDRESS_MODE_WRAP
public static final int CU_TR_ADDRESS_MODE_CLAMP
public static final int CU_TR_ADDRESS_MODE_MIRROR
public static final int CU_TR_ADDRESS_MODE_BORDER
public static final int CU_TR_FILTER_MODE_POINT
public static final int CU_TR_FILTER_MODE_LINEAR
public static final int CU_DEVICE_ATTRIBUTE_MAX_THREADS_PER_BLOCK
public static final int CU_DEVICE_ATTRIBUTE_MAX_BLOCK_DIM_X
public static final int CU_DEVICE_ATTRIBUTE_MAX_BLOCK_DIM_Y
public static final int CU_DEVICE_ATTRIBUTE_MAX_BLOCK_DIM_Z
public static final int CU_DEVICE_ATTRIBUTE_MAX_GRID_DIM_X
public static final int CU_DEVICE_ATTRIBUTE_MAX_GRID_DIM_Y
public static final int CU_DEVICE_ATTRIBUTE_MAX_GRID_DIM_Z
public static final int CU_DEVICE_ATTRIBUTE_MAX_SHARED_MEMORY_PER_BLOCK
public static final int CU_DEVICE_ATTRIBUTE_SHARED_MEMORY_PER_BLOCK
public static final int CU_DEVICE_ATTRIBUTE_TOTAL_CONSTANT_MEMORY
public static final int CU_DEVICE_ATTRIBUTE_WARP_SIZE
public static final int CU_DEVICE_ATTRIBUTE_MAX_PITCH
public static final int CU_DEVICE_ATTRIBUTE_MAX_REGISTERS_PER_BLOCK
public static final int CU_DEVICE_ATTRIBUTE_REGISTERS_PER_BLOCK
public static final int CU_DEVICE_ATTRIBUTE_CLOCK_RATE
public static final int CU_DEVICE_ATTRIBUTE_TEXTURE_ALIGNMENT
public static final int CU_DEVICE_ATTRIBUTE_GPU_OVERLAP
public static final int CU_DEVICE_ATTRIBUTE_MULTIPROCESSOR_COUNT
public static final int CU_DEVICE_ATTRIBUTE_KERNEL_EXEC_TIMEOUT
public static final int CU_DEVICE_ATTRIBUTE_INTEGRATED
public static final int CU_DEVICE_ATTRIBUTE_CAN_MAP_HOST_MEMORY
public static final int CU_DEVICE_ATTRIBUTE_COMPUTE_MODE
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE1D_WIDTH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_WIDTH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_HEIGHT
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE3D_WIDTH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE3D_HEIGHT
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE3D_DEPTH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_LAYERED_WIDTH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_LAYERED_HEIGHT
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_LAYERED_LAYERS
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_ARRAY_WIDTH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_ARRAY_HEIGHT
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_ARRAY_NUMSLICES
public static final int CU_DEVICE_ATTRIBUTE_SURFACE_ALIGNMENT
public static final int CU_DEVICE_ATTRIBUTE_CONCURRENT_KERNELS
public static final int CU_DEVICE_ATTRIBUTE_ECC_ENABLED
public static final int CU_DEVICE_ATTRIBUTE_PCI_BUS_ID
public static final int CU_DEVICE_ATTRIBUTE_PCI_DEVICE_ID
public static final int CU_DEVICE_ATTRIBUTE_TCC_DRIVER
public static final int CU_DEVICE_ATTRIBUTE_MEMORY_CLOCK_RATE
public static final int CU_DEVICE_ATTRIBUTE_GLOBAL_MEMORY_BUS_WIDTH
public static final int CU_DEVICE_ATTRIBUTE_L2_CACHE_SIZE
public static final int CU_DEVICE_ATTRIBUTE_MAX_THREADS_PER_MULTIPROCESSOR
public static final int CU_DEVICE_ATTRIBUTE_ASYNC_ENGINE_COUNT
public static final int CU_DEVICE_ATTRIBUTE_UNIFIED_ADDRESSING
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE1D_LAYERED_WIDTH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE1D_LAYERED_LAYERS
public static final int CU_DEVICE_ATTRIBUTE_CAN_TEX2D_GATHER
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_GATHER_WIDTH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_GATHER_HEIGHT
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE3D_WIDTH_ALTERNATE
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE3D_HEIGHT_ALTERNATE
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE3D_DEPTH_ALTERNATE
public static final int CU_DEVICE_ATTRIBUTE_PCI_DOMAIN_ID
public static final int CU_DEVICE_ATTRIBUTE_TEXTURE_PITCH_ALIGNMENT
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURECUBEMAP_WIDTH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURECUBEMAP_LAYERED_WIDTH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURECUBEMAP_LAYERED_LAYERS
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE1D_WIDTH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE2D_WIDTH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE2D_HEIGHT
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE3D_WIDTH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE3D_HEIGHT
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE3D_DEPTH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE1D_LAYERED_WIDTH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE1D_LAYERED_LAYERS
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE2D_LAYERED_WIDTH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE2D_LAYERED_HEIGHT
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACE2D_LAYERED_LAYERS
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACECUBEMAP_WIDTH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACECUBEMAP_LAYERED_WIDTH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_SURFACECUBEMAP_LAYERED_LAYERS
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE1D_LINEAR_WIDTH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_LINEAR_WIDTH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_LINEAR_HEIGHT
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_LINEAR_PITCH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_MIPMAPPED_WIDTH
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_MIPMAPPED_HEIGHT
public static final int CU_DEVICE_ATTRIBUTE_COMPUTE_CAPABILITY_MAJOR
public static final int CU_DEVICE_ATTRIBUTE_COMPUTE_CAPABILITY_MINOR
public static final int CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE1D_MIPMAPPED_WIDTH
public static final int CU_DEVICE_ATTRIBUTE_STREAM_PRIORITIES_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_GLOBAL_L1_CACHE_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_LOCAL_L1_CACHE_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_MAX_SHARED_MEMORY_PER_MULTIPROCESSOR
public static final int CU_DEVICE_ATTRIBUTE_MAX_REGISTERS_PER_MULTIPROCESSOR
public static final int CU_DEVICE_ATTRIBUTE_MANAGED_MEMORY
public static final int CU_DEVICE_ATTRIBUTE_MULTI_GPU_BOARD
public static final int CU_DEVICE_ATTRIBUTE_MULTI_GPU_BOARD_GROUP_ID
public static final int CU_DEVICE_ATTRIBUTE_HOST_NATIVE_ATOMIC_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_SINGLE_TO_DOUBLE_PRECISION_PERF_RATIO
public static final int CU_DEVICE_ATTRIBUTE_PAGEABLE_MEMORY_ACCESS
public static final int CU_DEVICE_ATTRIBUTE_CONCURRENT_MANAGED_ACCESS
public static final int CU_DEVICE_ATTRIBUTE_COMPUTE_PREEMPTION_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_CAN_USE_HOST_POINTER_FOR_REGISTERED_MEM
public static final int CU_DEVICE_ATTRIBUTE_CAN_USE_STREAM_MEM_OPS_V1
public static final int CU_DEVICE_ATTRIBUTE_CAN_USE_64_BIT_STREAM_MEM_OPS_V1
public static final int CU_DEVICE_ATTRIBUTE_CAN_USE_STREAM_WAIT_VALUE_NOR_V1
public static final int CU_DEVICE_ATTRIBUTE_COOPERATIVE_LAUNCH
public static final int CU_DEVICE_ATTRIBUTE_COOPERATIVE_MULTI_DEVICE_LAUNCH
public static final int CU_DEVICE_ATTRIBUTE_MAX_SHARED_MEMORY_PER_BLOCK_OPTIN
public static final int CU_DEVICE_ATTRIBUTE_CAN_FLUSH_REMOTE_WRITES
public static final int CU_DEVICE_ATTRIBUTE_HOST_REGISTER_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_PAGEABLE_MEMORY_ACCESS_USES_HOST_PAGE_TABLES
public static final int CU_DEVICE_ATTRIBUTE_DIRECT_MANAGED_MEM_ACCESS_FROM_HOST
public static final int CU_DEVICE_ATTRIBUTE_VIRTUAL_ADDRESS_MANAGEMENT_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_VIRTUAL_MEMORY_MANAGEMENT_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_HANDLE_TYPE_POSIX_FILE_DESCRIPTOR_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_HANDLE_TYPE_WIN32_HANDLE_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_HANDLE_TYPE_WIN32_KMT_HANDLE_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_MAX_BLOCKS_PER_MULTIPROCESSOR
public static final int CU_DEVICE_ATTRIBUTE_GENERIC_COMPRESSION_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_MAX_PERSISTING_L2_CACHE_SIZE
public static final int CU_DEVICE_ATTRIBUTE_MAX_ACCESS_POLICY_WINDOW_SIZE
public static final int CU_DEVICE_ATTRIBUTE_GPU_DIRECT_RDMA_WITH_CUDA_VMM_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_RESERVED_SHARED_MEMORY_PER_BLOCK
public static final int CU_DEVICE_ATTRIBUTE_SPARSE_CUDA_ARRAY_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_READ_ONLY_HOST_REGISTER_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_TIMELINE_SEMAPHORE_INTEROP_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_MEMORY_POOLS_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_GPU_DIRECT_RDMA_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_GPU_DIRECT_RDMA_FLUSH_WRITES_OPTIONS
public static final int CU_DEVICE_ATTRIBUTE_GPU_DIRECT_RDMA_WRITES_ORDERING
public static final int CU_DEVICE_ATTRIBUTE_MEMPOOL_SUPPORTED_HANDLE_TYPES
public static final int CU_DEVICE_ATTRIBUTE_CLUSTER_LAUNCH
public static final int CU_DEVICE_ATTRIBUTE_DEFERRED_MAPPING_CUDA_ARRAY_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_CAN_USE_64_BIT_STREAM_MEM_OPS
public static final int CU_DEVICE_ATTRIBUTE_CAN_USE_STREAM_WAIT_VALUE_NOR
public static final int CU_DEVICE_ATTRIBUTE_DMA_BUF_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_IPC_EVENT_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_MEM_SYNC_DOMAIN_COUNT
public static final int CU_DEVICE_ATTRIBUTE_TENSOR_MAP_ACCESS_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_HANDLE_TYPE_FABRIC_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_UNIFIED_FUNCTION_POINTERS
public static final int CU_DEVICE_ATTRIBUTE_NUMA_CONFIG
public static final int CU_DEVICE_ATTRIBUTE_NUMA_ID
public static final int CU_DEVICE_ATTRIBUTE_MULTICAST_SUPPORTED
public static final int CU_DEVICE_ATTRIBUTE_MPS_ENABLED
public static final int CU_DEVICE_ATTRIBUTE_HOST_NUMA_ID
public static final int CU_DEVICE_ATTRIBUTE_MAX
public static final int CU_POINTER_ATTRIBUTE_CONTEXT
public static final int CU_POINTER_ATTRIBUTE_MEMORY_TYPE
public static final int CU_POINTER_ATTRIBUTE_DEVICE_POINTER
public static final int CU_POINTER_ATTRIBUTE_HOST_POINTER
public static final int CU_POINTER_ATTRIBUTE_P2P_TOKENS
public static final int CU_POINTER_ATTRIBUTE_SYNC_MEMOPS
public static final int CU_POINTER_ATTRIBUTE_BUFFER_ID
public static final int CU_POINTER_ATTRIBUTE_IS_MANAGED
public static final int CU_POINTER_ATTRIBUTE_DEVICE_ORDINAL
public static final int CU_POINTER_ATTRIBUTE_IS_LEGACY_CUDA_IPC_CAPABLE
public static final int CU_POINTER_ATTRIBUTE_RANGE_START_ADDR
public static final int CU_POINTER_ATTRIBUTE_RANGE_SIZE
public static final int CU_POINTER_ATTRIBUTE_MAPPED
public static final int CU_POINTER_ATTRIBUTE_ALLOWED_HANDLE_TYPES
public static final int CU_POINTER_ATTRIBUTE_IS_GPU_DIRECT_RDMA_CAPABLE
public static final int CU_POINTER_ATTRIBUTE_ACCESS_FLAGS
public static final int CU_POINTER_ATTRIBUTE_MEMPOOL_HANDLE
public static final int CU_POINTER_ATTRIBUTE_MAPPING_SIZE
public static final int CU_POINTER_ATTRIBUTE_MAPPING_BASE_ADDR
public static final int CU_POINTER_ATTRIBUTE_MEMORY_BLOCK_ID
public static final int CU_FUNC_ATTRIBUTE_MAX_THREADS_PER_BLOCK
public static final int CU_FUNC_ATTRIBUTE_SHARED_SIZE_BYTES
public static final int CU_FUNC_ATTRIBUTE_CONST_SIZE_BYTES
public static final int CU_FUNC_ATTRIBUTE_LOCAL_SIZE_BYTES
public static final int CU_FUNC_ATTRIBUTE_NUM_REGS
public static final int CU_FUNC_ATTRIBUTE_PTX_VERSION
public static final int CU_FUNC_ATTRIBUTE_BINARY_VERSION
public static final int CU_FUNC_ATTRIBUTE_CACHE_MODE_CA
public static final int CU_FUNC_ATTRIBUTE_MAX_DYNAMIC_SHARED_SIZE_BYTES
public static final int CU_FUNC_ATTRIBUTE_PREFERRED_SHARED_MEMORY_CARVEOUT
public static final int CU_FUNC_ATTRIBUTE_CLUSTER_SIZE_MUST_BE_SET
public static final int CU_FUNC_ATTRIBUTE_REQUIRED_CLUSTER_WIDTH
public static final int CU_FUNC_ATTRIBUTE_REQUIRED_CLUSTER_HEIGHT
public static final int CU_FUNC_ATTRIBUTE_REQUIRED_CLUSTER_DEPTH
public static final int CU_FUNC_ATTRIBUTE_NON_PORTABLE_CLUSTER_SIZE_ALLOWED
public static final int CU_FUNC_ATTRIBUTE_CLUSTER_SCHEDULING_POLICY_PREFERENCE
public static final int CU_FUNC_ATTRIBUTE_MAX
public static final int CU_FUNC_CACHE_PREFER_NONE
public static final int CU_FUNC_CACHE_PREFER_SHARED
public static final int CU_FUNC_CACHE_PREFER_L1
public static final int CU_FUNC_CACHE_PREFER_EQUAL
public static final int CU_SHARED_MEM_CONFIG_DEFAULT_BANK_SIZE
public static final int CU_SHARED_MEM_CONFIG_FOUR_BYTE_BANK_SIZE
public static final int CU_SHARED_MEM_CONFIG_EIGHT_BYTE_BANK_SIZE
public static final int CU_SHAREDMEM_CARVEOUT_DEFAULT
public static final int CU_SHAREDMEM_CARVEOUT_MAX_SHARED
public static final int CU_SHAREDMEM_CARVEOUT_MAX_L1
public static final int CU_MEMORYTYPE_HOST
public static final int CU_MEMORYTYPE_DEVICE
public static final int CU_MEMORYTYPE_ARRAY
public static final int CU_MEMORYTYPE_UNIFIED
public static final int CU_COMPUTEMODE_DEFAULT
public static final int CU_COMPUTEMODE_PROHIBITED
public static final int CU_COMPUTEMODE_EXCLUSIVE_PROCESS
public static final int CU_MEM_ADVISE_SET_READ_MOSTLY
public static final int CU_MEM_ADVISE_UNSET_READ_MOSTLY
public static final int CU_MEM_ADVISE_SET_PREFERRED_LOCATION
public static final int CU_MEM_ADVISE_UNSET_PREFERRED_LOCATION
public static final int CU_MEM_ADVISE_SET_ACCESSED_BY
public static final int CU_MEM_ADVISE_UNSET_ACCESSED_BY
public static final int CU_MEM_RANGE_ATTRIBUTE_READ_MOSTLY
public static final int CU_MEM_RANGE_ATTRIBUTE_PREFERRED_LOCATION
public static final int CU_MEM_RANGE_ATTRIBUTE_ACCESSED_BY
public static final int CU_MEM_RANGE_ATTRIBUTE_LAST_PREFETCH_LOCATION
public static final int CU_MEM_RANGE_ATTRIBUTE_PREFERRED_LOCATION_TYPE
public static final int CU_MEM_RANGE_ATTRIBUTE_PREFERRED_LOCATION_ID
public static final int CU_MEM_RANGE_ATTRIBUTE_LAST_PREFETCH_LOCATION_TYPE
public static final int CU_MEM_RANGE_ATTRIBUTE_LAST_PREFETCH_LOCATION_ID
public static final int CU_JIT_MAX_REGISTERS
public static final int CU_JIT_THREADS_PER_BLOCK
public static final int CU_JIT_WALL_TIME
public static final int CU_JIT_INFO_LOG_BUFFER
public static final int CU_JIT_INFO_LOG_BUFFER_SIZE_BYTES
public static final int CU_JIT_ERROR_LOG_BUFFER
public static final int CU_JIT_ERROR_LOG_BUFFER_SIZE_BYTES
public static final int CU_JIT_OPTIMIZATION_LEVEL
public static final int CU_JIT_TARGET_FROM_CUCONTEXT
public static final int CU_JIT_TARGET
public static final int CU_JIT_FALLBACK_STRATEGY
public static final int CU_JIT_GENERATE_DEBUG_INFO
public static final int CU_JIT_LOG_VERBOSE
public static final int CU_JIT_GENERATE_LINE_INFO
public static final int CU_JIT_CACHE_MODE
public static final int CU_JIT_NEW_SM3X_OPT
public static final int CU_JIT_FAST_COMPILE
public static final int CU_JIT_GLOBAL_SYMBOL_NAMES
public static final int CU_JIT_GLOBAL_SYMBOL_ADDRESSES
public static final int CU_JIT_GLOBAL_SYMBOL_COUNT
public static final int CU_JIT_LTO
public static final int CU_JIT_FTZ
public static final int CU_JIT_PREC_DIV
public static final int CU_JIT_PREC_SQRT
public static final int CU_JIT_FMA
public static final int CU_JIT_REFERENCED_KERNEL_NAMES
public static final int CU_JIT_REFERENCED_KERNEL_COUNT
public static final int CU_JIT_REFERENCED_VARIABLE_NAMES
public static final int CU_JIT_REFERENCED_VARIABLE_COUNT
public static final int CU_JIT_OPTIMIZE_UNUSED_DEVICE_VARIABLES
public static final int CU_JIT_POSITION_INDEPENDENT_CODE
public static final int CU_JIT_MIN_CTA_PER_SM
public static final int CU_JIT_NUM_OPTIONS
public static final int CU_COMPUTE_ACCELERATED_TARGET_BASE
public static final int CU_TARGET_COMPUTE_30
public static final int CU_TARGET_COMPUTE_32
public static final int CU_TARGET_COMPUTE_35
public static final int CU_TARGET_COMPUTE_37
public static final int CU_TARGET_COMPUTE_50
public static final int CU_TARGET_COMPUTE_52
public static final int CU_TARGET_COMPUTE_53
public static final int CU_TARGET_COMPUTE_60
public static final int CU_TARGET_COMPUTE_61
public static final int CU_TARGET_COMPUTE_62
public static final int CU_TARGET_COMPUTE_70
public static final int CU_TARGET_COMPUTE_72
public static final int CU_TARGET_COMPUTE_75
public static final int CU_TARGET_COMPUTE_80
public static final int CU_TARGET_COMPUTE_86
public static final int CU_TARGET_COMPUTE_87
public static final int CU_TARGET_COMPUTE_89
public static final int CU_TARGET_COMPUTE_90
public static final int CU_TARGET_COMPUTE_90A
public static final int CU_PREFER_PTX
public static final int CU_PREFER_BINARY
public static final int CU_JIT_CACHE_OPTION_NONE
public static final int CU_JIT_CACHE_OPTION_CG
public static final int CU_JIT_CACHE_OPTION_CA
public static final int CU_JIT_INPUT_CUBIN
public static final int CU_JIT_INPUT_PTX
public static final int CU_JIT_INPUT_FATBINARY
public static final int CU_JIT_INPUT_OBJECT
public static final int CU_JIT_INPUT_LIBRARY
public static final int CU_JIT_INPUT_NVVM
public static final int CU_JIT_NUM_INPUT_TYPES
public static final int CU_GRAPHICS_REGISTER_FLAGS_NONE
public static final int CU_GRAPHICS_REGISTER_FLAGS_READ_ONLY
public static final int CU_GRAPHICS_REGISTER_FLAGS_WRITE_DISCARD
public static final int CU_GRAPHICS_REGISTER_FLAGS_SURFACE_LDST
public static final int CU_GRAPHICS_REGISTER_FLAGS_TEXTURE_GATHER
public static final int CU_GRAPHICS_MAP_RESOURCE_FLAGS_NONE
public static final int CU_GRAPHICS_MAP_RESOURCE_FLAGS_READ_ONLY
public static final int CU_GRAPHICS_MAP_RESOURCE_FLAGS_WRITE_DISCARD
public static final int CU_CUBEMAP_FACE_POSITIVE_X
public static final int CU_CUBEMAP_FACE_NEGATIVE_X
public static final int CU_CUBEMAP_FACE_POSITIVE_Y
public static final int CU_CUBEMAP_FACE_NEGATIVE_Y
public static final int CU_CUBEMAP_FACE_POSITIVE_Z
public static final int CU_CUBEMAP_FACE_NEGATIVE_Z
public static final int CU_LIMIT_STACK_SIZE
public static final int CU_LIMIT_PRINTF_FIFO_SIZE
public static final int CU_LIMIT_MALLOC_HEAP_SIZE
public static final int CU_LIMIT_DEV_RUNTIME_SYNC_DEPTH
public static final int CU_LIMIT_DEV_RUNTIME_PENDING_LAUNCH_COUNT
public static final int CU_LIMIT_MAX_L2_FETCH_GRANULARITY
public static final int CU_LIMIT_PERSISTING_L2_CACHE_SIZE
public static final int CU_LIMIT_MAX
public static final int CU_RESOURCE_TYPE_ARRAY
public static final int CU_RESOURCE_TYPE_MIPMAPPED_ARRAY
public static final int CU_RESOURCE_TYPE_LINEAR
public static final int CU_RESOURCE_TYPE_PITCH2D
public static final int CU_ACCESS_PROPERTY_NORMAL
public static final int CU_ACCESS_PROPERTY_STREAMING
public static final int CU_ACCESS_PROPERTY_PERSISTING
public static final int CU_GRAPH_COND_ASSIGN_DEFAULT
public static final int CU_GRAPH_COND_TYPE_IF
public static final int CU_GRAPH_COND_TYPE_WHILE
public static final int CU_GRAPH_NODE_TYPE_KERNEL
public static final int CU_GRAPH_NODE_TYPE_MEMCPY
public static final int CU_GRAPH_NODE_TYPE_MEMSET
public static final int CU_GRAPH_NODE_TYPE_HOST
public static final int CU_GRAPH_NODE_TYPE_GRAPH
public static final int CU_GRAPH_NODE_TYPE_EMPTY
public static final int CU_GRAPH_NODE_TYPE_WAIT_EVENT
public static final int CU_GRAPH_NODE_TYPE_EVENT_RECORD
public static final int CU_GRAPH_NODE_TYPE_EXT_SEMAS_SIGNAL
public static final int CU_GRAPH_NODE_TYPE_EXT_SEMAS_WAIT
public static final int CU_GRAPH_NODE_TYPE_MEM_ALLOC
public static final int CU_GRAPH_NODE_TYPE_MEM_FREE
public static final int CU_GRAPH_NODE_TYPE_BATCH_MEM_OP
public static final int CU_GRAPH_NODE_TYPE_CONDITIONAL
public static final int CU_GRAPH_DEPENDENCY_TYPE_DEFAULT
public static final int CU_GRAPH_DEPENDENCY_TYPE_PROGRAMMATIC
public static final int CU_GRAPH_KERNEL_NODE_PORT_DEFAULT
public static final int CU_GRAPH_KERNEL_NODE_PORT_PROGRAMMATIC
public static final int CU_GRAPH_KERNEL_NODE_PORT_LAUNCH_ORDER
public static final int CUDA_GRAPH_INSTANTIATE_SUCCESS
public static final int CUDA_GRAPH_INSTANTIATE_ERROR
public static final int CUDA_GRAPH_INSTANTIATE_INVALID_STRUCTURE
public static final int CUDA_GRAPH_INSTANTIATE_NODE_OPERATION_NOT_SUPPORTED
public static final int CUDA_GRAPH_INSTANTIATE_MULTIPLE_CTXS_NOT_SUPPORTED
public static final int CU_SYNC_POLICY_AUTO
public static final int CU_SYNC_POLICY_SPIN
public static final int CU_SYNC_POLICY_YIELD
public static final int CU_SYNC_POLICY_BLOCKING_SYNC
public static final int CU_CLUSTER_SCHEDULING_POLICY_DEFAULT
public static final int CU_CLUSTER_SCHEDULING_POLICY_SPREAD
public static final int CU_CLUSTER_SCHEDULING_POLICY_LOAD_BALANCING
public static final int CU_LAUNCH_MEM_SYNC_DOMAIN_DEFAULT
public static final int CU_LAUNCH_MEM_SYNC_DOMAIN_REMOTE
public static final int CU_LAUNCH_ATTRIBUTE_IGNORE
public static final int CU_LAUNCH_ATTRIBUTE_ACCESS_POLICY_WINDOW
public static final int CU_LAUNCH_ATTRIBUTE_COOPERATIVE
public static final int CU_LAUNCH_ATTRIBUTE_SYNCHRONIZATION_POLICY
public static final int CU_LAUNCH_ATTRIBUTE_CLUSTER_DIMENSION
public static final int CU_LAUNCH_ATTRIBUTE_CLUSTER_SCHEDULING_POLICY_PREFERENCE
public static final int CU_LAUNCH_ATTRIBUTE_PROGRAMMATIC_STREAM_SERIALIZATION
public static final int CU_LAUNCH_ATTRIBUTE_PROGRAMMATIC_EVENT
public static final int CU_LAUNCH_ATTRIBUTE_PRIORITY
public static final int CU_LAUNCH_ATTRIBUTE_MEM_SYNC_DOMAIN_MAP
public static final int CU_LAUNCH_ATTRIBUTE_MEM_SYNC_DOMAIN
public static final int CU_LAUNCH_ATTRIBUTE_LAUNCH_COMPLETION_EVENT
public static final int CU_LAUNCH_ATTRIBUTE_MAX
public static final int CU_KERNEL_NODE_ATTRIBUTE_ACCESS_POLICY_WINDOW
public static final int CU_KERNEL_NODE_ATTRIBUTE_COOPERATIVE
public static final int CU_KERNEL_NODE_ATTRIBUTE_CLUSTER_DIMENSION
public static final int CU_KERNEL_NODE_ATTRIBUTE_CLUSTER_SCHEDULING_POLICY_PREFERENCE
public static final int CU_KERNEL_NODE_ATTRIBUTE_PRIORITY
public static final int CU_KERNEL_NODE_ATTRIBUTE_MEM_SYNC_DOMAIN_MAP
public static final int CU_KERNEL_NODE_ATTRIBUTE_MEM_SYNC_DOMAIN
public static final int CU_STREAM_CAPTURE_STATUS_NONE
public static final int CU_STREAM_CAPTURE_STATUS_ACTIVE
public static final int CU_STREAM_CAPTURE_STATUS_INVALIDATED
public static final int CU_STREAM_CAPTURE_MODE_GLOBAL
public static final int CU_STREAM_CAPTURE_MODE_THREAD_LOCAL
public static final int CU_STREAM_CAPTURE_MODE_RELAXED
public static final int CU_STREAM_ATTRIBUTE_ACCESS_POLICY_WINDOW
public static final int CU_STREAM_ATTRIBUTE_SYNCHRONIZATION_POLICY
public static final int CU_STREAM_ATTRIBUTE_PRIORITY
public static final int CU_STREAM_ATTRIBUTE_MEM_SYNC_DOMAIN_MAP
public static final int CU_STREAM_ATTRIBUTE_MEM_SYNC_DOMAIN
public static final int CU_GET_PROC_ADDRESS_DEFAULT
public static final int CU_GET_PROC_ADDRESS_LEGACY_STREAM
public static final int CU_GET_PROC_ADDRESS_PER_THREAD_DEFAULT_STREAM
public static final int CU_GET_PROC_ADDRESS_SUCCESS
public static final int CU_GET_PROC_ADDRESS_SYMBOL_NOT_FOUND
public static final int CU_GET_PROC_ADDRESS_VERSION_NOT_SUFFICIENT
public static final int CU_EXEC_AFFINITY_TYPE_SM_COUNT
public static final int CU_EXEC_AFFINITY_TYPE_MAX
public static final int CU_LIBRARY_HOST_UNIVERSAL_FUNCTION_AND_DATA_TABLE
public static final int CU_LIBRARY_BINARY_IS_PRESERVED
public static final int CU_LIBRARY_NUM_OPTIONS
public static final int CUDA_SUCCESS
public static final int CUDA_ERROR_INVALID_VALUE
public static final int CUDA_ERROR_OUT_OF_MEMORY
public static final int CUDA_ERROR_NOT_INITIALIZED
public static final int CUDA_ERROR_DEINITIALIZED
public static final int CUDA_ERROR_PROFILER_DISABLED
public static final int CUDA_ERROR_PROFILER_NOT_INITIALIZED
public static final int CUDA_ERROR_PROFILER_ALREADY_STARTED
public static final int CUDA_ERROR_PROFILER_ALREADY_STOPPED
public static final int CUDA_ERROR_STUB_LIBRARY
public static final int CUDA_ERROR_DEVICE_UNAVAILABLE
public static final int CUDA_ERROR_NO_DEVICE
public static final int CUDA_ERROR_INVALID_DEVICE
public static final int CUDA_ERROR_DEVICE_NOT_LICENSED
public static final int CUDA_ERROR_INVALID_IMAGE
public static final int CUDA_ERROR_INVALID_CONTEXT
public static final int CUDA_ERROR_CONTEXT_ALREADY_CURRENT
public static final int CUDA_ERROR_MAP_FAILED
public static final int CUDA_ERROR_UNMAP_FAILED
public static final int CUDA_ERROR_ARRAY_IS_MAPPED
public static final int CUDA_ERROR_ALREADY_MAPPED
public static final int CUDA_ERROR_NO_BINARY_FOR_GPU
public static final int CUDA_ERROR_ALREADY_ACQUIRED
public static final int CUDA_ERROR_NOT_MAPPED
public static final int CUDA_ERROR_NOT_MAPPED_AS_ARRAY
public static final int CUDA_ERROR_NOT_MAPPED_AS_POINTER
public static final int CUDA_ERROR_ECC_UNCORRECTABLE
public static final int CUDA_ERROR_UNSUPPORTED_LIMIT
public static final int CUDA_ERROR_CONTEXT_ALREADY_IN_USE
public static final int CUDA_ERROR_PEER_ACCESS_UNSUPPORTED
public static final int CUDA_ERROR_INVALID_PTX
public static final int CUDA_ERROR_INVALID_GRAPHICS_CONTEXT
public static final int CUDA_ERROR_NVLINK_UNCORRECTABLE
public static final int CUDA_ERROR_JIT_COMPILER_NOT_FOUND
public static final int CUDA_ERROR_UNSUPPORTED_PTX_VERSION
public static final int CUDA_ERROR_JIT_COMPILATION_DISABLED
public static final int CUDA_ERROR_UNSUPPORTED_EXEC_AFFINITY
public static final int CUDA_ERROR_UNSUPPORTED_DEVSIDE_SYNC
public static final int CUDA_ERROR_INVALID_SOURCE
public static final int CUDA_ERROR_FILE_NOT_FOUND
public static final int CUDA_ERROR_SHARED_OBJECT_SYMBOL_NOT_FOUND
public static final int CUDA_ERROR_SHARED_OBJECT_INIT_FAILED
public static final int CUDA_ERROR_OPERATING_SYSTEM
public static final int CUDA_ERROR_INVALID_HANDLE
public static final int CUDA_ERROR_ILLEGAL_STATE
public static final int CUDA_ERROR_LOSSY_QUERY
public static final int CUDA_ERROR_NOT_FOUND
public static final int CUDA_ERROR_NOT_READY
public static final int CUDA_ERROR_ILLEGAL_ADDRESS
public static final int CUDA_ERROR_LAUNCH_OUT_OF_RESOURCES
public static final int CUDA_ERROR_LAUNCH_TIMEOUT
public static final int CUDA_ERROR_LAUNCH_INCOMPATIBLE_TEXTURING
public static final int CUDA_ERROR_PEER_ACCESS_ALREADY_ENABLED
public static final int CUDA_ERROR_PEER_ACCESS_NOT_ENABLED
public static final int CUDA_ERROR_PRIMARY_CONTEXT_ACTIVE
public static final int CUDA_ERROR_CONTEXT_IS_DESTROYED
public static final int CUDA_ERROR_ASSERT
public static final int CUDA_ERROR_TOO_MANY_PEERS
public static final int CUDA_ERROR_HOST_MEMORY_ALREADY_REGISTERED
public static final int CUDA_ERROR_HOST_MEMORY_NOT_REGISTERED
public static final int CUDA_ERROR_HARDWARE_STACK_ERROR
public static final int CUDA_ERROR_ILLEGAL_INSTRUCTION
public static final int CUDA_ERROR_MISALIGNED_ADDRESS
public static final int CUDA_ERROR_INVALID_ADDRESS_SPACE
public static final int CUDA_ERROR_INVALID_PC
public static final int CUDA_ERROR_LAUNCH_FAILED
public static final int CUDA_ERROR_COOPERATIVE_LAUNCH_TOO_LARGE
public static final int CUDA_ERROR_NOT_PERMITTED
public static final int CUDA_ERROR_NOT_SUPPORTED
public static final int CUDA_ERROR_SYSTEM_NOT_READY
public static final int CUDA_ERROR_SYSTEM_DRIVER_MISMATCH
public static final int CUDA_ERROR_COMPAT_NOT_SUPPORTED_ON_DEVICE
public static final int CUDA_ERROR_MPS_CONNECTION_FAILED
public static final int CUDA_ERROR_MPS_RPC_FAILURE
public static final int CUDA_ERROR_MPS_SERVER_NOT_READY
public static final int CUDA_ERROR_MPS_MAX_CLIENTS_REACHED
public static final int CUDA_ERROR_MPS_MAX_CONNECTIONS_REACHED
public static final int CUDA_ERROR_MPS_CLIENT_TERMINATED
public static final int CUDA_ERROR_CDP_NOT_SUPPORTED
public static final int CUDA_ERROR_CDP_VERSION_MISMATCH
public static final int CUDA_ERROR_STREAM_CAPTURE_UNSUPPORTED
public static final int CUDA_ERROR_STREAM_CAPTURE_INVALIDATED
public static final int CUDA_ERROR_STREAM_CAPTURE_MERGE
public static final int CUDA_ERROR_STREAM_CAPTURE_UNMATCHED
public static final int CUDA_ERROR_STREAM_CAPTURE_UNJOINED
public static final int CUDA_ERROR_STREAM_CAPTURE_ISOLATION
public static final int CUDA_ERROR_STREAM_CAPTURE_IMPLICIT
public static final int CUDA_ERROR_CAPTURED_EVENT
public static final int CUDA_ERROR_STREAM_CAPTURE_WRONG_THREAD
public static final int CUDA_ERROR_TIMEOUT
public static final int CUDA_ERROR_GRAPH_EXEC_UPDATE_FAILURE
public static final int CUDA_ERROR_EXTERNAL_DEVICE
public static final int CUDA_ERROR_INVALID_CLUSTER_SIZE
public static final int CUDA_ERROR_UNKNOWN
public static final int CU_DEVICE_P2P_ATTRIBUTE_PERFORMANCE_RANK
public static final int CU_DEVICE_P2P_ATTRIBUTE_ACCESS_SUPPORTED
public static final int CU_DEVICE_P2P_ATTRIBUTE_NATIVE_ATOMIC_SUPPORTED
public static final int CU_DEVICE_P2P_ATTRIBUTE_ACCESS_ACCESS_SUPPORTED
public static final int CU_DEVICE_P2P_ATTRIBUTE_CUDA_ARRAY_ACCESS_SUPPORTED
public static final int CU_MEMHOSTALLOC_PORTABLE
public static final int CU_MEMHOSTALLOC_DEVICEMAP
public static final int CU_MEMHOSTALLOC_WRITECOMBINED
public static final int CU_MEMHOSTREGISTER_PORTABLE
public static final int CU_MEMHOSTREGISTER_DEVICEMAP
public static final int CU_MEMHOSTREGISTER_IOMEMORY
public static final int CU_MEMHOSTREGISTER_READ_ONLY
public static final int CU_ARRAY_SPARSE_PROPERTIES_SINGLE_MIPTAIL
public static final int CU_RES_VIEW_FORMAT_NONE
public static final int CU_RES_VIEW_FORMAT_UINT_1X8
public static final int CU_RES_VIEW_FORMAT_UINT_2X8
public static final int CU_RES_VIEW_FORMAT_UINT_4X8
public static final int CU_RES_VIEW_FORMAT_SINT_1X8
public static final int CU_RES_VIEW_FORMAT_SINT_2X8
public static final int CU_RES_VIEW_FORMAT_SINT_4X8
public static final int CU_RES_VIEW_FORMAT_UINT_1X16
public static final int CU_RES_VIEW_FORMAT_UINT_2X16
public static final int CU_RES_VIEW_FORMAT_UINT_4X16
public static final int CU_RES_VIEW_FORMAT_SINT_1X16
public static final int CU_RES_VIEW_FORMAT_SINT_2X16
public static final int CU_RES_VIEW_FORMAT_SINT_4X16
public static final int CU_RES_VIEW_FORMAT_UINT_1X32
public static final int CU_RES_VIEW_FORMAT_UINT_2X32
public static final int CU_RES_VIEW_FORMAT_UINT_4X32
public static final int CU_RES_VIEW_FORMAT_SINT_1X32
public static final int CU_RES_VIEW_FORMAT_SINT_2X32
public static final int CU_RES_VIEW_FORMAT_SINT_4X32
public static final int CU_RES_VIEW_FORMAT_FLOAT_1X16
public static final int CU_RES_VIEW_FORMAT_FLOAT_2X16
public static final int CU_RES_VIEW_FORMAT_FLOAT_4X16
public static final int CU_RES_VIEW_FORMAT_FLOAT_1X32
public static final int CU_RES_VIEW_FORMAT_FLOAT_2X32
public static final int CU_RES_VIEW_FORMAT_FLOAT_4X32
public static final int CU_RES_VIEW_FORMAT_UNSIGNED_BC1
public static final int CU_RES_VIEW_FORMAT_UNSIGNED_BC2
public static final int CU_RES_VIEW_FORMAT_UNSIGNED_BC3
public static final int CU_RES_VIEW_FORMAT_UNSIGNED_BC4
public static final int CU_RES_VIEW_FORMAT_SIGNED_BC4
public static final int CU_RES_VIEW_FORMAT_UNSIGNED_BC5
public static final int CU_RES_VIEW_FORMAT_SIGNED_BC5
public static final int CU_RES_VIEW_FORMAT_UNSIGNED_BC6H
public static final int CU_RES_VIEW_FORMAT_SIGNED_BC6H
public static final int CU_RES_VIEW_FORMAT_UNSIGNED_BC7
public static final int CU_TENSOR_MAP_NUM_QWORDS
public static final int CU_TENSOR_MAP_DATA_TYPE_UINT8
public static final int CU_TENSOR_MAP_DATA_TYPE_UINT16
public static final int CU_TENSOR_MAP_DATA_TYPE_UINT32
public static final int CU_TENSOR_MAP_DATA_TYPE_INT32
public static final int CU_TENSOR_MAP_DATA_TYPE_UINT64
public static final int CU_TENSOR_MAP_DATA_TYPE_INT64
public static final int CU_TENSOR_MAP_DATA_TYPE_FLOAT16
public static final int CU_TENSOR_MAP_DATA_TYPE_FLOAT32
public static final int CU_TENSOR_MAP_DATA_TYPE_FLOAT64
public static final int CU_TENSOR_MAP_DATA_TYPE_BFLOAT16
public static final int CU_TENSOR_MAP_DATA_TYPE_FLOAT32_FTZ
public static final int CU_TENSOR_MAP_DATA_TYPE_TFLOAT32
public static final int CU_TENSOR_MAP_DATA_TYPE_TFLOAT32_FTZ
public static final int CU_TENSOR_MAP_INTERLEAVE_NONE
public static final int CU_TENSOR_MAP_INTERLEAVE_16B
public static final int CU_TENSOR_MAP_INTERLEAVE_32B
public static final int CU_TENSOR_MAP_SWIZZLE_NONE
public static final int CU_TENSOR_MAP_SWIZZLE_32B
public static final int CU_TENSOR_MAP_SWIZZLE_64B
public static final int CU_TENSOR_MAP_SWIZZLE_128B
public static final int CU_TENSOR_MAP_L2_PROMOTION_NONE
public static final int CU_TENSOR_MAP_L2_PROMOTION_L2_64B
public static final int CU_TENSOR_MAP_L2_PROMOTION_L2_128B
public static final int CU_TENSOR_MAP_L2_PROMOTION_L2_256B
public static final int CU_TENSOR_MAP_FLOAT_OOB_FILL_NONE
public static final int CU_TENSOR_MAP_FLOAT_OOB_FILL_NAN_REQUEST_ZERO_FMA
public static final int CU_POINTER_ATTRIBUTE_ACCESS_FLAG_NONE
public static final int CU_POINTER_ATTRIBUTE_ACCESS_FLAG_READ
public static final int CU_POINTER_ATTRIBUTE_ACCESS_FLAG_READWRITE
public static final int CU_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_FD
public static final int CU_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32
public static final int CU_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_KMT
public static final int CU_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_HEAP
public static final int CU_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_RESOURCE
public static final int CU_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_RESOURCE
public static final int CU_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_RESOURCE_KMT
public static final int CU_EXTERNAL_MEMORY_HANDLE_TYPE_NVSCIBUF
public static final int CUDA_EXTERNAL_MEMORY_DEDICATED
public static final int CUDA_EXTERNAL_SEMAPHORE_SIGNAL_SKIP_NVSCIBUF_MEMSYNC
public static final int CUDA_EXTERNAL_SEMAPHORE_WAIT_SKIP_NVSCIBUF_MEMSYNC
public static final int CUDA_NVSCISYNC_ATTR_SIGNAL
public static final int CUDA_NVSCISYNC_ATTR_WAIT
public static final int CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_FD
public static final int CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32
public static final int CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_KMT
public static final int CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D12_FENCE
public static final int CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D11_FENCE
public static final int CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_NVSCISYNC
public static final int CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D11_KEYED_MUTEX
public static final int CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D11_KEYED_MUTEX_KMT
public static final int CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_TIMELINE_SEMAPHORE_FD
public static final int CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_TIMELINE_SEMAPHORE_WIN32
public static final int CU_MEM_HANDLE_TYPE_NONE
public static final int CU_MEM_HANDLE_TYPE_POSIX_FILE_DESCRIPTOR
public static final int CU_MEM_HANDLE_TYPE_WIN32
public static final int CU_MEM_HANDLE_TYPE_WIN32_KMT
public static final int CU_MEM_HANDLE_TYPE_FABRIC
public static final int CU_MEM_HANDLE_TYPE_MAX
public static final int CU_MEM_ACCESS_FLAGS_PROT_NONE
public static final int CU_MEM_ACCESS_FLAGS_PROT_READ
public static final int CU_MEM_ACCESS_FLAGS_PROT_READWRITE
public static final int CU_MEM_ACCESS_FLAGS_PROT_MAX
public static final int CU_MEM_LOCATION_TYPE_INVALID
public static final int CU_MEM_LOCATION_TYPE_DEVICE
public static final int CU_MEM_LOCATION_TYPE_HOST
public static final int CU_MEM_LOCATION_TYPE_HOST_NUMA
public static final int CU_MEM_LOCATION_TYPE_HOST_NUMA_CURRENT
public static final int CU_MEM_LOCATION_TYPE_MAX
public static final int CU_MEM_ALLOCATION_TYPE_INVALID
public static final int CU_MEM_ALLOCATION_TYPE_PINNED
public static final int CU_MEM_ALLOCATION_TYPE_MAX
public static final int CU_MEM_ALLOC_GRANULARITY_MINIMUM
public static final int CU_MEM_ALLOC_GRANULARITY_RECOMMENDED
public static final int CU_MEM_RANGE_HANDLE_TYPE_DMA_BUF_FD
public static final int CU_MEM_RANGE_HANDLE_TYPE_MAX
public static final int CU_ARRAY_SPARSE_SUBRESOURCE_TYPE_SPARSE_LEVEL
public static final int CU_ARRAY_SPARSE_SUBRESOURCE_TYPE_MIPTAIL
public static final int CU_MEM_OPERATION_TYPE_MAP
public static final int CU_MEM_OPERATION_TYPE_UNMAP
public static final int CU_MEM_HANDLE_TYPE_GENERIC
public static final int CU_MEM_ALLOCATION_COMP_NONE
public static final int CU_MEM_ALLOCATION_COMP_GENERIC
public static final int CU_MEM_CREATE_USAGE_TILE_POOL
public static final int CU_MULTICAST_GRANULARITY_MINIMUM
public static final int CU_MULTICAST_GRANULARITY_RECOMMENDED
public static final int CU_GRAPH_EXEC_UPDATE_SUCCESS
public static final int CU_GRAPH_EXEC_UPDATE_ERROR
public static final int CU_GRAPH_EXEC_UPDATE_ERROR_TOPOLOGY_CHANGED
public static final int CU_GRAPH_EXEC_UPDATE_ERROR_NODE_TYPE_CHANGED
public static final int CU_GRAPH_EXEC_UPDATE_ERROR_FUNCTION_CHANGED
public static final int CU_GRAPH_EXEC_UPDATE_ERROR_PARAMETERS_CHANGED
public static final int CU_GRAPH_EXEC_UPDATE_ERROR_NOT_SUPPORTED
public static final int CU_GRAPH_EXEC_UPDATE_ERROR_UNSUPPORTED_FUNCTION_CHANGE
public static final int CU_GRAPH_EXEC_UPDATE_ERROR_ATTRIBUTES_CHANGED
public static final int CU_MEMPOOL_ATTR_REUSE_FOLLOW_EVENT_DEPENDENCIES
public static final int CU_MEMPOOL_ATTR_REUSE_ALLOW_OPPORTUNISTIC
public static final int CU_MEMPOOL_ATTR_REUSE_ALLOW_INTERNAL_DEPENDENCIES
public static final int CU_MEMPOOL_ATTR_RELEASE_THRESHOLD
public static final int CU_MEMPOOL_ATTR_RESERVED_MEM_CURRENT
public static final int CU_MEMPOOL_ATTR_RESERVED_MEM_HIGH
public static final int CU_MEMPOOL_ATTR_USED_MEM_CURRENT
public static final int CU_MEMPOOL_ATTR_USED_MEM_HIGH
public static final int CU_GRAPH_MEM_ATTR_USED_MEM_CURRENT
public static final int CU_GRAPH_MEM_ATTR_USED_MEM_HIGH
public static final int CU_GRAPH_MEM_ATTR_RESERVED_MEM_CURRENT
public static final int CU_GRAPH_MEM_ATTR_RESERVED_MEM_HIGH
public static final int CUDA_COOPERATIVE_LAUNCH_MULTI_DEVICE_NO_PRE_LAUNCH_SYNC
public static final int CUDA_COOPERATIVE_LAUNCH_MULTI_DEVICE_NO_POST_LAUNCH_SYNC
public static final int CUDA_ARRAY3D_LAYERED
public static final int CUDA_ARRAY3D_2DARRAY
public static final int CUDA_ARRAY3D_SURFACE_LDST
public static final int CUDA_ARRAY3D_CUBEMAP
public static final int CUDA_ARRAY3D_TEXTURE_GATHER
public static final int CUDA_ARRAY3D_DEPTH_TEXTURE
public static final int CUDA_ARRAY3D_COLOR_ATTACHMENT
public static final int CUDA_ARRAY3D_SPARSE
public static final int CUDA_ARRAY3D_DEFERRED_MAPPING
public static final int CU_TRSA_OVERRIDE_FORMAT
public static final int CU_TRSF_READ_AS_INTEGER
public static final int CU_TRSF_NORMALIZED_COORDINATES
public static final int CU_TRSF_SRGB
public static final int CU_TRSF_DISABLE_TRILINEAR_OPTIMIZATION
public static final int CU_TRSF_SEAMLESS_CUBEMAP
public static final int CU_LAUNCH_PARAM_END_AS_INT
public static final Pointer CU_LAUNCH_PARAM_END
public static final int CU_LAUNCH_PARAM_BUFFER_POINTER_AS_INT
public static final Pointer CU_LAUNCH_PARAM_BUFFER_POINTER
public static final int CU_LAUNCH_PARAM_BUFFER_SIZE_AS_INT
public static final Pointer CU_LAUNCH_PARAM_BUFFER_SIZE
public static final int CU_PARAM_TR_DEFAULT
public static final int CU_DEVICE_CPU
public static final int CU_DEVICE_INVALID
public static final int CU_FLUSH_GPU_DIRECT_RDMA_WRITES_OPTION_HOST
public static final int CU_FLUSH_GPU_DIRECT_RDMA_WRITES_OPTION_MEMOPS
public static final int CU_GPU_DIRECT_RDMA_WRITES_ORDERING_NONE
public static final int CU_GPU_DIRECT_RDMA_WRITES_ORDERING_OWNER
public static final int CU_GPU_DIRECT_RDMA_WRITES_ORDERING_ALL_DEVICES
public static final int CU_FLUSH_GPU_DIRECT_RDMA_WRITES_TO_OWNER
public static final int CU_FLUSH_GPU_DIRECT_RDMA_WRITES_TO_ALL_DEVICES
public static final int CU_FLUSH_GPU_DIRECT_RDMA_WRITES_TARGET_CURRENT_CTX
public static final int CU_GRAPH_DEBUG_DOT_FLAGS_VERBOSE
public static final int CU_GRAPH_DEBUG_DOT_FLAGS_RUNTIME_TYPES
public static final int CU_GRAPH_DEBUG_DOT_FLAGS_KERNEL_NODE_PARAMS
public static final int CU_GRAPH_DEBUG_DOT_FLAGS_MEMCPY_NODE_PARAMS
public static final int CU_GRAPH_DEBUG_DOT_FLAGS_MEMSET_NODE_PARAMS
public static final int CU_GRAPH_DEBUG_DOT_FLAGS_HOST_NODE_PARAMS
public static final int CU_GRAPH_DEBUG_DOT_FLAGS_EVENT_NODE_PARAMS
public static final int CU_GRAPH_DEBUG_DOT_FLAGS_EXT_SEMAS_SIGNAL_NODE_PARAMS
public static final int CU_GRAPH_DEBUG_DOT_FLAGS_EXT_SEMAS_WAIT_NODE_PARAMS
public static final int CU_GRAPH_DEBUG_DOT_FLAGS_KERNEL_NODE_ATTRIBUTES
public static final int CU_GRAPH_DEBUG_DOT_FLAGS_HANDLES
public static final int CU_GRAPH_DEBUG_DOT_FLAGS_MEM_ALLOC_NODE_PARAMS
public static final int CU_GRAPH_DEBUG_DOT_FLAGS_MEM_FREE_NODE_PARAMS
public static final int CU_GRAPH_DEBUG_DOT_FLAGS_BATCH_MEM_OP_NODE_PARAMS
public static final int CU_GRAPH_DEBUG_DOT_FLAGS_EXTRA_TOPO_INFO
public static final int CU_GRAPH_DEBUG_DOT_FLAGS_CONDITIONAL_NODE_PARAMS
public static final int CU_USER_OBJECT_NO_DESTRUCTOR_SYNC
public static final int CU_GRAPH_USER_OBJECT_MOVE
public static final int CUDA_GRAPH_INSTANTIATE_FLAG_AUTO_FREE_ON_LAUNCH
public static final int CUDA_GRAPH_INSTANTIATE_FLAG_UPLOAD
public static final int CUDA_GRAPH_INSTANTIATE_FLAG_DEVICE_LAUNCH
public static final int CUDA_GRAPH_INSTANTIATE_FLAG_USE_NODE_PRIORITY
public static final int CU_DEVICE_NUMA_CONFIG_NONE
public static final int CU_DEVICE_NUMA_CONFIG_NUMA_NODE
public static final int CU_MODULE_EAGER_LOADING
public static final int CU_MODULE_LAZY_LOADING
public static final int CU_COREDUMP_ENABLE_ON_EXCEPTION
public static final int CU_COREDUMP_TRIGGER_HOST
public static final int CU_COREDUMP_LIGHTWEIGHT
public static final int CU_COREDUMP_ENABLE_USER_TRIGGER
public static final int CU_COREDUMP_FILE
public static final int CU_COREDUMP_PIPE
public static final int CU_COREDUMP_MAX
public static final int cudaRoundNearest
public static final int cudaRoundZero
public static final int cudaRoundPosInf
public static final int cudaRoundMinInf
public static final int cudaHostAllocDefault
public static final int cudaHostAllocPortable
public static final int cudaHostAllocMapped
public static final int cudaHostAllocWriteCombined
public static final int cudaHostRegisterDefault
public static final int cudaHostRegisterPortable
public static final int cudaHostRegisterMapped
public static final int cudaHostRegisterIoMemory
public static final int cudaHostRegisterReadOnly
public static final int cudaPeerAccessDefault
public static final int cudaStreamDefault
public static final int cudaStreamNonBlocking
public static final CUstream_st cudaStreamLegacy
public static final CUstream_st cudaStreamPerThread
public static final int cudaEventDefault
public static final int cudaEventBlockingSync
public static final int cudaEventDisableTiming
public static final int cudaEventInterprocess
public static final int cudaEventRecordDefault
public static final int cudaEventRecordExternal
public static final int cudaEventWaitDefault
public static final int cudaEventWaitExternal
public static final int cudaDeviceScheduleAuto
public static final int cudaDeviceScheduleSpin
public static final int cudaDeviceScheduleYield
public static final int cudaDeviceScheduleBlockingSync
public static final int cudaDeviceBlockingSync
public static final int cudaDeviceScheduleMask
public static final int cudaDeviceMapHost
public static final int cudaDeviceLmemResizeToMax
public static final int cudaDeviceSyncMemops
public static final int cudaDeviceMask
public static final int cudaArrayDefault
public static final int cudaArrayLayered
public static final int cudaArraySurfaceLoadStore
public static final int cudaArrayCubemap
public static final int cudaArrayTextureGather
public static final int cudaArrayColorAttachment
public static final int cudaArraySparse
public static final int cudaArrayDeferredMapping
public static final int cudaIpcMemLazyEnablePeerAccess
public static final int cudaMemAttachGlobal
public static final int cudaMemAttachHost
public static final int cudaMemAttachSingle
public static final int cudaOccupancyDefault
public static final int cudaOccupancyDisableCachingOverride
public static final int cudaCpuDeviceId
public static final int cudaInvalidDeviceId
public static final int cudaInitDeviceFlagsAreValid
public static final int cudaCooperativeLaunchMultiDeviceNoPreSync
public static final int cudaCooperativeLaunchMultiDeviceNoPostSync
public static final int cudaSuccess
public static final int cudaErrorInvalidValue
public static final int cudaErrorMemoryAllocation
public static final int cudaErrorInitializationError
public static final int cudaErrorCudartUnloading
public static final int cudaErrorProfilerDisabled
public static final int cudaErrorProfilerNotInitialized
public static final int cudaErrorProfilerAlreadyStarted
public static final int cudaErrorProfilerAlreadyStopped
public static final int cudaErrorInvalidConfiguration
public static final int cudaErrorInvalidPitchValue
public static final int cudaErrorInvalidSymbol
public static final int cudaErrorInvalidHostPointer
public static final int cudaErrorInvalidDevicePointer
public static final int cudaErrorInvalidTexture
public static final int cudaErrorInvalidTextureBinding
public static final int cudaErrorInvalidChannelDescriptor
public static final int cudaErrorInvalidMemcpyDirection
public static final int cudaErrorAddressOfConstant
public static final int cudaErrorTextureFetchFailed
public static final int cudaErrorTextureNotBound
public static final int cudaErrorSynchronizationError
public static final int cudaErrorInvalidFilterSetting
public static final int cudaErrorInvalidNormSetting
public static final int cudaErrorMixedDeviceExecution
public static final int cudaErrorNotYetImplemented
public static final int cudaErrorMemoryValueTooLarge
public static final int cudaErrorStubLibrary
public static final int cudaErrorInsufficientDriver
public static final int cudaErrorCallRequiresNewerDriver
public static final int cudaErrorInvalidSurface
public static final int cudaErrorDuplicateVariableName
public static final int cudaErrorDuplicateTextureName
public static final int cudaErrorDuplicateSurfaceName
public static final int cudaErrorDevicesUnavailable
public static final int cudaErrorIncompatibleDriverContext
public static final int cudaErrorMissingConfiguration
public static final int cudaErrorPriorLaunchFailure
public static final int cudaErrorLaunchMaxDepthExceeded
public static final int cudaErrorLaunchFileScopedTex
public static final int cudaErrorLaunchFileScopedSurf
public static final int cudaErrorSyncDepthExceeded
public static final int cudaErrorLaunchPendingCountExceeded
public static final int cudaErrorInvalidDeviceFunction
public static final int cudaErrorNoDevice
public static final int cudaErrorInvalidDevice
public static final int cudaErrorDeviceNotLicensed
public static final int cudaErrorSoftwareValidityNotEstablished
public static final int cudaErrorStartupFailure
public static final int cudaErrorInvalidKernelImage
public static final int cudaErrorDeviceUninitialized
public static final int cudaErrorMapBufferObjectFailed
public static final int cudaErrorUnmapBufferObjectFailed
public static final int cudaErrorArrayIsMapped
public static final int cudaErrorAlreadyMapped
public static final int cudaErrorNoKernelImageForDevice
public static final int cudaErrorAlreadyAcquired
public static final int cudaErrorNotMapped
public static final int cudaErrorNotMappedAsArray
public static final int cudaErrorNotMappedAsPointer
public static final int cudaErrorECCUncorrectable
public static final int cudaErrorUnsupportedLimit
public static final int cudaErrorDeviceAlreadyInUse
public static final int cudaErrorPeerAccessUnsupported
public static final int cudaErrorInvalidPtx
public static final int cudaErrorInvalidGraphicsContext
public static final int cudaErrorNvlinkUncorrectable
public static final int cudaErrorJitCompilerNotFound
public static final int cudaErrorUnsupportedPtxVersion
public static final int cudaErrorJitCompilationDisabled
public static final int cudaErrorUnsupportedExecAffinity
public static final int cudaErrorUnsupportedDevSideSync
public static final int cudaErrorInvalidSource
public static final int cudaErrorFileNotFound
public static final int cudaErrorSharedObjectSymbolNotFound
public static final int cudaErrorSharedObjectInitFailed
public static final int cudaErrorOperatingSystem
public static final int cudaErrorInvalidResourceHandle
public static final int cudaErrorIllegalState
public static final int cudaErrorLossyQuery
public static final int cudaErrorSymbolNotFound
public static final int cudaErrorNotReady
public static final int cudaErrorIllegalAddress
public static final int cudaErrorLaunchOutOfResources
public static final int cudaErrorLaunchTimeout
public static final int cudaErrorLaunchIncompatibleTexturing
public static final int cudaErrorPeerAccessAlreadyEnabled
public static final int cudaErrorPeerAccessNotEnabled
public static final int cudaErrorSetOnActiveProcess
public static final int cudaErrorContextIsDestroyed
public static final int cudaErrorAssert
public static final int cudaErrorTooManyPeers
public static final int cudaErrorHostMemoryAlreadyRegistered
public static final int cudaErrorHostMemoryNotRegistered
public static final int cudaErrorHardwareStackError
public static final int cudaErrorIllegalInstruction
public static final int cudaErrorMisalignedAddress
public static final int cudaErrorInvalidAddressSpace
public static final int cudaErrorInvalidPc
public static final int cudaErrorLaunchFailure
public static final int cudaErrorCooperativeLaunchTooLarge
public static final int cudaErrorNotPermitted
public static final int cudaErrorNotSupported
public static final int cudaErrorSystemNotReady
public static final int cudaErrorSystemDriverMismatch
public static final int cudaErrorCompatNotSupportedOnDevice
public static final int cudaErrorMpsConnectionFailed
public static final int cudaErrorMpsRpcFailure
public static final int cudaErrorMpsServerNotReady
public static final int cudaErrorMpsMaxClientsReached
public static final int cudaErrorMpsMaxConnectionsReached
public static final int cudaErrorMpsClientTerminated
public static final int cudaErrorCdpNotSupported
public static final int cudaErrorCdpVersionMismatch
public static final int cudaErrorStreamCaptureUnsupported
public static final int cudaErrorStreamCaptureInvalidated
public static final int cudaErrorStreamCaptureMerge
public static final int cudaErrorStreamCaptureUnmatched
public static final int cudaErrorStreamCaptureUnjoined
public static final int cudaErrorStreamCaptureIsolation
public static final int cudaErrorStreamCaptureImplicit
public static final int cudaErrorCapturedEvent
public static final int cudaErrorStreamCaptureWrongThread
public static final int cudaErrorTimeout
public static final int cudaErrorGraphExecUpdateFailure
public static final int cudaErrorExternalDevice
public static final int cudaErrorInvalidClusterSize
public static final int cudaErrorUnknown
public static final int cudaErrorApiFailureBase
public static final int cudaChannelFormatKindSigned
public static final int cudaChannelFormatKindUnsigned
public static final int cudaChannelFormatKindFloat
public static final int cudaChannelFormatKindNone
public static final int cudaChannelFormatKindNV12
public static final int cudaChannelFormatKindUnsignedNormalized8X1
public static final int cudaChannelFormatKindUnsignedNormalized8X2
public static final int cudaChannelFormatKindUnsignedNormalized8X4
public static final int cudaChannelFormatKindUnsignedNormalized16X1
public static final int cudaChannelFormatKindUnsignedNormalized16X2
public static final int cudaChannelFormatKindUnsignedNormalized16X4
public static final int cudaChannelFormatKindSignedNormalized8X1
public static final int cudaChannelFormatKindSignedNormalized8X2
public static final int cudaChannelFormatKindSignedNormalized8X4
public static final int cudaChannelFormatKindSignedNormalized16X1
public static final int cudaChannelFormatKindSignedNormalized16X2
public static final int cudaChannelFormatKindSignedNormalized16X4
public static final int cudaChannelFormatKindUnsignedBlockCompressed1
public static final int cudaChannelFormatKindUnsignedBlockCompressed1SRGB
public static final int cudaChannelFormatKindUnsignedBlockCompressed2
public static final int cudaChannelFormatKindUnsignedBlockCompressed2SRGB
public static final int cudaChannelFormatKindUnsignedBlockCompressed3
public static final int cudaChannelFormatKindUnsignedBlockCompressed3SRGB
public static final int cudaChannelFormatKindUnsignedBlockCompressed4
public static final int cudaChannelFormatKindSignedBlockCompressed4
public static final int cudaChannelFormatKindUnsignedBlockCompressed5
public static final int cudaChannelFormatKindSignedBlockCompressed5
public static final int cudaChannelFormatKindUnsignedBlockCompressed6H
public static final int cudaChannelFormatKindSignedBlockCompressed6H
public static final int cudaChannelFormatKindUnsignedBlockCompressed7
public static final int cudaChannelFormatKindUnsignedBlockCompressed7SRGB
public static final int cudaArraySparsePropertiesSingleMipTail
public static final int cudaMemoryTypeUnregistered
public static final int cudaMemoryTypeHost
public static final int cudaMemoryTypeDevice
public static final int cudaMemoryTypeManaged
public static final int cudaMemcpyHostToHost
public static final int cudaMemcpyHostToDevice
public static final int cudaMemcpyDeviceToHost
public static final int cudaMemcpyDeviceToDevice
public static final int cudaMemcpyDefault
public static final int cudaAccessPropertyNormal
public static final int cudaAccessPropertyStreaming
public static final int cudaAccessPropertyPersisting
public static final int cudaStreamCaptureStatusNone
public static final int cudaStreamCaptureStatusActive
public static final int cudaStreamCaptureStatusInvalidated
public static final int cudaStreamCaptureModeGlobal
public static final int cudaStreamCaptureModeThreadLocal
public static final int cudaStreamCaptureModeRelaxed
public static final int cudaSyncPolicyAuto
public static final int cudaSyncPolicySpin
public static final int cudaSyncPolicyYield
public static final int cudaSyncPolicyBlockingSync
public static final int cudaClusterSchedulingPolicyDefault
public static final int cudaClusterSchedulingPolicySpread
public static final int cudaClusterSchedulingPolicyLoadBalancing
public static final int cudaStreamAddCaptureDependencies
public static final int cudaStreamSetCaptureDependencies
public static final int cudaUserObjectNoDestructorSync
public static final int cudaGraphUserObjectMove
public static final int cudaGraphicsRegisterFlagsNone
public static final int cudaGraphicsRegisterFlagsReadOnly
public static final int cudaGraphicsRegisterFlagsWriteDiscard
public static final int cudaGraphicsRegisterFlagsSurfaceLoadStore
public static final int cudaGraphicsRegisterFlagsTextureGather
public static final int cudaGraphicsMapFlagsNone
public static final int cudaGraphicsMapFlagsReadOnly
public static final int cudaGraphicsMapFlagsWriteDiscard
public static final int cudaGraphicsCubeFacePositiveX
public static final int cudaGraphicsCubeFaceNegativeX
public static final int cudaGraphicsCubeFacePositiveY
public static final int cudaGraphicsCubeFaceNegativeY
public static final int cudaGraphicsCubeFacePositiveZ
public static final int cudaGraphicsCubeFaceNegativeZ
public static final int cudaResourceTypeArray
public static final int cudaResourceTypeMipmappedArray
public static final int cudaResourceTypeLinear
public static final int cudaResourceTypePitch2D
public static final int cudaResViewFormatNone
public static final int cudaResViewFormatUnsignedChar1
public static final int cudaResViewFormatUnsignedChar2
public static final int cudaResViewFormatUnsignedChar4
public static final int cudaResViewFormatSignedChar1
public static final int cudaResViewFormatSignedChar2
public static final int cudaResViewFormatSignedChar4
public static final int cudaResViewFormatUnsignedShort1
public static final int cudaResViewFormatUnsignedShort2
public static final int cudaResViewFormatUnsignedShort4
public static final int cudaResViewFormatSignedShort1
public static final int cudaResViewFormatSignedShort2
public static final int cudaResViewFormatSignedShort4
public static final int cudaResViewFormatUnsignedInt1
public static final int cudaResViewFormatUnsignedInt2
public static final int cudaResViewFormatUnsignedInt4
public static final int cudaResViewFormatSignedInt1
public static final int cudaResViewFormatSignedInt2
public static final int cudaResViewFormatSignedInt4
public static final int cudaResViewFormatHalf1
public static final int cudaResViewFormatHalf2
public static final int cudaResViewFormatHalf4
public static final int cudaResViewFormatFloat1
public static final int cudaResViewFormatFloat2
public static final int cudaResViewFormatFloat4
public static final int cudaResViewFormatUnsignedBlockCompressed1
public static final int cudaResViewFormatUnsignedBlockCompressed2
public static final int cudaResViewFormatUnsignedBlockCompressed3
public static final int cudaResViewFormatUnsignedBlockCompressed4
public static final int cudaResViewFormatSignedBlockCompressed4
public static final int cudaResViewFormatUnsignedBlockCompressed5
public static final int cudaResViewFormatSignedBlockCompressed5
public static final int cudaResViewFormatUnsignedBlockCompressed6H
public static final int cudaResViewFormatSignedBlockCompressed6H
public static final int cudaResViewFormatUnsignedBlockCompressed7
public static final int cudaFuncAttributeMaxDynamicSharedMemorySize
public static final int cudaFuncAttributePreferredSharedMemoryCarveout
public static final int cudaFuncAttributeClusterDimMustBeSet
public static final int cudaFuncAttributeRequiredClusterWidth
public static final int cudaFuncAttributeRequiredClusterHeight
public static final int cudaFuncAttributeRequiredClusterDepth
public static final int cudaFuncAttributeNonPortableClusterSizeAllowed
public static final int cudaFuncAttributeClusterSchedulingPolicyPreference
public static final int cudaFuncAttributeMax
public static final int cudaFuncCachePreferNone
public static final int cudaFuncCachePreferShared
public static final int cudaFuncCachePreferL1
public static final int cudaFuncCachePreferEqual
public static final int cudaSharedMemBankSizeDefault
public static final int cudaSharedMemBankSizeFourByte
public static final int cudaSharedMemBankSizeEightByte
public static final int cudaSharedmemCarveoutDefault
public static final int cudaSharedmemCarveoutMaxShared
public static final int cudaSharedmemCarveoutMaxL1
public static final int cudaComputeModeDefault
public static final int cudaComputeModeExclusive
public static final int cudaComputeModeProhibited
public static final int cudaComputeModeExclusiveProcess
public static final int cudaLimitStackSize
public static final int cudaLimitPrintfFifoSize
public static final int cudaLimitMallocHeapSize
public static final int cudaLimitDevRuntimeSyncDepth
public static final int cudaLimitDevRuntimePendingLaunchCount
public static final int cudaLimitMaxL2FetchGranularity
public static final int cudaLimitPersistingL2CacheSize
public static final int cudaMemAdviseSetReadMostly
public static final int cudaMemAdviseUnsetReadMostly
public static final int cudaMemAdviseSetPreferredLocation
public static final int cudaMemAdviseUnsetPreferredLocation
public static final int cudaMemAdviseSetAccessedBy
public static final int cudaMemAdviseUnsetAccessedBy
public static final int cudaMemRangeAttributeReadMostly
public static final int cudaMemRangeAttributePreferredLocation
public static final int cudaMemRangeAttributeAccessedBy
public static final int cudaMemRangeAttributeLastPrefetchLocation
public static final int cudaMemRangeAttributePreferredLocationType
public static final int cudaMemRangeAttributePreferredLocationId
public static final int cudaMemRangeAttributeLastPrefetchLocationType
public static final int cudaMemRangeAttributeLastPrefetchLocationId
public static final int cudaFlushGPUDirectRDMAWritesOptionHost
public static final int cudaFlushGPUDirectRDMAWritesOptionMemOps
public static final int cudaGPUDirectRDMAWritesOrderingNone
public static final int cudaGPUDirectRDMAWritesOrderingOwner
public static final int cudaGPUDirectRDMAWritesOrderingAllDevices
public static final int cudaFlushGPUDirectRDMAWritesToOwner
public static final int cudaFlushGPUDirectRDMAWritesToAllDevices
public static final int cudaFlushGPUDirectRDMAWritesTargetCurrentDevice
public static final int cudaDevAttrMaxThreadsPerBlock
public static final int cudaDevAttrMaxBlockDimX
public static final int cudaDevAttrMaxBlockDimY
public static final int cudaDevAttrMaxBlockDimZ
public static final int cudaDevAttrMaxGridDimX
public static final int cudaDevAttrMaxGridDimY
public static final int cudaDevAttrMaxGridDimZ
public static final int cudaDevAttrMaxSharedMemoryPerBlock
public static final int cudaDevAttrTotalConstantMemory
public static final int cudaDevAttrWarpSize
public static final int cudaDevAttrMaxPitch
public static final int cudaDevAttrMaxRegistersPerBlock
public static final int cudaDevAttrClockRate
public static final int cudaDevAttrTextureAlignment
public static final int cudaDevAttrGpuOverlap
public static final int cudaDevAttrMultiProcessorCount
public static final int cudaDevAttrKernelExecTimeout
public static final int cudaDevAttrIntegrated
public static final int cudaDevAttrCanMapHostMemory
public static final int cudaDevAttrComputeMode
public static final int cudaDevAttrMaxTexture1DWidth
public static final int cudaDevAttrMaxTexture2DWidth
public static final int cudaDevAttrMaxTexture2DHeight
public static final int cudaDevAttrMaxTexture3DWidth
public static final int cudaDevAttrMaxTexture3DHeight
public static final int cudaDevAttrMaxTexture3DDepth
public static final int cudaDevAttrMaxTexture2DLayeredWidth
public static final int cudaDevAttrMaxTexture2DLayeredHeight
public static final int cudaDevAttrMaxTexture2DLayeredLayers
public static final int cudaDevAttrSurfaceAlignment
public static final int cudaDevAttrConcurrentKernels
public static final int cudaDevAttrEccEnabled
public static final int cudaDevAttrPciBusId
public static final int cudaDevAttrPciDeviceId
public static final int cudaDevAttrTccDriver
public static final int cudaDevAttrMemoryClockRate
public static final int cudaDevAttrGlobalMemoryBusWidth
public static final int cudaDevAttrL2CacheSize
public static final int cudaDevAttrMaxThreadsPerMultiProcessor
public static final int cudaDevAttrAsyncEngineCount
public static final int cudaDevAttrUnifiedAddressing
public static final int cudaDevAttrMaxTexture1DLayeredWidth
public static final int cudaDevAttrMaxTexture1DLayeredLayers
public static final int cudaDevAttrMaxTexture2DGatherWidth
public static final int cudaDevAttrMaxTexture2DGatherHeight
public static final int cudaDevAttrMaxTexture3DWidthAlt
public static final int cudaDevAttrMaxTexture3DHeightAlt
public static final int cudaDevAttrMaxTexture3DDepthAlt
public static final int cudaDevAttrPciDomainId
public static final int cudaDevAttrTexturePitchAlignment
public static final int cudaDevAttrMaxTextureCubemapWidth
public static final int cudaDevAttrMaxTextureCubemapLayeredWidth
public static final int cudaDevAttrMaxTextureCubemapLayeredLayers
public static final int cudaDevAttrMaxSurface1DWidth
public static final int cudaDevAttrMaxSurface2DWidth
public static final int cudaDevAttrMaxSurface2DHeight
public static final int cudaDevAttrMaxSurface3DWidth
public static final int cudaDevAttrMaxSurface3DHeight
public static final int cudaDevAttrMaxSurface3DDepth
public static final int cudaDevAttrMaxSurface1DLayeredWidth
public static final int cudaDevAttrMaxSurface1DLayeredLayers
public static final int cudaDevAttrMaxSurface2DLayeredWidth
public static final int cudaDevAttrMaxSurface2DLayeredHeight
public static final int cudaDevAttrMaxSurface2DLayeredLayers
public static final int cudaDevAttrMaxSurfaceCubemapWidth
public static final int cudaDevAttrMaxSurfaceCubemapLayeredWidth
public static final int cudaDevAttrMaxSurfaceCubemapLayeredLayers
public static final int cudaDevAttrMaxTexture1DLinearWidth
public static final int cudaDevAttrMaxTexture2DLinearWidth
public static final int cudaDevAttrMaxTexture2DLinearHeight
public static final int cudaDevAttrMaxTexture2DLinearPitch
public static final int cudaDevAttrMaxTexture2DMipmappedWidth
public static final int cudaDevAttrMaxTexture2DMipmappedHeight
public static final int cudaDevAttrComputeCapabilityMajor
public static final int cudaDevAttrComputeCapabilityMinor
public static final int cudaDevAttrMaxTexture1DMipmappedWidth
public static final int cudaDevAttrStreamPrioritiesSupported
public static final int cudaDevAttrGlobalL1CacheSupported
public static final int cudaDevAttrLocalL1CacheSupported
public static final int cudaDevAttrMaxSharedMemoryPerMultiprocessor
public static final int cudaDevAttrMaxRegistersPerMultiprocessor
public static final int cudaDevAttrManagedMemory
public static final int cudaDevAttrIsMultiGpuBoard
public static final int cudaDevAttrMultiGpuBoardGroupID
public static final int cudaDevAttrHostNativeAtomicSupported
public static final int cudaDevAttrSingleToDoublePrecisionPerfRatio
public static final int cudaDevAttrPageableMemoryAccess
public static final int cudaDevAttrConcurrentManagedAccess
public static final int cudaDevAttrComputePreemptionSupported
public static final int cudaDevAttrCanUseHostPointerForRegisteredMem
public static final int cudaDevAttrReserved92
public static final int cudaDevAttrReserved93
public static final int cudaDevAttrReserved94
public static final int cudaDevAttrCooperativeLaunch
public static final int cudaDevAttrCooperativeMultiDeviceLaunch
public static final int cudaDevAttrMaxSharedMemoryPerBlockOptin
public static final int cudaDevAttrCanFlushRemoteWrites
public static final int cudaDevAttrHostRegisterSupported
public static final int cudaDevAttrPageableMemoryAccessUsesHostPageTables
public static final int cudaDevAttrDirectManagedMemAccessFromHost
public static final int cudaDevAttrMaxBlocksPerMultiprocessor
public static final int cudaDevAttrMaxPersistingL2CacheSize
public static final int cudaDevAttrMaxAccessPolicyWindowSize
public static final int cudaDevAttrReservedSharedMemoryPerBlock
public static final int cudaDevAttrSparseCudaArraySupported
public static final int cudaDevAttrHostRegisterReadOnlySupported
public static final int cudaDevAttrTimelineSemaphoreInteropSupported
public static final int cudaDevAttrMaxTimelineSemaphoreInteropSupported
public static final int cudaDevAttrMemoryPoolsSupported
public static final int cudaDevAttrGPUDirectRDMASupported
public static final int cudaDevAttrGPUDirectRDMAFlushWritesOptions
public static final int cudaDevAttrGPUDirectRDMAWritesOrdering
public static final int cudaDevAttrMemoryPoolSupportedHandleTypes
public static final int cudaDevAttrClusterLaunch
public static final int cudaDevAttrDeferredMappingCudaArraySupported
public static final int cudaDevAttrReserved122
public static final int cudaDevAttrReserved123
public static final int cudaDevAttrReserved124
public static final int cudaDevAttrIpcEventSupport
public static final int cudaDevAttrMemSyncDomainCount
public static final int cudaDevAttrReserved127
public static final int cudaDevAttrReserved128
public static final int cudaDevAttrReserved129
public static final int cudaDevAttrNumaConfig
public static final int cudaDevAttrNumaId
public static final int cudaDevAttrReserved132
public static final int cudaDevAttrMpsEnabled
public static final int cudaDevAttrHostNumaId
public static final int cudaDevAttrMax
public static final int cudaMemPoolReuseFollowEventDependencies
public static final int cudaMemPoolReuseAllowOpportunistic
public static final int cudaMemPoolReuseAllowInternalDependencies
public static final int cudaMemPoolAttrReleaseThreshold
public static final int cudaMemPoolAttrReservedMemCurrent
public static final int cudaMemPoolAttrReservedMemHigh
public static final int cudaMemPoolAttrUsedMemCurrent
public static final int cudaMemPoolAttrUsedMemHigh
public static final int cudaMemLocationTypeInvalid
public static final int cudaMemLocationTypeDevice
public static final int cudaMemLocationTypeHost
public static final int cudaMemLocationTypeHostNuma
public static final int cudaMemLocationTypeHostNumaCurrent
public static final int cudaMemAccessFlagsProtNone
public static final int cudaMemAccessFlagsProtRead
public static final int cudaMemAccessFlagsProtReadWrite
public static final int cudaMemAllocationTypeInvalid
public static final int cudaMemAllocationTypePinned
public static final int cudaMemAllocationTypeMax
public static final int cudaMemHandleTypeNone
public static final int cudaMemHandleTypePosixFileDescriptor
public static final int cudaMemHandleTypeWin32
public static final int cudaMemHandleTypeWin32Kmt
public static final int cudaGraphMemAttrUsedMemCurrent
public static final int cudaGraphMemAttrUsedMemHigh
public static final int cudaGraphMemAttrReservedMemCurrent
public static final int cudaGraphMemAttrReservedMemHigh
public static final int cudaDevP2PAttrPerformanceRank
public static final int cudaDevP2PAttrAccessSupported
public static final int cudaDevP2PAttrNativeAtomicSupported
public static final int cudaDevP2PAttrCudaArrayAccessSupported
public static final int CUDA_IPC_HANDLE_SIZE
public static final int cudaExternalMemoryHandleTypeOpaqueFd
public static final int cudaExternalMemoryHandleTypeOpaqueWin32
public static final int cudaExternalMemoryHandleTypeOpaqueWin32Kmt
public static final int cudaExternalMemoryHandleTypeD3D12Heap
public static final int cudaExternalMemoryHandleTypeD3D12Resource
public static final int cudaExternalMemoryHandleTypeD3D11Resource
public static final int cudaExternalMemoryHandleTypeD3D11ResourceKmt
public static final int cudaExternalMemoryHandleTypeNvSciBuf
public static final int cudaExternalMemoryDedicated
public static final int cudaExternalSemaphoreSignalSkipNvSciBufMemSync
public static final int cudaExternalSemaphoreWaitSkipNvSciBufMemSync
public static final int cudaNvSciSyncAttrSignal
public static final int cudaNvSciSyncAttrWait
public static final int cudaExternalSemaphoreHandleTypeOpaqueFd
public static final int cudaExternalSemaphoreHandleTypeOpaqueWin32
public static final int cudaExternalSemaphoreHandleTypeOpaqueWin32Kmt
public static final int cudaExternalSemaphoreHandleTypeD3D12Fence
public static final int cudaExternalSemaphoreHandleTypeD3D11Fence
public static final int cudaExternalSemaphoreHandleTypeNvSciSync
public static final int cudaExternalSemaphoreHandleTypeKeyedMutex
public static final int cudaExternalSemaphoreHandleTypeKeyedMutexKmt
public static final int cudaExternalSemaphoreHandleTypeTimelineSemaphoreFd
public static final int cudaExternalSemaphoreHandleTypeTimelineSemaphoreWin32
public static final int cudaCGScopeInvalid
public static final int cudaCGScopeGrid
public static final int cudaCGScopeMultiGrid
public static final int cudaGraphCondAssignDefault
public static final int cudaGraphCondTypeIf
public static final int cudaGraphCondTypeWhile
public static final int cudaGraphNodeTypeKernel
public static final int cudaGraphNodeTypeMemcpy
public static final int cudaGraphNodeTypeMemset
public static final int cudaGraphNodeTypeHost
public static final int cudaGraphNodeTypeGraph
public static final int cudaGraphNodeTypeEmpty
public static final int cudaGraphNodeTypeWaitEvent
public static final int cudaGraphNodeTypeEventRecord
public static final int cudaGraphNodeTypeExtSemaphoreSignal
public static final int cudaGraphNodeTypeExtSemaphoreWait
public static final int cudaGraphNodeTypeMemAlloc
public static final int cudaGraphNodeTypeMemFree
public static final int cudaGraphNodeTypeConditional
public static final int cudaGraphNodeTypeCount
public static final int cudaGraphDependencyTypeDefault
public static final int cudaGraphDependencyTypeProgrammatic
public static final int cudaGraphKernelNodePortDefault
public static final int cudaGraphKernelNodePortProgrammatic
public static final int cudaGraphKernelNodePortLaunchCompletion
public static final int cudaGraphExecUpdateSuccess
public static final int cudaGraphExecUpdateError
public static final int cudaGraphExecUpdateErrorTopologyChanged
public static final int cudaGraphExecUpdateErrorNodeTypeChanged
public static final int cudaGraphExecUpdateErrorFunctionChanged
public static final int cudaGraphExecUpdateErrorParametersChanged
public static final int cudaGraphExecUpdateErrorNotSupported
public static final int cudaGraphExecUpdateErrorUnsupportedFunctionChange
public static final int cudaGraphExecUpdateErrorAttributesChanged
public static final int cudaGraphInstantiateSuccess
public static final int cudaGraphInstantiateError
public static final int cudaGraphInstantiateInvalidStructure
public static final int cudaGraphInstantiateNodeOperationNotSupported
public static final int cudaGraphInstantiateMultipleDevicesNotSupported
public static final int cudaEnableDefault
public static final int cudaEnableLegacyStream
public static final int cudaEnablePerThreadDefaultStream
public static final int cudaDriverEntryPointSuccess
public static final int cudaDriverEntryPointSymbolNotFound
public static final int cudaDriverEntryPointVersionNotSufficent
public static final int cudaGraphDebugDotFlagsVerbose
public static final int cudaGraphDebugDotFlagsKernelNodeParams
public static final int cudaGraphDebugDotFlagsMemcpyNodeParams
public static final int cudaGraphDebugDotFlagsMemsetNodeParams
public static final int cudaGraphDebugDotFlagsHostNodeParams
public static final int cudaGraphDebugDotFlagsEventNodeParams
public static final int cudaGraphDebugDotFlagsExtSemasSignalNodeParams
public static final int cudaGraphDebugDotFlagsExtSemasWaitNodeParams
public static final int cudaGraphDebugDotFlagsKernelNodeAttributes
public static final int cudaGraphDebugDotFlagsHandles
public static final int cudaGraphDebugDotFlagsConditionalNodeParams
public static final int cudaGraphInstantiateFlagAutoFreeOnLaunch
public static final int cudaGraphInstantiateFlagUpload
public static final int cudaGraphInstantiateFlagDeviceLaunch
public static final int cudaGraphInstantiateFlagUseNodePriority
public static final int cudaLaunchMemSyncDomainDefault
public static final int cudaLaunchMemSyncDomainRemote
public static final int cudaLaunchAttributeIgnore
public static final int cudaLaunchAttributeAccessPolicyWindow
public static final int cudaLaunchAttributeCooperative
public static final int cudaLaunchAttributeSynchronizationPolicy
public static final int cudaLaunchAttributeClusterDimension
public static final int cudaLaunchAttributeClusterSchedulingPolicyPreference
public static final int cudaLaunchAttributeProgrammaticStreamSerialization
public static final int cudaLaunchAttributeProgrammaticEvent
public static final int cudaLaunchAttributePriority
public static final int cudaLaunchAttributeMemSyncDomainMap
public static final int cudaLaunchAttributeMemSyncDomain
public static final int cudaLaunchAttributeLaunchCompletionEvent
public static final int cudaStreamAttributeMemSyncDomainMap
public static final int cudaStreamAttributeMemSyncDomain
public static final int cudaStreamAttributePriority
public static final int cudaKernelNodeAttributeMemSyncDomainMap
public static final int cudaKernelNodeAttributeMemSyncDomain
public static final int cudaDeviceNumaConfigNone
public static final int cudaDeviceNumaConfigNumaNode
public static final int cudaSurfaceType1D
public static final int cudaSurfaceType2D
public static final int cudaSurfaceType3D
public static final int cudaSurfaceTypeCubemap
public static final int cudaSurfaceType1DLayered
public static final int cudaSurfaceType2DLayered
public static final int cudaSurfaceTypeCubemapLayered
public static final int cudaBoundaryModeZero
public static final int cudaBoundaryModeClamp
public static final int cudaBoundaryModeTrap
public static final int cudaFormatModeForced
public static final int cudaFormatModeAuto
public static final int cudaTextureType1D
public static final int cudaTextureType2D
public static final int cudaTextureType3D
public static final int cudaTextureTypeCubemap
public static final int cudaTextureType1DLayered
public static final int cudaTextureType2DLayered
public static final int cudaTextureTypeCubemapLayered
public static final int cudaAddressModeWrap
public static final int cudaAddressModeClamp
public static final int cudaAddressModeMirror
public static final int cudaAddressModeBorder
public static final int cudaFilterModePoint
public static final int cudaFilterModeLinear
public static final int cudaReadModeElementType
public static final int cudaReadModeNormalizedFloat
public static final int CUDART_VERSION
public static final int __CUDART_API_VERSION
public static final int CUDART_INF_FP16
public static final int CUDART_NAN_FP16
public static final int CUDART_MIN_DENORM_FP16
public static final int CUDART_MAX_NORMAL_FP16
public static final int CUDART_NEG_ZERO_FP16
public static final int CUDART_ZERO_FP16
public static final int CUDART_ONE_FP16
public static final int warpSize
public static final int CUDART_INF_BF16
public static final int CUDART_NAN_BF16
public static final int CUDART_MIN_DENORM_BF16
public static final int CUDART_MAX_NORMAL_BF16
public static final int CUDART_NEG_ZERO_BF16
public static final int CUDART_ZERO_BF16
public static final int CUDART_ONE_BF16
public static final int CUDA_R_16F
public static final int CUDA_C_16F
public static final int CUDA_R_16BF
public static final int CUDA_C_16BF
public static final int CUDA_R_32F
public static final int CUDA_C_32F
public static final int CUDA_R_64F
public static final int CUDA_C_64F
public static final int CUDA_R_4I
public static final int CUDA_C_4I
public static final int CUDA_R_4U
public static final int CUDA_C_4U
public static final int CUDA_R_8I
public static final int CUDA_C_8I
public static final int CUDA_R_8U
public static final int CUDA_C_8U
public static final int CUDA_R_16I
public static final int CUDA_C_16I
public static final int CUDA_R_16U
public static final int CUDA_C_16U
public static final int CUDA_R_32I
public static final int CUDA_C_32I
public static final int CUDA_R_32U
public static final int CUDA_C_32U
public static final int CUDA_R_64I
public static final int CUDA_C_64I
public static final int CUDA_R_64U
public static final int CUDA_C_64U
public static final int CUDA_R_8F_E4M3
public static final int CUDA_R_8F_E5M2
public static final int MAJOR_VERSION
public static final int MINOR_VERSION
public static final int PATCH_LEVEL
public static final int CU_GL_DEVICE_LIST_ALL
public static final int CU_GL_DEVICE_LIST_CURRENT_FRAME
public static final int CU_GL_DEVICE_LIST_NEXT_FRAME
public static final int CU_GL_MAP_RESOURCE_FLAGS_NONE
public static final int CU_GL_MAP_RESOURCE_FLAGS_READ_ONLY
public static final int CU_GL_MAP_RESOURCE_FLAGS_WRITE_DISCARD
public static final int cudaGLDeviceListAll
public static final int cudaGLDeviceListCurrentFrame
public static final int cudaGLDeviceListNextFrame
public static final int cudaGLMapFlagsNone
public static final int cudaGLMapFlagsReadOnly
public static final int cudaGLMapFlagsWriteDiscard
@MemberGetter public static CUstream_st CU_STREAM_LEGACY()
@MemberGetter public static CUstream_st CU_STREAM_PER_THREAD()
@MemberGetter public static Pointer CU_LAUNCH_PARAM_END()
@MemberGetter public static Pointer CU_LAUNCH_PARAM_BUFFER_POINTER()
@MemberGetter public static Pointer CU_LAUNCH_PARAM_BUFFER_SIZE()
@MemberGetter public static int CU_DEVICE_CPU()
@MemberGetter public static int CU_DEVICE_INVALID()
@Cast(value="CUresult") public static int cuGetErrorString(@Cast(value="CUresult") int error, @Cast(value="const char**") PointerPointer pStr)
error
- - Error code to convert to stringpStr
- - Address of the string pointer.::CUresult,
::cudaGetErrorString
@Cast(value="CUresult") public static int cuGetErrorString(@Cast(value="CUresult") int error, @Cast(value="const char**") @ByPtrPtr BytePointer pStr)
@Cast(value="CUresult") public static int cuGetErrorString(@Cast(value="CUresult") int error, @Cast(value="const char**") @ByPtrPtr ByteBuffer pStr)
@Cast(value="CUresult") public static int cuGetErrorString(@Cast(value="CUresult") int error, @Cast(value="const char**") @ByPtrPtr byte[] pStr)
@Cast(value="CUresult") public static int cuGetErrorName(@Cast(value="CUresult") int error, @Cast(value="const char**") PointerPointer pStr)
error
- - Error code to convert to stringpStr
- - Address of the string pointer.::CUresult,
::cudaGetErrorName
@Cast(value="CUresult") public static int cuGetErrorName(@Cast(value="CUresult") int error, @Cast(value="const char**") @ByPtrPtr BytePointer pStr)
@Cast(value="CUresult") public static int cuGetErrorName(@Cast(value="CUresult") int error, @Cast(value="const char**") @ByPtrPtr ByteBuffer pStr)
@Cast(value="CUresult") public static int cuGetErrorName(@Cast(value="CUresult") int error, @Cast(value="const char**") @ByPtrPtr byte[] pStr)
@Cast(value="CUresult") public static int cuInit(@Cast(value="unsigned int") int Flags)
Flags
- - Initialization flag for CUDA.@Cast(value="CUresult") public static int cuDriverGetVersion(IntPointer driverVersion)
driverVersion
- - Returns the CUDA driver version::cudaDriverGetVersion,
::cudaRuntimeGetVersion
@Cast(value="CUresult") public static int cuDriverGetVersion(IntBuffer driverVersion)
@Cast(value="CUresult") public static int cuDriverGetVersion(int[] driverVersion)
@Cast(value="CUresult") public static int cuDeviceGet(@Cast(value="CUdevice*") IntPointer device, int ordinal)
device
- - Returned device handleordinal
- - Device number to get handle for::cuDeviceGetAttribute,
::cuDeviceGetCount,
::cuDeviceGetName,
::cuDeviceGetUuid,
::cuDeviceGetLuid,
::cuDeviceTotalMem,
::cuDeviceGetExecAffinitySupport
@Cast(value="CUresult") public static int cuDeviceGet(@Cast(value="CUdevice*") IntBuffer device, int ordinal)
@Cast(value="CUresult") public static int cuDeviceGet(@Cast(value="CUdevice*") int[] device, int ordinal)
@Cast(value="CUresult") public static int cuDeviceGetCount(IntPointer count)
count
- - Returned number of compute-capable devices::cuDeviceGetAttribute,
::cuDeviceGetName,
::cuDeviceGetUuid,
::cuDeviceGetLuid,
::cuDeviceGet,
::cuDeviceTotalMem,
::cuDeviceGetExecAffinitySupport,
::cudaGetDeviceCount
@Cast(value="CUresult") public static int cuDeviceGetName(@Cast(value="char*") BytePointer name, int len, @Cast(value="CUdevice") int dev)
name
- - Returned identifier string for the devicelen
- - Maximum length of string to store in \p namedev
- - Device to get identifier string for::cuDeviceGetAttribute,
::cuDeviceGetUuid,
::cuDeviceGetLuid,
::cuDeviceGetCount,
::cuDeviceGet,
::cuDeviceTotalMem,
::cuDeviceGetExecAffinitySupport,
::cudaGetDeviceProperties
@Cast(value="CUresult") public static int cuDeviceGetName(@Cast(value="char*") ByteBuffer name, int len, @Cast(value="CUdevice") int dev)
@Cast(value="CUresult") public static int cuDeviceGetName(@Cast(value="char*") byte[] name, int len, @Cast(value="CUdevice") int dev)
@Cast(value="CUresult") public static int cuDeviceGetUuid(CUuuid uuid, @Cast(value="CUdevice") int dev)
uuid
- - Returned UUIDdev
- - Device to get identifier string for::cuDeviceGetAttribute,
::cuDeviceGetCount,
::cuDeviceGetName,
::cuDeviceGetLuid,
::cuDeviceGet,
::cuDeviceTotalMem,
::cuDeviceGetExecAffinitySupport,
::cudaGetDeviceProperties
@Cast(value="CUresult") public static int cuDeviceGetUuid_v2(CUuuid uuid, @Cast(value="CUdevice") int dev)
uuid
- - Returned UUIDdev
- - Device to get identifier string for::cuDeviceGetAttribute,
::cuDeviceGetCount,
::cuDeviceGetName,
::cuDeviceGetLuid,
::cuDeviceGet,
::cuDeviceTotalMem,
::cudaGetDeviceProperties
@Cast(value="CUresult") public static int cuDeviceTotalMem(@Cast(value="size_t*") SizeTPointer bytes, @Cast(value="CUdevice") int dev)
bytes
- - Returned memory available on device in bytesdev
- - Device handle::cuDeviceGetAttribute,
::cuDeviceGetCount,
::cuDeviceGetName,
::cuDeviceGetUuid,
::cuDeviceGet,
::cuDeviceGetExecAffinitySupport,
::cudaMemGetInfo
@Cast(value="CUresult") public static int cuDeviceGetTexture1DLinearMaxWidth(@Cast(value="size_t*") SizeTPointer maxWidthInElements, @Cast(value="CUarray_format") int format, @Cast(value="unsigned") int numChannels, @Cast(value="CUdevice") int dev)
maxWidthInElements
- - Returned maximum number of texture elements allocatable for given \p format and \p numChannels.format
- - Texture format.numChannels
- - Number of channels per texture element.dev
- - Device handle.::cuDeviceGetAttribute,
::cuDeviceGetCount,
::cuDeviceGetName,
::cuDeviceGetUuid,
::cuDeviceGet,
::cudaMemGetInfo,
::cuDeviceTotalMem
@Cast(value="CUresult") public static int cuDeviceGetAttribute(IntPointer pi, @Cast(value="CUdevice_attribute") int attrib, @Cast(value="CUdevice") int dev)
pi
- - Returned device attribute valueattrib
- - Device attribute to querydev
- - Device handle::cuDeviceGetCount,
::cuDeviceGetName,
::cuDeviceGetUuid,
::cuDeviceGet,
::cuDeviceTotalMem,
::cuDeviceGetExecAffinitySupport,
::cudaDeviceGetAttribute,
::cudaGetDeviceProperties
@Cast(value="CUresult") public static int cuDeviceGetAttribute(IntBuffer pi, @Cast(value="CUdevice_attribute") int attrib, @Cast(value="CUdevice") int dev)
@Cast(value="CUresult") public static int cuDeviceGetAttribute(int[] pi, @Cast(value="CUdevice_attribute") int attrib, @Cast(value="CUdevice") int dev)
@Cast(value="CUresult") public static int cuDeviceSetMemPool(@Cast(value="CUdevice") int dev, CUmemPoolHandle_st pool)
::cuDeviceGetDefaultMemPool, ::cuDeviceGetMemPool, ::cuMemPoolCreate, ::cuMemPoolDestroy, ::cuMemAllocFromPoolAsync
@Cast(value="CUresult") public static int cuDeviceGetMemPool(@ByPtrPtr CUmemPoolHandle_st pool, @Cast(value="CUdevice") int dev)
::cuDeviceGetDefaultMemPool, ::cuMemPoolCreate, ::cuDeviceSetMemPool
@Cast(value="CUresult") public static int cuDeviceGetDefaultMemPool(@ByPtrPtr CUmemPoolHandle_st pool_out, @Cast(value="CUdevice") int dev)
::cuMemAllocAsync, ::cuMemPoolTrimTo, ::cuMemPoolGetAttribute, ::cuMemPoolSetAttribute, cuMemPoolSetAccess, ::cuDeviceGetMemPool, ::cuMemPoolCreate
@Cast(value="CUresult") public static int cuDeviceGetExecAffinitySupport(IntPointer pi, @Cast(value="CUexecAffinityType") int type, @Cast(value="CUdevice") int dev)
pi
- - 1 if the execution affinity type \p type is supported by the device, or 0 if nottype
- - Execution affinity type to querydev
- - Device handle::cuDeviceGetAttribute,
::cuDeviceGetCount,
::cuDeviceGetName,
::cuDeviceGetUuid,
::cuDeviceGet,
::cuDeviceTotalMem
@Cast(value="CUresult") public static int cuDeviceGetExecAffinitySupport(IntBuffer pi, @Cast(value="CUexecAffinityType") int type, @Cast(value="CUdevice") int dev)
@Cast(value="CUresult") public static int cuDeviceGetExecAffinitySupport(int[] pi, @Cast(value="CUexecAffinityType") int type, @Cast(value="CUdevice") int dev)
@Cast(value="CUresult") public static int cuFlushGPUDirectRDMAWrites(@Cast(value="CUflushGPUDirectRDMAWritesTarget") int target, @Cast(value="CUflushGPUDirectRDMAWritesScope") int scope)
target
- - The target of the operation, see ::CUflushGPUDirectRDMAWritesTargetscope
- - The scope of the operation, see ::CUflushGPUDirectRDMAWritesScope@Cast(value="CUresult") @Deprecated public static int cuDeviceGetProperties(@Cast(value="CUdevprop*") CUdevprop_v1 prop, @Cast(value="CUdevice") int dev)
typedef struct CUdevprop_st {
int maxThreadsPerBlock;
int maxThreadsDim[3];
int maxGridSize[3];
int sharedMemPerBlock;
int totalConstantMemory;
int SIMDWidth;
int memPitch;
int regsPerBlock;
int clockRate;
int textureAlign
} CUdevprop;
where:
- ::maxThreadsPerBlock is the maximum number of threads per block;
- ::maxThreadsDim[3] is the maximum sizes of each dimension of a block;
- ::maxGridSize[3] is the maximum sizes of each dimension of a grid;
- ::sharedMemPerBlock is the total amount of shared memory available per
block in bytes;
- ::totalConstantMemory is the total amount of constant memory available on
the device in bytes;
- ::SIMDWidth is the warp size;
- ::memPitch is the maximum pitch allowed by the memory copy functions that
involve memory regions allocated through ::cuMemAllocPitch();
- ::regsPerBlock is the total number of registers available per block;
- ::clockRate is the clock frequency in kilohertz;
- ::textureAlign is the alignment requirement; texture base addresses that
are aligned to ::textureAlign bytes do not need an offset applied to
texture fetches.prop
- - Returned properties of devicedev
- - Device to get properties for::cuDeviceGetAttribute,
::cuDeviceGetCount,
::cuDeviceGetName,
::cuDeviceGetUuid,
::cuDeviceGet,
::cuDeviceTotalMem
@Cast(value="CUresult") @Deprecated public static int cuDeviceComputeCapability(IntPointer major, IntPointer minor, @Cast(value="CUdevice") int dev)
major
- - Major revision numberminor
- - Minor revision numberdev
- - Device handle::cuDeviceGetAttribute,
::cuDeviceGetCount,
::cuDeviceGetName,
::cuDeviceGetUuid,
::cuDeviceGet,
::cuDeviceTotalMem
@Cast(value="CUresult") @Deprecated public static int cuDeviceComputeCapability(IntBuffer major, IntBuffer minor, @Cast(value="CUdevice") int dev)
@Cast(value="CUresult") @Deprecated public static int cuDeviceComputeCapability(int[] major, int[] minor, @Cast(value="CUdevice") int dev)
@Cast(value="CUresult") public static int cuDevicePrimaryCtxRetain(@ByPtrPtr CUctx_st pctx, @Cast(value="CUdevice") int dev)
pctx
- - Returned context handle of the new contextdev
- - Device for which primary context is requested::cuDevicePrimaryCtxRelease,
::cuDevicePrimaryCtxSetFlags,
::cuCtxCreate,
::cuCtxGetApiVersion,
::cuCtxGetCacheConfig,
::cuCtxGetDevice,
::cuCtxGetFlags,
::cuCtxGetLimit,
::cuCtxPopCurrent,
::cuCtxPushCurrent,
::cuCtxSetCacheConfig,
::cuCtxSetLimit,
::cuCtxSynchronize
@Cast(value="CUresult") public static int cuDevicePrimaryCtxRelease(@Cast(value="CUdevice") int dev)
dev
- - Device which primary context is released::cuDevicePrimaryCtxRetain,
::cuCtxDestroy,
::cuCtxGetApiVersion,
::cuCtxGetCacheConfig,
::cuCtxGetDevice,
::cuCtxGetFlags,
::cuCtxGetLimit,
::cuCtxPopCurrent,
::cuCtxPushCurrent,
::cuCtxSetCacheConfig,
::cuCtxSetLimit,
::cuCtxSynchronize
@Cast(value="CUresult") public static int cuDevicePrimaryCtxSetFlags(@Cast(value="CUdevice") int dev, @Cast(value="unsigned int") int flags)
dev
- - Device for which the primary context flags are setflags
- - New flags for the device::cuDevicePrimaryCtxRetain,
::cuDevicePrimaryCtxGetState,
::cuCtxCreate,
::cuCtxGetFlags,
::cuCtxSetFlags,
::cudaSetDeviceFlags
@Cast(value="CUresult") public static int cuDevicePrimaryCtxGetState(@Cast(value="CUdevice") int dev, @Cast(value="unsigned int*") IntPointer flags, IntPointer active)
dev
- - Device to get primary context flags forflags
- - Pointer to store flagsactive
- - Pointer to store context state; 0 = inactive, 1 = active::cuDevicePrimaryCtxSetFlags,
::cuCtxGetFlags,
::cuCtxSetFlags,
::cudaGetDeviceFlags
@Cast(value="CUresult") public static int cuDevicePrimaryCtxGetState(@Cast(value="CUdevice") int dev, @Cast(value="unsigned int*") IntBuffer flags, IntBuffer active)
@Cast(value="CUresult") public static int cuDevicePrimaryCtxGetState(@Cast(value="CUdevice") int dev, @Cast(value="unsigned int*") int[] flags, int[] active)
@Cast(value="CUresult") public static int cuDevicePrimaryCtxReset(@Cast(value="CUdevice") int dev)
dev
- - Device for which primary context is destroyed::cuDevicePrimaryCtxRetain,
::cuDevicePrimaryCtxRelease,
::cuCtxGetApiVersion,
::cuCtxGetCacheConfig,
::cuCtxGetDevice,
::cuCtxGetFlags,
::cuCtxGetLimit,
::cuCtxPopCurrent,
::cuCtxPushCurrent,
::cuCtxSetCacheConfig,
::cuCtxSetLimit,
::cuCtxSynchronize,
::cudaDeviceReset
@Cast(value="CUresult") public static int cuCtxCreate(@ByPtrPtr CUctx_st pctx, @Cast(value="unsigned int") int flags, @Cast(value="CUdevice") int dev)
pctx
- - Returned context handle of the new contextflags
- - Context creation flagsdev
- - Device to create context on::cuCtxDestroy,
::cuCtxGetApiVersion,
::cuCtxGetCacheConfig,
::cuCtxGetDevice,
::cuCtxGetFlags,
::cuCtxGetLimit,
::cuCtxPopCurrent,
::cuCtxPushCurrent,
::cuCtxSetCacheConfig,
::cuCtxSetLimit,
::cuCoredumpSetAttributeGlobal,
::cuCoredumpSetAttribute,
::cuCtxSynchronize
@Cast(value="CUresult") public static int cuCtxCreate_v3(@ByPtrPtr CUctx_st pctx, @Cast(value="CUexecAffinityParam*") CUexecAffinityParam_v1 paramsArray, int numParams, @Cast(value="unsigned int") int flags, @Cast(value="CUdevice") int dev)
pctx
- - Returned context handle of the new contextparamsArray
- - Execution affinity parametersnumParams
- - Number of execution affinity parametersflags
- - Context creation flagsdev
- - Device to create context on::cuCtxDestroy,
::cuCtxGetApiVersion,
::cuCtxGetCacheConfig,
::cuCtxGetDevice,
::cuCtxGetFlags,
::cuCtxGetLimit,
::cuCtxPopCurrent,
::cuCtxPushCurrent,
::cuCtxSetCacheConfig,
::cuCtxSetLimit,
::cuCtxSynchronize,
::cuCoredumpSetAttributeGlobal,
::cuCoredumpSetAttribute,
::CUexecAffinityParam
@Cast(value="CUresult") public static int cuCtxDestroy(CUctx_st ctx)
ctx
- - Context to destroy::cuCtxCreate,
::cuCtxGetApiVersion,
::cuCtxGetCacheConfig,
::cuCtxGetDevice,
::cuCtxGetFlags,
::cuCtxGetLimit,
::cuCtxPopCurrent,
::cuCtxPushCurrent,
::cuCtxSetCacheConfig,
::cuCtxSetLimit,
::cuCtxSynchronize
@Cast(value="CUresult") public static int cuCtxPushCurrent(CUctx_st ctx)
ctx
- - Context to push::cuCtxCreate,
::cuCtxDestroy,
::cuCtxGetApiVersion,
::cuCtxGetCacheConfig,
::cuCtxGetDevice,
::cuCtxGetFlags,
::cuCtxGetLimit,
::cuCtxPopCurrent,
::cuCtxSetCacheConfig,
::cuCtxSetLimit,
::cuCtxSynchronize
@Cast(value="CUresult") public static int cuCtxPopCurrent(@ByPtrPtr CUctx_st pctx)
pctx
- - Returned popped context handle::cuCtxCreate,
::cuCtxDestroy,
::cuCtxGetApiVersion,
::cuCtxGetCacheConfig,
::cuCtxGetDevice,
::cuCtxGetFlags,
::cuCtxGetLimit,
::cuCtxPushCurrent,
::cuCtxSetCacheConfig,
::cuCtxSetLimit,
::cuCtxSynchronize
@Cast(value="CUresult") public static int cuCtxSetCurrent(CUctx_st ctx)
ctx
- - Context to bind to the calling CPU thread::cuCtxGetCurrent,
::cuCtxCreate,
::cuCtxDestroy,
::cudaSetDevice
@Cast(value="CUresult") public static int cuCtxGetCurrent(@ByPtrPtr CUctx_st pctx)
pctx
- - Returned context handle::cuCtxSetCurrent,
::cuCtxCreate,
::cuCtxDestroy,
::cudaGetDevice
@Cast(value="CUresult") public static int cuCtxGetDevice(@Cast(value="CUdevice*") IntPointer device)
device
- - Returned device ID for the current context::cuCtxCreate,
::cuCtxDestroy,
::cuCtxGetApiVersion,
::cuCtxGetCacheConfig,
::cuCtxGetFlags,
::cuCtxGetLimit,
::cuCtxPopCurrent,
::cuCtxPushCurrent,
::cuCtxSetCacheConfig,
::cuCtxSetLimit,
::cuCtxSynchronize,
::cudaGetDevice
@Cast(value="CUresult") public static int cuCtxGetDevice(@Cast(value="CUdevice*") IntBuffer device)
@Cast(value="CUresult") public static int cuCtxGetDevice(@Cast(value="CUdevice*") int[] device)
@Cast(value="CUresult") public static int cuCtxGetFlags(@Cast(value="unsigned int*") IntPointer flags)
flags
- - Pointer to store flags of current context::cuCtxCreate,
::cuCtxGetApiVersion,
::cuCtxGetCacheConfig,
::cuCtxGetCurrent,
::cuCtxGetDevice,
::cuCtxGetLimit,
::cuCtxGetSharedMemConfig,
::cuCtxGetStreamPriorityRange,
::cuCtxSetFlags,
::cudaGetDeviceFlags
@Cast(value="CUresult") public static int cuCtxGetFlags(@Cast(value="unsigned int*") IntBuffer flags)
@Cast(value="CUresult") public static int cuCtxGetFlags(@Cast(value="unsigned int*") int[] flags)
@Cast(value="CUresult") public static int cuCtxSetFlags(@Cast(value="unsigned int") int flags)
flags
- - Flags to set on the current context::cuCtxCreate,
::cuCtxGetApiVersion,
::cuCtxGetCacheConfig,
::cuCtxGetCurrent,
::cuCtxGetDevice,
::cuCtxGetLimit,
::cuCtxGetSharedMemConfig,
::cuCtxGetStreamPriorityRange,
::cuCtxGetFlags,
::cudaGetDeviceFlags,
::cuDevicePrimaryCtxSetFlags,
@Cast(value="CUresult") public static int cuCtxGetId(CUctx_st ctx, @Cast(value="unsigned long long*") LongPointer ctxId)
ctx
- - Context for which to obtain the IdctxId
- - Pointer to store the Id of the context::cuCtxCreate,
::cuCtxDestroy,
::cuCtxGetApiVersion,
::cuCtxGetCacheConfig,
::cuCtxGetDevice,
::cuCtxGetFlags,
::cuCtxGetLimit,
::cuCtxPushCurrent
@Cast(value="CUresult") public static int cuCtxGetId(CUctx_st ctx, @Cast(value="unsigned long long*") LongBuffer ctxId)
@Cast(value="CUresult") public static int cuCtxGetId(CUctx_st ctx, @Cast(value="unsigned long long*") long[] ctxId)
@Cast(value="CUresult") public static int cuCtxSynchronize()
::cuCtxCreate,
::cuCtxDestroy,
::cuCtxGetApiVersion,
::cuCtxGetCacheConfig,
::cuCtxGetDevice,
::cuCtxGetFlags,
::cuCtxGetLimit,
::cuCtxPopCurrent,
::cuCtxPushCurrent,
::cuCtxSetCacheConfig,
::cuCtxSetLimit,
::cudaDeviceSynchronize
@Cast(value="CUresult") public static int cuCtxSetLimit(@Cast(value="CUlimit") int _limit, @Cast(value="size_t") long value)
limit
- - Limit to setvalue
- - Size of limit::cuCtxCreate,
::cuCtxDestroy,
::cuCtxGetApiVersion,
::cuCtxGetCacheConfig,
::cuCtxGetDevice,
::cuCtxGetFlags,
::cuCtxGetLimit,
::cuCtxPopCurrent,
::cuCtxPushCurrent,
::cuCtxSetCacheConfig,
::cuCtxSynchronize,
::cudaDeviceSetLimit
@Cast(value="CUresult") public static int cuCtxGetLimit(@Cast(value="size_t*") SizeTPointer pvalue, @Cast(value="CUlimit") int _limit)
limit
- - Limit to querypvalue
- - Returned size of limit::cuCtxCreate,
::cuCtxDestroy,
::cuCtxGetApiVersion,
::cuCtxGetCacheConfig,
::cuCtxGetDevice,
::cuCtxGetFlags,
::cuCtxPopCurrent,
::cuCtxPushCurrent,
::cuCtxSetCacheConfig,
::cuCtxSetLimit,
::cuCtxSynchronize,
::cudaDeviceGetLimit
@Cast(value="CUresult") public static int cuCtxGetCacheConfig(@Cast(value="CUfunc_cache*") IntPointer pconfig)
pconfig
- - Returned cache configuration::cuCtxCreate,
::cuCtxDestroy,
::cuCtxGetApiVersion,
::cuCtxGetDevice,
::cuCtxGetFlags,
::cuCtxGetLimit,
::cuCtxPopCurrent,
::cuCtxPushCurrent,
::cuCtxSetCacheConfig,
::cuCtxSetLimit,
::cuCtxSynchronize,
::cuFuncSetCacheConfig,
::cudaDeviceGetCacheConfig
@Cast(value="CUresult") public static int cuCtxGetCacheConfig(@Cast(value="CUfunc_cache*") IntBuffer pconfig)
@Cast(value="CUresult") public static int cuCtxGetCacheConfig(@Cast(value="CUfunc_cache*") int[] pconfig)
@Cast(value="CUresult") public static int cuCtxSetCacheConfig(@Cast(value="CUfunc_cache") int config)
config
- - Requested cache configuration::cuCtxCreate,
::cuCtxDestroy,
::cuCtxGetApiVersion,
::cuCtxGetCacheConfig,
::cuCtxGetDevice,
::cuCtxGetFlags,
::cuCtxGetLimit,
::cuCtxPopCurrent,
::cuCtxPushCurrent,
::cuCtxSetLimit,
::cuCtxSynchronize,
::cuFuncSetCacheConfig,
::cudaDeviceSetCacheConfig,
::cuKernelSetCacheConfig
@Cast(value="CUresult") public static int cuCtxGetSharedMemConfig(@Cast(value="CUsharedconfig*") IntPointer pConfig)
pConfig
- - returned shared memory configuration::cuCtxCreate,
::cuCtxDestroy,
::cuCtxGetApiVersion,
::cuCtxGetCacheConfig,
::cuCtxGetDevice,
::cuCtxGetFlags,
::cuCtxGetLimit,
::cuCtxPopCurrent,
::cuCtxPushCurrent,
::cuCtxSetLimit,
::cuCtxSynchronize,
::cuCtxGetSharedMemConfig,
::cuFuncSetCacheConfig,
::cudaDeviceGetSharedMemConfig
@Cast(value="CUresult") public static int cuCtxGetSharedMemConfig(@Cast(value="CUsharedconfig*") IntBuffer pConfig)
@Cast(value="CUresult") public static int cuCtxGetSharedMemConfig(@Cast(value="CUsharedconfig*") int[] pConfig)
@Cast(value="CUresult") public static int cuCtxSetSharedMemConfig(@Cast(value="CUsharedconfig") int config)
config
- - requested shared memory configuration::cuCtxCreate,
::cuCtxDestroy,
::cuCtxGetApiVersion,
::cuCtxGetCacheConfig,
::cuCtxGetDevice,
::cuCtxGetFlags,
::cuCtxGetLimit,
::cuCtxPopCurrent,
::cuCtxPushCurrent,
::cuCtxSetLimit,
::cuCtxSynchronize,
::cuCtxGetSharedMemConfig,
::cuFuncSetCacheConfig,
::cudaDeviceSetSharedMemConfig
@Cast(value="CUresult") public static int cuCtxGetApiVersion(CUctx_st ctx, @Cast(value="unsigned int*") IntPointer version)
ctx
- - Context to checkversion
- - Pointer to version::cuCtxCreate,
::cuCtxDestroy,
::cuCtxGetDevice,
::cuCtxGetFlags,
::cuCtxGetLimit,
::cuCtxPopCurrent,
::cuCtxPushCurrent,
::cuCtxSetCacheConfig,
::cuCtxSetLimit,
::cuCtxSynchronize
@Cast(value="CUresult") public static int cuCtxGetApiVersion(CUctx_st ctx, @Cast(value="unsigned int*") IntBuffer version)
@Cast(value="CUresult") public static int cuCtxGetApiVersion(CUctx_st ctx, @Cast(value="unsigned int*") int[] version)
@Cast(value="CUresult") public static int cuCtxGetStreamPriorityRange(IntPointer leastPriority, IntPointer greatestPriority)
leastPriority
- - Pointer to an int in which the numerical value for least
stream priority is returnedgreatestPriority
- - Pointer to an int in which the numerical value for greatest
stream priority is returned::cuStreamCreateWithPriority,
::cuStreamGetPriority,
::cuCtxGetDevice,
::cuCtxGetFlags,
::cuCtxSetLimit,
::cuCtxSynchronize,
::cudaDeviceGetStreamPriorityRange
@Cast(value="CUresult") public static int cuCtxGetStreamPriorityRange(IntBuffer leastPriority, IntBuffer greatestPriority)
@Cast(value="CUresult") public static int cuCtxGetStreamPriorityRange(int[] leastPriority, int[] greatestPriority)
@Cast(value="CUresult") public static int cuCtxResetPersistingL2Cache()
::CUaccessPolicyWindow
@Cast(value="CUresult") public static int cuCtxGetExecAffinity(@Cast(value="CUexecAffinityParam*") CUexecAffinityParam_v1 pExecAffinity, @Cast(value="CUexecAffinityType") int type)
type
- - Execution affinity type to querypExecAffinity
- - Returned execution affinity::CUexecAffinityParam
@Cast(value="CUresult") @Deprecated public static int cuCtxAttach(@ByPtrPtr CUctx_st pctx, @Cast(value="unsigned int") int flags)
pctx
- - Returned context handle of the current contextflags
- - Context attach flags (must be 0)::cuCtxCreate,
::cuCtxDestroy,
::cuCtxDetach,
::cuCtxGetApiVersion,
::cuCtxGetCacheConfig,
::cuCtxGetDevice,
::cuCtxGetFlags,
::cuCtxGetLimit,
::cuCtxPopCurrent,
::cuCtxPushCurrent,
::cuCtxSetCacheConfig,
::cuCtxSetLimit,
::cuCtxSynchronize
@Cast(value="CUresult") @Deprecated public static int cuCtxDetach(CUctx_st ctx)
ctx
- - Context to destroy::cuCtxCreate,
::cuCtxDestroy,
::cuCtxGetApiVersion,
::cuCtxGetCacheConfig,
::cuCtxGetDevice,
::cuCtxGetFlags,
::cuCtxGetLimit,
::cuCtxPopCurrent,
::cuCtxPushCurrent,
::cuCtxSetCacheConfig,
::cuCtxSetLimit,
::cuCtxSynchronize
@Cast(value="CUresult") public static int cuModuleLoad(@ByPtrPtr CUmod_st module, @Cast(value="const char*") BytePointer fname)
module
- - Returned modulefname
- - Filename of module to load::cuModuleGetFunction,
::cuModuleGetGlobal,
::cuModuleGetTexRef,
::cuModuleLoadData,
::cuModuleLoadDataEx,
::cuModuleLoadFatBinary,
::cuModuleUnload
@Cast(value="CUresult") public static int cuModuleLoad(@ByPtrPtr CUmod_st module, String fname)
@Cast(value="CUresult") public static int cuModuleLoadData(@ByPtrPtr CUmod_st module, @Const Pointer image)
module
- - Returned moduleimage
- - Module data to load::cuModuleGetFunction,
::cuModuleGetGlobal,
::cuModuleGetTexRef,
::cuModuleLoad,
::cuModuleLoadDataEx,
::cuModuleLoadFatBinary,
::cuModuleUnload
@Cast(value="CUresult") public static int cuModuleLoadDataEx(@ByPtrPtr CUmod_st module, @Const Pointer image, @Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") IntPointer options, @Cast(value="void**") PointerPointer optionValues)
module
- - Returned moduleimage
- - Module data to loadnumOptions
- - Number of optionsoptions
- - Options for JIToptionValues
- - Option values for JIT::cuModuleGetFunction,
::cuModuleGetGlobal,
::cuModuleGetTexRef,
::cuModuleLoad,
::cuModuleLoadData,
::cuModuleLoadFatBinary,
::cuModuleUnload
@Cast(value="CUresult") public static int cuModuleLoadDataEx(@ByPtrPtr CUmod_st module, @Const Pointer image, @Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") IntPointer options, @Cast(value="void**") @ByPtrPtr Pointer optionValues)
@Cast(value="CUresult") public static int cuModuleLoadDataEx(@ByPtrPtr CUmod_st module, @Const Pointer image, @Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") IntBuffer options, @Cast(value="void**") @ByPtrPtr Pointer optionValues)
@Cast(value="CUresult") public static int cuModuleLoadDataEx(@ByPtrPtr CUmod_st module, @Const Pointer image, @Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") int[] options, @Cast(value="void**") @ByPtrPtr Pointer optionValues)
@Cast(value="CUresult") public static int cuModuleLoadFatBinary(@ByPtrPtr CUmod_st module, @Const Pointer fatCubin)
module
- - Returned modulefatCubin
- - Fat binary to load::cuModuleGetFunction,
::cuModuleGetGlobal,
::cuModuleGetTexRef,
::cuModuleLoad,
::cuModuleLoadData,
::cuModuleLoadDataEx,
::cuModuleUnload
@Cast(value="CUresult") public static int cuModuleUnload(CUmod_st hmod)
hmod
- - Module to unload::cuModuleGetFunction,
::cuModuleGetGlobal,
::cuModuleGetTexRef,
::cuModuleLoad,
::cuModuleLoadData,
::cuModuleLoadDataEx,
::cuModuleLoadFatBinary
@Cast(value="CUresult") public static int cuModuleGetLoadingMode(@Cast(value="CUmoduleLoadingMode*") IntPointer mode)
mode
- - Returns the lazy loading mode::cuModuleLoad,
@Cast(value="CUresult") public static int cuModuleGetLoadingMode(@Cast(value="CUmoduleLoadingMode*") IntBuffer mode)
@Cast(value="CUresult") public static int cuModuleGetLoadingMode(@Cast(value="CUmoduleLoadingMode*") int[] mode)
@Cast(value="CUresult") public static int cuModuleGetFunction(@ByPtrPtr CUfunc_st hfunc, CUmod_st hmod, @Cast(value="const char*") BytePointer name)
hfunc
- - Returned function handlehmod
- - Module to retrieve function fromname
- - Name of function to retrieve::cuModuleGetGlobal,
::cuModuleGetTexRef,
::cuModuleLoad,
::cuModuleLoadData,
::cuModuleLoadDataEx,
::cuModuleLoadFatBinary,
::cuModuleUnload
@Cast(value="CUresult") public static int cuModuleGetFunction(@ByPtrPtr CUfunc_st hfunc, CUmod_st hmod, String name)
@Cast(value="CUresult") public static int cuModuleGetGlobal(@Cast(value="CUdeviceptr*") LongPointer dptr, @Cast(value="size_t*") SizeTPointer bytes, CUmod_st hmod, @Cast(value="const char*") BytePointer name)
dptr
- - Returned global device pointerbytes
- - Returned global size in byteshmod
- - Module to retrieve global fromname
- - Name of global to retrieve::cuModuleGetFunction,
::cuModuleGetTexRef,
::cuModuleLoad,
::cuModuleLoadData,
::cuModuleLoadDataEx,
::cuModuleLoadFatBinary,
::cuModuleUnload,
::cudaGetSymbolAddress,
::cudaGetSymbolSize
@Cast(value="CUresult") public static int cuModuleGetGlobal(@Cast(value="CUdeviceptr*") LongBuffer dptr, @Cast(value="size_t*") SizeTPointer bytes, CUmod_st hmod, String name)
@Cast(value="CUresult") public static int cuModuleGetGlobal(@Cast(value="CUdeviceptr*") long[] dptr, @Cast(value="size_t*") SizeTPointer bytes, CUmod_st hmod, @Cast(value="const char*") BytePointer name)
@Cast(value="CUresult") public static int cuModuleGetGlobal(@Cast(value="CUdeviceptr*") LongPointer dptr, @Cast(value="size_t*") SizeTPointer bytes, CUmod_st hmod, String name)
@Cast(value="CUresult") public static int cuModuleGetGlobal(@Cast(value="CUdeviceptr*") LongBuffer dptr, @Cast(value="size_t*") SizeTPointer bytes, CUmod_st hmod, @Cast(value="const char*") BytePointer name)
@Cast(value="CUresult") public static int cuModuleGetGlobal(@Cast(value="CUdeviceptr*") long[] dptr, @Cast(value="size_t*") SizeTPointer bytes, CUmod_st hmod, String name)
@Cast(value="CUresult") public static int cuLinkCreate(@Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") IntPointer options, @Cast(value="void**") PointerPointer optionValues, @ByPtrPtr CUlinkState_st stateOut)
numOptions
- Size of options arraysoptions
- Array of linker and compiler optionsoptionValues
- Array of option values, each cast to void *stateOut
- On success, this will contain a CUlinkState to specify
and complete this action::cuLinkAddData,
::cuLinkAddFile,
::cuLinkComplete,
::cuLinkDestroy
@Cast(value="CUresult") public static int cuLinkCreate(@Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") IntPointer options, @Cast(value="void**") @ByPtrPtr Pointer optionValues, @ByPtrPtr CUlinkState_st stateOut)
@Cast(value="CUresult") public static int cuLinkCreate(@Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") IntBuffer options, @Cast(value="void**") @ByPtrPtr Pointer optionValues, @ByPtrPtr CUlinkState_st stateOut)
@Cast(value="CUresult") public static int cuLinkCreate(@Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") int[] options, @Cast(value="void**") @ByPtrPtr Pointer optionValues, @ByPtrPtr CUlinkState_st stateOut)
@Cast(value="CUresult") public static int cuLinkAddData(CUlinkState_st state, @Cast(value="CUjitInputType") int type, Pointer data, @Cast(value="size_t") long size, @Cast(value="const char*") BytePointer name, @Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") IntPointer options, @Cast(value="void**") PointerPointer optionValues)
state
- A pending linker action.type
- The type of the input data.data
- The input data. PTX must be NULL-terminated.size
- The length of the input data.name
- An optional name for this input in log messages.numOptions
- Size of options.options
- Options to be applied only for this input (overrides options from ::cuLinkCreate).optionValues
- Array of option values, each cast to void *.::cuLinkCreate,
::cuLinkAddFile,
::cuLinkComplete,
::cuLinkDestroy
@Cast(value="CUresult") public static int cuLinkAddData(CUlinkState_st state, @Cast(value="CUjitInputType") int type, Pointer data, @Cast(value="size_t") long size, @Cast(value="const char*") BytePointer name, @Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") IntPointer options, @Cast(value="void**") @ByPtrPtr Pointer optionValues)
@Cast(value="CUresult") public static int cuLinkAddData(CUlinkState_st state, @Cast(value="CUjitInputType") int type, Pointer data, @Cast(value="size_t") long size, String name, @Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") IntBuffer options, @Cast(value="void**") @ByPtrPtr Pointer optionValues)
@Cast(value="CUresult") public static int cuLinkAddData(CUlinkState_st state, @Cast(value="CUjitInputType") int type, Pointer data, @Cast(value="size_t") long size, @Cast(value="const char*") BytePointer name, @Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") int[] options, @Cast(value="void**") @ByPtrPtr Pointer optionValues)
@Cast(value="CUresult") public static int cuLinkAddData(CUlinkState_st state, @Cast(value="CUjitInputType") int type, Pointer data, @Cast(value="size_t") long size, String name, @Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") IntPointer options, @Cast(value="void**") @ByPtrPtr Pointer optionValues)
@Cast(value="CUresult") public static int cuLinkAddData(CUlinkState_st state, @Cast(value="CUjitInputType") int type, Pointer data, @Cast(value="size_t") long size, @Cast(value="const char*") BytePointer name, @Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") IntBuffer options, @Cast(value="void**") @ByPtrPtr Pointer optionValues)
@Cast(value="CUresult") public static int cuLinkAddData(CUlinkState_st state, @Cast(value="CUjitInputType") int type, Pointer data, @Cast(value="size_t") long size, String name, @Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") int[] options, @Cast(value="void**") @ByPtrPtr Pointer optionValues)
@Cast(value="CUresult") public static int cuLinkAddFile(CUlinkState_st state, @Cast(value="CUjitInputType") int type, @Cast(value="const char*") BytePointer path, @Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") IntPointer options, @Cast(value="void**") PointerPointer optionValues)
state
- A pending linker actiontype
- The type of the input datapath
- Path to the input filenumOptions
- Size of optionsoptions
- Options to be applied only for this input (overrides options from ::cuLinkCreate)optionValues
- Array of option values, each cast to void *::cuLinkCreate,
::cuLinkAddData,
::cuLinkComplete,
::cuLinkDestroy
@Cast(value="CUresult") public static int cuLinkAddFile(CUlinkState_st state, @Cast(value="CUjitInputType") int type, @Cast(value="const char*") BytePointer path, @Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") IntPointer options, @Cast(value="void**") @ByPtrPtr Pointer optionValues)
@Cast(value="CUresult") public static int cuLinkAddFile(CUlinkState_st state, @Cast(value="CUjitInputType") int type, String path, @Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") IntBuffer options, @Cast(value="void**") @ByPtrPtr Pointer optionValues)
@Cast(value="CUresult") public static int cuLinkAddFile(CUlinkState_st state, @Cast(value="CUjitInputType") int type, @Cast(value="const char*") BytePointer path, @Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") int[] options, @Cast(value="void**") @ByPtrPtr Pointer optionValues)
@Cast(value="CUresult") public static int cuLinkAddFile(CUlinkState_st state, @Cast(value="CUjitInputType") int type, String path, @Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") IntPointer options, @Cast(value="void**") @ByPtrPtr Pointer optionValues)
@Cast(value="CUresult") public static int cuLinkAddFile(CUlinkState_st state, @Cast(value="CUjitInputType") int type, @Cast(value="const char*") BytePointer path, @Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") IntBuffer options, @Cast(value="void**") @ByPtrPtr Pointer optionValues)
@Cast(value="CUresult") public static int cuLinkAddFile(CUlinkState_st state, @Cast(value="CUjitInputType") int type, String path, @Cast(value="unsigned int") int numOptions, @Cast(value="CUjit_option*") int[] options, @Cast(value="void**") @ByPtrPtr Pointer optionValues)
@Cast(value="CUresult") public static int cuLinkComplete(CUlinkState_st state, @Cast(value="void**") PointerPointer cubinOut, @Cast(value="size_t*") SizeTPointer sizeOut)
state
- A pending linker invocationcubinOut
- On success, this will point to the output imagesizeOut
- Optional parameter to receive the size of the generated image::cuLinkCreate,
::cuLinkAddData,
::cuLinkAddFile,
::cuLinkDestroy,
::cuModuleLoadData
@Cast(value="CUresult") public static int cuLinkComplete(CUlinkState_st state, @Cast(value="void**") @ByPtrPtr Pointer cubinOut, @Cast(value="size_t*") SizeTPointer sizeOut)
@Cast(value="CUresult") public static int cuLinkDestroy(CUlinkState_st state)
state
- State object for the linker invocation::cuLinkCreate
@Cast(value="CUresult") @Deprecated public static int cuModuleGetTexRef(@ByPtrPtr CUtexref_st pTexRef, CUmod_st hmod, @Cast(value="const char*") BytePointer name)
pTexRef
- - Returned texture referencehmod
- - Module to retrieve texture reference fromname
- - Name of texture reference to retrieve::cuModuleGetFunction,
::cuModuleGetGlobal,
::cuModuleGetSurfRef,
::cuModuleLoad,
::cuModuleLoadData,
::cuModuleLoadDataEx,
::cuModuleLoadFatBinary,
::cuModuleUnload
@Cast(value="CUresult") @Deprecated public static int cuModuleGetTexRef(@ByPtrPtr CUtexref_st pTexRef, CUmod_st hmod, String name)
@Cast(value="CUresult") @Deprecated public static int cuModuleGetSurfRef(@ByPtrPtr CUsurfref_st pSurfRef, CUmod_st hmod, @Cast(value="const char*") BytePointer name)
pSurfRef
- - Returned surface referencehmod
- - Module to retrieve surface reference fromname
- - Name of surface reference to retrieve::cuModuleGetFunction,
::cuModuleGetGlobal,
::cuModuleGetTexRef,
::cuModuleLoad,
::cuModuleLoadData,
::cuModuleLoadDataEx,
::cuModuleLoadFatBinary,
::cuModuleUnload
@Cast(value="CUresult") @Deprecated public static int cuModuleGetSurfRef(@ByPtrPtr CUsurfref_st pSurfRef, CUmod_st hmod, String name)
@Cast(value="CUresult") public static int cuLibraryLoadData(@ByPtrPtr CUlib_st library, @Const Pointer code, @Cast(value="CUjit_option*") IntPointer jitOptions, @Cast(value="void**") PointerPointer jitOptionsValues, @Cast(value="unsigned int") int numJitOptions, @Cast(value="CUlibraryOption*") IntPointer libraryOptions, @Cast(value="void**") PointerPointer libraryOptionValues, @Cast(value="unsigned int") int numLibraryOptions)
library
- - Returned librarycode
- - Code to loadjitOptions
- - Options for JITjitOptionsValues
- - Option values for JITnumJitOptions
- - Number of optionslibraryOptions
- - Options for loadinglibraryOptionValues
- - Option values for loadingnumLibraryOptions
- - Number of options for loading::cuLibraryLoadFromFile,
::cuLibraryUnload,
::cuModuleLoad,
::cuModuleLoadData,
::cuModuleLoadDataEx
@Cast(value="CUresult") public static int cuLibraryLoadData(@ByPtrPtr CUlib_st library, @Const Pointer code, @Cast(value="CUjit_option*") IntPointer jitOptions, @Cast(value="void**") @ByPtrPtr Pointer jitOptionsValues, @Cast(value="unsigned int") int numJitOptions, @Cast(value="CUlibraryOption*") IntPointer libraryOptions, @Cast(value="void**") @ByPtrPtr Pointer libraryOptionValues, @Cast(value="unsigned int") int numLibraryOptions)
@Cast(value="CUresult") public static int cuLibraryLoadData(@ByPtrPtr CUlib_st library, @Const Pointer code, @Cast(value="CUjit_option*") IntBuffer jitOptions, @Cast(value="void**") @ByPtrPtr Pointer jitOptionsValues, @Cast(value="unsigned int") int numJitOptions, @Cast(value="CUlibraryOption*") IntBuffer libraryOptions, @Cast(value="void**") @ByPtrPtr Pointer libraryOptionValues, @Cast(value="unsigned int") int numLibraryOptions)
@Cast(value="CUresult") public static int cuLibraryLoadData(@ByPtrPtr CUlib_st library, @Const Pointer code, @Cast(value="CUjit_option*") int[] jitOptions, @Cast(value="void**") @ByPtrPtr Pointer jitOptionsValues, @Cast(value="unsigned int") int numJitOptions, @Cast(value="CUlibraryOption*") int[] libraryOptions, @Cast(value="void**") @ByPtrPtr Pointer libraryOptionValues, @Cast(value="unsigned int") int numLibraryOptions)
@Cast(value="CUresult") public static int cuLibraryLoadFromFile(@ByPtrPtr CUlib_st library, @Cast(value="const char*") BytePointer fileName, @Cast(value="CUjit_option*") IntPointer jitOptions, @Cast(value="void**") PointerPointer jitOptionsValues, @Cast(value="unsigned int") int numJitOptions, @Cast(value="CUlibraryOption*") IntPointer libraryOptions, @Cast(value="void**") PointerPointer libraryOptionValues, @Cast(value="unsigned int") int numLibraryOptions)
library
- - Returned libraryfileName
- - File to load fromjitOptions
- - Options for JITjitOptionsValues
- - Option values for JITnumJitOptions
- - Number of optionslibraryOptions
- - Options for loadinglibraryOptionValues
- - Option values for loadingnumLibraryOptions
- - Number of options for loading::cuLibraryLoadData,
::cuLibraryUnload,
::cuModuleLoad,
::cuModuleLoadData,
::cuModuleLoadDataEx
@Cast(value="CUresult") public static int cuLibraryLoadFromFile(@ByPtrPtr CUlib_st library, @Cast(value="const char*") BytePointer fileName, @Cast(value="CUjit_option*") IntPointer jitOptions, @Cast(value="void**") @ByPtrPtr Pointer jitOptionsValues, @Cast(value="unsigned int") int numJitOptions, @Cast(value="CUlibraryOption*") IntPointer libraryOptions, @Cast(value="void**") @ByPtrPtr Pointer libraryOptionValues, @Cast(value="unsigned int") int numLibraryOptions)
@Cast(value="CUresult") public static int cuLibraryLoadFromFile(@ByPtrPtr CUlib_st library, String fileName, @Cast(value="CUjit_option*") IntBuffer jitOptions, @Cast(value="void**") @ByPtrPtr Pointer jitOptionsValues, @Cast(value="unsigned int") int numJitOptions, @Cast(value="CUlibraryOption*") IntBuffer libraryOptions, @Cast(value="void**") @ByPtrPtr Pointer libraryOptionValues, @Cast(value="unsigned int") int numLibraryOptions)
@Cast(value="CUresult") public static int cuLibraryLoadFromFile(@ByPtrPtr CUlib_st library, @Cast(value="const char*") BytePointer fileName, @Cast(value="CUjit_option*") int[] jitOptions, @Cast(value="void**") @ByPtrPtr Pointer jitOptionsValues, @Cast(value="unsigned int") int numJitOptions, @Cast(value="CUlibraryOption*") int[] libraryOptions, @Cast(value="void**") @ByPtrPtr Pointer libraryOptionValues, @Cast(value="unsigned int") int numLibraryOptions)
@Cast(value="CUresult") public static int cuLibraryLoadFromFile(@ByPtrPtr CUlib_st library, String fileName, @Cast(value="CUjit_option*") IntPointer jitOptions, @Cast(value="void**") @ByPtrPtr Pointer jitOptionsValues, @Cast(value="unsigned int") int numJitOptions, @Cast(value="CUlibraryOption*") IntPointer libraryOptions, @Cast(value="void**") @ByPtrPtr Pointer libraryOptionValues, @Cast(value="unsigned int") int numLibraryOptions)
@Cast(value="CUresult") public static int cuLibraryLoadFromFile(@ByPtrPtr CUlib_st library, @Cast(value="const char*") BytePointer fileName, @Cast(value="CUjit_option*") IntBuffer jitOptions, @Cast(value="void**") @ByPtrPtr Pointer jitOptionsValues, @Cast(value="unsigned int") int numJitOptions, @Cast(value="CUlibraryOption*") IntBuffer libraryOptions, @Cast(value="void**") @ByPtrPtr Pointer libraryOptionValues, @Cast(value="unsigned int") int numLibraryOptions)
@Cast(value="CUresult") public static int cuLibraryLoadFromFile(@ByPtrPtr CUlib_st library, String fileName, @Cast(value="CUjit_option*") int[] jitOptions, @Cast(value="void**") @ByPtrPtr Pointer jitOptionsValues, @Cast(value="unsigned int") int numJitOptions, @Cast(value="CUlibraryOption*") int[] libraryOptions, @Cast(value="void**") @ByPtrPtr Pointer libraryOptionValues, @Cast(value="unsigned int") int numLibraryOptions)
@Cast(value="CUresult") public static int cuLibraryUnload(CUlib_st library)
library
- - Library to unload::cuLibraryLoadData,
::cuLibraryLoadFromFile,
::cuModuleUnload
@Cast(value="CUresult") public static int cuLibraryGetKernel(@ByPtrPtr CUkern_st pKernel, CUlib_st library, @Cast(value="const char*") BytePointer name)
pKernel
- - Returned kernel handlelibrary
- - Library to retrieve kernel fromname
- - Name of kernel to retrieve::cuLibraryLoadData,
::cuLibraryLoadFromFile,
::cuLibraryUnload,
::cuKernelGetFunction,
::cuLibraryGetModule,
::cuModuleGetFunction
@Cast(value="CUresult") public static int cuLibraryGetKernel(@ByPtrPtr CUkern_st pKernel, CUlib_st library, String name)
@Cast(value="CUresult") public static int cuLibraryGetModule(@ByPtrPtr CUmod_st pMod, CUlib_st library)
pMod
- - Returned module handlelibrary
- - Library to retrieve module from::cuLibraryLoadData,
::cuLibraryLoadFromFile,
::cuLibraryUnload,
::cuModuleGetFunction
@Cast(value="CUresult") public static int cuKernelGetFunction(@ByPtrPtr CUfunc_st pFunc, CUkern_st kernel)
pFunc
- - Returned function handlekernel
- - Kernel to retrieve function for the requested context::cuLibraryLoadData,
::cuLibraryLoadFromFile,
::cuLibraryUnload,
::cuLibraryGetKernel,
::cuLibraryGetModule,
::cuModuleGetFunction
@Cast(value="CUresult") public static int cuLibraryGetGlobal(@Cast(value="CUdeviceptr*") LongPointer dptr, @Cast(value="size_t*") SizeTPointer bytes, CUlib_st library, @Cast(value="const char*") BytePointer name)
dptr
- - Returned global device pointer for the requested contextbytes
- - Returned global size in byteslibrary
- - Library to retrieve global fromname
- - Name of global to retrieve::cuLibraryLoadData,
::cuLibraryLoadFromFile,
::cuLibraryUnload,
::cuLibraryGetModule,
cuModuleGetGlobal
@Cast(value="CUresult") public static int cuLibraryGetGlobal(@Cast(value="CUdeviceptr*") LongBuffer dptr, @Cast(value="size_t*") SizeTPointer bytes, CUlib_st library, String name)
@Cast(value="CUresult") public static int cuLibraryGetGlobal(@Cast(value="CUdeviceptr*") long[] dptr, @Cast(value="size_t*") SizeTPointer bytes, CUlib_st library, @Cast(value="const char*") BytePointer name)
@Cast(value="CUresult") public static int cuLibraryGetGlobal(@Cast(value="CUdeviceptr*") LongPointer dptr, @Cast(value="size_t*") SizeTPointer bytes, CUlib_st library, String name)
@Cast(value="CUresult") public static int cuLibraryGetGlobal(@Cast(value="CUdeviceptr*") LongBuffer dptr, @Cast(value="size_t*") SizeTPointer bytes, CUlib_st library, @Cast(value="const char*") BytePointer name)
@Cast(value="CUresult") public static int cuLibraryGetGlobal(@Cast(value="CUdeviceptr*") long[] dptr, @Cast(value="size_t*") SizeTPointer bytes, CUlib_st library, String name)
@Cast(value="CUresult") public static int cuLibraryGetManaged(@Cast(value="CUdeviceptr*") LongPointer dptr, @Cast(value="size_t*") SizeTPointer bytes, CUlib_st library, @Cast(value="const char*") BytePointer name)
dptr
- - Returned pointer to the managed memorybytes
- - Returned memory size in byteslibrary
- - Library to retrieve managed memory fromname
- - Name of managed memory to retrieve::cuLibraryLoadData,
::cuLibraryLoadFromFile,
::cuLibraryUnload,
@Cast(value="CUresult") public static int cuLibraryGetManaged(@Cast(value="CUdeviceptr*") LongBuffer dptr, @Cast(value="size_t*") SizeTPointer bytes, CUlib_st library, String name)
@Cast(value="CUresult") public static int cuLibraryGetManaged(@Cast(value="CUdeviceptr*") long[] dptr, @Cast(value="size_t*") SizeTPointer bytes, CUlib_st library, @Cast(value="const char*") BytePointer name)
@Cast(value="CUresult") public static int cuLibraryGetManaged(@Cast(value="CUdeviceptr*") LongPointer dptr, @Cast(value="size_t*") SizeTPointer bytes, CUlib_st library, String name)
@Cast(value="CUresult") public static int cuLibraryGetManaged(@Cast(value="CUdeviceptr*") LongBuffer dptr, @Cast(value="size_t*") SizeTPointer bytes, CUlib_st library, @Cast(value="const char*") BytePointer name)
@Cast(value="CUresult") public static int cuLibraryGetManaged(@Cast(value="CUdeviceptr*") long[] dptr, @Cast(value="size_t*") SizeTPointer bytes, CUlib_st library, String name)
@Cast(value="CUresult") public static int cuLibraryGetUnifiedFunction(@Cast(value="void**") PointerPointer fptr, CUlib_st library, @Cast(value="const char*") BytePointer symbol)
fptr
- - Returned pointer to a unified functionlibrary
- - Library to retrieve function pointer memory fromsymbol
- - Name of function pointer to retrieve::cuLibraryLoadData,
::cuLibraryLoadFromFile,
::cuLibraryUnload,
@Cast(value="CUresult") public static int cuLibraryGetUnifiedFunction(@Cast(value="void**") @ByPtrPtr Pointer fptr, CUlib_st library, @Cast(value="const char*") BytePointer symbol)
@Cast(value="CUresult") public static int cuLibraryGetUnifiedFunction(@Cast(value="void**") @ByPtrPtr Pointer fptr, CUlib_st library, String symbol)
@Cast(value="CUresult") public static int cuKernelGetAttribute(IntPointer pi, @Cast(value="CUfunction_attribute") int attrib, CUkern_st kernel, @Cast(value="CUdevice") int dev)
pi
- - Returned attribute valueattrib
- - Attribute requestedkernel
- - Kernel to query attribute ofdev
- - Device to query attribute of::cuLibraryLoadData,
::cuLibraryLoadFromFile,
::cuLibraryUnload,
::cuKernelSetAttribute,
::cuLibraryGetKernel,
::cuLaunchKernel,
::cuKernelGetFunction,
::cuLibraryGetModule,
::cuModuleGetFunction,
::cuFuncGetAttribute
@Cast(value="CUresult") public static int cuKernelGetAttribute(IntBuffer pi, @Cast(value="CUfunction_attribute") int attrib, CUkern_st kernel, @Cast(value="CUdevice") int dev)
@Cast(value="CUresult") public static int cuKernelGetAttribute(int[] pi, @Cast(value="CUfunction_attribute") int attrib, CUkern_st kernel, @Cast(value="CUdevice") int dev)
@Cast(value="CUresult") public static int cuKernelSetAttribute(@Cast(value="CUfunction_attribute") int attrib, int val, CUkern_st kernel, @Cast(value="CUdevice") int dev)
attrib
- - Attribute requestedval
- - Value to setkernel
- - Kernel to set attribute ofdev
- - Device to set attribute of::cuLibraryLoadData,
::cuLibraryLoadFromFile,
::cuLibraryUnload,
::cuKernelGetAttribute,
::cuLibraryGetKernel,
::cuLaunchKernel,
::cuKernelGetFunction,
::cuLibraryGetModule,
::cuModuleGetFunction,
::cuFuncSetAttribute
@Cast(value="CUresult") public static int cuKernelSetCacheConfig(CUkern_st kernel, @Cast(value="CUfunc_cache") int config, @Cast(value="CUdevice") int dev)
kernel
- - Kernel to configure cache forconfig
- - Requested cache configurationdev
- - Device to set attribute of::cuLibraryLoadData,
::cuLibraryLoadFromFile,
::cuLibraryUnload,
::cuLibraryGetKernel,
::cuKernelGetFunction,
::cuLibraryGetModule,
::cuModuleGetFunction,
::cuFuncSetCacheConfig,
::cuCtxSetCacheConfig,
::cuLaunchKernel
@Cast(value="CUresult") public static int cuKernelGetName(@Cast(value="const char**") PointerPointer name, CUkern_st hfunc)
name
- - The returned name of the functionhfunc
- - The function handle to retrieve the name for@Cast(value="CUresult") public static int cuKernelGetName(@Cast(value="const char**") @ByPtrPtr BytePointer name, CUkern_st hfunc)
@Cast(value="CUresult") public static int cuKernelGetName(@Cast(value="const char**") @ByPtrPtr ByteBuffer name, CUkern_st hfunc)
@Cast(value="CUresult") public static int cuKernelGetName(@Cast(value="const char**") @ByPtrPtr byte[] name, CUkern_st hfunc)
@Cast(value="CUresult") public static int cuMemGetInfo(@Cast(value="size_t*") SizeTPointer _free, @Cast(value="size_t*") SizeTPointer total)
free
- - Returned free memory in bytestotal
- - Returned total memory in bytes::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaMemGetInfo
@Cast(value="CUresult") public static int cuMemAlloc(@Cast(value="CUdeviceptr*") LongPointer dptr, @Cast(value="size_t") long bytesize)
dptr
- - Returned device pointerbytesize
- - Requested allocation size in bytes::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaMalloc
@Cast(value="CUresult") public static int cuMemAlloc(@Cast(value="CUdeviceptr*") LongBuffer dptr, @Cast(value="size_t") long bytesize)
@Cast(value="CUresult") public static int cuMemAlloc(@Cast(value="CUdeviceptr*") long[] dptr, @Cast(value="size_t") long bytesize)
@Cast(value="CUresult") public static int cuMemAllocPitch(@Cast(value="CUdeviceptr*") LongPointer dptr, @Cast(value="size_t*") SizeTPointer pPitch, @Cast(value="size_t") long WidthInBytes, @Cast(value="size_t") long Height, @Cast(value="unsigned int") int ElementSizeBytes)
T* pElement = (T*)((char*)BaseAddress + Row * Pitch) + Column;
The pitch returned by ::cuMemAllocPitch() is guaranteed to work with
::cuMemcpy2D() under all circumstances. For allocations of 2D arrays, it is
recommended that programmers consider performing pitch allocations using
::cuMemAllocPitch(). Due to alignment restrictions in the hardware, this is
especially true if the application will be performing 2D memory copies
between different regions of device memory (whether linear memory or CUDA
arrays).
The byte alignment of the pitch returned by ::cuMemAllocPitch() is guaranteed
to match or exceed the alignment requirement for texture binding with
::cuTexRefSetAddress2D().dptr
- - Returned device pointerpPitch
- - Returned pitch of allocation in bytesWidthInBytes
- - Requested allocation width in bytesHeight
- - Requested allocation height in rowsElementSizeBytes
- - Size of largest reads/writes for range::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaMallocPitch
@Cast(value="CUresult") public static int cuMemAllocPitch(@Cast(value="CUdeviceptr*") LongBuffer dptr, @Cast(value="size_t*") SizeTPointer pPitch, @Cast(value="size_t") long WidthInBytes, @Cast(value="size_t") long Height, @Cast(value="unsigned int") int ElementSizeBytes)
@Cast(value="CUresult") public static int cuMemAllocPitch(@Cast(value="CUdeviceptr*") long[] dptr, @Cast(value="size_t*") SizeTPointer pPitch, @Cast(value="size_t") long WidthInBytes, @Cast(value="size_t") long Height, @Cast(value="unsigned int") int ElementSizeBytes)
@Cast(value="CUresult") public static int cuMemFree(@Cast(value="CUdeviceptr") long dptr)
dptr
- - Pointer to memory to free::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemAllocManaged, ::cuMemAllocAsync, ::cuMemAllocFromPoolAsync,
::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned, ::cuMemcpy3D, ::cuMemcpy3DAsync,
::cuMemcpyAtoA, ::cuMemcpyAtoD, ::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA,
::cuMemcpyDtoD, ::cuMemcpyDtoDAsync, ::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA,
::cuMemcpyHtoAAsync, ::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc, ::cuMemFreeAsync,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaFree
@Cast(value="CUresult") public static int cuMemGetAddressRange(@Cast(value="CUdeviceptr*") LongPointer pbase, @Cast(value="size_t*") SizeTPointer psize, @Cast(value="CUdeviceptr") long dptr)
pbase
- - Returned base addresspsize
- - Returned size of device memory allocationdptr
- - Device pointer to query::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32
@Cast(value="CUresult") public static int cuMemGetAddressRange(@Cast(value="CUdeviceptr*") LongBuffer pbase, @Cast(value="size_t*") SizeTPointer psize, @Cast(value="CUdeviceptr") long dptr)
@Cast(value="CUresult") public static int cuMemGetAddressRange(@Cast(value="CUdeviceptr*") long[] pbase, @Cast(value="size_t*") SizeTPointer psize, @Cast(value="CUdeviceptr") long dptr)
@Cast(value="CUresult") public static int cuMemAllocHost(@Cast(value="void**") PointerPointer pp, @Cast(value="size_t") long bytesize)
pp
- - Returned pointer to host memorybytesize
- - Requested allocation size in bytes::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaMallocHost
@Cast(value="CUresult") public static int cuMemAllocHost(@Cast(value="void**") @ByPtrPtr Pointer pp, @Cast(value="size_t") long bytesize)
@Cast(value="CUresult") public static int cuMemFreeHost(Pointer p)
p
- - Pointer to memory to free::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaFreeHost
@Cast(value="CUresult") public static int cuMemHostAlloc(@Cast(value="void**") PointerPointer pp, @Cast(value="size_t") long bytesize, @Cast(value="unsigned int") int Flags)
pp
- - Returned pointer to host memorybytesize
- - Requested allocation size in bytesFlags
- - Flags for allocation request::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaHostAlloc
@Cast(value="CUresult") public static int cuMemHostAlloc(@Cast(value="void**") @ByPtrPtr Pointer pp, @Cast(value="size_t") long bytesize, @Cast(value="unsigned int") int Flags)
@Cast(value="CUresult") public static int cuMemHostGetDevicePointer(@Cast(value="CUdeviceptr*") LongPointer pdptr, Pointer p, @Cast(value="unsigned int") int Flags)
pdptr
- - Returned device pointerp
- - Host pointerFlags
- - Options (must be 0)::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaHostGetDevicePointer
@Cast(value="CUresult") public static int cuMemHostGetDevicePointer(@Cast(value="CUdeviceptr*") LongBuffer pdptr, Pointer p, @Cast(value="unsigned int") int Flags)
@Cast(value="CUresult") public static int cuMemHostGetDevicePointer(@Cast(value="CUdeviceptr*") long[] pdptr, Pointer p, @Cast(value="unsigned int") int Flags)
@Cast(value="CUresult") public static int cuMemHostGetFlags(@Cast(value="unsigned int*") IntPointer pFlags, Pointer p)
pFlags
- - Returned flags wordp
- - Host pointer::cuMemAllocHost,
::cuMemHostAlloc,
::cudaHostGetFlags
@Cast(value="CUresult") public static int cuMemHostGetFlags(@Cast(value="unsigned int*") IntBuffer pFlags, Pointer p)
@Cast(value="CUresult") public static int cuMemHostGetFlags(@Cast(value="unsigned int*") int[] pFlags, Pointer p)
@Cast(value="CUresult") public static int cuMemAllocManaged(@Cast(value="CUdeviceptr*") LongPointer dptr, @Cast(value="size_t") long bytesize, @Cast(value="unsigned int") int flags)
dptr
- - Returned device pointerbytesize
- - Requested allocation size in bytesflags
- - Must be one of ::CU_MEM_ATTACH_GLOBAL or ::CU_MEM_ATTACH_HOST::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cuDeviceGetAttribute, ::cuStreamAttachMemAsync,
::cudaMallocManaged
@Cast(value="CUresult") public static int cuMemAllocManaged(@Cast(value="CUdeviceptr*") LongBuffer dptr, @Cast(value="size_t") long bytesize, @Cast(value="unsigned int") int flags)
@Cast(value="CUresult") public static int cuMemAllocManaged(@Cast(value="CUdeviceptr*") long[] dptr, @Cast(value="size_t") long bytesize, @Cast(value="unsigned int") int flags)
@Cast(value="CUresult") public static int cuDeviceGetByPCIBusId(@Cast(value="CUdevice*") IntPointer dev, @Cast(value="const char*") BytePointer pciBusId)
dev
- - Returned device handlepciBusId
- - String in one of the following forms:
[domain]:[bus]:[device].[function]
[domain]:[bus]:[device]
[bus]:[device].[function]
where \p domain, \p bus, \p device, and \p function are all hexadecimal values::cuDeviceGet,
::cuDeviceGetAttribute,
::cuDeviceGetPCIBusId,
::cudaDeviceGetByPCIBusId
@Cast(value="CUresult") public static int cuDeviceGetByPCIBusId(@Cast(value="CUdevice*") IntBuffer dev, String pciBusId)
@Cast(value="CUresult") public static int cuDeviceGetByPCIBusId(@Cast(value="CUdevice*") int[] dev, @Cast(value="const char*") BytePointer pciBusId)
@Cast(value="CUresult") public static int cuDeviceGetByPCIBusId(@Cast(value="CUdevice*") IntPointer dev, String pciBusId)
@Cast(value="CUresult") public static int cuDeviceGetByPCIBusId(@Cast(value="CUdevice*") IntBuffer dev, @Cast(value="const char*") BytePointer pciBusId)
@Cast(value="CUresult") public static int cuDeviceGetByPCIBusId(@Cast(value="CUdevice*") int[] dev, String pciBusId)
@Cast(value="CUresult") public static int cuDeviceGetPCIBusId(@Cast(value="char*") BytePointer pciBusId, int len, @Cast(value="CUdevice") int dev)
pciBusId
- - Returned identifier string for the device in the following format
[domain]:[bus]:[device].[function]
where \p domain, \p bus, \p device, and \p function are all hexadecimal values.
pciBusId should be large enough to store 13 characters including the NULL-terminator.len
- - Maximum length of string to store in \p namedev
- - Device to get identifier string for::cuDeviceGet,
::cuDeviceGetAttribute,
::cuDeviceGetByPCIBusId,
::cudaDeviceGetPCIBusId
@Cast(value="CUresult") public static int cuDeviceGetPCIBusId(@Cast(value="char*") ByteBuffer pciBusId, int len, @Cast(value="CUdevice") int dev)
@Cast(value="CUresult") public static int cuDeviceGetPCIBusId(@Cast(value="char*") byte[] pciBusId, int len, @Cast(value="CUdevice") int dev)
@Cast(value="CUresult") public static int cuIpcGetEventHandle(@Cast(value="CUipcEventHandle*") CUipcEventHandle_v1 pHandle, CUevent_st event)
pHandle
- - Pointer to a user allocated CUipcEventHandle
in which to return the opaque event handleevent
- - Event allocated with ::CU_EVENT_INTERPROCESS and
::CU_EVENT_DISABLE_TIMING flags.::cuEventCreate,
::cuEventDestroy,
::cuEventSynchronize,
::cuEventQuery,
::cuStreamWaitEvent,
::cuIpcOpenEventHandle,
::cuIpcGetMemHandle,
::cuIpcOpenMemHandle,
::cuIpcCloseMemHandle,
::cudaIpcGetEventHandle
@Cast(value="CUresult") public static int cuIpcOpenEventHandle(@ByPtrPtr CUevent_st phEvent, @ByVal @Cast(value="CUipcEventHandle*") CUipcEventHandle_v1 handle)
phEvent
- - Returns the imported eventhandle
- - Interprocess handle to open::cuEventCreate,
::cuEventDestroy,
::cuEventSynchronize,
::cuEventQuery,
::cuStreamWaitEvent,
::cuIpcGetEventHandle,
::cuIpcGetMemHandle,
::cuIpcOpenMemHandle,
::cuIpcCloseMemHandle,
::cudaIpcOpenEventHandle
@Cast(value="CUresult") public static int cuIpcGetMemHandle(@Cast(value="CUipcMemHandle*") CUipcMemHandle_v1 pHandle, @Cast(value="CUdeviceptr") long dptr)
pHandle
- - Pointer to user allocated ::CUipcMemHandle to return
the handle in.dptr
- - Base pointer to previously allocated device memory::cuMemAlloc,
::cuMemFree,
::cuIpcGetEventHandle,
::cuIpcOpenEventHandle,
::cuIpcOpenMemHandle,
::cuIpcCloseMemHandle,
::cudaIpcGetMemHandle
@Cast(value="CUresult") public static int cuIpcOpenMemHandle(@Cast(value="CUdeviceptr*") LongPointer pdptr, @ByVal @Cast(value="CUipcMemHandle*") CUipcMemHandle_v1 handle, @Cast(value="unsigned int") int Flags)
pdptr
- - Returned device pointerhandle
- - ::CUipcMemHandle to openFlags
- - Flags for this operation. Must be specified as ::CU_IPC_MEM_LAZY_ENABLE_PEER_ACCESS::cuMemAlloc,
::cuMemFree,
::cuIpcGetEventHandle,
::cuIpcOpenEventHandle,
::cuIpcGetMemHandle,
::cuIpcCloseMemHandle,
::cuCtxEnablePeerAccess,
::cuDeviceCanAccessPeer,
::cudaIpcOpenMemHandle
@Cast(value="CUresult") public static int cuIpcOpenMemHandle(@Cast(value="CUdeviceptr*") LongBuffer pdptr, @ByVal @Cast(value="CUipcMemHandle*") CUipcMemHandle_v1 handle, @Cast(value="unsigned int") int Flags)
@Cast(value="CUresult") public static int cuIpcOpenMemHandle(@Cast(value="CUdeviceptr*") long[] pdptr, @ByVal @Cast(value="CUipcMemHandle*") CUipcMemHandle_v1 handle, @Cast(value="unsigned int") int Flags)
@Cast(value="CUresult") public static int cuIpcCloseMemHandle(@Cast(value="CUdeviceptr") long dptr)
dptr
- - Device pointer returned by ::cuIpcOpenMemHandle::cuMemAlloc,
::cuMemFree,
::cuIpcGetEventHandle,
::cuIpcOpenEventHandle,
::cuIpcGetMemHandle,
::cuIpcOpenMemHandle,
::cudaIpcCloseMemHandle
@Cast(value="CUresult") public static int cuMemHostRegister(Pointer p, @Cast(value="size_t") long bytesize, @Cast(value="unsigned int") int Flags)
p
- - Host pointer to memory to page-lockbytesize
- - Size in bytes of the address range to page-lockFlags
- - Flags for allocation request::cuMemHostUnregister,
::cuMemHostGetFlags,
::cuMemHostGetDevicePointer,
::cudaHostRegister
@Cast(value="CUresult") public static int cuMemHostUnregister(Pointer p)
p
- - Host pointer to memory to unregister::cuMemHostRegister,
::cudaHostUnregister
@Cast(value="CUresult") public static int cuMemcpy(@Cast(value="CUdeviceptr") long dst, @Cast(value="CUdeviceptr") long src, @Cast(value="size_t") long ByteCount)
dst
- - Destination unified virtual address space pointersrc
- - Source unified virtual address space pointerByteCount
- - Size of memory copy in bytes::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaMemcpy,
::cudaMemcpyToSymbol,
::cudaMemcpyFromSymbol
@Cast(value="CUresult") public static int cuMemcpyPeer(@Cast(value="CUdeviceptr") long dstDevice, CUctx_st dstContext, @Cast(value="CUdeviceptr") long srcDevice, CUctx_st srcContext, @Cast(value="size_t") long ByteCount)
dstDevice
- - Destination device pointerdstContext
- - Destination contextsrcDevice
- - Source device pointersrcContext
- - Source contextByteCount
- - Size of memory copy in bytes::cuMemcpyDtoD, ::cuMemcpy3DPeer, ::cuMemcpyDtoDAsync, ::cuMemcpyPeerAsync,
::cuMemcpy3DPeerAsync,
::cudaMemcpyPeer
@Cast(value="CUresult") public static int cuMemcpyHtoD(@Cast(value="CUdeviceptr") long dstDevice, @Const Pointer srcHost, @Cast(value="size_t") long ByteCount)
dstDevice
- - Destination device pointersrcHost
- - Source host pointerByteCount
- - Size of memory copy in bytes::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaMemcpy,
::cudaMemcpyToSymbol
@Cast(value="CUresult") public static int cuMemcpyDtoH(Pointer dstHost, @Cast(value="CUdeviceptr") long srcDevice, @Cast(value="size_t") long ByteCount)
dstHost
- - Destination host pointersrcDevice
- - Source device pointerByteCount
- - Size of memory copy in bytes::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaMemcpy,
::cudaMemcpyFromSymbol
@Cast(value="CUresult") public static int cuMemcpyDtoD(@Cast(value="CUdeviceptr") long dstDevice, @Cast(value="CUdeviceptr") long srcDevice, @Cast(value="size_t") long ByteCount)
dstDevice
- - Destination device pointersrcDevice
- - Source device pointerByteCount
- - Size of memory copy in bytes::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaMemcpy,
::cudaMemcpyToSymbol,
::cudaMemcpyFromSymbol
@Cast(value="CUresult") public static int cuMemcpyDtoA(CUarray_st dstArray, @Cast(value="size_t") long dstOffset, @Cast(value="CUdeviceptr") long srcDevice, @Cast(value="size_t") long ByteCount)
dstArray
- - Destination arraydstOffset
- - Offset in bytes of destination arraysrcDevice
- - Source device pointerByteCount
- - Size of memory copy in bytes::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaMemcpyToArray
@Cast(value="CUresult") public static int cuMemcpyAtoD(@Cast(value="CUdeviceptr") long dstDevice, CUarray_st srcArray, @Cast(value="size_t") long srcOffset, @Cast(value="size_t") long ByteCount)
dstDevice
- - Destination device pointersrcArray
- - Source arraysrcOffset
- - Offset in bytes of source arrayByteCount
- - Size of memory copy in bytes::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaMemcpyFromArray
@Cast(value="CUresult") public static int cuMemcpyHtoA(CUarray_st dstArray, @Cast(value="size_t") long dstOffset, @Const Pointer srcHost, @Cast(value="size_t") long ByteCount)
dstArray
- - Destination arraydstOffset
- - Offset in bytes of destination arraysrcHost
- - Source host pointerByteCount
- - Size of memory copy in bytes::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaMemcpyToArray
@Cast(value="CUresult") public static int cuMemcpyAtoH(Pointer dstHost, CUarray_st srcArray, @Cast(value="size_t") long srcOffset, @Cast(value="size_t") long ByteCount)
dstHost
- - Destination device pointersrcArray
- - Source arraysrcOffset
- - Offset in bytes of source arrayByteCount
- - Size of memory copy in bytes::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaMemcpyFromArray
@Cast(value="CUresult") public static int cuMemcpyAtoA(CUarray_st dstArray, @Cast(value="size_t") long dstOffset, CUarray_st srcArray, @Cast(value="size_t") long srcOffset, @Cast(value="size_t") long ByteCount)
dstArray
- - Destination arraydstOffset
- - Offset in bytes of destination arraysrcArray
- - Source arraysrcOffset
- - Offset in bytes of source arrayByteCount
- - Size of memory copy in bytes::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaMemcpyArrayToArray
@Cast(value="CUresult") public static int cuMemcpy2D(@Cast(value="const CUDA_MEMCPY2D*") CUDA_MEMCPY2D_v2 pCopy)
typedef struct CUDA_MEMCPY2D_st {
unsigned int srcXInBytes, srcY;
CUmemorytype srcMemoryType;
const void *srcHost;
CUdeviceptr srcDevice;
CUarray srcArray;
unsigned int srcPitch;
unsigned int dstXInBytes, dstY;
CUmemorytype dstMemoryType;
void *dstHost;
CUdeviceptr dstDevice;
CUarray dstArray;
unsigned int dstPitch;
unsigned int WidthInBytes;
unsigned int Height;
} CUDA_MEMCPY2D;
where:
- ::srcMemoryType and ::dstMemoryType specify the type of memory of the
source and destination, respectively; ::CUmemorytype_enum is defined as:
typedef enum CUmemorytype_enum {
CU_MEMORYTYPE_HOST = 0x01,
CU_MEMORYTYPE_DEVICE = 0x02,
CU_MEMORYTYPE_ARRAY = 0x03,
CU_MEMORYTYPE_UNIFIED = 0x04
} CUmemorytype;
\par
If ::srcMemoryType is ::CU_MEMORYTYPE_UNIFIED, ::srcDevice and ::srcPitch
specify the (unified virtual address space) base address of the source data
and the bytes per row to apply. ::srcArray is ignored.
This value may be used only if unified addressing is supported in the calling
context.
\par
If ::srcMemoryType is ::CU_MEMORYTYPE_HOST, ::srcHost and ::srcPitch
specify the (host) base address of the source data and the bytes per row to
apply. ::srcArray is ignored.
\par
If ::srcMemoryType is ::CU_MEMORYTYPE_DEVICE, ::srcDevice and ::srcPitch
specify the (device) base address of the source data and the bytes per row
to apply. ::srcArray is ignored.
\par
If ::srcMemoryType is ::CU_MEMORYTYPE_ARRAY, ::srcArray specifies the
handle of the source data. ::srcHost, ::srcDevice and ::srcPitch are
ignored.
\par
If ::dstMemoryType is ::CU_MEMORYTYPE_HOST, ::dstHost and ::dstPitch
specify the (host) base address of the destination data and the bytes per
row to apply. ::dstArray is ignored.
\par
If ::dstMemoryType is ::CU_MEMORYTYPE_UNIFIED, ::dstDevice and ::dstPitch
specify the (unified virtual address space) base address of the source data
and the bytes per row to apply. ::dstArray is ignored.
This value may be used only if unified addressing is supported in the calling
context.
\par
If ::dstMemoryType is ::CU_MEMORYTYPE_DEVICE, ::dstDevice and ::dstPitch
specify the (device) base address of the destination data and the bytes per
row to apply. ::dstArray is ignored.
\par
If ::dstMemoryType is ::CU_MEMORYTYPE_ARRAY, ::dstArray specifies the
handle of the destination data. ::dstHost, ::dstDevice and ::dstPitch are
ignored.
- ::srcXInBytes and ::srcY specify the base address of the source data for
the copy.
\par
For host pointers, the starting address is
void* Start = (void*)((char*)srcHost+srcY*srcPitch + srcXInBytes);
\par
For device pointers, the starting address is
CUdeviceptr Start = srcDevice+srcY*srcPitch+srcXInBytes;
\par
For CUDA arrays, ::srcXInBytes must be evenly divisible by the array
element size.
- ::dstXInBytes and ::dstY specify the base address of the destination data
for the copy.
\par
For host pointers, the base address is
void* dstStart = (void*)((char*)dstHost+dstY*dstPitch + dstXInBytes);
\par
For device pointers, the starting address is
CUdeviceptr dstStart = dstDevice+dstY*dstPitch+dstXInBytes;
\par
For CUDA arrays, ::dstXInBytes must be evenly divisible by the array
element size.
- ::WidthInBytes and ::Height specify the width (in bytes) and height of
the 2D copy being performed.
- If specified, ::srcPitch must be greater than or equal to ::WidthInBytes +
::srcXInBytes, and ::dstPitch must be greater than or equal to
::WidthInBytes + dstXInBytes.
\par
::cuMemcpy2D() returns an error if any pitch is greater than the maximum
allowed (::CU_DEVICE_ATTRIBUTE_MAX_PITCH). ::cuMemAllocPitch() passes back
pitches that always work with ::cuMemcpy2D(). On intra-device memory copies
(device to device, CUDA array to device, CUDA array to CUDA array),
::cuMemcpy2D() may fail for pitches not computed by ::cuMemAllocPitch().
::cuMemcpy2DUnaligned() does not have this restriction, but may run
significantly slower in the cases where ::cuMemcpy2D() would have returned
an error code.pCopy
- - Parameters for the memory copy::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaMemcpy2D,
::cudaMemcpy2DToArray,
::cudaMemcpy2DFromArray
@Cast(value="CUresult") public static int cuMemcpy2DUnaligned(@Cast(value="const CUDA_MEMCPY2D*") CUDA_MEMCPY2D_v2 pCopy)
typedef struct CUDA_MEMCPY2D_st {
unsigned int srcXInBytes, srcY;
CUmemorytype srcMemoryType;
const void *srcHost;
CUdeviceptr srcDevice;
CUarray srcArray;
unsigned int srcPitch;
unsigned int dstXInBytes, dstY;
CUmemorytype dstMemoryType;
void *dstHost;
CUdeviceptr dstDevice;
CUarray dstArray;
unsigned int dstPitch;
unsigned int WidthInBytes;
unsigned int Height;
} CUDA_MEMCPY2D;
where:
- ::srcMemoryType and ::dstMemoryType specify the type of memory of the
source and destination, respectively; ::CUmemorytype_enum is defined as:
typedef enum CUmemorytype_enum {
CU_MEMORYTYPE_HOST = 0x01,
CU_MEMORYTYPE_DEVICE = 0x02,
CU_MEMORYTYPE_ARRAY = 0x03,
CU_MEMORYTYPE_UNIFIED = 0x04
} CUmemorytype;
\par
If ::srcMemoryType is ::CU_MEMORYTYPE_UNIFIED, ::srcDevice and ::srcPitch
specify the (unified virtual address space) base address of the source data
and the bytes per row to apply. ::srcArray is ignored.
This value may be used only if unified addressing is supported in the calling
context.
\par
If ::srcMemoryType is ::CU_MEMORYTYPE_HOST, ::srcHost and ::srcPitch
specify the (host) base address of the source data and the bytes per row to
apply. ::srcArray is ignored.
\par
If ::srcMemoryType is ::CU_MEMORYTYPE_DEVICE, ::srcDevice and ::srcPitch
specify the (device) base address of the source data and the bytes per row
to apply. ::srcArray is ignored.
\par
If ::srcMemoryType is ::CU_MEMORYTYPE_ARRAY, ::srcArray specifies the
handle of the source data. ::srcHost, ::srcDevice and ::srcPitch are
ignored.
\par
If ::dstMemoryType is ::CU_MEMORYTYPE_UNIFIED, ::dstDevice and ::dstPitch
specify the (unified virtual address space) base address of the source data
and the bytes per row to apply. ::dstArray is ignored.
This value may be used only if unified addressing is supported in the calling
context.
\par
If ::dstMemoryType is ::CU_MEMORYTYPE_HOST, ::dstHost and ::dstPitch
specify the (host) base address of the destination data and the bytes per
row to apply. ::dstArray is ignored.
\par
If ::dstMemoryType is ::CU_MEMORYTYPE_DEVICE, ::dstDevice and ::dstPitch
specify the (device) base address of the destination data and the bytes per
row to apply. ::dstArray is ignored.
\par
If ::dstMemoryType is ::CU_MEMORYTYPE_ARRAY, ::dstArray specifies the
handle of the destination data. ::dstHost, ::dstDevice and ::dstPitch are
ignored.
- ::srcXInBytes and ::srcY specify the base address of the source data for
the copy.
\par
For host pointers, the starting address is
void* Start = (void*)((char*)srcHost+srcY*srcPitch + srcXInBytes);
\par
For device pointers, the starting address is
CUdeviceptr Start = srcDevice+srcY*srcPitch+srcXInBytes;
\par
For CUDA arrays, ::srcXInBytes must be evenly divisible by the array
element size.
- ::dstXInBytes and ::dstY specify the base address of the destination data
for the copy.
\par
For host pointers, the base address is
void* dstStart = (void*)((char*)dstHost+dstY*dstPitch + dstXInBytes);
\par
For device pointers, the starting address is
CUdeviceptr dstStart = dstDevice+dstY*dstPitch+dstXInBytes;
\par
For CUDA arrays, ::dstXInBytes must be evenly divisible by the array
element size.
- ::WidthInBytes and ::Height specify the width (in bytes) and height of
the 2D copy being performed.
- If specified, ::srcPitch must be greater than or equal to ::WidthInBytes +
::srcXInBytes, and ::dstPitch must be greater than or equal to
::WidthInBytes + dstXInBytes.
\par
::cuMemcpy2D() returns an error if any pitch is greater than the maximum
allowed (::CU_DEVICE_ATTRIBUTE_MAX_PITCH). ::cuMemAllocPitch() passes back
pitches that always work with ::cuMemcpy2D(). On intra-device memory copies
(device to device, CUDA array to device, CUDA array to CUDA array),
::cuMemcpy2D() may fail for pitches not computed by ::cuMemAllocPitch().
::cuMemcpy2DUnaligned() does not have this restriction, but may run
significantly slower in the cases where ::cuMemcpy2D() would have returned
an error code.pCopy
- - Parameters for the memory copy::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaMemcpy2D,
::cudaMemcpy2DToArray,
::cudaMemcpy2DFromArray
@Cast(value="CUresult") public static int cuMemcpy3D(@Cast(value="const CUDA_MEMCPY3D*") CUDA_MEMCPY3D_v2 pCopy)
typedef struct CUDA_MEMCPY3D_st {
unsigned int srcXInBytes, srcY, srcZ;
unsigned int srcLOD;
CUmemorytype srcMemoryType;
const void *srcHost;
CUdeviceptr srcDevice;
CUarray srcArray;
unsigned int srcPitch; // ignored when src is array
unsigned int srcHeight; // ignored when src is array; may be 0 if Depth==1
unsigned int dstXInBytes, dstY, dstZ;
unsigned int dstLOD;
CUmemorytype dstMemoryType;
void *dstHost;
CUdeviceptr dstDevice;
CUarray dstArray;
unsigned int dstPitch; // ignored when dst is array
unsigned int dstHeight; // ignored when dst is array; may be 0 if Depth==1
unsigned int WidthInBytes;
unsigned int Height;
unsigned int Depth;
} CUDA_MEMCPY3D;
where:
- ::srcMemoryType and ::dstMemoryType specify the type of memory of the
source and destination, respectively; ::CUmemorytype_enum is defined as:
typedef enum CUmemorytype_enum {
CU_MEMORYTYPE_HOST = 0x01,
CU_MEMORYTYPE_DEVICE = 0x02,
CU_MEMORYTYPE_ARRAY = 0x03,
CU_MEMORYTYPE_UNIFIED = 0x04
} CUmemorytype;
\par
If ::srcMemoryType is ::CU_MEMORYTYPE_UNIFIED, ::srcDevice and ::srcPitch
specify the (unified virtual address space) base address of the source data
and the bytes per row to apply. ::srcArray is ignored.
This value may be used only if unified addressing is supported in the calling
context.
\par
If ::srcMemoryType is ::CU_MEMORYTYPE_HOST, ::srcHost, ::srcPitch and
::srcHeight specify the (host) base address of the source data, the bytes
per row, and the height of each 2D slice of the 3D array. ::srcArray is
ignored.
\par
If ::srcMemoryType is ::CU_MEMORYTYPE_DEVICE, ::srcDevice, ::srcPitch and
::srcHeight specify the (device) base address of the source data, the bytes
per row, and the height of each 2D slice of the 3D array. ::srcArray is
ignored.
\par
If ::srcMemoryType is ::CU_MEMORYTYPE_ARRAY, ::srcArray specifies the
handle of the source data. ::srcHost, ::srcDevice, ::srcPitch and
::srcHeight are ignored.
\par
If ::dstMemoryType is ::CU_MEMORYTYPE_UNIFIED, ::dstDevice and ::dstPitch
specify the (unified virtual address space) base address of the source data
and the bytes per row to apply. ::dstArray is ignored.
This value may be used only if unified addressing is supported in the calling
context.
\par
If ::dstMemoryType is ::CU_MEMORYTYPE_HOST, ::dstHost and ::dstPitch
specify the (host) base address of the destination data, the bytes per row,
and the height of each 2D slice of the 3D array. ::dstArray is ignored.
\par
If ::dstMemoryType is ::CU_MEMORYTYPE_DEVICE, ::dstDevice and ::dstPitch
specify the (device) base address of the destination data, the bytes per
row, and the height of each 2D slice of the 3D array. ::dstArray is ignored.
\par
If ::dstMemoryType is ::CU_MEMORYTYPE_ARRAY, ::dstArray specifies the
handle of the destination data. ::dstHost, ::dstDevice, ::dstPitch and
::dstHeight are ignored.
- ::srcXInBytes, ::srcY and ::srcZ specify the base address of the source
data for the copy.
\par
For host pointers, the starting address is
void* Start = (void*)((char*)srcHost+(srcZ*srcHeight+srcY)*srcPitch + srcXInBytes);
\par
For device pointers, the starting address is
CUdeviceptr Start = srcDevice+(srcZ*srcHeight+srcY)*srcPitch+srcXInBytes;
\par
For CUDA arrays, ::srcXInBytes must be evenly divisible by the array
element size.
- dstXInBytes, ::dstY and ::dstZ specify the base address of the
destination data for the copy.
\par
For host pointers, the base address is
void* dstStart = (void*)((char*)dstHost+(dstZ*dstHeight+dstY)*dstPitch + dstXInBytes);
\par
For device pointers, the starting address is
CUdeviceptr dstStart = dstDevice+(dstZ*dstHeight+dstY)*dstPitch+dstXInBytes;
\par
For CUDA arrays, ::dstXInBytes must be evenly divisible by the array
element size.
- ::WidthInBytes, ::Height and ::Depth specify the width (in bytes), height
and depth of the 3D copy being performed.
- If specified, ::srcPitch must be greater than or equal to ::WidthInBytes +
::srcXInBytes, and ::dstPitch must be greater than or equal to
::WidthInBytes + dstXInBytes.
- If specified, ::srcHeight must be greater than or equal to ::Height +
::srcY, and ::dstHeight must be greater than or equal to ::Height + ::dstY.
\par
::cuMemcpy3D() returns an error if any pitch is greater than the maximum
allowed (::CU_DEVICE_ATTRIBUTE_MAX_PITCH).
The ::srcLOD and ::dstLOD members of the ::CUDA_MEMCPY3D structure must be
set to 0.pCopy
- - Parameters for the memory copy::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaMemcpy3D
@Cast(value="CUresult") public static int cuMemcpy3DPeer(@Cast(value="const CUDA_MEMCPY3D_PEER*") CUDA_MEMCPY3D_PEER_v1 pCopy)
pCopy
- - Parameters for the memory copy::cuMemcpyDtoD, ::cuMemcpyPeer, ::cuMemcpyDtoDAsync, ::cuMemcpyPeerAsync,
::cuMemcpy3DPeerAsync,
::cudaMemcpy3DPeer
@Cast(value="CUresult") public static int cuMemcpyAsync(@Cast(value="CUdeviceptr") long dst, @Cast(value="CUdeviceptr") long src, @Cast(value="size_t") long ByteCount, CUstream_st hStream)
dst
- - Destination unified virtual address space pointersrc
- - Source unified virtual address space pointerByteCount
- - Size of memory copy in byteshStream
- - Stream identifier::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D8Async,
::cuMemsetD2D16, ::cuMemsetD2D16Async, ::cuMemsetD2D32, ::cuMemsetD2D32Async,
::cuMemsetD8, ::cuMemsetD8Async, ::cuMemsetD16, ::cuMemsetD16Async,
::cuMemsetD32, ::cuMemsetD32Async,
::cudaMemcpyAsync,
::cudaMemcpyToSymbolAsync,
::cudaMemcpyFromSymbolAsync
@Cast(value="CUresult") public static int cuMemcpyPeerAsync(@Cast(value="CUdeviceptr") long dstDevice, CUctx_st dstContext, @Cast(value="CUdeviceptr") long srcDevice, CUctx_st srcContext, @Cast(value="size_t") long ByteCount, CUstream_st hStream)
dstDevice
- - Destination device pointerdstContext
- - Destination contextsrcDevice
- - Source device pointersrcContext
- - Source contextByteCount
- - Size of memory copy in byteshStream
- - Stream identifier::cuMemcpyDtoD, ::cuMemcpyPeer, ::cuMemcpy3DPeer, ::cuMemcpyDtoDAsync,
::cuMemcpy3DPeerAsync,
::cudaMemcpyPeerAsync
@Cast(value="CUresult") public static int cuMemcpyHtoDAsync(@Cast(value="CUdeviceptr") long dstDevice, @Const Pointer srcHost, @Cast(value="size_t") long ByteCount, CUstream_st hStream)
dstDevice
- - Destination device pointersrcHost
- - Source host pointerByteCount
- - Size of memory copy in byteshStream
- - Stream identifier::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D8Async,
::cuMemsetD2D16, ::cuMemsetD2D16Async, ::cuMemsetD2D32, ::cuMemsetD2D32Async,
::cuMemsetD8, ::cuMemsetD8Async, ::cuMemsetD16, ::cuMemsetD16Async,
::cuMemsetD32, ::cuMemsetD32Async,
::cudaMemcpyAsync,
::cudaMemcpyToSymbolAsync
@Cast(value="CUresult") public static int cuMemcpyDtoHAsync(Pointer dstHost, @Cast(value="CUdeviceptr") long srcDevice, @Cast(value="size_t") long ByteCount, CUstream_st hStream)
dstHost
- - Destination host pointersrcDevice
- - Source device pointerByteCount
- - Size of memory copy in byteshStream
- - Stream identifier::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D8Async,
::cuMemsetD2D16, ::cuMemsetD2D16Async, ::cuMemsetD2D32, ::cuMemsetD2D32Async,
::cuMemsetD8, ::cuMemsetD8Async, ::cuMemsetD16, ::cuMemsetD16Async,
::cuMemsetD32, ::cuMemsetD32Async,
::cudaMemcpyAsync,
::cudaMemcpyFromSymbolAsync
@Cast(value="CUresult") public static int cuMemcpyDtoDAsync(@Cast(value="CUdeviceptr") long dstDevice, @Cast(value="CUdeviceptr") long srcDevice, @Cast(value="size_t") long ByteCount, CUstream_st hStream)
dstDevice
- - Destination device pointersrcDevice
- - Source device pointerByteCount
- - Size of memory copy in byteshStream
- - Stream identifier::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D8Async,
::cuMemsetD2D16, ::cuMemsetD2D16Async, ::cuMemsetD2D32, ::cuMemsetD2D32Async,
::cuMemsetD8, ::cuMemsetD8Async, ::cuMemsetD16, ::cuMemsetD16Async,
::cuMemsetD32, ::cuMemsetD32Async,
::cudaMemcpyAsync,
::cudaMemcpyToSymbolAsync,
::cudaMemcpyFromSymbolAsync
@Cast(value="CUresult") public static int cuMemcpyHtoAAsync(CUarray_st dstArray, @Cast(value="size_t") long dstOffset, @Const Pointer srcHost, @Cast(value="size_t") long ByteCount, CUstream_st hStream)
dstArray
- - Destination arraydstOffset
- - Offset in bytes of destination arraysrcHost
- - Source host pointerByteCount
- - Size of memory copy in byteshStream
- - Stream identifier::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D8Async,
::cuMemsetD2D16, ::cuMemsetD2D16Async, ::cuMemsetD2D32, ::cuMemsetD2D32Async,
::cuMemsetD8, ::cuMemsetD8Async, ::cuMemsetD16, ::cuMemsetD16Async,
::cuMemsetD32, ::cuMemsetD32Async,
::cudaMemcpyToArrayAsync
@Cast(value="CUresult") public static int cuMemcpyAtoHAsync(Pointer dstHost, CUarray_st srcArray, @Cast(value="size_t") long srcOffset, @Cast(value="size_t") long ByteCount, CUstream_st hStream)
dstHost
- - Destination pointersrcArray
- - Source arraysrcOffset
- - Offset in bytes of source arrayByteCount
- - Size of memory copy in byteshStream
- - Stream identifier::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D8Async,
::cuMemsetD2D16, ::cuMemsetD2D16Async, ::cuMemsetD2D32, ::cuMemsetD2D32Async,
::cuMemsetD8, ::cuMemsetD8Async, ::cuMemsetD16, ::cuMemsetD16Async,
::cuMemsetD32, ::cuMemsetD32Async,
::cudaMemcpyFromArrayAsync
@Cast(value="CUresult") public static int cuMemcpy2DAsync(@Cast(value="const CUDA_MEMCPY2D*") CUDA_MEMCPY2D_v2 pCopy, CUstream_st hStream)
typedef struct CUDA_MEMCPY2D_st {
unsigned int srcXInBytes, srcY;
CUmemorytype srcMemoryType;
const void *srcHost;
CUdeviceptr srcDevice;
CUarray srcArray;
unsigned int srcPitch;
unsigned int dstXInBytes, dstY;
CUmemorytype dstMemoryType;
void *dstHost;
CUdeviceptr dstDevice;
CUarray dstArray;
unsigned int dstPitch;
unsigned int WidthInBytes;
unsigned int Height;
} CUDA_MEMCPY2D;
where:
- ::srcMemoryType and ::dstMemoryType specify the type of memory of the
source and destination, respectively; ::CUmemorytype_enum is defined as:
typedef enum CUmemorytype_enum {
CU_MEMORYTYPE_HOST = 0x01,
CU_MEMORYTYPE_DEVICE = 0x02,
CU_MEMORYTYPE_ARRAY = 0x03,
CU_MEMORYTYPE_UNIFIED = 0x04
} CUmemorytype;
\par
If ::srcMemoryType is ::CU_MEMORYTYPE_HOST, ::srcHost and ::srcPitch
specify the (host) base address of the source data and the bytes per row to
apply. ::srcArray is ignored.
\par
If ::srcMemoryType is ::CU_MEMORYTYPE_UNIFIED, ::srcDevice and ::srcPitch
specify the (unified virtual address space) base address of the source data
and the bytes per row to apply. ::srcArray is ignored.
This value may be used only if unified addressing is supported in the calling
context.
\par
If ::srcMemoryType is ::CU_MEMORYTYPE_DEVICE, ::srcDevice and ::srcPitch
specify the (device) base address of the source data and the bytes per row
to apply. ::srcArray is ignored.
\par
If ::srcMemoryType is ::CU_MEMORYTYPE_ARRAY, ::srcArray specifies the
handle of the source data. ::srcHost, ::srcDevice and ::srcPitch are
ignored.
\par
If ::dstMemoryType is ::CU_MEMORYTYPE_UNIFIED, ::dstDevice and ::dstPitch
specify the (unified virtual address space) base address of the source data
and the bytes per row to apply. ::dstArray is ignored.
This value may be used only if unified addressing is supported in the calling
context.
\par
If ::dstMemoryType is ::CU_MEMORYTYPE_HOST, ::dstHost and ::dstPitch
specify the (host) base address of the destination data and the bytes per
row to apply. ::dstArray is ignored.
\par
If ::dstMemoryType is ::CU_MEMORYTYPE_DEVICE, ::dstDevice and ::dstPitch
specify the (device) base address of the destination data and the bytes per
row to apply. ::dstArray is ignored.
\par
If ::dstMemoryType is ::CU_MEMORYTYPE_ARRAY, ::dstArray specifies the
handle of the destination data. ::dstHost, ::dstDevice and ::dstPitch are
ignored.
- ::srcXInBytes and ::srcY specify the base address of the source data for
the copy.
\par
For host pointers, the starting address is
void* Start = (void*)((char*)srcHost+srcY*srcPitch + srcXInBytes);
\par
For device pointers, the starting address is
CUdeviceptr Start = srcDevice+srcY*srcPitch+srcXInBytes;
\par
For CUDA arrays, ::srcXInBytes must be evenly divisible by the array
element size.
- ::dstXInBytes and ::dstY specify the base address of the destination data
for the copy.
\par
For host pointers, the base address is
void* dstStart = (void*)((char*)dstHost+dstY*dstPitch + dstXInBytes);
\par
For device pointers, the starting address is
CUdeviceptr dstStart = dstDevice+dstY*dstPitch+dstXInBytes;
\par
For CUDA arrays, ::dstXInBytes must be evenly divisible by the array
element size.
- ::WidthInBytes and ::Height specify the width (in bytes) and height of
the 2D copy being performed.
- If specified, ::srcPitch must be greater than or equal to ::WidthInBytes +
::srcXInBytes, and ::dstPitch must be greater than or equal to
::WidthInBytes + dstXInBytes.
- If specified, ::srcPitch must be greater than or equal to ::WidthInBytes +
::srcXInBytes, and ::dstPitch must be greater than or equal to
::WidthInBytes + dstXInBytes.
- If specified, ::srcHeight must be greater than or equal to ::Height +
::srcY, and ::dstHeight must be greater than or equal to ::Height + ::dstY.
\par
::cuMemcpy2DAsync() returns an error if any pitch is greater than the maximum
allowed (::CU_DEVICE_ATTRIBUTE_MAX_PITCH). ::cuMemAllocPitch() passes back
pitches that always work with ::cuMemcpy2D(). On intra-device memory copies
(device to device, CUDA array to device, CUDA array to CUDA array),
::cuMemcpy2DAsync() may fail for pitches not computed by ::cuMemAllocPitch().pCopy
- - Parameters for the memory copyhStream
- - Stream identifier::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D8Async,
::cuMemsetD2D16, ::cuMemsetD2D16Async, ::cuMemsetD2D32, ::cuMemsetD2D32Async,
::cuMemsetD8, ::cuMemsetD8Async, ::cuMemsetD16, ::cuMemsetD16Async,
::cuMemsetD32, ::cuMemsetD32Async,
::cudaMemcpy2DAsync,
::cudaMemcpy2DToArrayAsync,
::cudaMemcpy2DFromArrayAsync
@Cast(value="CUresult") public static int cuMemcpy3DAsync(@Cast(value="const CUDA_MEMCPY3D*") CUDA_MEMCPY3D_v2 pCopy, CUstream_st hStream)
typedef struct CUDA_MEMCPY3D_st {
unsigned int srcXInBytes, srcY, srcZ;
unsigned int srcLOD;
CUmemorytype srcMemoryType;
const void *srcHost;
CUdeviceptr srcDevice;
CUarray srcArray;
unsigned int srcPitch; // ignored when src is array
unsigned int srcHeight; // ignored when src is array; may be 0 if Depth==1
unsigned int dstXInBytes, dstY, dstZ;
unsigned int dstLOD;
CUmemorytype dstMemoryType;
void *dstHost;
CUdeviceptr dstDevice;
CUarray dstArray;
unsigned int dstPitch; // ignored when dst is array
unsigned int dstHeight; // ignored when dst is array; may be 0 if Depth==1
unsigned int WidthInBytes;
unsigned int Height;
unsigned int Depth;
} CUDA_MEMCPY3D;
where:
- ::srcMemoryType and ::dstMemoryType specify the type of memory of the
source and destination, respectively; ::CUmemorytype_enum is defined as:
typedef enum CUmemorytype_enum {
CU_MEMORYTYPE_HOST = 0x01,
CU_MEMORYTYPE_DEVICE = 0x02,
CU_MEMORYTYPE_ARRAY = 0x03,
CU_MEMORYTYPE_UNIFIED = 0x04
} CUmemorytype;
\par
If ::srcMemoryType is ::CU_MEMORYTYPE_UNIFIED, ::srcDevice and ::srcPitch
specify the (unified virtual address space) base address of the source data
and the bytes per row to apply. ::srcArray is ignored.
This value may be used only if unified addressing is supported in the calling
context.
\par
If ::srcMemoryType is ::CU_MEMORYTYPE_HOST, ::srcHost, ::srcPitch and
::srcHeight specify the (host) base address of the source data, the bytes
per row, and the height of each 2D slice of the 3D array. ::srcArray is
ignored.
\par
If ::srcMemoryType is ::CU_MEMORYTYPE_DEVICE, ::srcDevice, ::srcPitch and
::srcHeight specify the (device) base address of the source data, the bytes
per row, and the height of each 2D slice of the 3D array. ::srcArray is
ignored.
\par
If ::srcMemoryType is ::CU_MEMORYTYPE_ARRAY, ::srcArray specifies the
handle of the source data. ::srcHost, ::srcDevice, ::srcPitch and
::srcHeight are ignored.
\par
If ::dstMemoryType is ::CU_MEMORYTYPE_UNIFIED, ::dstDevice and ::dstPitch
specify the (unified virtual address space) base address of the source data
and the bytes per row to apply. ::dstArray is ignored.
This value may be used only if unified addressing is supported in the calling
context.
\par
If ::dstMemoryType is ::CU_MEMORYTYPE_HOST, ::dstHost and ::dstPitch
specify the (host) base address of the destination data, the bytes per row,
and the height of each 2D slice of the 3D array. ::dstArray is ignored.
\par
If ::dstMemoryType is ::CU_MEMORYTYPE_DEVICE, ::dstDevice and ::dstPitch
specify the (device) base address of the destination data, the bytes per
row, and the height of each 2D slice of the 3D array. ::dstArray is ignored.
\par
If ::dstMemoryType is ::CU_MEMORYTYPE_ARRAY, ::dstArray specifies the
handle of the destination data. ::dstHost, ::dstDevice, ::dstPitch and
::dstHeight are ignored.
- ::srcXInBytes, ::srcY and ::srcZ specify the base address of the source
data for the copy.
\par
For host pointers, the starting address is
void* Start = (void*)((char*)srcHost+(srcZ*srcHeight+srcY)*srcPitch + srcXInBytes);
\par
For device pointers, the starting address is
CUdeviceptr Start = srcDevice+(srcZ*srcHeight+srcY)*srcPitch+srcXInBytes;
\par
For CUDA arrays, ::srcXInBytes must be evenly divisible by the array
element size.
- dstXInBytes, ::dstY and ::dstZ specify the base address of the
destination data for the copy.
\par
For host pointers, the base address is
void* dstStart = (void*)((char*)dstHost+(dstZ*dstHeight+dstY)*dstPitch + dstXInBytes);
\par
For device pointers, the starting address is
CUdeviceptr dstStart = dstDevice+(dstZ*dstHeight+dstY)*dstPitch+dstXInBytes;
\par
For CUDA arrays, ::dstXInBytes must be evenly divisible by the array
element size.
- ::WidthInBytes, ::Height and ::Depth specify the width (in bytes), height
and depth of the 3D copy being performed.
- If specified, ::srcPitch must be greater than or equal to ::WidthInBytes +
::srcXInBytes, and ::dstPitch must be greater than or equal to
::WidthInBytes + dstXInBytes.
- If specified, ::srcHeight must be greater than or equal to ::Height +
::srcY, and ::dstHeight must be greater than or equal to ::Height + ::dstY.
\par
::cuMemcpy3DAsync() returns an error if any pitch is greater than the maximum
allowed (::CU_DEVICE_ATTRIBUTE_MAX_PITCH).
The ::srcLOD and ::dstLOD members of the ::CUDA_MEMCPY3D structure must be
set to 0.pCopy
- - Parameters for the memory copyhStream
- - Stream identifier::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D8Async,
::cuMemsetD2D16, ::cuMemsetD2D16Async, ::cuMemsetD2D32, ::cuMemsetD2D32Async,
::cuMemsetD8, ::cuMemsetD8Async, ::cuMemsetD16, ::cuMemsetD16Async,
::cuMemsetD32, ::cuMemsetD32Async,
::cudaMemcpy3DAsync
@Cast(value="CUresult") public static int cuMemcpy3DPeerAsync(@Cast(value="const CUDA_MEMCPY3D_PEER*") CUDA_MEMCPY3D_PEER_v1 pCopy, CUstream_st hStream)
pCopy
- - Parameters for the memory copyhStream
- - Stream identifier::cuMemcpyDtoD, ::cuMemcpyPeer, ::cuMemcpyDtoDAsync, ::cuMemcpyPeerAsync,
::cuMemcpy3DPeerAsync,
::cudaMemcpy3DPeerAsync
@Cast(value="CUresult") public static int cuMemsetD8(@Cast(value="CUdeviceptr") long dstDevice, @Cast(value="unsigned char") byte uc, @Cast(value="size_t") long N)
dstDevice
- - Destination device pointeruc
- - Value to setN
- - Number of elements::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D8Async,
::cuMemsetD2D16, ::cuMemsetD2D16Async, ::cuMemsetD2D32, ::cuMemsetD2D32Async,
::cuMemsetD8Async, ::cuMemsetD16, ::cuMemsetD16Async,
::cuMemsetD32, ::cuMemsetD32Async,
::cudaMemset
@Cast(value="CUresult") public static int cuMemsetD16(@Cast(value="CUdeviceptr") long dstDevice, @Cast(value="unsigned short") short us, @Cast(value="size_t") long N)
dstDevice
- - Destination device pointerus
- - Value to setN
- - Number of elements::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D8Async,
::cuMemsetD2D16, ::cuMemsetD2D16Async, ::cuMemsetD2D32, ::cuMemsetD2D32Async,
::cuMemsetD8, ::cuMemsetD8Async, ::cuMemsetD16Async,
::cuMemsetD32, ::cuMemsetD32Async,
::cudaMemset
@Cast(value="CUresult") public static int cuMemsetD32(@Cast(value="CUdeviceptr") long dstDevice, @Cast(value="unsigned int") int ui, @Cast(value="size_t") long N)
dstDevice
- - Destination device pointerui
- - Value to setN
- - Number of elements::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D8Async,
::cuMemsetD2D16, ::cuMemsetD2D16Async, ::cuMemsetD2D32, ::cuMemsetD2D32Async,
::cuMemsetD8, ::cuMemsetD8Async, ::cuMemsetD16, ::cuMemsetD16Async,
::cuMemsetD32Async,
::cudaMemset
@Cast(value="CUresult") public static int cuMemsetD2D8(@Cast(value="CUdeviceptr") long dstDevice, @Cast(value="size_t") long dstPitch, @Cast(value="unsigned char") byte uc, @Cast(value="size_t") long Width, @Cast(value="size_t") long Height)
dstDevice
- - Destination device pointerdstPitch
- - Pitch of destination device pointer(Unused if \p Height is 1)uc
- - Value to setWidth
- - Width of rowHeight
- - Number of rows::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8Async,
::cuMemsetD2D16, ::cuMemsetD2D16Async, ::cuMemsetD2D32, ::cuMemsetD2D32Async,
::cuMemsetD8, ::cuMemsetD8Async, ::cuMemsetD16, ::cuMemsetD16Async,
::cuMemsetD32, ::cuMemsetD32Async,
::cudaMemset2D
@Cast(value="CUresult") public static int cuMemsetD2D16(@Cast(value="CUdeviceptr") long dstDevice, @Cast(value="size_t") long dstPitch, @Cast(value="unsigned short") short us, @Cast(value="size_t") long Width, @Cast(value="size_t") long Height)
dstDevice
- - Destination device pointerdstPitch
- - Pitch of destination device pointer(Unused if \p Height is 1)us
- - Value to setWidth
- - Width of rowHeight
- - Number of rows::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D8Async,
::cuMemsetD2D16Async, ::cuMemsetD2D32, ::cuMemsetD2D32Async,
::cuMemsetD8, ::cuMemsetD8Async, ::cuMemsetD16, ::cuMemsetD16Async,
::cuMemsetD32, ::cuMemsetD32Async,
::cudaMemset2D
@Cast(value="CUresult") public static int cuMemsetD2D32(@Cast(value="CUdeviceptr") long dstDevice, @Cast(value="size_t") long dstPitch, @Cast(value="unsigned int") int ui, @Cast(value="size_t") long Width, @Cast(value="size_t") long Height)
dstDevice
- - Destination device pointerdstPitch
- - Pitch of destination device pointer(Unused if \p Height is 1)ui
- - Value to setWidth
- - Width of rowHeight
- - Number of rows::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D8Async,
::cuMemsetD2D16, ::cuMemsetD2D16Async, ::cuMemsetD2D32Async,
::cuMemsetD8, ::cuMemsetD8Async, ::cuMemsetD16, ::cuMemsetD16Async,
::cuMemsetD32, ::cuMemsetD32Async,
::cudaMemset2D
@Cast(value="CUresult") public static int cuMemsetD8Async(@Cast(value="CUdeviceptr") long dstDevice, @Cast(value="unsigned char") byte uc, @Cast(value="size_t") long N, CUstream_st hStream)
dstDevice
- - Destination device pointeruc
- - Value to setN
- - Number of elementshStream
- - Stream identifier::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D8Async,
::cuMemsetD2D16, ::cuMemsetD2D16Async, ::cuMemsetD2D32, ::cuMemsetD2D32Async,
::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD16Async,
::cuMemsetD32, ::cuMemsetD32Async,
::cudaMemsetAsync
@Cast(value="CUresult") public static int cuMemsetD16Async(@Cast(value="CUdeviceptr") long dstDevice, @Cast(value="unsigned short") short us, @Cast(value="size_t") long N, CUstream_st hStream)
dstDevice
- - Destination device pointerus
- - Value to setN
- - Number of elementshStream
- - Stream identifier::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D8Async,
::cuMemsetD2D16, ::cuMemsetD2D16Async, ::cuMemsetD2D32, ::cuMemsetD2D32Async,
::cuMemsetD8, ::cuMemsetD8Async, ::cuMemsetD16,
::cuMemsetD32, ::cuMemsetD32Async,
::cudaMemsetAsync
@Cast(value="CUresult") public static int cuMemsetD32Async(@Cast(value="CUdeviceptr") long dstDevice, @Cast(value="unsigned int") int ui, @Cast(value="size_t") long N, CUstream_st hStream)
dstDevice
- - Destination device pointerui
- - Value to setN
- - Number of elementshStream
- - Stream identifier::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D8Async,
::cuMemsetD2D16, ::cuMemsetD2D16Async, ::cuMemsetD2D32, ::cuMemsetD2D32Async,
::cuMemsetD8, ::cuMemsetD8Async, ::cuMemsetD16, ::cuMemsetD16Async, ::cuMemsetD32,
::cudaMemsetAsync
@Cast(value="CUresult") public static int cuMemsetD2D8Async(@Cast(value="CUdeviceptr") long dstDevice, @Cast(value="size_t") long dstPitch, @Cast(value="unsigned char") byte uc, @Cast(value="size_t") long Width, @Cast(value="size_t") long Height, CUstream_st hStream)
dstDevice
- - Destination device pointerdstPitch
- - Pitch of destination device pointer(Unused if \p Height is 1)uc
- - Value to setWidth
- - Width of rowHeight
- - Number of rowshStream
- - Stream identifier::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8,
::cuMemsetD2D16, ::cuMemsetD2D16Async, ::cuMemsetD2D32, ::cuMemsetD2D32Async,
::cuMemsetD8, ::cuMemsetD8Async, ::cuMemsetD16, ::cuMemsetD16Async,
::cuMemsetD32, ::cuMemsetD32Async,
::cudaMemset2DAsync
@Cast(value="CUresult") public static int cuMemsetD2D16Async(@Cast(value="CUdeviceptr") long dstDevice, @Cast(value="size_t") long dstPitch, @Cast(value="unsigned short") short us, @Cast(value="size_t") long Width, @Cast(value="size_t") long Height, CUstream_st hStream)
dstDevice
- - Destination device pointerdstPitch
- - Pitch of destination device pointer(Unused if \p Height is 1)us
- - Value to setWidth
- - Width of rowHeight
- - Number of rowshStream
- - Stream identifier::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D8Async,
::cuMemsetD2D16, ::cuMemsetD2D32, ::cuMemsetD2D32Async,
::cuMemsetD8, ::cuMemsetD8Async, ::cuMemsetD16, ::cuMemsetD16Async,
::cuMemsetD32, ::cuMemsetD32Async,
::cudaMemset2DAsync
@Cast(value="CUresult") public static int cuMemsetD2D32Async(@Cast(value="CUdeviceptr") long dstDevice, @Cast(value="size_t") long dstPitch, @Cast(value="unsigned int") int ui, @Cast(value="size_t") long Width, @Cast(value="size_t") long Height, CUstream_st hStream)
dstDevice
- - Destination device pointerdstPitch
- - Pitch of destination device pointer(Unused if \p Height is 1)ui
- - Value to setWidth
- - Width of rowHeight
- - Number of rowshStream
- - Stream identifier::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D8Async,
::cuMemsetD2D16, ::cuMemsetD2D16Async, ::cuMemsetD2D32,
::cuMemsetD8, ::cuMemsetD8Async, ::cuMemsetD16, ::cuMemsetD16Async,
::cuMemsetD32, ::cuMemsetD32Async,
::cudaMemset2DAsync
@Cast(value="CUresult") public static int cuArrayCreate(@ByPtrPtr CUarray_st pHandle, @Cast(value="const CUDA_ARRAY_DESCRIPTOR*") CUDA_ARRAY_DESCRIPTOR_v2 pAllocateArray)
typedef struct {
unsigned int Width;
unsigned int Height;
CUarray_format Format;
unsigned int NumChannels;
} CUDA_ARRAY_DESCRIPTOR;
where:
- \p Width, and \p Height are the width, and height of the CUDA array (in
elements); the CUDA array is one-dimensional if height is 0, two-dimensional
otherwise;
- ::Format specifies the format of the elements; ::CUarray_format is
defined as:
typedef enum CUarray_format_enum {
CU_AD_FORMAT_UNSIGNED_INT8 = 0x01,
CU_AD_FORMAT_UNSIGNED_INT16 = 0x02,
CU_AD_FORMAT_UNSIGNED_INT32 = 0x03,
CU_AD_FORMAT_SIGNED_INT8 = 0x08,
CU_AD_FORMAT_SIGNED_INT16 = 0x09,
CU_AD_FORMAT_SIGNED_INT32 = 0x0a,
CU_AD_FORMAT_HALF = 0x10,
CU_AD_FORMAT_FLOAT = 0x20
} CUarray_format;
- \p NumChannels specifies the number of packed components per CUDA array
element; it may be 1, 2, or 4;
Here are examples of CUDA array descriptions:
Description for a CUDA array of 2048 floats:
CUDA_ARRAY_DESCRIPTOR desc;
desc.Format = CU_AD_FORMAT_FLOAT;
desc.NumChannels = 1;
desc.Width = 2048;
desc.Height = 1;
Description for a 64 x 64 CUDA array of floats:
CUDA_ARRAY_DESCRIPTOR desc;
desc.Format = CU_AD_FORMAT_FLOAT;
desc.NumChannels = 1;
desc.Width = 64;
desc.Height = 64;
Description for a \p width x \p height CUDA array of 64-bit, 4x16-bit
float16's:
CUDA_ARRAY_DESCRIPTOR desc;
desc.Format = CU_AD_FORMAT_HALF;
desc.NumChannels = 4;
desc.Width = width;
desc.Height = height;
Description for a \p width x \p height CUDA array of 16-bit elements, each
of which is two 8-bit unsigned chars:
CUDA_ARRAY_DESCRIPTOR arrayDesc;
desc.Format = CU_AD_FORMAT_UNSIGNED_INT8;
desc.NumChannels = 2;
desc.Width = width;
desc.Height = height;
pHandle
- - Returned arraypAllocateArray
- - Array descriptor::cuArray3DCreate, ::cuArray3DGetDescriptor,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaMallocArray
@Cast(value="CUresult") public static int cuArrayGetDescriptor(@Cast(value="CUDA_ARRAY_DESCRIPTOR*") CUDA_ARRAY_DESCRIPTOR_v2 pArrayDescriptor, CUarray_st hArray)
pArrayDescriptor
- - Returned array descriptorhArray
- - Array to get descriptor of::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaArrayGetInfo
@Cast(value="CUresult") public static int cuArrayGetSparseProperties(@Cast(value="CUDA_ARRAY_SPARSE_PROPERTIES*") CUDA_ARRAY_SPARSE_PROPERTIES_v1 sparseProperties, CUarray_st array)
sparseProperties
- [out] - Pointer to ::CUDA_ARRAY_SPARSE_PROPERTIESarray
- [in] - CUDA array to get the sparse properties of::cuMipmappedArrayGetSparseProperties, ::cuMemMapArrayAsync
@Cast(value="CUresult") public static int cuMipmappedArrayGetSparseProperties(@Cast(value="CUDA_ARRAY_SPARSE_PROPERTIES*") CUDA_ARRAY_SPARSE_PROPERTIES_v1 sparseProperties, CUmipmappedArray_st mipmap)
sparseProperties
- [out] - Pointer to ::CUDA_ARRAY_SPARSE_PROPERTIESmipmap
- [in] - CUDA mipmapped array to get the sparse properties of::cuArrayGetSparseProperties, ::cuMemMapArrayAsync
@Cast(value="CUresult") public static int cuArrayGetMemoryRequirements(@Cast(value="CUDA_ARRAY_MEMORY_REQUIREMENTS*") CUDA_ARRAY_MEMORY_REQUIREMENTS_v1 memoryRequirements, CUarray_st array, @Cast(value="CUdevice") int device)
memoryRequirements
- [out] - Pointer to ::CUDA_ARRAY_MEMORY_REQUIREMENTSarray
- [in] - CUDA array to get the memory requirements ofdevice
- [in] - Device to get the memory requirements for::cuMipmappedArrayGetMemoryRequirements, ::cuMemMapArrayAsync
@Cast(value="CUresult") public static int cuMipmappedArrayGetMemoryRequirements(@Cast(value="CUDA_ARRAY_MEMORY_REQUIREMENTS*") CUDA_ARRAY_MEMORY_REQUIREMENTS_v1 memoryRequirements, CUmipmappedArray_st mipmap, @Cast(value="CUdevice") int device)
memoryRequirements
- [out] - Pointer to ::CUDA_ARRAY_MEMORY_REQUIREMENTSmipmap
- [in] - CUDA mipmapped array to get the memory requirements ofdevice
- [in] - Device to get the memory requirements for::cuArrayGetMemoryRequirements, ::cuMemMapArrayAsync
@Cast(value="CUresult") public static int cuArrayGetPlane(@ByPtrPtr CUarray_st pPlaneArray, CUarray_st hArray, @Cast(value="unsigned int") int planeIdx)
pPlaneArray
- - Returned CUDA array referenced by the \p planeIdxhArray
- - Multiplanar CUDA arrayplaneIdx
- - Plane index::cuArrayCreate,
::cudaArrayGetPlane
@Cast(value="CUresult") public static int cuArrayDestroy(CUarray_st hArray)
hArray
- - Array to destroy::cuArray3DCreate, ::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaFreeArray
@Cast(value="CUresult") public static int cuArray3DCreate(@ByPtrPtr CUarray_st pHandle, @Cast(value="const CUDA_ARRAY3D_DESCRIPTOR*") CUDA_ARRAY3D_DESCRIPTOR_v2 pAllocateArray)
typedef struct {
unsigned int Width;
unsigned int Height;
unsigned int Depth;
CUarray_format Format;
unsigned int NumChannels;
unsigned int Flags;
} CUDA_ARRAY3D_DESCRIPTOR;
where:
- \p Width, \p Height, and \p Depth are the width, height, and depth of the
CUDA array (in elements); the following types of CUDA arrays can be allocated:
- A 1D array is allocated if \p Height and \p Depth extents are both zero.
- A 2D array is allocated if only \p Depth extent is zero.
- A 3D array is allocated if all three extents are non-zero.
- A 1D layered CUDA array is allocated if only \p Height is zero and the
::CUDA_ARRAY3D_LAYERED flag is set. Each layer is a 1D array. The number
of layers is determined by the depth extent.
- A 2D layered CUDA array is allocated if all three extents are non-zero and
the ::CUDA_ARRAY3D_LAYERED flag is set. Each layer is a 2D array. The number
of layers is determined by the depth extent.
- A cubemap CUDA array is allocated if all three extents are non-zero and the
::CUDA_ARRAY3D_CUBEMAP flag is set. \p Width must be equal to \p Height, and
\p Depth must be six. A cubemap is a special type of 2D layered CUDA array,
where the six layers represent the six faces of a cube. The order of the six
layers in memory is the same as that listed in ::CUarray_cubemap_face.
- A cubemap layered CUDA array is allocated if all three extents are non-zero,
and both, ::CUDA_ARRAY3D_CUBEMAP and ::CUDA_ARRAY3D_LAYERED flags are set.
\p Width must be equal to \p Height, and \p Depth must be a multiple of six.
A cubemap layered CUDA array is a special type of 2D layered CUDA array that
consists of a collection of cubemaps. The first six layers represent the first
cubemap, the next six layers form the second cubemap, and so on.
- ::Format specifies the format of the elements; ::CUarray_format is
defined as:
typedef enum CUarray_format_enum {
CU_AD_FORMAT_UNSIGNED_INT8 = 0x01,
CU_AD_FORMAT_UNSIGNED_INT16 = 0x02,
CU_AD_FORMAT_UNSIGNED_INT32 = 0x03,
CU_AD_FORMAT_SIGNED_INT8 = 0x08,
CU_AD_FORMAT_SIGNED_INT16 = 0x09,
CU_AD_FORMAT_SIGNED_INT32 = 0x0a,
CU_AD_FORMAT_HALF = 0x10,
CU_AD_FORMAT_FLOAT = 0x20
} CUarray_format;
- \p NumChannels specifies the number of packed components per CUDA array
element; it may be 1, 2, or 4;
- ::Flags may be set to
- ::CUDA_ARRAY3D_LAYERED to enable creation of layered CUDA arrays. If this flag is set,
\p Depth specifies the number of layers, not the depth of a 3D array.
- ::CUDA_ARRAY3D_SURFACE_LDST to enable surface references to be bound to the CUDA array.
If this flag is not set, ::cuSurfRefSetArray will fail when attempting to bind the CUDA array
to a surface reference.
- ::CUDA_ARRAY3D_CUBEMAP to enable creation of cubemaps. If this flag is set, \p Width must be
equal to \p Height, and \p Depth must be six. If the ::CUDA_ARRAY3D_LAYERED flag is also set,
then \p Depth must be a multiple of six.
- ::CUDA_ARRAY3D_TEXTURE_GATHER to indicate that the CUDA array will be used for texture gather.
Texture gather can only be performed on 2D CUDA arrays.
\p Width, \p Height and \p Depth must meet certain size requirements as listed in the following table.
All values are specified in elements. Note that for brevity's sake, the full name of the device attribute
is not specified. For ex., TEXTURE1D_WIDTH refers to the device attribute
::CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE1D_WIDTH.
Note that 2D CUDA arrays have different size requirements if the ::CUDA_ARRAY3D_TEXTURE_GATHER flag
is set. \p Width and \p Height must not be greater than ::CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_GATHER_WIDTH
and ::CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_GATHER_HEIGHT respectively, in that case.
CUDA array type | Valid extents that must always be met {(width range in elements), (height range), (depth range)} |
Valid extents with CUDA_ARRAY3D_SURFACE_LDST set {(width range in elements), (height range), (depth range)} |
1D | { (1,TEXTURE1D_WIDTH), 0, 0 } | { (1,SURFACE1D_WIDTH), 0, 0 } |
2D | { (1,TEXTURE2D_WIDTH), (1,TEXTURE2D_HEIGHT), 0 } | { (1,SURFACE2D_WIDTH), (1,SURFACE2D_HEIGHT), 0 } |
3D | { (1,TEXTURE3D_WIDTH), (1,TEXTURE3D_HEIGHT), (1,TEXTURE3D_DEPTH) }
OR { (1,TEXTURE3D_WIDTH_ALTERNATE), (1,TEXTURE3D_HEIGHT_ALTERNATE), (1,TEXTURE3D_DEPTH_ALTERNATE) } |
{ (1,SURFACE3D_WIDTH), (1,SURFACE3D_HEIGHT), (1,SURFACE3D_DEPTH) } |
1D Layered | { (1,TEXTURE1D_LAYERED_WIDTH), 0, (1,TEXTURE1D_LAYERED_LAYERS) } | { (1,SURFACE1D_LAYERED_WIDTH), 0, (1,SURFACE1D_LAYERED_LAYERS) } |
2D Layered | { (1,TEXTURE2D_LAYERED_WIDTH), (1,TEXTURE2D_LAYERED_HEIGHT), (1,TEXTURE2D_LAYERED_LAYERS) } | { (1,SURFACE2D_LAYERED_WIDTH), (1,SURFACE2D_LAYERED_HEIGHT), (1,SURFACE2D_LAYERED_LAYERS) } |
Cubemap | { (1,TEXTURECUBEMAP_WIDTH), (1,TEXTURECUBEMAP_WIDTH), 6 } | { (1,SURFACECUBEMAP_WIDTH), (1,SURFACECUBEMAP_WIDTH), 6 } |
Cubemap Layered | { (1,TEXTURECUBEMAP_LAYERED_WIDTH), (1,TEXTURECUBEMAP_LAYERED_WIDTH), (1,TEXTURECUBEMAP_LAYERED_LAYERS) } | { (1,SURFACECUBEMAP_LAYERED_WIDTH), (1,SURFACECUBEMAP_LAYERED_WIDTH), (1,SURFACECUBEMAP_LAYERED_LAYERS) } |
CUDA_ARRAY3D_DESCRIPTOR desc;
desc.Format = CU_AD_FORMAT_FLOAT;
desc.NumChannels = 1;
desc.Width = 2048;
desc.Height = 0;
desc.Depth = 0;
Description for a 64 x 64 CUDA array of floats:
CUDA_ARRAY3D_DESCRIPTOR desc;
desc.Format = CU_AD_FORMAT_FLOAT;
desc.NumChannels = 1;
desc.Width = 64;
desc.Height = 64;
desc.Depth = 0;
Description for a \p width x \p height x \p depth CUDA array of 64-bit,
4x16-bit float16's:
CUDA_ARRAY3D_DESCRIPTOR desc;
desc.Format = CU_AD_FORMAT_HALF;
desc.NumChannels = 4;
desc.Width = width;
desc.Height = height;
desc.Depth = depth;
pHandle
- - Returned arraypAllocateArray
- - 3D array descriptor::cuArray3DGetDescriptor, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaMalloc3DArray
@Cast(value="CUresult") public static int cuArray3DGetDescriptor(@Cast(value="CUDA_ARRAY3D_DESCRIPTOR*") CUDA_ARRAY3D_DESCRIPTOR_v2 pArrayDescriptor, CUarray_st hArray)
pArrayDescriptor
- - Returned 3D array descriptorhArray
- - 3D array to get descriptor of::cuArray3DCreate, ::cuArrayCreate,
::cuArrayDestroy, ::cuArrayGetDescriptor, ::cuMemAlloc, ::cuMemAllocHost,
::cuMemAllocPitch, ::cuMemcpy2D, ::cuMemcpy2DAsync, ::cuMemcpy2DUnaligned,
::cuMemcpy3D, ::cuMemcpy3DAsync, ::cuMemcpyAtoA, ::cuMemcpyAtoD,
::cuMemcpyAtoH, ::cuMemcpyAtoHAsync, ::cuMemcpyDtoA, ::cuMemcpyDtoD, ::cuMemcpyDtoDAsync,
::cuMemcpyDtoH, ::cuMemcpyDtoHAsync, ::cuMemcpyHtoA, ::cuMemcpyHtoAAsync,
::cuMemcpyHtoD, ::cuMemcpyHtoDAsync, ::cuMemFree, ::cuMemFreeHost,
::cuMemGetAddressRange, ::cuMemGetInfo, ::cuMemHostAlloc,
::cuMemHostGetDevicePointer, ::cuMemsetD2D8, ::cuMemsetD2D16,
::cuMemsetD2D32, ::cuMemsetD8, ::cuMemsetD16, ::cuMemsetD32,
::cudaArrayGetInfo
@Cast(value="CUresult") public static int cuMipmappedArrayCreate(@ByPtrPtr CUmipmappedArray_st pHandle, @Cast(value="const CUDA_ARRAY3D_DESCRIPTOR*") CUDA_ARRAY3D_DESCRIPTOR_v2 pMipmappedArrayDesc, @Cast(value="unsigned int") int numMipmapLevels)
typedef struct {
unsigned int Width;
unsigned int Height;
unsigned int Depth;
CUarray_format Format;
unsigned int NumChannels;
unsigned int Flags;
} CUDA_ARRAY3D_DESCRIPTOR;
where:
- \p Width, \p Height, and \p Depth are the width, height, and depth of the
CUDA array (in elements); the following types of CUDA arrays can be allocated:
- A 1D mipmapped array is allocated if \p Height and \p Depth extents are both zero.
- A 2D mipmapped array is allocated if only \p Depth extent is zero.
- A 3D mipmapped array is allocated if all three extents are non-zero.
- A 1D layered CUDA mipmapped array is allocated if only \p Height is zero and the
::CUDA_ARRAY3D_LAYERED flag is set. Each layer is a 1D array. The number
of layers is determined by the depth extent.
- A 2D layered CUDA mipmapped array is allocated if all three extents are non-zero and
the ::CUDA_ARRAY3D_LAYERED flag is set. Each layer is a 2D array. The number
of layers is determined by the depth extent.
- A cubemap CUDA mipmapped array is allocated if all three extents are non-zero and the
::CUDA_ARRAY3D_CUBEMAP flag is set. \p Width must be equal to \p Height, and
\p Depth must be six. A cubemap is a special type of 2D layered CUDA array,
where the six layers represent the six faces of a cube. The order of the six
layers in memory is the same as that listed in ::CUarray_cubemap_face.
- A cubemap layered CUDA mipmapped array is allocated if all three extents are non-zero,
and both, ::CUDA_ARRAY3D_CUBEMAP and ::CUDA_ARRAY3D_LAYERED flags are set.
\p Width must be equal to \p Height, and \p Depth must be a multiple of six.
A cubemap layered CUDA array is a special type of 2D layered CUDA array that
consists of a collection of cubemaps. The first six layers represent the first
cubemap, the next six layers form the second cubemap, and so on.
- ::Format specifies the format of the elements; ::CUarray_format is
defined as:
typedef enum CUarray_format_enum {
CU_AD_FORMAT_UNSIGNED_INT8 = 0x01,
CU_AD_FORMAT_UNSIGNED_INT16 = 0x02,
CU_AD_FORMAT_UNSIGNED_INT32 = 0x03,
CU_AD_FORMAT_SIGNED_INT8 = 0x08,
CU_AD_FORMAT_SIGNED_INT16 = 0x09,
CU_AD_FORMAT_SIGNED_INT32 = 0x0a,
CU_AD_FORMAT_HALF = 0x10,
CU_AD_FORMAT_FLOAT = 0x20
} CUarray_format;
- \p NumChannels specifies the number of packed components per CUDA array
element; it may be 1, 2, or 4;
- ::Flags may be set to
- ::CUDA_ARRAY3D_LAYERED to enable creation of layered CUDA mipmapped arrays. If this flag is set,
\p Depth specifies the number of layers, not the depth of a 3D array.
- ::CUDA_ARRAY3D_SURFACE_LDST to enable surface references to be bound to individual mipmap levels of
the CUDA mipmapped array. If this flag is not set, ::cuSurfRefSetArray will fail when attempting to
bind a mipmap level of the CUDA mipmapped array to a surface reference.
- ::CUDA_ARRAY3D_CUBEMAP to enable creation of mipmapped cubemaps. If this flag is set, \p Width must be
equal to \p Height, and \p Depth must be six. If the ::CUDA_ARRAY3D_LAYERED flag is also set,
then \p Depth must be a multiple of six.
- ::CUDA_ARRAY3D_TEXTURE_GATHER to indicate that the CUDA mipmapped array will be used for texture gather.
Texture gather can only be performed on 2D CUDA mipmapped arrays.
\p Width, \p Height and \p Depth must meet certain size requirements as listed in the following table.
All values are specified in elements. Note that for brevity's sake, the full name of the device attribute
is not specified. For ex., TEXTURE1D_MIPMAPPED_WIDTH refers to the device attribute
::CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE1D_MIPMAPPED_WIDTH.
CUDA array type | Valid extents that must always be met {(width range in elements), (height range), (depth range)} |
Valid extents with CUDA_ARRAY3D_SURFACE_LDST set {(width range in elements), (height range), (depth range)} |
1D | { (1,TEXTURE1D_MIPMAPPED_WIDTH), 0, 0 } | { (1,SURFACE1D_WIDTH), 0, 0 } |
2D | { (1,TEXTURE2D_MIPMAPPED_WIDTH), (1,TEXTURE2D_MIPMAPPED_HEIGHT), 0 } | { (1,SURFACE2D_WIDTH), (1,SURFACE2D_HEIGHT), 0 } |
3D | { (1,TEXTURE3D_WIDTH), (1,TEXTURE3D_HEIGHT), (1,TEXTURE3D_DEPTH) }
OR { (1,TEXTURE3D_WIDTH_ALTERNATE), (1,TEXTURE3D_HEIGHT_ALTERNATE), (1,TEXTURE3D_DEPTH_ALTERNATE) } |
{ (1,SURFACE3D_WIDTH), (1,SURFACE3D_HEIGHT), (1,SURFACE3D_DEPTH) } |
1D Layered | { (1,TEXTURE1D_LAYERED_WIDTH), 0, (1,TEXTURE1D_LAYERED_LAYERS) } | { (1,SURFACE1D_LAYERED_WIDTH), 0, (1,SURFACE1D_LAYERED_LAYERS) } |
2D Layered | { (1,TEXTURE2D_LAYERED_WIDTH), (1,TEXTURE2D_LAYERED_HEIGHT), (1,TEXTURE2D_LAYERED_LAYERS) } | { (1,SURFACE2D_LAYERED_WIDTH), (1,SURFACE2D_LAYERED_HEIGHT), (1,SURFACE2D_LAYERED_LAYERS) } |
Cubemap | { (1,TEXTURECUBEMAP_WIDTH), (1,TEXTURECUBEMAP_WIDTH), 6 } | { (1,SURFACECUBEMAP_WIDTH), (1,SURFACECUBEMAP_WIDTH), 6 } |
Cubemap Layered | { (1,TEXTURECUBEMAP_LAYERED_WIDTH), (1,TEXTURECUBEMAP_LAYERED_WIDTH), (1,TEXTURECUBEMAP_LAYERED_LAYERS) } | { (1,SURFACECUBEMAP_LAYERED_WIDTH), (1,SURFACECUBEMAP_LAYERED_WIDTH), (1,SURFACECUBEMAP_LAYERED_LAYERS) } |
pHandle
- - Returned mipmapped arraypMipmappedArrayDesc
- - mipmapped array descriptornumMipmapLevels
- - Number of mipmap levels::cuMipmappedArrayDestroy,
::cuMipmappedArrayGetLevel,
::cuArrayCreate,
::cudaMallocMipmappedArray
@Cast(value="CUresult") public static int cuMipmappedArrayGetLevel(@ByPtrPtr CUarray_st pLevelArray, CUmipmappedArray_st hMipmappedArray, @Cast(value="unsigned int") int level)
pLevelArray
- - Returned mipmap level CUDA arrayhMipmappedArray
- - CUDA mipmapped arraylevel
- - Mipmap level::cuMipmappedArrayCreate,
::cuMipmappedArrayDestroy,
::cuArrayCreate,
::cudaGetMipmappedArrayLevel
@Cast(value="CUresult") public static int cuMipmappedArrayDestroy(CUmipmappedArray_st hMipmappedArray)
hMipmappedArray
- - Mipmapped array to destroy::cuMipmappedArrayCreate,
::cuMipmappedArrayGetLevel,
::cuArrayCreate,
::cudaFreeMipmappedArray
@Cast(value="CUresult") public static int cuMemGetHandleForAddressRange(Pointer handle, @Cast(value="CUdeviceptr") long dptr, @Cast(value="size_t") long size, @Cast(value="CUmemRangeHandleType") int handleType, @Cast(value="unsigned long long") long flags)
handle
- [out] - Pointer to the location where the returned handle will be stored.dptr
- [in] - Pointer to a valid CUDA device allocation. Must be aligned to host page size.size
- [in] - Length of the address range. Must be aligned to host page size.handleType
- [in] - Type of handle requested (defines type and size of the \p handle output parameter)flags
- [in] - Reserved, must be zero@Cast(value="CUresult") public static int cuMemAddressReserve(@Cast(value="CUdeviceptr*") LongPointer ptr, @Cast(value="size_t") long size, @Cast(value="size_t") long alignment, @Cast(value="CUdeviceptr") long addr, @Cast(value="unsigned long long") long flags)
ptr
- [out] - Resulting pointer to start of virtual address range allocatedsize
- [in] - Size of the reserved virtual address range requestedalignment
- [in] - Alignment of the reserved virtual address range requestedaddr
- [in] - Fixed starting address range requestedflags
- [in] - Currently unused, must be zero::cuMemAddressFree
@Cast(value="CUresult") public static int cuMemAddressReserve(@Cast(value="CUdeviceptr*") LongBuffer ptr, @Cast(value="size_t") long size, @Cast(value="size_t") long alignment, @Cast(value="CUdeviceptr") long addr, @Cast(value="unsigned long long") long flags)
@Cast(value="CUresult") public static int cuMemAddressReserve(@Cast(value="CUdeviceptr*") long[] ptr, @Cast(value="size_t") long size, @Cast(value="size_t") long alignment, @Cast(value="CUdeviceptr") long addr, @Cast(value="unsigned long long") long flags)
@Cast(value="CUresult") public static int cuMemAddressFree(@Cast(value="CUdeviceptr") long ptr, @Cast(value="size_t") long size)
ptr
- [in] - Starting address of the virtual address range to freesize
- [in] - Size of the virtual address region to free::cuMemAddressReserve
@Cast(value="CUresult") public static int cuMemCreate(@Cast(value="CUmemGenericAllocationHandle*") LongPointer handle, @Cast(value="size_t") long size, @Cast(value="const CUmemAllocationProp*") CUmemAllocationProp_v1 prop, @Cast(value="unsigned long long") long flags)
handle
- [out] - Value of handle returned. All operations on this allocation are to be performed using this handle.size
- [in] - Size of the allocation requestedprop
- [in] - Properties of the allocation to create.flags
- [in] - flags for future use, must be zero now.::cuMemRelease, ::cuMemExportToShareableHandle, ::cuMemImportFromShareableHandle
@Cast(value="CUresult") public static int cuMemCreate(@Cast(value="CUmemGenericAllocationHandle*") LongBuffer handle, @Cast(value="size_t") long size, @Cast(value="const CUmemAllocationProp*") CUmemAllocationProp_v1 prop, @Cast(value="unsigned long long") long flags)
@Cast(value="CUresult") public static int cuMemCreate(@Cast(value="CUmemGenericAllocationHandle*") long[] handle, @Cast(value="size_t") long size, @Cast(value="const CUmemAllocationProp*") CUmemAllocationProp_v1 prop, @Cast(value="unsigned long long") long flags)
@Cast(value="CUresult") public static int cuMemRelease(@Cast(value="CUmemGenericAllocationHandle") long handle)
handle
- [in] Value of handle which was returned previously by cuMemCreate.::cuMemCreate
@Cast(value="CUresult") public static int cuMemMap(@Cast(value="CUdeviceptr") long ptr, @Cast(value="size_t") long size, @Cast(value="size_t") long offset, @Cast(value="CUmemGenericAllocationHandle") long handle, @Cast(value="unsigned long long") long flags)
ptr
- [in] - Address where memory will be mapped.size
- [in] - Size of the memory mapping.offset
- [in] - Offset into the memory represented by
- \p handle from which to start mapping
- Note: currently must be zero.handle
- [in] - Handle to a shareable memoryflags
- [in] - flags for future use, must be zero now.::cuMemUnmap, ::cuMemSetAccess, ::cuMemCreate, ::cuMemAddressReserve, ::cuMemImportFromShareableHandle
@Cast(value="CUresult") public static int cuMemMapArrayAsync(@Cast(value="CUarrayMapInfo*") CUarrayMapInfo_v1 mapInfoList, @Cast(value="unsigned int") int count, CUstream_st hStream)
typedef struct CUarrayMapInfo_st {
CUresourcetype resourceType;
union {
CUmipmappedArray mipmap;
CUarray array;
} resource;
CUarraySparseSubresourceType subresourceType;
union {
struct {
unsigned int level;
unsigned int layer;
unsigned int offsetX;
unsigned int offsetY;
unsigned int offsetZ;
unsigned int extentWidth;
unsigned int extentHeight;
unsigned int extentDepth;
} sparseLevel;
struct {
unsigned int layer;
unsigned long long offset;
unsigned long long size;
} miptail;
} subresource;
CUmemOperationType memOperationType;
CUmemHandleType memHandleType;
union {
CUmemGenericAllocationHandle memHandle;
} memHandle;
unsigned long long offset;
unsigned int deviceBitMask;
unsigned int flags;
unsigned int reserved[2];
} CUarrayMapInfo;
where ::CUarrayMapInfo::resourceType specifies the type of resource to be operated on.
If ::CUarrayMapInfo::resourceType is set to ::CUresourcetype::CU_RESOURCE_TYPE_ARRAY then
::CUarrayMapInfo::resource::array must be set to a valid sparse CUDA array handle.
The CUDA array must be either a 2D, 2D layered or 3D CUDA array and must have been allocated using
::cuArrayCreate or ::cuArray3DCreate with the flag ::CUDA_ARRAY3D_SPARSE
or ::CUDA_ARRAY3D_DEFERRED_MAPPING.
For CUDA arrays obtained using ::cuMipmappedArrayGetLevel, ::CUDA_ERROR_INVALID_VALUE will be returned.
If ::CUarrayMapInfo::resourceType is set to ::CUresourcetype::CU_RESOURCE_TYPE_MIPMAPPED_ARRAY
then ::CUarrayMapInfo::resource::mipmap must be set to a valid sparse CUDA mipmapped array handle.
The CUDA mipmapped array must be either a 2D, 2D layered or 3D CUDA mipmapped array and must have been
allocated using ::cuMipmappedArrayCreate with the flag ::CUDA_ARRAY3D_SPARSE
or ::CUDA_ARRAY3D_DEFERRED_MAPPING.
::CUarrayMapInfo::subresourceType specifies the type of subresource within the resource.
::CUarraySparseSubresourceType_enum is defined as:
typedef enum CUarraySparseSubresourceType_enum {
CU_ARRAY_SPARSE_SUBRESOURCE_TYPE_SPARSE_LEVEL = 0,
CU_ARRAY_SPARSE_SUBRESOURCE_TYPE_MIPTAIL = 1
} CUarraySparseSubresourceType;
where ::CUarraySparseSubresourceType::CU_ARRAY_SPARSE_SUBRESOURCE_TYPE_SPARSE_LEVEL indicates a
sparse-miplevel which spans at least one tile in every dimension. The remaining miplevels which
are too small to span at least one tile in any dimension constitute the mip tail region as indicated by
::CUarraySparseSubresourceType::CU_ARRAY_SPARSE_SUBRESOURCE_TYPE_MIPTAIL subresource type.
If ::CUarrayMapInfo::subresourceType is set to ::CUarraySparseSubresourceType::CU_ARRAY_SPARSE_SUBRESOURCE_TYPE_SPARSE_LEVEL
then ::CUarrayMapInfo::subresource::sparseLevel struct must contain valid array subregion offsets and extents.
The ::CUarrayMapInfo::subresource::sparseLevel::offsetX, ::CUarrayMapInfo::subresource::sparseLevel::offsetY
and ::CUarrayMapInfo::subresource::sparseLevel::offsetZ must specify valid X, Y and Z offsets respectively.
The ::CUarrayMapInfo::subresource::sparseLevel::extentWidth, ::CUarrayMapInfo::subresource::sparseLevel::extentHeight
and ::CUarrayMapInfo::subresource::sparseLevel::extentDepth must specify valid width, height and depth extents respectively.
These offsets and extents must be aligned to the corresponding tile dimension.
For CUDA mipmapped arrays ::CUarrayMapInfo::subresource::sparseLevel::level must specify a valid mip level index. Otherwise,
must be zero.
For layered CUDA arrays and layered CUDA mipmapped arrays ::CUarrayMapInfo::subresource::sparseLevel::layer must specify a valid layer index. Otherwise,
must be zero.
::CUarrayMapInfo::subresource::sparseLevel::offsetZ must be zero and ::CUarrayMapInfo::subresource::sparseLevel::extentDepth
must be set to 1 for 2D and 2D layered CUDA arrays and CUDA mipmapped arrays.
Tile extents can be obtained by calling ::cuArrayGetSparseProperties and ::cuMipmappedArrayGetSparseProperties
If ::CUarrayMapInfo::subresourceType is set to ::CUarraySparseSubresourceType::CU_ARRAY_SPARSE_SUBRESOURCE_TYPE_MIPTAIL
then ::CUarrayMapInfo::subresource::miptail struct must contain valid mip tail offset in
::CUarrayMapInfo::subresource::miptail::offset and size in ::CUarrayMapInfo::subresource::miptail::size.
Both, mip tail offset and mip tail size must be aligned to the tile size.
For layered CUDA mipmapped arrays which don't have the flag ::CU_ARRAY_SPARSE_PROPERTIES_SINGLE_MIPTAIL set in ::CUDA_ARRAY_SPARSE_PROPERTIES::flags
as returned by ::cuMipmappedArrayGetSparseProperties, ::CUarrayMapInfo::subresource::miptail::layer must specify a valid layer index.
Otherwise, must be zero.
If ::CUarrayMapInfo::resource::array or ::CUarrayMapInfo::resource::mipmap was created with ::CUDA_ARRAY3D_DEFERRED_MAPPING
flag set the ::CUarrayMapInfo::subresourceType and the contents of ::CUarrayMapInfo::subresource will be ignored.
::CUarrayMapInfo::memOperationType specifies the type of operation. ::CUmemOperationType is defined as:
typedef enum CUmemOperationType_enum {
CU_MEM_OPERATION_TYPE_MAP = 1,
CU_MEM_OPERATION_TYPE_UNMAP = 2
} CUmemOperationType;
If ::CUarrayMapInfo::memOperationType is set to ::CUmemOperationType::CU_MEM_OPERATION_TYPE_MAP then the subresource
will be mapped onto the tile pool memory specified by ::CUarrayMapInfo::memHandle at offset ::CUarrayMapInfo::offset.
The tile pool allocation has to be created by specifying the ::CU_MEM_CREATE_USAGE_TILE_POOL flag when calling ::cuMemCreate. Also,
::CUarrayMapInfo::memHandleType must be set to ::CUmemHandleType::CU_MEM_HANDLE_TYPE_GENERIC.
If ::CUarrayMapInfo::memOperationType is set to ::CUmemOperationType::CU_MEM_OPERATION_TYPE_UNMAP then an unmapping operation
is performed. ::CUarrayMapInfo::memHandle must be NULL.
::CUarrayMapInfo::deviceBitMask specifies the list of devices that must map or unmap physical memory.
Currently, this mask must have exactly one bit set, and the corresponding device must match the device associated with the stream.
If ::CUarrayMapInfo::memOperationType is set to ::CUmemOperationType::CU_MEM_OPERATION_TYPE_MAP, the device must also match
the device associated with the tile pool memory allocation as specified by ::CUarrayMapInfo::memHandle.
::CUarrayMapInfo::flags and ::CUarrayMapInfo::reserved[] are unused and must be set to zero.mapInfoList
- [in] - List of ::CUarrayMapInfocount
- [in] - Count of ::CUarrayMapInfo in \p mapInfoListhStream
- [in] - Stream identifier for the stream to use for map or unmap operations::cuMipmappedArrayCreate, ::cuArrayCreate, ::cuArray3DCreate, ::cuMemCreate, ::cuArrayGetSparseProperties, ::cuMipmappedArrayGetSparseProperties
@Cast(value="CUresult") public static int cuMemUnmap(@Cast(value="CUdeviceptr") long ptr, @Cast(value="size_t") long size)
ptr
- [in] - Starting address for the virtual address range to unmapsize
- [in] - Size of the virtual address range to unmap::cuMemCreate, ::cuMemAddressReserve
@Cast(value="CUresult") public static int cuMemSetAccess(@Cast(value="CUdeviceptr") long ptr, @Cast(value="size_t") long size, @Cast(value="const CUmemAccessDesc*") CUmemAccessDesc_v1 desc, @Cast(value="size_t") long count)
ptr
- [in] - Starting address for the virtual address rangesize
- [in] - Length of the virtual address rangedesc
- [in] - Array of ::CUmemAccessDesc that describe how to change the
- mapping for each location specifiedcount
- [in] - Number of ::CUmemAccessDesc in \p desc::cuMemSetAccess, ::cuMemCreate, :cuMemMap
@Cast(value="CUresult") public static int cuMemGetAccess(@Cast(value="unsigned long long*") LongPointer flags, @Cast(value="const CUmemLocation*") CUmemLocation_v1 location, @Cast(value="CUdeviceptr") long ptr)
flags
- [out] - Flags set for this locationlocation
- [in] - Location in which to check the flags forptr
- [in] - Address in which to check the access flags for::cuMemSetAccess
@Cast(value="CUresult") public static int cuMemGetAccess(@Cast(value="unsigned long long*") LongBuffer flags, @Cast(value="const CUmemLocation*") CUmemLocation_v1 location, @Cast(value="CUdeviceptr") long ptr)
@Cast(value="CUresult") public static int cuMemGetAccess(@Cast(value="unsigned long long*") long[] flags, @Cast(value="const CUmemLocation*") CUmemLocation_v1 location, @Cast(value="CUdeviceptr") long ptr)
@Cast(value="CUresult") public static int cuMemExportToShareableHandle(Pointer shareableHandle, @Cast(value="CUmemGenericAllocationHandle") long handle, @Cast(value="CUmemAllocationHandleType") int handleType, @Cast(value="unsigned long long") long flags)
shareableHandle
- [out] - Pointer to the location in which to store the requested handle typehandle
- [in] - CUDA handle for the memory allocationhandleType
- [in] - Type of shareable handle requested (defines type and size of the \p shareableHandle output parameter)flags
- [in] - Reserved, must be zero::cuMemImportFromShareableHandle
@Cast(value="CUresult") public static int cuMemImportFromShareableHandle(@Cast(value="CUmemGenericAllocationHandle*") LongPointer handle, Pointer osHandle, @Cast(value="CUmemAllocationHandleType") int shHandleType)
handle
- [out] - CUDA Memory handle for the memory allocation.osHandle
- [in] - Shareable Handle representing the memory allocation that is to be imported.shHandleType
- [in] - handle type of the exported handle ::CUmemAllocationHandleType.::cuMemExportToShareableHandle, ::cuMemMap, ::cuMemRelease
@Cast(value="CUresult") public static int cuMemImportFromShareableHandle(@Cast(value="CUmemGenericAllocationHandle*") LongBuffer handle, Pointer osHandle, @Cast(value="CUmemAllocationHandleType") int shHandleType)
@Cast(value="CUresult") public static int cuMemImportFromShareableHandle(@Cast(value="CUmemGenericAllocationHandle*") long[] handle, Pointer osHandle, @Cast(value="CUmemAllocationHandleType") int shHandleType)
@Cast(value="CUresult") public static int cuMemGetAllocationGranularity(@Cast(value="size_t*") SizeTPointer granularity, @Cast(value="const CUmemAllocationProp*") CUmemAllocationProp_v1 prop, @Cast(value="CUmemAllocationGranularity_flags") int option)
granularity
- [out] Returned granularity.prop
- [in] Property for which to determine the granularity foroption
- [in] Determines which granularity to return::cuMemCreate, ::cuMemMap
@Cast(value="CUresult") public static int cuMemGetAllocationPropertiesFromHandle(@Cast(value="CUmemAllocationProp*") CUmemAllocationProp_v1 prop, @Cast(value="CUmemGenericAllocationHandle") long handle)
prop
- [out] - Pointer to a properties structure which will hold the information about this handlehandle
- [in] - Handle which to perform the query on::cuMemCreate, ::cuMemImportFromShareableHandle
@Cast(value="CUresult") public static int cuMemFreeAsync(@Cast(value="CUdeviceptr") long dptr, CUstream_st hStream)
dptr
- - memory to freehStream
- - The stream establishing the stream ordering contract.@Cast(value="CUresult") public static int cuMemAllocAsync(@Cast(value="CUdeviceptr*") LongPointer dptr, @Cast(value="size_t") long bytesize, CUstream_st hStream)
dptr
- [out] - Returned device pointerbytesize
- [in] - Number of bytes to allocatehStream
- [in] - The stream establishing the stream ordering contract and the memory pool to allocate from::cuMemAllocFromPoolAsync, ::cuMemFreeAsync, ::cuDeviceSetMemPool,
::cuDeviceGetDefaultMemPool, ::cuDeviceGetMemPool, ::cuMemPoolCreate,
::cuMemPoolSetAccess, ::cuMemPoolSetAttribute
@Cast(value="CUresult") public static int cuMemAllocAsync(@Cast(value="CUdeviceptr*") LongBuffer dptr, @Cast(value="size_t") long bytesize, CUstream_st hStream)
@Cast(value="CUresult") public static int cuMemAllocAsync(@Cast(value="CUdeviceptr*") long[] dptr, @Cast(value="size_t") long bytesize, CUstream_st hStream)
@Cast(value="CUresult") public static int cuMemPoolTrimTo(CUmemPoolHandle_st pool, @Cast(value="size_t") long minBytesToKeep)
pool
- [in] - The memory pool to trimminBytesToKeep
- [in] - If the pool has less than minBytesToKeep reserved,
the TrimTo operation is a no-op. Otherwise the pool will be guaranteed to have
at least minBytesToKeep bytes reserved after the operation.::cuMemAllocAsync, ::cuMemFreeAsync, ::cuDeviceGetDefaultMemPool,
::cuDeviceGetMemPool, ::cuMemPoolCreate
@Cast(value="CUresult") public static int cuMemPoolSetAttribute(CUmemPoolHandle_st pool, @Cast(value="CUmemPool_attribute") int attr, Pointer value)
pool
- [in] - The memory pool to modifyattr
- [in] - The attribute to modifyvalue
- [in] - Pointer to the value to assign::cuMemAllocAsync, ::cuMemFreeAsync, ::cuDeviceGetDefaultMemPool,
::cuDeviceGetMemPool, ::cuMemPoolCreate
@Cast(value="CUresult") public static int cuMemPoolGetAttribute(CUmemPoolHandle_st pool, @Cast(value="CUmemPool_attribute") int attr, Pointer value)
pool
- [in] - The memory pool to get attributes ofattr
- [in] - The attribute to getvalue
- [out] - Retrieved value::cuMemAllocAsync, ::cuMemFreeAsync, ::cuDeviceGetDefaultMemPool,
::cuDeviceGetMemPool, ::cuMemPoolCreate
@Cast(value="CUresult") public static int cuMemPoolSetAccess(CUmemPoolHandle_st pool, @Cast(value="const CUmemAccessDesc*") CUmemAccessDesc_v1 map, @Cast(value="size_t") long count)
pool
- [in] - The pool being modifiedmap
- [in] - Array of access descriptors. Each descriptor instructs the access to enable for a single gpu.count
- [in] - Number of descriptors in the map array.::cuMemAllocAsync, ::cuMemFreeAsync, ::cuDeviceGetDefaultMemPool,
::cuDeviceGetMemPool, ::cuMemPoolCreate
@Cast(value="CUresult") public static int cuMemPoolGetAccess(@Cast(value="CUmemAccess_flags*") IntPointer flags, CUmemPoolHandle_st memPool, @Cast(value="CUmemLocation*") CUmemLocation_v1 location)
flags
- [out] - the accessibility of the pool from the specified locationmemPool
- [in] - the pool being queriedlocation
- [in] - the location accessing the pool::cuMemAllocAsync, ::cuMemFreeAsync, ::cuDeviceGetDefaultMemPool,
::cuDeviceGetMemPool, ::cuMemPoolCreate
@Cast(value="CUresult") public static int cuMemPoolGetAccess(@Cast(value="CUmemAccess_flags*") IntBuffer flags, CUmemPoolHandle_st memPool, @Cast(value="CUmemLocation*") CUmemLocation_v1 location)
@Cast(value="CUresult") public static int cuMemPoolGetAccess(@Cast(value="CUmemAccess_flags*") int[] flags, CUmemPoolHandle_st memPool, @Cast(value="CUmemLocation*") CUmemLocation_v1 location)
@Cast(value="CUresult") public static int cuMemPoolCreate(@ByPtrPtr CUmemPoolHandle_st pool, @Cast(value="const CUmemPoolProps*") CUmemPoolProps_v1 poolProps)
::cuDeviceSetMemPool, ::cuDeviceGetMemPool, ::cuDeviceGetDefaultMemPool,
::cuMemAllocFromPoolAsync, ::cuMemPoolExportToShareableHandle
@Cast(value="CUresult") public static int cuMemPoolDestroy(CUmemPoolHandle_st pool)
::cuMemFreeAsync, ::cuDeviceSetMemPool, ::cuDeviceGetMemPool,
::cuDeviceGetDefaultMemPool, ::cuMemPoolCreate
@Cast(value="CUresult") public static int cuMemAllocFromPoolAsync(@Cast(value="CUdeviceptr*") LongPointer dptr, @Cast(value="size_t") long bytesize, CUmemPoolHandle_st pool, CUstream_st hStream)
dptr
- [out] - Returned device pointerbytesize
- [in] - Number of bytes to allocatepool
- [in] - The pool to allocate fromhStream
- [in] - The stream establishing the stream ordering semantic::cuMemAllocAsync, ::cuMemFreeAsync, ::cuDeviceGetDefaultMemPool,
::cuDeviceGetMemPool, ::cuMemPoolCreate, ::cuMemPoolSetAccess,
::cuMemPoolSetAttribute
@Cast(value="CUresult") public static int cuMemAllocFromPoolAsync(@Cast(value="CUdeviceptr*") LongBuffer dptr, @Cast(value="size_t") long bytesize, CUmemPoolHandle_st pool, CUstream_st hStream)
@Cast(value="CUresult") public static int cuMemAllocFromPoolAsync(@Cast(value="CUdeviceptr*") long[] dptr, @Cast(value="size_t") long bytesize, CUmemPoolHandle_st pool, CUstream_st hStream)
@Cast(value="CUresult") public static int cuMemPoolExportToShareableHandle(Pointer handle_out, CUmemPoolHandle_st pool, @Cast(value="CUmemAllocationHandleType") int handleType, @Cast(value="unsigned long long") long flags)
handle_out
- [out] - Returned OS handlepool
- [in] - pool to exporthandleType
- [in] - the type of handle to createflags
- [in] - must be 0::cuMemPoolImportFromShareableHandle, ::cuMemPoolExportPointer,
::cuMemPoolImportPointer, ::cuMemAllocAsync, ::cuMemFreeAsync,
::cuDeviceGetDefaultMemPool, ::cuDeviceGetMemPool, ::cuMemPoolCreate,
::cuMemPoolSetAccess, ::cuMemPoolSetAttribute
@Cast(value="CUresult") public static int cuMemPoolImportFromShareableHandle(@ByPtrPtr CUmemPoolHandle_st pool_out, Pointer handle, @Cast(value="CUmemAllocationHandleType") int handleType, @Cast(value="unsigned long long") long flags)
pool_out
- [out] - Returned memory poolhandle
- [in] - OS handle of the pool to openhandleType
- [in] - The type of handle being importedflags
- [in] - must be 0::cuMemPoolExportToShareableHandle, ::cuMemPoolExportPointer, ::cuMemPoolImportPointer
@Cast(value="CUresult") public static int cuMemPoolExportPointer(@Cast(value="CUmemPoolPtrExportData*") CUmemPoolPtrExportData_v1 shareData_out, @Cast(value="CUdeviceptr") long ptr)
shareData_out
- [out] - Returned export dataptr
- [in] - pointer to memory being exported::cuMemPoolExportToShareableHandle, ::cuMemPoolImportFromShareableHandle, ::cuMemPoolImportPointer
@Cast(value="CUresult") public static int cuMemPoolImportPointer(@Cast(value="CUdeviceptr*") LongPointer ptr_out, CUmemPoolHandle_st pool, @Cast(value="CUmemPoolPtrExportData*") CUmemPoolPtrExportData_v1 shareData)
ptr_out
- [out] - pointer to imported memorypool
- [in] - pool from which to importshareData
- [in] - data specifying the memory to import::cuMemPoolExportToShareableHandle, ::cuMemPoolImportFromShareableHandle, ::cuMemPoolExportPointer
@Cast(value="CUresult") public static int cuMemPoolImportPointer(@Cast(value="CUdeviceptr*") LongBuffer ptr_out, CUmemPoolHandle_st pool, @Cast(value="CUmemPoolPtrExportData*") CUmemPoolPtrExportData_v1 shareData)
@Cast(value="CUresult") public static int cuMemPoolImportPointer(@Cast(value="CUdeviceptr*") long[] ptr_out, CUmemPoolHandle_st pool, @Cast(value="CUmemPoolPtrExportData*") CUmemPoolPtrExportData_v1 shareData)
@Cast(value="CUresult") public static int cuMulticastCreate(@Cast(value="CUmemGenericAllocationHandle*") LongPointer mcHandle, @Cast(value="const CUmulticastObjectProp*") CUmulticastObjectProp_v1 prop)
mcHandle
- [out] Value of handle returned.prop
- [in] Properties of the multicast object to create.::cuMulticastAddDevice, ::cuMulticastBindMem, ::cuMulticastBindAddr, ::cuMulticastUnbind
,
::cuMemCreate, ::cuMemRelease, ::cuMemExportToShareableHandle, ::cuMemImportFromShareableHandle
@Cast(value="CUresult") public static int cuMulticastCreate(@Cast(value="CUmemGenericAllocationHandle*") LongBuffer mcHandle, @Cast(value="const CUmulticastObjectProp*") CUmulticastObjectProp_v1 prop)
@Cast(value="CUresult") public static int cuMulticastCreate(@Cast(value="CUmemGenericAllocationHandle*") long[] mcHandle, @Cast(value="const CUmulticastObjectProp*") CUmulticastObjectProp_v1 prop)
@Cast(value="CUresult") public static int cuMulticastAddDevice(@Cast(value="CUmemGenericAllocationHandle") long mcHandle, @Cast(value="CUdevice") int dev)
mcHandle
- [in] Handle representing a multicast object.dev
- [in] Device that will be associated to the multicast
object.::cuMulticastCreate, ::cuMulticastBindMem, ::cuMulticastBindAddr
@Cast(value="CUresult") public static int cuMulticastBindMem(@Cast(value="CUmemGenericAllocationHandle") long mcHandle, @Cast(value="size_t") long mcOffset, @Cast(value="CUmemGenericAllocationHandle") long memHandle, @Cast(value="size_t") long memOffset, @Cast(value="size_t") long size, @Cast(value="unsigned long long") long flags)
mcHandle
- [in] Handle representing a multicast object.mcOffset
- [in] Offset into the multicast object for attachment.memHandle
- [in] Handle representing a memory allocation.memOffset
- [in] Offset into the memory for attachment.size
- [in] Size of the memory that will be bound to the
multicast object.flags
- [in] Flags for future use, must be zero for now.::cuMulticastCreate, ::cuMulticastAddDevice, ::cuMemCreate
@Cast(value="CUresult") public static int cuMulticastBindAddr(@Cast(value="CUmemGenericAllocationHandle") long mcHandle, @Cast(value="size_t") long mcOffset, @Cast(value="CUdeviceptr") long memptr, @Cast(value="size_t") long size, @Cast(value="unsigned long long") long flags)
mcHandle
- [in] Handle representing a multicast object.mcOffset
- [in] Offset into multicast va range for attachment.memptr
- [in] Virtual address of the memory allocation.size
- [in] Size of memory that will be bound to the
multicast object.flags
- [in] Flags for future use, must be zero now.::cuMulticastCreate, ::cuMulticastAddDevice, ::cuMemCreate
@Cast(value="CUresult") public static int cuMulticastUnbind(@Cast(value="CUmemGenericAllocationHandle") long mcHandle, @Cast(value="CUdevice") int dev, @Cast(value="size_t") long mcOffset, @Cast(value="size_t") long size)
mcHandle
- [in] Handle representing a multicast object.dev
- [in] Device that hosts the memory allocation.mcOffset
- [in] Offset into the multicast object.size
- [in] Desired size to unbind.::cuMulticastBindMem, ::cuMulticastBindAddr
@Cast(value="CUresult") public static int cuMulticastGetGranularity(@Cast(value="size_t*") SizeTPointer granularity, @Cast(value="const CUmulticastObjectProp*") CUmulticastObjectProp_v1 prop, @Cast(value="CUmulticastGranularity_flags") int option)
granularity
- [out] Returned granularity.prop
- [in] Properties of the multicast object.option
- [in] Determines which granularity to return.::cuMulticastCreate, ::cuMulticastBindMem, ::cuMulticastBindAddr, ::cuMulticastUnbind
@Cast(value="CUresult") public static int cuPointerGetAttribute(Pointer data, @Cast(value="CUpointer_attribute") int attribute, @Cast(value="CUdeviceptr") long ptr)
data
- - Returned pointer attribute valueattribute
- - Pointer attribute to queryptr
- - Pointer::cuPointerSetAttribute,
::cuMemAlloc,
::cuMemFree,
::cuMemAllocHost,
::cuMemFreeHost,
::cuMemHostAlloc,
::cuMemHostRegister,
::cuMemHostUnregister,
::cudaPointerGetAttributes
@Cast(value="CUresult") public static int cuMemPrefetchAsync(@Cast(value="CUdeviceptr") long devPtr, @Cast(value="size_t") long count, @Cast(value="CUdevice") int dstDevice, CUstream_st hStream)
devPtr
- - Pointer to be prefetchedcount
- - Size in bytesdstDevice
- - Destination device to prefetch tohStream
- - Stream to enqueue prefetch operation::cuMemcpy, ::cuMemcpyPeer, ::cuMemcpyAsync,
::cuMemcpy3DPeerAsync, ::cuMemAdvise, ::cuMemPrefetchAsync
::cudaMemPrefetchAsync_v2
@Cast(value="CUresult") public static int cuMemPrefetchAsync_v2(@Cast(value="CUdeviceptr") long devPtr, @Cast(value="size_t") long count, @ByVal @Cast(value="CUmemLocation*") CUmemLocation_v1 location, @Cast(value="unsigned int") int flags, CUstream_st hStream)
devPtr
- - Pointer to be prefetchedcount
- - Size in bytesdstDevice
- - Destination device to prefetch toflags
- - flags for future use, must be zero now.hStream
- - Stream to enqueue prefetch operation::cuMemcpy, ::cuMemcpyPeer, ::cuMemcpyAsync,
::cuMemcpy3DPeerAsync, ::cuMemAdvise, ::cuMemPrefetchAsync
::cudaMemPrefetchAsync_v2
@Cast(value="CUresult") public static int cuMemAdvise(@Cast(value="CUdeviceptr") long devPtr, @Cast(value="size_t") long count, @Cast(value="CUmem_advise") int advice, @Cast(value="CUdevice") int device)
devPtr
- - Pointer to memory to set the advice forcount
- - Size in bytes of the memory rangeadvice
- - Advice to be applied for the specified memory rangedevice
- - Device to apply the advice for::cuMemcpy, ::cuMemcpyPeer, ::cuMemcpyAsync,
::cuMemcpy3DPeerAsync, ::cuMemPrefetchAsync, ::cuMemAdvise_v2
::cudaMemAdvise
@Cast(value="CUresult") public static int cuMemAdvise_v2(@Cast(value="CUdeviceptr") long devPtr, @Cast(value="size_t") long count, @Cast(value="CUmem_advise") int advice, @ByVal @Cast(value="CUmemLocation*") CUmemLocation_v1 location)
devPtr
- - Pointer to memory to set the advice forcount
- - Size in bytes of the memory rangeadvice
- - Advice to be applied for the specified memory rangelocation
- - location to apply the advice for::cuMemcpy, ::cuMemcpyPeer, ::cuMemcpyAsync,
::cuMemcpy3DPeerAsync, ::cuMemPrefetchAsync, ::cuMemAdvise
::cudaMemAdvise
@Cast(value="CUresult") public static int cuPointerSetAttribute(@Const Pointer value, @Cast(value="CUpointer_attribute") int attribute, @Cast(value="CUdeviceptr") long ptr)
value
- - Pointer to memory containing the value to be setattribute
- - Pointer attribute to setptr
- - Pointer to a memory region allocated using CUDA memory allocation APIs::cuPointerGetAttribute,
::cuPointerGetAttributes,
::cuMemAlloc,
::cuMemFree,
::cuMemAllocHost,
::cuMemFreeHost,
::cuMemHostAlloc,
::cuMemHostRegister,
::cuMemHostUnregister
@Cast(value="CUresult") public static int cuPointerGetAttributes(@Cast(value="unsigned int") int numAttributes, @Cast(value="CUpointer_attribute*") IntPointer attributes, @Cast(value="void**") PointerPointer data, @Cast(value="CUdeviceptr") long ptr)
numAttributes
- - Number of attributes to queryattributes
- - An array of attributes to query
(numAttributes and the number of attributes in this array should match)data
- - A two-dimensional array containing pointers to memory
locations where the result of each attribute query will be written to.ptr
- - Pointer to query
Unlike ::cuPointerGetAttribute, this function will not return an error when the \p ptr
encountered is not a valid CUDA pointer. Instead, the attributes are assigned default NULL values
and CUDA_SUCCESS is returned.
If \p ptr was not allocated by, mapped by, or registered with a ::CUcontext which uses UVA
(Unified Virtual Addressing), ::CUDA_ERROR_INVALID_CONTEXT is returned.::cuPointerGetAttribute,
::cuPointerSetAttribute,
::cudaPointerGetAttributes
@Cast(value="CUresult") public static int cuPointerGetAttributes(@Cast(value="unsigned int") int numAttributes, @Cast(value="CUpointer_attribute*") IntPointer attributes, @Cast(value="void**") @ByPtrPtr Pointer data, @Cast(value="CUdeviceptr") long ptr)
@Cast(value="CUresult") public static int cuPointerGetAttributes(@Cast(value="unsigned int") int numAttributes, @Cast(value="CUpointer_attribute*") IntBuffer attributes, @Cast(value="void**") @ByPtrPtr Pointer data, @Cast(value="CUdeviceptr") long ptr)
@Cast(value="CUresult") public static int cuPointerGetAttributes(@Cast(value="unsigned int") int numAttributes, @Cast(value="CUpointer_attribute*") int[] attributes, @Cast(value="void**") @ByPtrPtr Pointer data, @Cast(value="CUdeviceptr") long ptr)
@Cast(value="CUresult") public static int cuStreamCreate(@ByPtrPtr CUstream_st phStream, @Cast(value="unsigned int") int Flags)
phStream
- - Returned newly created streamFlags
- - Parameters for stream creation::cuStreamDestroy,
::cuStreamCreateWithPriority,
::cuStreamGetPriority,
::cuStreamGetFlags,
::cuStreamWaitEvent,
::cuStreamQuery,
::cuStreamSynchronize,
::cuStreamAddCallback,
::cudaStreamCreate,
::cudaStreamCreateWithFlags
@Cast(value="CUresult") public static int cuStreamCreateWithPriority(@ByPtrPtr CUstream_st phStream, @Cast(value="unsigned int") int flags, int priority)
phStream
- - Returned newly created streamflags
- - Flags for stream creation. See ::cuStreamCreate for a list of
valid flagspriority
- - Stream priority. Lower numbers represent higher priorities.
See ::cuCtxGetStreamPriorityRange for more information about
meaningful stream priorities that can be passed.::cuStreamDestroy,
::cuStreamCreate,
::cuStreamGetPriority,
::cuCtxGetStreamPriorityRange,
::cuStreamGetFlags,
::cuStreamWaitEvent,
::cuStreamQuery,
::cuStreamSynchronize,
::cuStreamAddCallback,
::cudaStreamCreateWithPriority
@Cast(value="CUresult") public static int cuStreamGetPriority(CUstream_st hStream, IntPointer priority)
hStream
- - Handle to the stream to be queriedpriority
- - Pointer to a signed integer in which the stream's priority is returned::cuStreamDestroy,
::cuStreamCreate,
::cuStreamCreateWithPriority,
::cuCtxGetStreamPriorityRange,
::cuStreamGetFlags,
::cudaStreamGetPriority
@Cast(value="CUresult") public static int cuStreamGetPriority(CUstream_st hStream, IntBuffer priority)
@Cast(value="CUresult") public static int cuStreamGetPriority(CUstream_st hStream, int[] priority)
@Cast(value="CUresult") public static int cuStreamGetFlags(CUstream_st hStream, @Cast(value="unsigned int*") IntPointer flags)
hStream
- - Handle to the stream to be queriedflags
- - Pointer to an unsigned integer in which the stream's flags are returned
The value returned in \p flags is a logical 'OR' of all flags that
were used while creating this stream. See ::cuStreamCreate for the list
of valid flags::cuStreamDestroy,
::cuStreamCreate,
::cuStreamGetPriority,
::cudaStreamGetFlags
@Cast(value="CUresult") public static int cuStreamGetFlags(CUstream_st hStream, @Cast(value="unsigned int*") IntBuffer flags)
@Cast(value="CUresult") public static int cuStreamGetFlags(CUstream_st hStream, @Cast(value="unsigned int*") int[] flags)
@Cast(value="CUresult") public static int cuStreamGetId(CUstream_st hStream, @Cast(value="unsigned long long*") LongPointer streamId)
hStream
- - Handle to the stream to be queriedstreamId
- - Pointer to store the Id of the stream::cuStreamDestroy,
::cuStreamCreate,
::cuStreamGetPriority,
::cudaStreamGetId
@Cast(value="CUresult") public static int cuStreamGetId(CUstream_st hStream, @Cast(value="unsigned long long*") LongBuffer streamId)
@Cast(value="CUresult") public static int cuStreamGetId(CUstream_st hStream, @Cast(value="unsigned long long*") long[] streamId)
@Cast(value="CUresult") public static int cuStreamGetCtx(CUstream_st hStream, @ByPtrPtr CUctx_st pctx)
hStream
- - Handle to the stream to be queriedpctx
- - Returned context associated with the stream::cuStreamDestroy,
::cuStreamCreateWithPriority,
::cuStreamGetPriority,
::cuStreamGetFlags,
::cuStreamWaitEvent,
::cuStreamQuery,
::cuStreamSynchronize,
::cuStreamAddCallback,
::cudaStreamCreate,
::cudaStreamCreateWithFlags
@Cast(value="CUresult") public static int cuStreamWaitEvent(CUstream_st hStream, CUevent_st hEvent, @Cast(value="unsigned int") int Flags)
hStream
- - Stream to waithEvent
- - Event to wait on (may not be NULL)Flags
- - See ::CUevent_capture_flags::cuStreamCreate,
::cuEventRecord,
::cuStreamQuery,
::cuStreamSynchronize,
::cuStreamAddCallback,
::cuStreamDestroy,
::cudaStreamWaitEvent
@Cast(value="CUresult") public static int cuStreamAddCallback(CUstream_st hStream, CUstreamCallback callback, Pointer userData, @Cast(value="unsigned int") int flags)
hStream
- - Stream to add callback tocallback
- - The function to call once preceding stream operations are completeuserData
- - User specified data to be passed to the callback functionflags
- - Reserved for future use, must be 0::cuStreamCreate,
::cuStreamQuery,
::cuStreamSynchronize,
::cuStreamWaitEvent,
::cuStreamDestroy,
::cuMemAllocManaged,
::cuStreamAttachMemAsync,
::cuLaunchHostFunc,
::cudaStreamAddCallback
@Cast(value="CUresult") public static int cuStreamBeginCapture(CUstream_st hStream, @Cast(value="CUstreamCaptureMode") int mode)
hStream
- - Stream in which to initiate capturemode
- - Controls the interaction of this capture sequence with other API
calls that are potentially unsafe. For more details see
::cuThreadExchangeStreamCaptureMode.
\note Kernels captured using this API must not use texture and surface references.
Reading or writing through any texture or surface reference is undefined
behavior. This restriction does not apply to texture and surface objects.::cuStreamCreate,
::cuStreamIsCapturing,
::cuStreamEndCapture,
::cuThreadExchangeStreamCaptureMode
@Cast(value="CUresult") public static int cuStreamBeginCaptureToGraph(CUstream_st hStream, CUgraph_st hGraph, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st dependencies, @Const CUgraphEdgeData dependencyData, @Cast(value="size_t") long numDependencies, @Cast(value="CUstreamCaptureMode") int mode)
hStream
- - Stream in which to initiate capture.hGraph
- - Graph to capture into.dependencies
- - Dependencies of the first node captured in the stream. Can be NULL if numDependencies is 0.dependencyData
- - Optional array of data associated with each dependency.numDependencies
- - Number of dependencies.mode
- - Controls the interaction of this capture sequence with other API
calls that are potentially unsafe. For more details see
::cuThreadExchangeStreamCaptureMode.
\note Kernels captured using this API must not use texture and surface references.
Reading or writing through any texture or surface reference is undefined
behavior. This restriction does not apply to texture and surface objects.::cuStreamBeginCapture,
::cuStreamCreate,
::cuStreamIsCapturing,
::cuStreamEndCapture,
::cuThreadExchangeStreamCaptureMode,
::cuGraphAddNode,
@Cast(value="CUresult") public static int cuThreadExchangeStreamCaptureMode(@Cast(value="CUstreamCaptureMode*") IntPointer mode)
CUstreamCaptureMode mode = desiredMode;
cuThreadExchangeStreamCaptureMode(&mode);
...
cuThreadExchangeStreamCaptureMode(&mode); // restore previous mode
During stream capture (see ::cuStreamBeginCapture), some actions, such as a call
to ::cudaMalloc, may be unsafe. In the case of ::cudaMalloc, the operation is
not enqueued asynchronously to a stream, and is not observed by stream capture.
Therefore, if the sequence of operations captured via ::cuStreamBeginCapture
depended on the allocation being replayed whenever the graph is launched, the
captured graph would be invalid.
Therefore, stream capture places restrictions on API calls that can be made within
or concurrently to a ::cuStreamBeginCapture-::cuStreamEndCapture sequence. This
behavior can be controlled via this API and flags to ::cuStreamBeginCapture.
A thread's mode is one of the following:
- \p CU_STREAM_CAPTURE_MODE_GLOBAL: This is the default mode. If the local thread has
an ongoing capture sequence that was not initiated with
\p CU_STREAM_CAPTURE_MODE_RELAXED at \p cuStreamBeginCapture, or if any other thread
has a concurrent capture sequence initiated with \p CU_STREAM_CAPTURE_MODE_GLOBAL,
this thread is prohibited from potentially unsafe API calls.
- \p CU_STREAM_CAPTURE_MODE_THREAD_LOCAL: If the local thread has an ongoing capture
sequence not initiated with \p CU_STREAM_CAPTURE_MODE_RELAXED, it is prohibited
from potentially unsafe API calls. Concurrent capture sequences in other threads
are ignored.
- \p CU_STREAM_CAPTURE_MODE_RELAXED: The local thread is not prohibited from potentially
unsafe API calls. Note that the thread is still prohibited from API calls which
necessarily conflict with stream capture, for example, attempting ::cuEventQuery
on an event that was last recorded inside a capture sequence.mode
- - Pointer to mode value to swap with the current mode::cuStreamBeginCapture
@Cast(value="CUresult") public static int cuThreadExchangeStreamCaptureMode(@Cast(value="CUstreamCaptureMode*") IntBuffer mode)
@Cast(value="CUresult") public static int cuThreadExchangeStreamCaptureMode(@Cast(value="CUstreamCaptureMode*") int[] mode)
@Cast(value="CUresult") public static int cuStreamEndCapture(CUstream_st hStream, @ByPtrPtr CUgraph_st phGraph)
hStream
- - Stream to queryphGraph
- - The captured graph::cuStreamCreate,
::cuStreamBeginCapture,
::cuStreamIsCapturing,
::cuGraphDestroy
@Cast(value="CUresult") public static int cuStreamIsCapturing(CUstream_st hStream, @Cast(value="CUstreamCaptureStatus*") IntPointer captureStatus)
hStream
- - Stream to querycaptureStatus
- - Returns the stream's capture status::cuStreamCreate,
::cuStreamBeginCapture,
::cuStreamEndCapture
@Cast(value="CUresult") public static int cuStreamIsCapturing(CUstream_st hStream, @Cast(value="CUstreamCaptureStatus*") IntBuffer captureStatus)
@Cast(value="CUresult") public static int cuStreamIsCapturing(CUstream_st hStream, @Cast(value="CUstreamCaptureStatus*") int[] captureStatus)
@Cast(value="CUresult") public static int cuStreamGetCaptureInfo(CUstream_st hStream, @Cast(value="CUstreamCaptureStatus*") IntPointer captureStatus_out, @Cast(value="cuuint64_t*") LongPointer id_out, @ByPtrPtr CUgraph_st graph_out, @Cast(value="const CUgraphNode**") @ByPtrPtr PointerPointer dependencies_out, @Cast(value="size_t*") SizeTPointer numDependencies_out)
hStream
- - The stream to querycaptureStatus_out
- - Location to return the capture status of the stream; requiredid_out
- - Optional location to return an id for the capture sequence, which is
unique over the lifetime of the processgraph_out
- - Optional location to return the graph being captured into. All
operations other than destroy and node removal are permitted on the graph
while the capture sequence is in progress. This API does not transfer
ownership of the graph, which is transferred or destroyed at
::cuStreamEndCapture. Note that the graph handle may be invalidated before
end of capture for certain errors. Nodes that are or become
unreachable from the original stream at ::cuStreamEndCapture due to direct
actions on the graph do not trigger ::CUDA_ERROR_STREAM_CAPTURE_UNJOINED.dependencies_out
- - Optional location to store a pointer to an array of nodes.
The next node to be captured in the stream will depend on this set of nodes,
absent operations such as event wait which modify this set. The array pointer
is valid until the next API call which operates on the stream or until the
capture is terminated. The node handles may be copied out and are valid until
they or the graph is destroyed. The driver-owned array may also be passed
directly to APIs that operate on the graph (not the stream) without copying.numDependencies_out
- - Optional location to store the size of the array
returned in dependencies_out.::cuStreamBeginCapture,
::cuStreamIsCapturing,
::cuStreamUpdateCaptureDependencies
@Cast(value="CUresult") public static int cuStreamGetCaptureInfo(CUstream_st hStream, @Cast(value="CUstreamCaptureStatus*") IntBuffer captureStatus_out, @Cast(value="cuuint64_t*") LongBuffer id_out, @ByPtrPtr CUgraph_st graph_out, @Cast(value="const CUgraphNode**") @ByPtrPtr PointerPointer dependencies_out, @Cast(value="size_t*") SizeTPointer numDependencies_out)
@Cast(value="CUresult") public static int cuStreamGetCaptureInfo(CUstream_st hStream, @Cast(value="CUstreamCaptureStatus*") int[] captureStatus_out, @Cast(value="cuuint64_t*") long[] id_out, @ByPtrPtr CUgraph_st graph_out, @Cast(value="const CUgraphNode**") @ByPtrPtr PointerPointer dependencies_out, @Cast(value="size_t*") SizeTPointer numDependencies_out)
@Cast(value="CUresult") public static int cuStreamGetCaptureInfo_v3(CUstream_st hStream, @Cast(value="CUstreamCaptureStatus*") IntPointer captureStatus_out, @Cast(value="cuuint64_t*") LongPointer id_out, @ByPtrPtr CUgraph_st graph_out, @Cast(value="const CUgraphNode**") @ByPtrPtr PointerPointer dependencies_out, @Cast(value="const CUgraphEdgeData**") PointerPointer edgeData_out, @Cast(value="size_t*") SizeTPointer numDependencies_out)
hStream
- - The stream to querycaptureStatus_out
- - Location to return the capture status of the stream; requiredid_out
- - Optional location to return an id for the capture sequence, which is
unique over the lifetime of the processgraph_out
- - Optional location to return the graph being captured into. All
operations other than destroy and node removal are permitted on the graph
while the capture sequence is in progress. This API does not transfer
ownership of the graph, which is transferred or destroyed at
::cuStreamEndCapture. Note that the graph handle may be invalidated before
end of capture for certain errors. Nodes that are or become
unreachable from the original stream at ::cuStreamEndCapture due to direct
actions on the graph do not trigger ::CUDA_ERROR_STREAM_CAPTURE_UNJOINED.dependencies_out
- - Optional location to store a pointer to an array of nodes.
The next node to be captured in the stream will depend on this set of nodes,
absent operations such as event wait which modify this set. The array pointer
is valid until the next API call which operates on the stream or until the
capture is terminated. The node handles may be copied out and are valid until
they or the graph is destroyed. The driver-owned array may also be passed
directly to APIs that operate on the graph (not the stream) without copying.edgeData_out
- - Optional location to store a pointer to an array of graph edge
data. This array parallels \c dependencies_out; the next node to be added
has an edge to \c dependencies_out[i] with annotation \c edgeData_out[i] for
each \c i. The array pointer is valid until the next API call which operates
on the stream or until the capture is terminated.numDependencies_out
- - Optional location to store the size of the array
returned in dependencies_out.::cuStreamBeginCapture,
::cuStreamIsCapturing,
::cuStreamUpdateCaptureDependencies
@Cast(value="CUresult") public static int cuStreamGetCaptureInfo_v3(CUstream_st hStream, @Cast(value="CUstreamCaptureStatus*") IntPointer captureStatus_out, @Cast(value="cuuint64_t*") LongPointer id_out, @ByPtrPtr CUgraph_st graph_out, @Cast(value="const CUgraphNode**") @ByPtrPtr PointerPointer dependencies_out, @Const @ByPtrPtr CUgraphEdgeData edgeData_out, @Cast(value="size_t*") SizeTPointer numDependencies_out)
@Cast(value="CUresult") public static int cuStreamGetCaptureInfo_v3(CUstream_st hStream, @Cast(value="CUstreamCaptureStatus*") IntBuffer captureStatus_out, @Cast(value="cuuint64_t*") LongBuffer id_out, @ByPtrPtr CUgraph_st graph_out, @Cast(value="const CUgraphNode**") @ByPtrPtr PointerPointer dependencies_out, @Const @ByPtrPtr CUgraphEdgeData edgeData_out, @Cast(value="size_t*") SizeTPointer numDependencies_out)
@Cast(value="CUresult") public static int cuStreamGetCaptureInfo_v3(CUstream_st hStream, @Cast(value="CUstreamCaptureStatus*") int[] captureStatus_out, @Cast(value="cuuint64_t*") long[] id_out, @ByPtrPtr CUgraph_st graph_out, @Cast(value="const CUgraphNode**") @ByPtrPtr PointerPointer dependencies_out, @Const @ByPtrPtr CUgraphEdgeData edgeData_out, @Cast(value="size_t*") SizeTPointer numDependencies_out)
@Cast(value="CUresult") public static int cuStreamUpdateCaptureDependencies(CUstream_st hStream, @ByPtrPtr CUgraphNode_st dependencies, @Cast(value="size_t") long numDependencies, @Cast(value="unsigned int") int flags)
hStream
- - The stream to updatedependencies
- - The set of dependencies to addnumDependencies
- - The size of the dependencies arrayflags
- - See above::cuStreamBeginCapture,
::cuStreamGetCaptureInfo,
@Cast(value="CUresult") public static int cuStreamUpdateCaptureDependencies_v2(CUstream_st hStream, @ByPtrPtr CUgraphNode_st dependencies, @Const CUgraphEdgeData dependencyData, @Cast(value="size_t") long numDependencies, @Cast(value="unsigned int") int flags)
hStream
- - The stream to updatedependencies
- - The set of dependencies to adddependencyData
- - Optional array of data associated with each dependency.numDependencies
- - The size of the dependencies arrayflags
- - See above::cuStreamBeginCapture,
::cuStreamGetCaptureInfo,
@Cast(value="CUresult") public static int cuStreamAttachMemAsync(CUstream_st hStream, @Cast(value="CUdeviceptr") long dptr, @Cast(value="size_t") long length, @Cast(value="unsigned int") int flags)
hStream
- - Stream in which to enqueue the attach operationdptr
- - Pointer to memory (must be a pointer to managed memory or
to a valid host-accessible region of system-allocated
pageable memory)length
- - Length of memoryflags
- - Must be one of ::CUmemAttach_flags::cuStreamCreate,
::cuStreamQuery,
::cuStreamSynchronize,
::cuStreamWaitEvent,
::cuStreamDestroy,
::cuMemAllocManaged,
::cudaStreamAttachMemAsync
@Cast(value="CUresult") public static int cuStreamQuery(CUstream_st hStream)
hStream
- - Stream to query status of::cuStreamCreate,
::cuStreamWaitEvent,
::cuStreamDestroy,
::cuStreamSynchronize,
::cuStreamAddCallback,
::cudaStreamQuery
@Cast(value="CUresult") public static int cuStreamSynchronize(CUstream_st hStream)
hStream
- - Stream to wait for\note_null_stream \notefnerr
::cuStreamCreate,
::cuStreamDestroy,
::cuStreamWaitEvent,
::cuStreamQuery,
::cuStreamAddCallback,
::cudaStreamSynchronize
@Cast(value="CUresult") public static int cuStreamDestroy(CUstream_st hStream)
hStream
- - Stream to destroy::cuStreamCreate,
::cuStreamWaitEvent,
::cuStreamQuery,
::cuStreamSynchronize,
::cuStreamAddCallback,
::cudaStreamDestroy
@Cast(value="CUresult") public static int cuStreamCopyAttributes(CUstream_st dst, CUstream_st src)
dst
- [out] Destination streamsrc
- [in] Source stream
For list of attributes see ::CUstreamAttrID::CUaccessPolicyWindow
@Cast(value="CUresult") public static int cuStreamGetAttribute(CUstream_st hStream, @Cast(value="CUstreamAttrID") int attr, @Cast(value="CUstreamAttrValue*") CUlaunchAttributeValue value_out)
hStream
- [in]attr
- [in]value_out
- [out]::CUaccessPolicyWindow
@Cast(value="CUresult") public static int cuStreamSetAttribute(CUstream_st hStream, @Cast(value="CUstreamAttrID") int attr, @Cast(value="const CUstreamAttrValue*") CUlaunchAttributeValue value)
hStream
- [out]attr
- [in]value
- [in]::CUaccessPolicyWindow
@Cast(value="CUresult") public static int cuEventCreate(@ByPtrPtr CUevent_st phEvent, @Cast(value="unsigned int") int Flags)
phEvent
- - Returns newly created eventFlags
- - Event creation flags::cuEventRecord,
::cuEventQuery,
::cuEventSynchronize,
::cuEventDestroy,
::cuEventElapsedTime,
::cudaEventCreate,
::cudaEventCreateWithFlags
@Cast(value="CUresult") public static int cuEventRecord(CUevent_st hEvent, CUstream_st hStream)
hEvent
- - Event to recordhStream
- - Stream to record event for::cuEventCreate,
::cuEventQuery,
::cuEventSynchronize,
::cuStreamWaitEvent,
::cuEventDestroy,
::cuEventElapsedTime,
::cudaEventRecord,
::cuEventRecordWithFlags
@Cast(value="CUresult") public static int cuEventRecordWithFlags(CUevent_st hEvent, CUstream_st hStream, @Cast(value="unsigned int") int flags)
hEvent
- - Event to recordhStream
- - Stream to record event forflags
- - See ::CUevent_capture_flags::cuEventCreate,
::cuEventQuery,
::cuEventSynchronize,
::cuStreamWaitEvent,
::cuEventDestroy,
::cuEventElapsedTime,
::cuEventRecord,
::cudaEventRecord
@Cast(value="CUresult") public static int cuEventQuery(CUevent_st hEvent)
hEvent
- - Event to query::cuEventCreate,
::cuEventRecord,
::cuEventSynchronize,
::cuEventDestroy,
::cuEventElapsedTime,
::cudaEventQuery
@Cast(value="CUresult") public static int cuEventSynchronize(CUevent_st hEvent)
hEvent
- - Event to wait for::cuEventCreate,
::cuEventRecord,
::cuEventQuery,
::cuEventDestroy,
::cuEventElapsedTime,
::cudaEventSynchronize
@Cast(value="CUresult") public static int cuEventDestroy(CUevent_st hEvent)
hEvent
- - Event to destroy::cuEventCreate,
::cuEventRecord,
::cuEventQuery,
::cuEventSynchronize,
::cuEventElapsedTime,
::cudaEventDestroy
@Cast(value="CUresult") public static int cuEventElapsedTime(FloatPointer pMilliseconds, CUevent_st hStart, CUevent_st hEnd)
pMilliseconds
- - Time between \p hStart and \p hEnd in mshStart
- - Starting eventhEnd
- - Ending event::cuEventCreate,
::cuEventRecord,
::cuEventQuery,
::cuEventSynchronize,
::cuEventDestroy,
::cudaEventElapsedTime
@Cast(value="CUresult") public static int cuEventElapsedTime(FloatBuffer pMilliseconds, CUevent_st hStart, CUevent_st hEnd)
@Cast(value="CUresult") public static int cuEventElapsedTime(float[] pMilliseconds, CUevent_st hStart, CUevent_st hEnd)
@Cast(value="CUresult") public static int cuImportExternalMemory(@ByPtrPtr CUextMemory_st extMem_out, @Cast(value="const CUDA_EXTERNAL_MEMORY_HANDLE_DESC*") CUDA_EXTERNAL_MEMORY_HANDLE_DESC_v1 memHandleDesc)
typedef struct CUDA_EXTERNAL_MEMORY_HANDLE_DESC_st {
CUexternalMemoryHandleType type;
union {
int fd;
struct {
void *handle;
const void *name;
} win32;
const void *nvSciBufObject;
} handle;
unsigned long long size;
unsigned int flags;
} CUDA_EXTERNAL_MEMORY_HANDLE_DESC;
where ::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::type specifies the type
of handle being imported. ::CUexternalMemoryHandleType is
defined as:
typedef enum CUexternalMemoryHandleType_enum {
CU_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_FD = 1,
CU_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32 = 2,
CU_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_KMT = 3,
CU_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_HEAP = 4,
CU_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_RESOURCE = 5,
CU_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_RESOURCE = 6,
CU_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_RESOURCE_KMT = 7,
CU_EXTERNAL_MEMORY_HANDLE_TYPE_NVSCIBUF = 8
} CUexternalMemoryHandleType;
If ::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::type is
::CU_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_FD, then
::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::handle::fd must be a valid
file descriptor referencing a memory object. Ownership of
the file descriptor is transferred to the CUDA driver when the
handle is imported successfully. Performing any operations on the
file descriptor after it is imported results in undefined behavior.
If ::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::type is
::CU_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32, then exactly one
of ::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::handle::win32::handle and
::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::handle::win32::name must not be
NULL. If ::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::handle::win32::handle
is not NULL, then it must represent a valid shared NT handle that
references a memory object. Ownership of this handle is
not transferred to CUDA after the import operation, so the
application must release the handle using the appropriate system
call. If ::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::handle::win32::name
is not NULL, then it must point to a NULL-terminated array of
UTF-16 characters that refers to a memory object.
If ::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::type is
::CU_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_WIN32_KMT, then
::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::handle::win32::handle must
be non-NULL and
::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::handle::win32::name
must be NULL. The handle specified must be a globally shared KMT
handle. This handle does not hold a reference to the underlying
object, and thus will be invalid when all references to the
memory object are destroyed.
If ::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::type is
::CU_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_HEAP, then exactly one
of ::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::handle::win32::handle and
::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::handle::win32::name must not be
NULL. If ::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::handle::win32::handle
is not NULL, then it must represent a valid shared NT handle that
is returned by ID3D12Device::CreateSharedHandle when referring to a
ID3D12Heap object. This handle holds a reference to the underlying
object. If ::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::handle::win32::name
is not NULL, then it must point to a NULL-terminated array of
UTF-16 characters that refers to a ID3D12Heap object.
If ::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::type is
::CU_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_RESOURCE, then exactly one
of ::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::handle::win32::handle and
::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::handle::win32::name must not be
NULL. If ::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::handle::win32::handle
is not NULL, then it must represent a valid shared NT handle that
is returned by ID3D12Device::CreateSharedHandle when referring to a
ID3D12Resource object. This handle holds a reference to the
underlying object. If
::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::handle::win32::name
is not NULL, then it must point to a NULL-terminated array of
UTF-16 characters that refers to a ID3D12Resource object.
If ::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::type is
::CU_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_RESOURCE, then
::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::handle::win32::handle must
represent a valid shared NT handle that is returned by
IDXGIResource1::CreateSharedHandle when referring to a
ID3D11Resource object. If
::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::handle::win32::name
is not NULL, then it must point to a NULL-terminated array of
UTF-16 characters that refers to a ID3D11Resource object.
If ::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::type is
::CU_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_RESOURCE_KMT, then
::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::handle::win32::handle must
represent a valid shared KMT handle that is returned by
IDXGIResource::GetSharedHandle when referring to a
ID3D11Resource object and
::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::handle::win32::name
must be NULL.
If ::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::type is
::CU_EXTERNAL_MEMORY_HANDLE_TYPE_NVSCIBUF, then
::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::handle::nvSciBufObject must be non-NULL
and reference a valid NvSciBuf object.
If the NvSciBuf object imported into CUDA is also mapped by other drivers, then the
application must use ::cuWaitExternalSemaphoresAsync or ::cuSignalExternalSemaphoresAsync
as appropriate barriers to maintain coherence between CUDA and the other drivers.
See ::CUDA_EXTERNAL_SEMAPHORE_SIGNAL_SKIP_NVSCIBUF_MEMSYNC and ::CUDA_EXTERNAL_SEMAPHORE_WAIT_SKIP_NVSCIBUF_MEMSYNC
for memory synchronization.
The size of the memory object must be specified in
::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::size.
Specifying the flag ::CUDA_EXTERNAL_MEMORY_DEDICATED in
::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::flags indicates that the
resource is a dedicated resource. The definition of what a
dedicated resource is outside the scope of this extension.
This flag must be set if ::CUDA_EXTERNAL_MEMORY_HANDLE_DESC::type
is one of the following:
::CU_EXTERNAL_MEMORY_HANDLE_TYPE_D3D12_RESOURCE
::CU_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_RESOURCE
::CU_EXTERNAL_MEMORY_HANDLE_TYPE_D3D11_RESOURCE_KMTextMem_out
- - Returned handle to an external memory objectmemHandleDesc
- - Memory import handle descriptor::cuDestroyExternalMemory,
::cuExternalMemoryGetMappedBuffer,
::cuExternalMemoryGetMappedMipmappedArray
@Cast(value="CUresult") public static int cuExternalMemoryGetMappedBuffer(@Cast(value="CUdeviceptr*") LongPointer devPtr, CUextMemory_st extMem, @Cast(value="const CUDA_EXTERNAL_MEMORY_BUFFER_DESC*") CUDA_EXTERNAL_MEMORY_BUFFER_DESC_v1 bufferDesc)
typedef struct CUDA_EXTERNAL_MEMORY_BUFFER_DESC_st {
unsigned long long offset;
unsigned long long size;
unsigned int flags;
} CUDA_EXTERNAL_MEMORY_BUFFER_DESC;
where ::CUDA_EXTERNAL_MEMORY_BUFFER_DESC::offset is the offset in
the memory object where the buffer's base address is.
::CUDA_EXTERNAL_MEMORY_BUFFER_DESC::size is the size of the buffer.
::CUDA_EXTERNAL_MEMORY_BUFFER_DESC::flags must be zero.
The offset and size have to be suitably aligned to match the
requirements of the external API. Mapping two buffers whose ranges
overlap may or may not result in the same virtual address being
returned for the overlapped portion. In such cases, the application
must ensure that all accesses to that region from the GPU are
volatile. Otherwise writes made via one address are not guaranteed
to be visible via the other address, even if they're issued by the
same thread. It is recommended that applications map the combined
range instead of mapping separate buffers and then apply the
appropriate offsets to the returned pointer to derive the
individual buffers.
The returned pointer \p devPtr must be freed using ::cuMemFree.devPtr
- - Returned device pointer to bufferextMem
- - Handle to external memory objectbufferDesc
- - Buffer descriptor::cuImportExternalMemory,
::cuDestroyExternalMemory,
::cuExternalMemoryGetMappedMipmappedArray
@Cast(value="CUresult") public static int cuExternalMemoryGetMappedBuffer(@Cast(value="CUdeviceptr*") LongBuffer devPtr, CUextMemory_st extMem, @Cast(value="const CUDA_EXTERNAL_MEMORY_BUFFER_DESC*") CUDA_EXTERNAL_MEMORY_BUFFER_DESC_v1 bufferDesc)
@Cast(value="CUresult") public static int cuExternalMemoryGetMappedBuffer(@Cast(value="CUdeviceptr*") long[] devPtr, CUextMemory_st extMem, @Cast(value="const CUDA_EXTERNAL_MEMORY_BUFFER_DESC*") CUDA_EXTERNAL_MEMORY_BUFFER_DESC_v1 bufferDesc)
@Cast(value="CUresult") public static int cuExternalMemoryGetMappedMipmappedArray(@ByPtrPtr CUmipmappedArray_st mipmap, CUextMemory_st extMem, @Cast(value="const CUDA_EXTERNAL_MEMORY_MIPMAPPED_ARRAY_DESC*") CUDA_EXTERNAL_MEMORY_MIPMAPPED_ARRAY_DESC_v1 mipmapDesc)
typedef struct CUDA_EXTERNAL_MEMORY_MIPMAPPED_ARRAY_DESC_st {
unsigned long long offset;
CUDA_ARRAY3D_DESCRIPTOR arrayDesc;
unsigned int numLevels;
} CUDA_EXTERNAL_MEMORY_MIPMAPPED_ARRAY_DESC;
where ::CUDA_EXTERNAL_MEMORY_MIPMAPPED_ARRAY_DESC::offset is the
offset in the memory object where the base level of the mipmap
chain is.
::CUDA_EXTERNAL_MEMORY_MIPMAPPED_ARRAY_DESC::arrayDesc describes
the format, dimensions and type of the base level of the mipmap
chain. For further details on these parameters, please refer to the
documentation for ::cuMipmappedArrayCreate. Note that if the mipmapped
array is bound as a color target in the graphics API, then the flag
::CUDA_ARRAY3D_COLOR_ATTACHMENT must be specified in
::CUDA_EXTERNAL_MEMORY_MIPMAPPED_ARRAY_DESC::arrayDesc::Flags.
::CUDA_EXTERNAL_MEMORY_MIPMAPPED_ARRAY_DESC::numLevels specifies
the total number of levels in the mipmap chain.
If \p extMem was imported from a handle of type ::CU_EXTERNAL_MEMORY_HANDLE_TYPE_NVSCIBUF, then
::CUDA_EXTERNAL_MEMORY_MIPMAPPED_ARRAY_DESC::numLevels must be equal to 1.
The returned CUDA mipmapped array must be freed using ::cuMipmappedArrayDestroy.mipmap
- - Returned CUDA mipmapped arrayextMem
- - Handle to external memory objectmipmapDesc
- - CUDA array descriptor::cuImportExternalMemory,
::cuDestroyExternalMemory,
::cuExternalMemoryGetMappedBuffer
@Cast(value="CUresult") public static int cuDestroyExternalMemory(CUextMemory_st extMem)
extMem
- - External memory object to be destroyed::cuImportExternalMemory,
::cuExternalMemoryGetMappedBuffer,
::cuExternalMemoryGetMappedMipmappedArray
@Cast(value="CUresult") public static int cuImportExternalSemaphore(@ByPtrPtr CUextSemaphore_st extSem_out, @Cast(value="const CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC*") CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC_v1 semHandleDesc)
typedef struct CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC_st {
CUexternalSemaphoreHandleType type;
union {
int fd;
struct {
void *handle;
const void *name;
} win32;
const void* NvSciSyncObj;
} handle;
unsigned int flags;
} CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC;
where ::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::type specifies the type of
handle being imported. ::CUexternalSemaphoreHandleType is defined
as:
typedef enum CUexternalSemaphoreHandleType_enum {
CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_FD = 1,
CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32 = 2,
CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_KMT = 3,
CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D12_FENCE = 4,
CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D11_FENCE = 5,
CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_NVSCISYNC = 6,
CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D11_KEYED_MUTEX = 7,
CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D11_KEYED_MUTEX_KMT = 8,
CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_TIMELINE_SEMAPHORE_FD = 9,
CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_TIMELINE_SEMAPHORE_WIN32 = 10
} CUexternalSemaphoreHandleType;
If ::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::type is
::CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_FD, then
::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::fd must be a valid
file descriptor referencing a synchronization object. Ownership of
the file descriptor is transferred to the CUDA driver when the
handle is imported successfully. Performing any operations on the
file descriptor after it is imported results in undefined behavior.
If ::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::type is
::CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32, then exactly one
of ::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::win32::handle and
::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::win32::name must not be
NULL. If
::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::win32::handle
is not NULL, then it must represent a valid shared NT handle that
references a synchronization object. Ownership of this handle is
not transferred to CUDA after the import operation, so the
application must release the handle using the appropriate system
call. If ::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::win32::name
is not NULL, then it must name a valid synchronization object.
If ::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::type is
::CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_OPAQUE_WIN32_KMT, then
::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::win32::handle must
be non-NULL and
::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::win32::name
must be NULL. The handle specified must be a globally shared KMT
handle. This handle does not hold a reference to the underlying
object, and thus will be invalid when all references to the
synchronization object are destroyed.
If ::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::type is
::CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D12_FENCE, then exactly one
of ::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::win32::handle and
::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::win32::name must not be
NULL. If
::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::win32::handle
is not NULL, then it must represent a valid shared NT handle that
is returned by ID3D12Device::CreateSharedHandle when referring to a
ID3D12Fence object. This handle holds a reference to the underlying
object. If
::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::win32::name
is not NULL, then it must name a valid synchronization object that
refers to a valid ID3D12Fence object.
If ::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::type is
::CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D11_FENCE, then
::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::win32::handle
represents a valid shared NT handle that is returned by
ID3D11Fence::CreateSharedHandle. If
::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::win32::name
is not NULL, then it must name a valid synchronization object that
refers to a valid ID3D11Fence object.
If ::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::type is
::CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_NVSCISYNC, then
::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::nvSciSyncObj
represents a valid NvSciSyncObj.
::CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D11_KEYED_MUTEX, then
::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::win32::handle
represents a valid shared NT handle that
is returned by IDXGIResource1::CreateSharedHandle when referring to
a IDXGIKeyedMutex object. If
::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::win32::name
is not NULL, then it must name a valid synchronization object that
refers to a valid IDXGIKeyedMutex object.
If ::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::type is
::CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_D3D11_KEYED_MUTEX_KMT, then
::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::win32::handle
represents a valid shared KMT handle that
is returned by IDXGIResource::GetSharedHandle when referring to
a IDXGIKeyedMutex object and
::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::win32::name must be NULL.
If ::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::type is
::CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_TIMELINE_SEMAPHORE_FD, then
::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::fd must be a valid
file descriptor referencing a synchronization object. Ownership of
the file descriptor is transferred to the CUDA driver when the
handle is imported successfully. Performing any operations on the
file descriptor after it is imported results in undefined behavior.
If ::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::type is
::CU_EXTERNAL_SEMAPHORE_HANDLE_TYPE_TIMELINE_SEMAPHORE_WIN32, then exactly one
of ::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::win32::handle and
::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::win32::name must not be
NULL. If
::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::win32::handle
is not NULL, then it must represent a valid shared NT handle that
references a synchronization object. Ownership of this handle is
not transferred to CUDA after the import operation, so the
application must release the handle using the appropriate system
call. If ::CUDA_EXTERNAL_SEMAPHORE_HANDLE_DESC::handle::win32::name
is not NULL, then it must name a valid synchronization object.extSem_out
- - Returned handle to an external semaphoresemHandleDesc
- - Semaphore import handle descriptor::cuDestroyExternalSemaphore,
::cuSignalExternalSemaphoresAsync,
::cuWaitExternalSemaphoresAsync
@Cast(value="CUresult") public static int cuSignalExternalSemaphoresAsync(@Cast(value="const CUexternalSemaphore*") @ByPtrPtr CUextSemaphore_st extSemArray, @Cast(value="const CUDA_EXTERNAL_SEMAPHORE_SIGNAL_PARAMS*") CUDA_EXTERNAL_SEMAPHORE_SIGNAL_PARAMS_v1 paramsArray, @Cast(value="unsigned int") int numExtSems, CUstream_st stream)
extSemArray
- - Set of external semaphores to be signaledparamsArray
- - Array of semaphore parametersnumExtSems
- - Number of semaphores to signalstream
- - Stream to enqueue the signal operations in::cuImportExternalSemaphore,
::cuDestroyExternalSemaphore,
::cuWaitExternalSemaphoresAsync
@Cast(value="CUresult") public static int cuWaitExternalSemaphoresAsync(@Cast(value="const CUexternalSemaphore*") @ByPtrPtr CUextSemaphore_st extSemArray, @Cast(value="const CUDA_EXTERNAL_SEMAPHORE_WAIT_PARAMS*") CUDA_EXTERNAL_SEMAPHORE_WAIT_PARAMS_v1 paramsArray, @Cast(value="unsigned int") int numExtSems, CUstream_st stream)
extSemArray
- - External semaphores to be waited onparamsArray
- - Array of semaphore parametersnumExtSems
- - Number of semaphores to wait onstream
- - Stream to enqueue the wait operations in::cuImportExternalSemaphore,
::cuDestroyExternalSemaphore,
::cuSignalExternalSemaphoresAsync
@Cast(value="CUresult") public static int cuDestroyExternalSemaphore(CUextSemaphore_st extSem)
extSem
- - External semaphore to be destroyed::cuImportExternalSemaphore,
::cuSignalExternalSemaphoresAsync,
::cuWaitExternalSemaphoresAsync
@Cast(value="CUresult") public static int cuStreamWaitValue32(CUstream_st stream, @Cast(value="CUdeviceptr") long addr, @Cast(value="cuuint32_t") int value, @Cast(value="unsigned int") int flags)
stream
- The stream to synchronize on the memory location.addr
- The memory location to wait on.value
- The value to compare with the memory location.flags
- See ::CUstreamWaitValue_flags.::cuStreamWaitValue64,
::cuStreamWriteValue32,
::cuStreamWriteValue64,
::cuStreamBatchMemOp,
::cuMemHostRegister,
::cuStreamWaitEvent
@Cast(value="CUresult") public static int cuStreamWaitValue64(CUstream_st stream, @Cast(value="CUdeviceptr") long addr, @Cast(value="cuuint64_t") long value, @Cast(value="unsigned int") int flags)
stream
- The stream to synchronize on the memory location.addr
- The memory location to wait on.value
- The value to compare with the memory location.flags
- See ::CUstreamWaitValue_flags.::cuStreamWaitValue32,
::cuStreamWriteValue32,
::cuStreamWriteValue64,
::cuStreamBatchMemOp,
::cuMemHostRegister,
::cuStreamWaitEvent
@Cast(value="CUresult") public static int cuStreamWriteValue32(CUstream_st stream, @Cast(value="CUdeviceptr") long addr, @Cast(value="cuuint32_t") int value, @Cast(value="unsigned int") int flags)
stream
- The stream to do the write in.addr
- The device address to write to.value
- The value to write.flags
- See ::CUstreamWriteValue_flags.::cuStreamWriteValue64,
::cuStreamWaitValue32,
::cuStreamWaitValue64,
::cuStreamBatchMemOp,
::cuMemHostRegister,
::cuEventRecord
@Cast(value="CUresult") public static int cuStreamWriteValue64(CUstream_st stream, @Cast(value="CUdeviceptr") long addr, @Cast(value="cuuint64_t") long value, @Cast(value="unsigned int") int flags)
stream
- The stream to do the write in.addr
- The device address to write to.value
- The value to write.flags
- See ::CUstreamWriteValue_flags.::cuStreamWriteValue32,
::cuStreamWaitValue32,
::cuStreamWaitValue64,
::cuStreamBatchMemOp,
::cuMemHostRegister,
::cuEventRecord
@Cast(value="CUresult") public static int cuStreamBatchMemOp(CUstream_st stream, @Cast(value="unsigned int") int count, @Cast(value="CUstreamBatchMemOpParams*") CUstreamBatchMemOpParams_v1 paramArray, @Cast(value="unsigned int") int flags)
stream
- The stream to enqueue the operations in.count
- The number of operations in the array. Must be less than 256.paramArray
- The types and parameters of the individual operations.flags
- Reserved for future expansion; must be 0.::cuStreamWaitValue32,
::cuStreamWaitValue64,
::cuStreamWriteValue32,
::cuStreamWriteValue64,
::cuMemHostRegister
@Cast(value="CUresult") public static int cuFuncGetAttribute(IntPointer pi, @Cast(value="CUfunction_attribute") int attrib, CUfunc_st hfunc)
pi
- - Returned attribute valueattrib
- - Attribute requestedhfunc
- - Function to query attribute of::cuCtxGetCacheConfig,
::cuCtxSetCacheConfig,
::cuFuncSetCacheConfig,
::cuLaunchKernel,
::cudaFuncGetAttributes,
::cudaFuncSetAttribute,
::cuKernelGetAttribute
@Cast(value="CUresult") public static int cuFuncGetAttribute(IntBuffer pi, @Cast(value="CUfunction_attribute") int attrib, CUfunc_st hfunc)
@Cast(value="CUresult") public static int cuFuncGetAttribute(int[] pi, @Cast(value="CUfunction_attribute") int attrib, CUfunc_st hfunc)
@Cast(value="CUresult") public static int cuFuncSetAttribute(CUfunc_st hfunc, @Cast(value="CUfunction_attribute") int attrib, int value)
hfunc
- - Function to query attribute ofattrib
- - Attribute requestedvalue
- - The value to set::cuCtxGetCacheConfig,
::cuCtxSetCacheConfig,
::cuFuncSetCacheConfig,
::cuLaunchKernel,
::cudaFuncGetAttributes,
::cudaFuncSetAttribute,
::cuKernelSetAttribute
@Cast(value="CUresult") public static int cuFuncSetCacheConfig(CUfunc_st hfunc, @Cast(value="CUfunc_cache") int config)
hfunc
- - Kernel to configure cache forconfig
- - Requested cache configuration::cuCtxGetCacheConfig,
::cuCtxSetCacheConfig,
::cuFuncGetAttribute,
::cuLaunchKernel,
::cudaFuncSetCacheConfig,
::cuKernelSetCacheConfig
@Cast(value="CUresult") public static int cuFuncSetSharedMemConfig(CUfunc_st hfunc, @Cast(value="CUsharedconfig") int config)
hfunc
- - kernel to be given a shared memory configconfig
- - requested shared memory configuration::cuCtxGetCacheConfig,
::cuCtxSetCacheConfig,
::cuCtxGetSharedMemConfig,
::cuCtxSetSharedMemConfig,
::cuFuncGetAttribute,
::cuLaunchKernel,
::cudaFuncSetSharedMemConfig
@Cast(value="CUresult") public static int cuFuncGetModule(@ByPtrPtr CUmod_st hmod, CUfunc_st hfunc)
hmod
- - Returned module handlehfunc
- - Function to retrieve module for@Cast(value="CUresult") public static int cuFuncGetName(@Cast(value="const char**") PointerPointer name, CUfunc_st hfunc)
name
- - The returned name of the functionhfunc
- - The function handle to retrieve the name for@Cast(value="CUresult") public static int cuFuncGetName(@Cast(value="const char**") @ByPtrPtr BytePointer name, CUfunc_st hfunc)
@Cast(value="CUresult") public static int cuFuncGetName(@Cast(value="const char**") @ByPtrPtr ByteBuffer name, CUfunc_st hfunc)
@Cast(value="CUresult") public static int cuFuncGetName(@Cast(value="const char**") @ByPtrPtr byte[] name, CUfunc_st hfunc)
@Cast(value="CUresult") public static int cuLaunchKernel(CUfunc_st f, @Cast(value="unsigned int") int gridDimX, @Cast(value="unsigned int") int gridDimY, @Cast(value="unsigned int") int gridDimZ, @Cast(value="unsigned int") int blockDimX, @Cast(value="unsigned int") int blockDimY, @Cast(value="unsigned int") int blockDimZ, @Cast(value="unsigned int") int sharedMemBytes, CUstream_st hStream, @Cast(value="void**") PointerPointer kernelParams, @Cast(value="void**") PointerPointer extra)
size_t argBufferSize;
char argBuffer[256];
// populate argBuffer and argBufferSize
void *config[] = {
CU_LAUNCH_PARAM_BUFFER_POINTER, argBuffer,
CU_LAUNCH_PARAM_BUFFER_SIZE, &argBufferSize,
CU_LAUNCH_PARAM_END
};
status = cuLaunchKernel(f, gx, gy, gz, bx, by, bz, sh, s, NULL, config);
The \p extra parameter exists to allow ::cuLaunchKernel to take
additional less commonly used arguments. \p extra specifies a list of
names of extra settings and their corresponding values. Each extra
setting name is immediately followed by the corresponding value. The
list must be terminated with either NULL or ::CU_LAUNCH_PARAM_END.
- ::CU_LAUNCH_PARAM_END, which indicates the end of the \p extra
array;
- ::CU_LAUNCH_PARAM_BUFFER_POINTER, which specifies that the next
value in \p extra will be a pointer to a buffer containing all
the kernel parameters for launching kernel \p f;
- ::CU_LAUNCH_PARAM_BUFFER_SIZE, which specifies that the next
value in \p extra will be a pointer to a size_t containing the
size of the buffer specified with ::CU_LAUNCH_PARAM_BUFFER_POINTER;
The error ::CUDA_ERROR_INVALID_VALUE will be returned if kernel
parameters are specified with both \p kernelParams and \p extra
(i.e. both \p kernelParams and \p extra are non-NULL).
Calling ::cuLaunchKernel() invalidates the persistent function state
set through the following deprecated APIs:
::cuFuncSetBlockShape(),
::cuFuncSetSharedSize(),
::cuParamSetSize(),
::cuParamSeti(),
::cuParamSetf(),
::cuParamSetv().
Note that to use ::cuLaunchKernel(), the kernel \p f must either have
been compiled with toolchain version 3.2 or later so that it will
contain kernel parameter information, or have no kernel parameters.
If either of these conditions is not met, then ::cuLaunchKernel() will
return ::CUDA_ERROR_INVALID_IMAGE.
Note that the API can also be used to launch context-less kernel ::CUkernel
by querying the handle using ::cuLibraryGetKernel() and then passing it
to the API by casting to ::CUfunction. Here, the context to launch
the kernel on will either be taken from the specified stream \p hStream
or the current context in case of NULL stream.f
- - Function ::CUfunction or Kernel ::CUkernel to launchgridDimX
- - Width of grid in blocksgridDimY
- - Height of grid in blocksgridDimZ
- - Depth of grid in blocksblockDimX
- - X dimension of each thread blockblockDimY
- - Y dimension of each thread blockblockDimZ
- - Z dimension of each thread blocksharedMemBytes
- - Dynamic shared-memory size per thread block in byteshStream
- - Stream identifierkernelParams
- - Array of pointers to kernel parametersextra
- - Extra options::cuCtxGetCacheConfig,
::cuCtxSetCacheConfig,
::cuFuncSetCacheConfig,
::cuFuncGetAttribute,
::cudaLaunchKernel,
::cuLibraryGetKernel,
::cuKernelSetCacheConfig,
::cuKernelGetAttribute,
::cuKernelSetAttribute
@Cast(value="CUresult") public static int cuLaunchKernel(CUfunc_st f, @Cast(value="unsigned int") int gridDimX, @Cast(value="unsigned int") int gridDimY, @Cast(value="unsigned int") int gridDimZ, @Cast(value="unsigned int") int blockDimX, @Cast(value="unsigned int") int blockDimY, @Cast(value="unsigned int") int blockDimZ, @Cast(value="unsigned int") int sharedMemBytes, CUstream_st hStream, @Cast(value="void**") @ByPtrPtr Pointer kernelParams, @Cast(value="void**") @ByPtrPtr Pointer extra)
@Cast(value="CUresult") public static int cuLaunchKernelEx(@Const CUlaunchConfig config, CUfunc_st f, @Cast(value="void**") PointerPointer kernelParams, @Cast(value="void**") PointerPointer extra)
typedef struct CUlaunchConfig_st {
unsigned int gridDimX;
unsigned int gridDimY;
unsigned int gridDimZ;
unsigned int blockDimX;
unsigned int blockDimY;
unsigned int blockDimZ;
unsigned int sharedMemBytes;
CUstream hStream;
CUlaunchAttribute *attrs;
unsigned int numAttrs;
} CUlaunchConfig;
where:
- ::CUlaunchConfig::gridDimX is the width of the grid in blocks.
- ::CUlaunchConfig::gridDimY is the height of the grid in blocks.
- ::CUlaunchConfig::gridDimZ is the depth of the grid in blocks.
- ::CUlaunchConfig::blockDimX is the X dimension of each thread block.
- ::CUlaunchConfig::blockDimX is the Y dimension of each thread block.
- ::CUlaunchConfig::blockDimZ is the Z dimension of each thread block.
- ::CUlaunchConfig::sharedMemBytes is the dynamic shared-memory size per
thread block in bytes.
- ::CUlaunchConfig::hStream is the handle to the stream to perform the launch
in. The CUDA context associated with this stream must match that associated
with function f.
- ::CUlaunchConfig::attrs is an array of ::CUlaunchConfig::numAttrs
continguous ::CUlaunchAttribute elements. The value of this pointer is not
considered if ::CUlaunchConfig::numAttrs is zero. However, in that case, it
is recommended to set the pointer to NULL.
- ::CUlaunchConfig::numAttrs is the number of attributes populating the
first ::CUlaunchConfig::numAttrs positions of the ::CUlaunchConfig::attrs
array.
Launch-time configuration is specified by adding entries to
::CUlaunchConfig::attrs. Each entry is an attribute ID and a corresponding
attribute value.
The ::CUlaunchAttribute structure is defined as:
typedef struct CUlaunchAttribute_st {
CUlaunchAttributeID id;
CUlaunchAttributeValue value;
} CUlaunchAttribute;
where:
- ::CUlaunchAttribute::id is a unique enum identifying the attribute.
- ::CUlaunchAttribute::value is a union that hold the attribute value.
An example of using the \p config parameter:
CUlaunchAttribute coopAttr = {.id = CU_LAUNCH_ATTRIBUTE_COOPERATIVE,
.value = 1};
CUlaunchConfig config = {... // set block and grid dimensions
.attrs = &coopAttr,
.numAttrs = 1};
cuLaunchKernelEx(&config, kernel, NULL, NULL);
The ::CUlaunchAttributeID enum is defined as:
typedef enum CUlaunchAttributeID_enum {
CU_LAUNCH_ATTRIBUTE_IGNORE = 0,
CU_LAUNCH_ATTRIBUTE_ACCESS_POLICY_WINDOW = 1,
CU_LAUNCH_ATTRIBUTE_COOPERATIVE = 2,
CU_LAUNCH_ATTRIBUTE_SYNCHRONIZATION_POLICY = 3,
CU_LAUNCH_ATTRIBUTE_CLUSTER_DIMENSION = 4,
CU_LAUNCH_ATTRIBUTE_CLUSTER_SCHEDULING_POLICY_PREFERENCE = 5,
CU_LAUNCH_ATTRIBUTE_PROGRAMMATIC_STREAM_SERIALIZATION = 6,
CU_LAUNCH_ATTRIBUTE_PROGRAMMATIC_EVENT = 7,
CU_LAUNCH_ATTRIBUTE_PRIORITY = 8,
CU_LAUNCH_ATTRIBUTE_MEM_SYNC_DOMAIN_MAP = 9,
CU_LAUNCH_ATTRIBUTE_MEM_SYNC_DOMAIN = 10,
CU_LAUNCH_ATTRIBUTE_LAUNCH_COMPLETION_EVENT = 12,
} CUlaunchAttributeID;
and the corresponding ::CUlaunchAttributeValue union as :
typedef union CUlaunchAttributeValue_union {
CUaccessPolicyWindow accessPolicyWindow;
int cooperative;
CUsynchronizationPolicy syncPolicy;
struct {
unsigned int x;
unsigned int y;
unsigned int z;
} clusterDim;
CUclusterSchedulingPolicy clusterSchedulingPolicyPreference;
int programmaticStreamSerializationAllowed;
struct {
CUevent event;
int flags;
int triggerAtBlockStart;
} programmaticEvent;
int priority;
CUlaunchMemSyncDomainMap memSyncDomainMap;
CUlaunchMemSyncDomain memSyncDomain;
struct {
CUevent event;
int flags;
} launchCompletionEvent;
} CUlaunchAttributeValue;
Setting ::CU_LAUNCH_ATTRIBUTE_COOPERATIVE to a non-zero value causes the
kernel launch to be a cooperative launch, with exactly the same usage and
semantics of ::cuLaunchCooperativeKernel.
Setting ::CU_LAUNCH_ATTRIBUTE_PROGRAMMATIC_STREAM_SERIALIZATION to a non-zero
values causes the kernel to use programmatic means to resolve its stream
dependency -- enabling the CUDA runtime to opportunistically allow the grid's
execution to overlap with the previous kernel in the stream, if that kernel
requests the overlap.
::CU_LAUNCH_ATTRIBUTE_PROGRAMMATIC_EVENT records an event along with the
kernel launch. Event recorded through this launch attribute is guaranteed to
only trigger after all block in the associated kernel trigger the event. A
block can trigger the event through PTX launchdep.release or CUDA builtin
function cudaTriggerProgrammaticLaunchCompletion(). A trigger can also be
inserted at the beginning of each block's execution if triggerAtBlockStart is
set to non-0. Note that dependents (including the CPU thread calling
cuEventSynchronize()) are not guaranteed to observe the release precisely
when it is released. For example, cuEventSynchronize() may only observe the
event trigger long after the associated kernel has completed. This recording
type is primarily meant for establishing programmatic dependency between
device tasks. The event supplied must not be an interprocess or interop
event. The event must disable timing (i.e. created with
::CU_EVENT_DISABLE_TIMING flag set).
::CU_LAUNCH_ATTRIBUTE_LAUNCH_COMPLETION_EVENT records an event along with
the kernel launch. Nominally, the event is triggered once all blocks of the
kernel have begun execution. Currently this is a best effort. If a kernel B
has a launch completion dependency on a kernel A, B may wait until A is
complete. Alternatively, blocks of B may begin before all blocks of A have
begun, for example:
- If B can claim execution resources unavaiable to A, for example if they
run on different GPUs.
- If B is a higher priority than A.
Exercise caution if such an ordering inversion could lead to deadlock. The
event supplied must not be an interprocess or interop event. The event must
disable timing (i.e. must be created with the ::CU_EVENT_DISABLE_TIMING flag
set).
The effect of other attributes is consistent with their effect when set via
persistent APIs.
See ::cuStreamSetAttribute for
- ::CU_LAUNCH_ATTRIBUTE_ACCESS_POLICY_WINDOW
- ::CU_LAUNCH_ATTRIBUTE_SYNCHRONIZATION_POLICY
See ::cuFuncSetAttribute for
- ::CU_LAUNCH_ATTRIBUTE_CLUSTER_DIMENSION
- ::CU_LAUNCH_ATTRIBUTE_CLUSTER_SCHEDULING_POLICY_PREFERENCE
Kernel parameters to \p f can be specified in the same ways that they can be
using ::cuLaunchKernel.
Note that the API can also be used to launch context-less kernel ::CUkernel
by querying the handle using ::cuLibraryGetKernel() and then passing it
to the API by casting to ::CUfunction. Here, the context to launch
the kernel on will either be taken from the specified stream ::CUlaunchConfig::hStream
or the current context in case of NULL stream.config
- - Config to launchf
- - Function ::CUfunction or Kernel ::CUkernel to launchkernelParams
- - Array of pointers to kernel parametersextra
- - Extra options::cuCtxGetCacheConfig,
::cuCtxSetCacheConfig,
::cuFuncSetCacheConfig,
::cuFuncGetAttribute,
::cudaLaunchKernel,
::cudaLaunchKernelEx,
::cuLibraryGetKernel,
::cuKernelSetCacheConfig,
::cuKernelGetAttribute,
::cuKernelSetAttribute
@Cast(value="CUresult") public static int cuLaunchKernelEx(@Const CUlaunchConfig config, CUfunc_st f, @Cast(value="void**") @ByPtrPtr Pointer kernelParams, @Cast(value="void**") @ByPtrPtr Pointer extra)
@Cast(value="CUresult") public static int cuLaunchCooperativeKernel(CUfunc_st f, @Cast(value="unsigned int") int gridDimX, @Cast(value="unsigned int") int gridDimY, @Cast(value="unsigned int") int gridDimZ, @Cast(value="unsigned int") int blockDimX, @Cast(value="unsigned int") int blockDimY, @Cast(value="unsigned int") int blockDimZ, @Cast(value="unsigned int") int sharedMemBytes, CUstream_st hStream, @Cast(value="void**") PointerPointer kernelParams)
f
- - Function ::CUfunction or Kernel ::CUkernel to launchgridDimX
- - Width of grid in blocksgridDimY
- - Height of grid in blocksgridDimZ
- - Depth of grid in blocksblockDimX
- - X dimension of each thread blockblockDimY
- - Y dimension of each thread blockblockDimZ
- - Z dimension of each thread blocksharedMemBytes
- - Dynamic shared-memory size per thread block in byteshStream
- - Stream identifierkernelParams
- - Array of pointers to kernel parameters::cuCtxGetCacheConfig,
::cuCtxSetCacheConfig,
::cuFuncSetCacheConfig,
::cuFuncGetAttribute,
::cuLaunchCooperativeKernelMultiDevice,
::cudaLaunchCooperativeKernel,
::cuLibraryGetKernel,
::cuKernelSetCacheConfig,
::cuKernelGetAttribute,
::cuKernelSetAttribute
@Cast(value="CUresult") public static int cuLaunchCooperativeKernel(CUfunc_st f, @Cast(value="unsigned int") int gridDimX, @Cast(value="unsigned int") int gridDimY, @Cast(value="unsigned int") int gridDimZ, @Cast(value="unsigned int") int blockDimX, @Cast(value="unsigned int") int blockDimY, @Cast(value="unsigned int") int blockDimZ, @Cast(value="unsigned int") int sharedMemBytes, CUstream_st hStream, @Cast(value="void**") @ByPtrPtr Pointer kernelParams)
@Cast(value="CUresult") @Deprecated public static int cuLaunchCooperativeKernelMultiDevice(@Cast(value="CUDA_LAUNCH_PARAMS*") CUDA_LAUNCH_PARAMS_v1 launchParamsList, @Cast(value="unsigned int") int numDevices, @Cast(value="unsigned int") int flags)
typedef struct CUDA_LAUNCH_PARAMS_st
{
CUfunction function;
unsigned int gridDimX;
unsigned int gridDimY;
unsigned int gridDimZ;
unsigned int blockDimX;
unsigned int blockDimY;
unsigned int blockDimZ;
unsigned int sharedMemBytes;
CUstream hStream;
void **kernelParams;
} CUDA_LAUNCH_PARAMS;
where:
- ::CUDA_LAUNCH_PARAMS::function specifies the kernel to be launched. All functions must
be identical with respect to the compiled code.
Note that you can also specify context-less kernel ::CUkernel by querying the handle
using ::cuLibraryGetKernel() and then casting to ::CUfunction. In this case, the context to
launch the kernel on be taken from the specified stream ::CUDA_LAUNCH_PARAMS::hStream.
- ::CUDA_LAUNCH_PARAMS::gridDimX is the width of the grid in blocks. This must match across
all kernels launched.
- ::CUDA_LAUNCH_PARAMS::gridDimY is the height of the grid in blocks. This must match across
all kernels launched.
- ::CUDA_LAUNCH_PARAMS::gridDimZ is the depth of the grid in blocks. This must match across
all kernels launched.
- ::CUDA_LAUNCH_PARAMS::blockDimX is the X dimension of each thread block. This must match across
all kernels launched.
- ::CUDA_LAUNCH_PARAMS::blockDimX is the Y dimension of each thread block. This must match across
all kernels launched.
- ::CUDA_LAUNCH_PARAMS::blockDimZ is the Z dimension of each thread block. This must match across
all kernels launched.
- ::CUDA_LAUNCH_PARAMS::sharedMemBytes is the dynamic shared-memory size per thread block in bytes.
This must match across all kernels launched.
- ::CUDA_LAUNCH_PARAMS::hStream is the handle to the stream to perform the launch in. This cannot
be the NULL stream or ::CU_STREAM_LEGACY or ::CU_STREAM_PER_THREAD. The CUDA context associated
with this stream must match that associated with ::CUDA_LAUNCH_PARAMS::function.
- ::CUDA_LAUNCH_PARAMS::kernelParams is an array of pointers to kernel parameters. If
::CUDA_LAUNCH_PARAMS::function has N parameters, then ::CUDA_LAUNCH_PARAMS::kernelParams
needs to be an array of N pointers. Each of ::CUDA_LAUNCH_PARAMS::kernelParams[0] through
::CUDA_LAUNCH_PARAMS::kernelParams[N-1] must point to a region of memory from which the actual
kernel parameter will be copied. The number of kernel parameters and their offsets and sizes
do not need to be specified as that information is retrieved directly from the kernel's image.
By default, the kernel won't begin execution on any GPU until all prior work in all the specified
streams has completed. This behavior can be overridden by specifying the flag
::CUDA_COOPERATIVE_LAUNCH_MULTI_DEVICE_NO_PRE_LAUNCH_SYNC. When this flag is specified, each kernel
will only wait for prior work in the stream corresponding to that GPU to complete before it begins
execution.
Similarly, by default, any subsequent work pushed in any of the specified streams will not begin
execution until the kernels on all GPUs have completed. This behavior can be overridden by specifying
the flag ::CUDA_COOPERATIVE_LAUNCH_MULTI_DEVICE_NO_POST_LAUNCH_SYNC. When this flag is specified,
any subsequent work pushed in any of the specified streams will only wait for the kernel launched
on the GPU corresponding to that stream to complete before it begins execution.
Calling ::cuLaunchCooperativeKernelMultiDevice() sets persistent function state that is
the same as function state set through ::cuLaunchKernel API when called individually for each
element in \p launchParamsList.
When kernels are launched via ::cuLaunchCooperativeKernelMultiDevice(), the previous
block shape, shared size and parameter info associated with each ::CUDA_LAUNCH_PARAMS::function
in \p launchParamsList is overwritten.
Note that to use ::cuLaunchCooperativeKernelMultiDevice(), the kernels must either have
been compiled with toolchain version 3.2 or later so that it will
contain kernel parameter information, or have no kernel parameters.
If either of these conditions is not met, then ::cuLaunchCooperativeKernelMultiDevice() will
return ::CUDA_ERROR_INVALID_IMAGE.launchParamsList
- - List of launch parameters, one per devicenumDevices
- - Size of the \p launchParamsList arrayflags
- - Flags to control launch behavior::cuCtxGetCacheConfig,
::cuCtxSetCacheConfig,
::cuFuncSetCacheConfig,
::cuFuncGetAttribute,
::cuLaunchCooperativeKernel,
::cudaLaunchCooperativeKernelMultiDevice
@Cast(value="CUresult") public static int cuLaunchHostFunc(CUstream_st hStream, CUhostFn fn, Pointer userData)
hStream
- - Stream to enqueue function call infn
- - The function to call once preceding stream operations are completeuserData
- - User-specified data to be passed to the function::cuStreamCreate,
::cuStreamQuery,
::cuStreamSynchronize,
::cuStreamWaitEvent,
::cuStreamDestroy,
::cuMemAllocManaged,
::cuStreamAttachMemAsync,
::cuStreamAddCallback
@Cast(value="CUresult") @Deprecated public static int cuFuncSetBlockShape(CUfunc_st hfunc, int x, int y, int z)
hfunc
- - Kernel to specify dimensions ofx
- - X dimensiony
- - Y dimensionz
- - Z dimension::cuFuncSetSharedSize,
::cuFuncSetCacheConfig,
::cuFuncGetAttribute,
::cuParamSetSize,
::cuParamSeti,
::cuParamSetf,
::cuParamSetv,
::cuLaunch,
::cuLaunchGrid,
::cuLaunchGridAsync,
::cuLaunchKernel
@Cast(value="CUresult") @Deprecated public static int cuFuncSetSharedSize(CUfunc_st hfunc, @Cast(value="unsigned int") int bytes)
hfunc
- - Kernel to specify dynamic shared-memory size forbytes
- - Dynamic shared-memory size per thread in bytes::cuFuncSetBlockShape,
::cuFuncSetCacheConfig,
::cuFuncGetAttribute,
::cuParamSetSize,
::cuParamSeti,
::cuParamSetf,
::cuParamSetv,
::cuLaunch,
::cuLaunchGrid,
::cuLaunchGridAsync,
::cuLaunchKernel
@Cast(value="CUresult") @Deprecated public static int cuParamSetSize(CUfunc_st hfunc, @Cast(value="unsigned int") int numbytes)
hfunc
- - Kernel to set parameter size fornumbytes
- - Size of parameter list in bytes::cuFuncSetBlockShape,
::cuFuncSetSharedSize,
::cuFuncGetAttribute,
::cuParamSetf,
::cuParamSeti,
::cuParamSetv,
::cuLaunch,
::cuLaunchGrid,
::cuLaunchGridAsync,
::cuLaunchKernel
@Cast(value="CUresult") @Deprecated public static int cuParamSeti(CUfunc_st hfunc, int offset, @Cast(value="unsigned int") int value)
hfunc
- - Kernel to add parameter tooffset
- - Offset to add parameter to argument listvalue
- - Value of parameter::cuFuncSetBlockShape,
::cuFuncSetSharedSize,
::cuFuncGetAttribute,
::cuParamSetSize,
::cuParamSetf,
::cuParamSetv,
::cuLaunch,
::cuLaunchGrid,
::cuLaunchGridAsync,
::cuLaunchKernel
@Cast(value="CUresult") @Deprecated public static int cuParamSetf(CUfunc_st hfunc, int offset, float value)
hfunc
- - Kernel to add parameter tooffset
- - Offset to add parameter to argument listvalue
- - Value of parameter::cuFuncSetBlockShape,
::cuFuncSetSharedSize,
::cuFuncGetAttribute,
::cuParamSetSize,
::cuParamSeti,
::cuParamSetv,
::cuLaunch,
::cuLaunchGrid,
::cuLaunchGridAsync,
::cuLaunchKernel
@Cast(value="CUresult") @Deprecated public static int cuParamSetv(CUfunc_st hfunc, int offset, Pointer ptr, @Cast(value="unsigned int") int numbytes)
hfunc
- - Kernel to add data tooffset
- - Offset to add data to argument listptr
- - Pointer to arbitrary datanumbytes
- - Size of data to copy in bytes::cuFuncSetBlockShape,
::cuFuncSetSharedSize,
::cuFuncGetAttribute,
::cuParamSetSize,
::cuParamSetf,
::cuParamSeti,
::cuLaunch,
::cuLaunchGrid,
::cuLaunchGridAsync,
::cuLaunchKernel
@Cast(value="CUresult") @Deprecated public static int cuLaunch(CUfunc_st f)
f
- - Kernel to launch::cuFuncSetBlockShape,
::cuFuncSetSharedSize,
::cuFuncGetAttribute,
::cuParamSetSize,
::cuParamSetf,
::cuParamSeti,
::cuParamSetv,
::cuLaunchGrid,
::cuLaunchGridAsync,
::cuLaunchKernel
@Cast(value="CUresult") @Deprecated public static int cuLaunchGrid(CUfunc_st f, int grid_width, int grid_height)
f
- - Kernel to launchgrid_width
- - Width of grid in blocksgrid_height
- - Height of grid in blocks::cuFuncSetBlockShape,
::cuFuncSetSharedSize,
::cuFuncGetAttribute,
::cuParamSetSize,
::cuParamSetf,
::cuParamSeti,
::cuParamSetv,
::cuLaunch,
::cuLaunchGridAsync,
::cuLaunchKernel
@Cast(value="CUresult") @Deprecated public static int cuLaunchGridAsync(CUfunc_st f, int grid_width, int grid_height, CUstream_st hStream)
f
- - Kernel to launchgrid_width
- - Width of grid in blocksgrid_height
- - Height of grid in blockshStream
- - Stream identifier::cuFuncSetBlockShape,
::cuFuncSetSharedSize,
::cuFuncGetAttribute,
::cuParamSetSize,
::cuParamSetf,
::cuParamSeti,
::cuParamSetv,
::cuLaunch,
::cuLaunchGrid,
::cuLaunchKernel
@Cast(value="CUresult") @Deprecated public static int cuParamSetTexRef(CUfunc_st hfunc, int texunit, CUtexref_st hTexRef)
hfunc
- - Kernel to add texture-reference totexunit
- - Texture unit (must be ::CU_PARAM_TR_DEFAULT)hTexRef
- - Texture-reference to add to argument list@Cast(value="CUresult") public static int cuGraphCreate(@ByPtrPtr CUgraph_st phGraph, @Cast(value="unsigned int") int flags)
phGraph
- - Returns newly created graphflags
- - Graph creation flags, must be 0::cuGraphAddChildGraphNode,
::cuGraphAddEmptyNode,
::cuGraphAddKernelNode,
::cuGraphAddHostNode,
::cuGraphAddMemcpyNode,
::cuGraphAddMemsetNode,
::cuGraphInstantiate,
::cuGraphDestroy,
::cuGraphGetNodes,
::cuGraphGetRootNodes,
::cuGraphGetEdges,
::cuGraphClone
@Cast(value="CUresult") public static int cuGraphAddKernelNode(@ByPtrPtr CUgraphNode_st phGraphNode, CUgraph_st hGraph, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st dependencies, @Cast(value="size_t") long numDependencies, @Cast(value="const CUDA_KERNEL_NODE_PARAMS*") CUDA_KERNEL_NODE_PARAMS_v2 nodeParams)
typedef struct CUDA_KERNEL_NODE_PARAMS_st {
CUfunction func;
unsigned int gridDimX;
unsigned int gridDimY;
unsigned int gridDimZ;
unsigned int blockDimX;
unsigned int blockDimY;
unsigned int blockDimZ;
unsigned int sharedMemBytes;
void **kernelParams;
void **extra;
CUkernel kern;
CUcontext ctx;
} CUDA_KERNEL_NODE_PARAMS;
When the graph is launched, the node will invoke kernel \p func on a (\p gridDimX x
\p gridDimY x \p gridDimZ) grid of blocks. Each block contains
(\p blockDimX x \p blockDimY x \p blockDimZ) threads.
\p sharedMemBytes sets the amount of dynamic shared memory that will be
available to each thread block.
Kernel parameters to \p func can be specified in one of two ways:
1) Kernel parameters can be specified via \p kernelParams. If the kernel has N
parameters, then \p kernelParams needs to be an array of N pointers. Each pointer,
from \p kernelParams[0] to \p kernelParams[N-1], points to the region of memory from which the actual
parameter will be copied. The number of kernel parameters and their offsets and sizes do not need
to be specified as that information is retrieved directly from the kernel's image.
2) Kernel parameters for non-cooperative kernels can also be packaged by the application into a single
buffer that is passed in via \p extra. This places the burden on the application of knowing each
kernel parameter's size and alignment/padding within the buffer. The \p extra parameter exists
to allow this function to take additional less commonly used arguments. \p extra specifies
a list of names of extra settings and their corresponding values. Each extra setting name is
immediately followed by the corresponding value. The list must be terminated with either NULL or
CU_LAUNCH_PARAM_END.
- ::CU_LAUNCH_PARAM_END, which indicates the end of the \p extra
array;
- ::CU_LAUNCH_PARAM_BUFFER_POINTER, which specifies that the next
value in \p extra will be a pointer to a buffer
containing all the kernel parameters for launching kernel
\p func;
- ::CU_LAUNCH_PARAM_BUFFER_SIZE, which specifies that the next
value in \p extra will be a pointer to a size_t
containing the size of the buffer specified with
::CU_LAUNCH_PARAM_BUFFER_POINTER;
The error ::CUDA_ERROR_INVALID_VALUE will be returned if kernel parameters are specified with both
\p kernelParams and \p extra (i.e. both \p kernelParams and \p extra are non-NULL).
::CUDA_ERROR_INVALID_VALUE will be returned if \p extra is used for a cooperative kernel.
The \p kernelParams or \p extra array, as well as the argument values it points to,
are copied during this call.
\note Kernels launched using graphs must not use texture and surface references. Reading or
writing through any texture or surface reference is undefined behavior.
This restriction does not apply to texture and surface objects.phGraphNode
- - Returns newly created nodehGraph
- - Graph to which to add the nodedependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciesnodeParams
- - Parameters for the GPU execution node::cuGraphAddNode,
::cuLaunchKernel,
::cuLaunchCooperativeKernel,
::cuGraphKernelNodeGetParams,
::cuGraphKernelNodeSetParams,
::cuGraphCreate,
::cuGraphDestroyNode,
::cuGraphAddChildGraphNode,
::cuGraphAddEmptyNode,
::cuGraphAddHostNode,
::cuGraphAddMemcpyNode,
::cuGraphAddMemsetNode
@Cast(value="CUresult") public static int cuGraphKernelNodeGetParams(CUgraphNode_st hNode, @Cast(value="CUDA_KERNEL_NODE_PARAMS*") CUDA_KERNEL_NODE_PARAMS_v2 nodeParams)
hNode
- - Node to get the parameters fornodeParams
- - Pointer to return the parameters::cuLaunchKernel,
::cuGraphAddKernelNode,
::cuGraphKernelNodeSetParams
@Cast(value="CUresult") public static int cuGraphKernelNodeSetParams(CUgraphNode_st hNode, @Cast(value="const CUDA_KERNEL_NODE_PARAMS*") CUDA_KERNEL_NODE_PARAMS_v2 nodeParams)
hNode
- - Node to set the parameters fornodeParams
- - Parameters to copy::cuGraphNodeSetParams,
::cuLaunchKernel,
::cuGraphAddKernelNode,
::cuGraphKernelNodeGetParams
@Cast(value="CUresult") public static int cuGraphAddMemcpyNode(@ByPtrPtr CUgraphNode_st phGraphNode, CUgraph_st hGraph, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st dependencies, @Cast(value="size_t") long numDependencies, @Cast(value="const CUDA_MEMCPY3D*") CUDA_MEMCPY3D_v2 copyParams, CUctx_st ctx)
phGraphNode
- - Returns newly created nodehGraph
- - Graph to which to add the nodedependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciescopyParams
- - Parameters for the memory copyctx
- - Context on which to run the node::cuGraphAddNode,
::cuMemcpy3D,
::cuGraphMemcpyNodeGetParams,
::cuGraphMemcpyNodeSetParams,
::cuGraphCreate,
::cuGraphDestroyNode,
::cuGraphAddChildGraphNode,
::cuGraphAddEmptyNode,
::cuGraphAddKernelNode,
::cuGraphAddHostNode,
::cuGraphAddMemsetNode
@Cast(value="CUresult") public static int cuGraphMemcpyNodeGetParams(CUgraphNode_st hNode, @Cast(value="CUDA_MEMCPY3D*") CUDA_MEMCPY3D_v2 nodeParams)
hNode
- - Node to get the parameters fornodeParams
- - Pointer to return the parameters::cuMemcpy3D,
::cuGraphAddMemcpyNode,
::cuGraphMemcpyNodeSetParams
@Cast(value="CUresult") public static int cuGraphMemcpyNodeSetParams(CUgraphNode_st hNode, @Cast(value="const CUDA_MEMCPY3D*") CUDA_MEMCPY3D_v2 nodeParams)
hNode
- - Node to set the parameters fornodeParams
- - Parameters to copy::cuGraphNodeSetParams,
::cuMemcpy3D,
::cuGraphAddMemcpyNode,
::cuGraphMemcpyNodeGetParams
@Cast(value="CUresult") public static int cuGraphAddMemsetNode(@ByPtrPtr CUgraphNode_st phGraphNode, CUgraph_st hGraph, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st dependencies, @Cast(value="size_t") long numDependencies, @Cast(value="const CUDA_MEMSET_NODE_PARAMS*") CUDA_MEMSET_NODE_PARAMS_v1 memsetParams, CUctx_st ctx)
phGraphNode
- - Returns newly created nodehGraph
- - Graph to which to add the nodedependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciesmemsetParams
- - Parameters for the memory setctx
- - Context on which to run the node::cuGraphAddNode,
::cuMemsetD2D32,
::cuGraphMemsetNodeGetParams,
::cuGraphMemsetNodeSetParams,
::cuGraphCreate,
::cuGraphDestroyNode,
::cuGraphAddChildGraphNode,
::cuGraphAddEmptyNode,
::cuGraphAddKernelNode,
::cuGraphAddHostNode,
::cuGraphAddMemcpyNode
@Cast(value="CUresult") public static int cuGraphMemsetNodeGetParams(CUgraphNode_st hNode, @Cast(value="CUDA_MEMSET_NODE_PARAMS*") CUDA_MEMSET_NODE_PARAMS_v1 nodeParams)
hNode
- - Node to get the parameters fornodeParams
- - Pointer to return the parameters::cuMemsetD2D32,
::cuGraphAddMemsetNode,
::cuGraphMemsetNodeSetParams
@Cast(value="CUresult") public static int cuGraphMemsetNodeSetParams(CUgraphNode_st hNode, @Cast(value="const CUDA_MEMSET_NODE_PARAMS*") CUDA_MEMSET_NODE_PARAMS_v1 nodeParams)
hNode
- - Node to set the parameters fornodeParams
- - Parameters to copy::cuGraphNodeSetParams,
::cuMemsetD2D32,
::cuGraphAddMemsetNode,
::cuGraphMemsetNodeGetParams
@Cast(value="CUresult") public static int cuGraphAddHostNode(@ByPtrPtr CUgraphNode_st phGraphNode, CUgraph_st hGraph, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st dependencies, @Cast(value="size_t") long numDependencies, @Cast(value="const CUDA_HOST_NODE_PARAMS*") CUDA_HOST_NODE_PARAMS_v1 nodeParams)
phGraphNode
- - Returns newly created nodehGraph
- - Graph to which to add the nodedependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciesnodeParams
- - Parameters for the host node::cuGraphAddNode,
::cuLaunchHostFunc,
::cuGraphHostNodeGetParams,
::cuGraphHostNodeSetParams,
::cuGraphCreate,
::cuGraphDestroyNode,
::cuGraphAddChildGraphNode,
::cuGraphAddEmptyNode,
::cuGraphAddKernelNode,
::cuGraphAddMemcpyNode,
::cuGraphAddMemsetNode
@Cast(value="CUresult") public static int cuGraphHostNodeGetParams(CUgraphNode_st hNode, @Cast(value="CUDA_HOST_NODE_PARAMS*") CUDA_HOST_NODE_PARAMS_v1 nodeParams)
hNode
- - Node to get the parameters fornodeParams
- - Pointer to return the parameters::cuLaunchHostFunc,
::cuGraphAddHostNode,
::cuGraphHostNodeSetParams
@Cast(value="CUresult") public static int cuGraphHostNodeSetParams(CUgraphNode_st hNode, @Cast(value="const CUDA_HOST_NODE_PARAMS*") CUDA_HOST_NODE_PARAMS_v1 nodeParams)
hNode
- - Node to set the parameters fornodeParams
- - Parameters to copy::cuGraphNodeSetParams,
::cuLaunchHostFunc,
::cuGraphAddHostNode,
::cuGraphHostNodeGetParams
@Cast(value="CUresult") public static int cuGraphAddChildGraphNode(@ByPtrPtr CUgraphNode_st phGraphNode, CUgraph_st hGraph, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st dependencies, @Cast(value="size_t") long numDependencies, CUgraph_st childGraph)
phGraphNode
- - Returns newly created nodehGraph
- - Graph to which to add the nodedependencies
- - Dependencies of the nodenumDependencies
- - Number of dependencieschildGraph
- - The graph to clone into this node::cuGraphAddNode,
::cuGraphChildGraphNodeGetGraph,
::cuGraphCreate,
::cuGraphDestroyNode,
::cuGraphAddEmptyNode,
::cuGraphAddKernelNode,
::cuGraphAddHostNode,
::cuGraphAddMemcpyNode,
::cuGraphAddMemsetNode,
::cuGraphClone
@Cast(value="CUresult") public static int cuGraphChildGraphNodeGetGraph(CUgraphNode_st hNode, @ByPtrPtr CUgraph_st phGraph)
hNode
- - Node to get the embedded graph forphGraph
- - Location to store a handle to the graph::cuGraphAddChildGraphNode,
::cuGraphNodeFindInClone
@Cast(value="CUresult") public static int cuGraphAddEmptyNode(@ByPtrPtr CUgraphNode_st phGraphNode, CUgraph_st hGraph, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st dependencies, @Cast(value="size_t") long numDependencies)
phGraphNode
- - Returns newly created nodehGraph
- - Graph to which to add the nodedependencies
- - Dependencies of the nodenumDependencies
- - Number of dependencies::cuGraphAddNode,
::cuGraphCreate,
::cuGraphDestroyNode,
::cuGraphAddChildGraphNode,
::cuGraphAddKernelNode,
::cuGraphAddHostNode,
::cuGraphAddMemcpyNode,
::cuGraphAddMemsetNode
@Cast(value="CUresult") public static int cuGraphAddEventRecordNode(@ByPtrPtr CUgraphNode_st phGraphNode, CUgraph_st hGraph, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st dependencies, @Cast(value="size_t") long numDependencies, CUevent_st event)
phGraphNode
- - Returns newly created nodehGraph
- - Graph to which to add the nodedependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciesevent
- - Event for the node::cuGraphAddNode,
::cuGraphAddEventWaitNode,
::cuEventRecordWithFlags,
::cuStreamWaitEvent,
::cuGraphCreate,
::cuGraphDestroyNode,
::cuGraphAddChildGraphNode,
::cuGraphAddEmptyNode,
::cuGraphAddKernelNode,
::cuGraphAddMemcpyNode,
::cuGraphAddMemsetNode
@Cast(value="CUresult") public static int cuGraphEventRecordNodeGetEvent(CUgraphNode_st hNode, @ByPtrPtr CUevent_st event_out)
hNode
- - Node to get the event forevent_out
- - Pointer to return the event::cuGraphAddEventRecordNode,
::cuGraphEventRecordNodeSetEvent,
::cuGraphEventWaitNodeGetEvent,
::cuEventRecordWithFlags,
::cuStreamWaitEvent
@Cast(value="CUresult") public static int cuGraphEventRecordNodeSetEvent(CUgraphNode_st hNode, CUevent_st event)
hNode
- - Node to set the event forevent
- - Event to use::cuGraphNodeSetParams,
::cuGraphAddEventRecordNode,
::cuGraphEventRecordNodeGetEvent,
::cuGraphEventWaitNodeSetEvent,
::cuEventRecordWithFlags,
::cuStreamWaitEvent
@Cast(value="CUresult") public static int cuGraphAddEventWaitNode(@ByPtrPtr CUgraphNode_st phGraphNode, CUgraph_st hGraph, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st dependencies, @Cast(value="size_t") long numDependencies, CUevent_st event)
phGraphNode
- - Returns newly created nodehGraph
- - Graph to which to add the nodedependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciesevent
- - Event for the node::cuGraphAddNode,
::cuGraphAddEventRecordNode,
::cuEventRecordWithFlags,
::cuStreamWaitEvent,
::cuGraphCreate,
::cuGraphDestroyNode,
::cuGraphAddChildGraphNode,
::cuGraphAddEmptyNode,
::cuGraphAddKernelNode,
::cuGraphAddMemcpyNode,
::cuGraphAddMemsetNode
@Cast(value="CUresult") public static int cuGraphEventWaitNodeGetEvent(CUgraphNode_st hNode, @ByPtrPtr CUevent_st event_out)
hNode
- - Node to get the event forevent_out
- - Pointer to return the event::cuGraphAddEventWaitNode,
::cuGraphEventWaitNodeSetEvent,
::cuGraphEventRecordNodeGetEvent,
::cuEventRecordWithFlags,
::cuStreamWaitEvent
@Cast(value="CUresult") public static int cuGraphEventWaitNodeSetEvent(CUgraphNode_st hNode, CUevent_st event)
hNode
- - Node to set the event forevent
- - Event to use::cuGraphNodeSetParams,
::cuGraphAddEventWaitNode,
::cuGraphEventWaitNodeGetEvent,
::cuGraphEventRecordNodeSetEvent,
::cuEventRecordWithFlags,
::cuStreamWaitEvent
@Cast(value="CUresult") public static int cuGraphAddExternalSemaphoresSignalNode(@ByPtrPtr CUgraphNode_st phGraphNode, CUgraph_st hGraph, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st dependencies, @Cast(value="size_t") long numDependencies, @Cast(value="const CUDA_EXT_SEM_SIGNAL_NODE_PARAMS*") CUDA_EXT_SEM_SIGNAL_NODE_PARAMS_v1 nodeParams)
phGraphNode
- - Returns newly created nodehGraph
- - Graph to which to add the nodedependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciesnodeParams
- - Parameters for the node::cuGraphAddNode,
::cuGraphExternalSemaphoresSignalNodeGetParams,
::cuGraphExternalSemaphoresSignalNodeSetParams,
::cuGraphExecExternalSemaphoresSignalNodeSetParams,
::cuGraphAddExternalSemaphoresWaitNode,
::cuImportExternalSemaphore,
::cuSignalExternalSemaphoresAsync,
::cuWaitExternalSemaphoresAsync,
::cuGraphCreate,
::cuGraphDestroyNode,
::cuGraphAddEventRecordNode,
::cuGraphAddEventWaitNode,
::cuGraphAddChildGraphNode,
::cuGraphAddEmptyNode,
::cuGraphAddKernelNode,
::cuGraphAddMemcpyNode,
::cuGraphAddMemsetNode
@Cast(value="CUresult") public static int cuGraphExternalSemaphoresSignalNodeGetParams(CUgraphNode_st hNode, @Cast(value="CUDA_EXT_SEM_SIGNAL_NODE_PARAMS*") CUDA_EXT_SEM_SIGNAL_NODE_PARAMS_v1 params_out)
hNode
- - Node to get the parameters forparams_out
- - Pointer to return the parameters::cuLaunchKernel,
::cuGraphAddExternalSemaphoresSignalNode,
::cuGraphExternalSemaphoresSignalNodeSetParams,
::cuGraphAddExternalSemaphoresWaitNode,
::cuSignalExternalSemaphoresAsync,
::cuWaitExternalSemaphoresAsync
@Cast(value="CUresult") public static int cuGraphExternalSemaphoresSignalNodeSetParams(CUgraphNode_st hNode, @Cast(value="const CUDA_EXT_SEM_SIGNAL_NODE_PARAMS*") CUDA_EXT_SEM_SIGNAL_NODE_PARAMS_v1 nodeParams)
hNode
- - Node to set the parameters fornodeParams
- - Parameters to copy::cuGraphNodeSetParams,
::cuGraphAddExternalSemaphoresSignalNode,
::cuGraphExternalSemaphoresSignalNodeSetParams,
::cuGraphAddExternalSemaphoresWaitNode,
::cuSignalExternalSemaphoresAsync,
::cuWaitExternalSemaphoresAsync
@Cast(value="CUresult") public static int cuGraphAddExternalSemaphoresWaitNode(@ByPtrPtr CUgraphNode_st phGraphNode, CUgraph_st hGraph, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st dependencies, @Cast(value="size_t") long numDependencies, @Cast(value="const CUDA_EXT_SEM_WAIT_NODE_PARAMS*") CUDA_EXT_SEM_WAIT_NODE_PARAMS_v1 nodeParams)
phGraphNode
- - Returns newly created nodehGraph
- - Graph to which to add the nodedependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciesnodeParams
- - Parameters for the node::cuGraphAddNode,
::cuGraphExternalSemaphoresWaitNodeGetParams,
::cuGraphExternalSemaphoresWaitNodeSetParams,
::cuGraphExecExternalSemaphoresWaitNodeSetParams,
::cuGraphAddExternalSemaphoresSignalNode,
::cuImportExternalSemaphore,
::cuSignalExternalSemaphoresAsync,
::cuWaitExternalSemaphoresAsync,
::cuGraphCreate,
::cuGraphDestroyNode,
::cuGraphAddEventRecordNode,
::cuGraphAddEventWaitNode,
::cuGraphAddChildGraphNode,
::cuGraphAddEmptyNode,
::cuGraphAddKernelNode,
::cuGraphAddMemcpyNode,
::cuGraphAddMemsetNode
@Cast(value="CUresult") public static int cuGraphExternalSemaphoresWaitNodeGetParams(CUgraphNode_st hNode, @Cast(value="CUDA_EXT_SEM_WAIT_NODE_PARAMS*") CUDA_EXT_SEM_WAIT_NODE_PARAMS_v1 params_out)
hNode
- - Node to get the parameters forparams_out
- - Pointer to return the parameters::cuLaunchKernel,
::cuGraphAddExternalSemaphoresWaitNode,
::cuGraphExternalSemaphoresWaitNodeSetParams,
::cuGraphAddExternalSemaphoresWaitNode,
::cuSignalExternalSemaphoresAsync,
::cuWaitExternalSemaphoresAsync
@Cast(value="CUresult") public static int cuGraphExternalSemaphoresWaitNodeSetParams(CUgraphNode_st hNode, @Cast(value="const CUDA_EXT_SEM_WAIT_NODE_PARAMS*") CUDA_EXT_SEM_WAIT_NODE_PARAMS_v1 nodeParams)
hNode
- - Node to set the parameters fornodeParams
- - Parameters to copy::cuGraphNodeSetParams,
::cuGraphAddExternalSemaphoresWaitNode,
::cuGraphExternalSemaphoresWaitNodeSetParams,
::cuGraphAddExternalSemaphoresWaitNode,
::cuSignalExternalSemaphoresAsync,
::cuWaitExternalSemaphoresAsync
@Cast(value="CUresult") public static int cuGraphAddBatchMemOpNode(@ByPtrPtr CUgraphNode_st phGraphNode, CUgraph_st hGraph, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st dependencies, @Cast(value="size_t") long numDependencies, @Cast(value="const CUDA_BATCH_MEM_OP_NODE_PARAMS*") CUDA_BATCH_MEM_OP_NODE_PARAMS_v1 nodeParams)
phGraphNode
- - Returns newly created nodehGraph
- - Graph to which to add the nodedependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciesnodeParams
- - Parameters for the node::cuGraphAddNode,
::cuStreamBatchMemOp,
::cuStreamWaitValue32,
::cuStreamWriteValue32,
::cuStreamWaitValue64,
::cuStreamWriteValue64,
::cuGraphBatchMemOpNodeGetParams,
::cuGraphBatchMemOpNodeSetParams,
::cuGraphCreate,
::cuGraphDestroyNode,
::cuGraphAddChildGraphNode,
::cuGraphAddEmptyNode,
::cuGraphAddKernelNode,
::cuGraphAddMemcpyNode,
::cuGraphAddMemsetNode
@Cast(value="CUresult") public static int cuGraphBatchMemOpNodeGetParams(CUgraphNode_st hNode, @Cast(value="CUDA_BATCH_MEM_OP_NODE_PARAMS*") CUDA_BATCH_MEM_OP_NODE_PARAMS_v1 nodeParams_out)
hNode
- - Node to get the parameters fornodeParams_out
- - Pointer to return the parameters::cuStreamBatchMemOp,
::cuGraphAddBatchMemOpNode,
::cuGraphBatchMemOpNodeSetParams
@Cast(value="CUresult") public static int cuGraphBatchMemOpNodeSetParams(CUgraphNode_st hNode, @Cast(value="const CUDA_BATCH_MEM_OP_NODE_PARAMS*") CUDA_BATCH_MEM_OP_NODE_PARAMS_v1 nodeParams)
hNode
- - Node to set the parameters fornodeParams
- - Parameters to copy::cuGraphNodeSetParams,
::cuStreamBatchMemOp,
::cuGraphAddBatchMemOpNode,
::cuGraphBatchMemOpNodeGetParams
@Cast(value="CUresult") public static int cuGraphExecBatchMemOpNodeSetParams(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, @Cast(value="const CUDA_BATCH_MEM_OP_NODE_PARAMS*") CUDA_BATCH_MEM_OP_NODE_PARAMS_v1 nodeParams)
hGraphExec
- - The executable graph in which to set the specified nodehNode
- - Batch mem op node from the graph from which graphExec was instantiatednodeParams
- - Updated Parameters to set::cuGraphExecNodeSetParams,
::cuStreamBatchMemOp,
::cuGraphAddBatchMemOpNode,
::cuGraphBatchMemOpNodeGetParams,
::cuGraphBatchMemOpNodeSetParams,
::cuGraphInstantiate
@Cast(value="CUresult") public static int cuGraphAddMemAllocNode(@ByPtrPtr CUgraphNode_st phGraphNode, CUgraph_st hGraph, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st dependencies, @Cast(value="size_t") long numDependencies, @Cast(value="CUDA_MEM_ALLOC_NODE_PARAMS*") CUDA_MEM_ALLOC_NODE_PARAMS_v1 nodeParams)
phGraphNode
- - Returns newly created nodehGraph
- - Graph to which to add the nodedependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciesnodeParams
- - Parameters for the node
When ::cuGraphAddMemAllocNode creates an allocation node, it returns the address of the allocation in
\p nodeParams.dptr. The allocation's address remains fixed across instantiations and launches.
If the allocation is freed in the same graph, by creating a free node using ::cuGraphAddMemFreeNode,
the allocation can be accessed by nodes ordered after the allocation node but before the free node.
These allocations cannot be freed outside the owning graph, and they can only be freed once in the
owning graph.
If the allocation is not freed in the same graph, then it can be accessed not only by nodes in the
graph which are ordered after the allocation node, but also by stream operations ordered after the
graph's execution but before the allocation is freed.
Allocations which are not freed in the same graph can be freed by:
- passing the allocation to ::cuMemFreeAsync or ::cuMemFree;
- launching a graph with a free node for that allocation; or
- specifying ::CUDA_GRAPH_INSTANTIATE_FLAG_AUTO_FREE_ON_LAUNCH during instantiation, which makes
each launch behave as though it called ::cuMemFreeAsync for every unfreed allocation.
It is not possible to free an allocation in both the owning graph and another graph. If the allocation
is freed in the same graph, a free node cannot be added to another graph. If the allocation is freed
in another graph, a free node can no longer be added to the owning graph.
The following restrictions apply to graphs which contain allocation and/or memory free nodes:
- Nodes and edges of the graph cannot be deleted.
- The graph cannot be used in a child node.
- Only one instantiation of the graph may exist at any point in time.
- The graph cannot be cloned.::cuGraphAddNode,
::cuGraphAddMemFreeNode,
::cuGraphMemAllocNodeGetParams,
::cuDeviceGraphMemTrim,
::cuDeviceGetGraphMemAttribute,
::cuDeviceSetGraphMemAttribute,
::cuMemAllocAsync,
::cuMemFreeAsync,
::cuGraphCreate,
::cuGraphDestroyNode,
::cuGraphAddChildGraphNode,
::cuGraphAddEmptyNode,
::cuGraphAddEventRecordNode,
::cuGraphAddEventWaitNode,
::cuGraphAddExternalSemaphoresSignalNode,
::cuGraphAddExternalSemaphoresWaitNode,
::cuGraphAddKernelNode,
::cuGraphAddMemcpyNode,
::cuGraphAddMemsetNode
@Cast(value="CUresult") public static int cuGraphMemAllocNodeGetParams(CUgraphNode_st hNode, @Cast(value="CUDA_MEM_ALLOC_NODE_PARAMS*") CUDA_MEM_ALLOC_NODE_PARAMS_v1 params_out)
hNode
- - Node to get the parameters forparams_out
- - Pointer to return the parameters::cuGraphAddMemAllocNode,
::cuGraphMemFreeNodeGetParams
@Cast(value="CUresult") public static int cuGraphAddMemFreeNode(@ByPtrPtr CUgraphNode_st phGraphNode, CUgraph_st hGraph, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st dependencies, @Cast(value="size_t") long numDependencies, @Cast(value="CUdeviceptr") long dptr)
phGraphNode
- - Returns newly created nodehGraph
- - Graph to which to add the nodedependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciesdptr
- - Address of memory to free
::cuGraphAddMemFreeNode will return ::CUDA_ERROR_INVALID_VALUE if the user attempts to free:
- an allocation twice in the same graph.
- an address that was not returned by an allocation node.
- an invalid address.
The following restrictions apply to graphs which contain allocation and/or memory free nodes:
- Nodes and edges of the graph cannot be deleted.
- The graph cannot be used in a child node.
- Only one instantiation of the graph may exist at any point in time.
- The graph cannot be cloned.::cuGraphAddNode,
::cuGraphAddMemAllocNode,
::cuGraphMemFreeNodeGetParams,
::cuDeviceGraphMemTrim,
::cuDeviceGetGraphMemAttribute,
::cuDeviceSetGraphMemAttribute,
::cuMemAllocAsync,
::cuMemFreeAsync,
::cuGraphCreate,
::cuGraphDestroyNode,
::cuGraphAddChildGraphNode,
::cuGraphAddEmptyNode,
::cuGraphAddEventRecordNode,
::cuGraphAddEventWaitNode,
::cuGraphAddExternalSemaphoresSignalNode,
::cuGraphAddExternalSemaphoresWaitNode,
::cuGraphAddKernelNode,
::cuGraphAddMemcpyNode,
::cuGraphAddMemsetNode
@Cast(value="CUresult") public static int cuGraphMemFreeNodeGetParams(CUgraphNode_st hNode, @Cast(value="CUdeviceptr*") LongPointer dptr_out)
hNode
- - Node to get the parameters fordptr_out
- - Pointer to return the device address::cuGraphAddMemFreeNode,
::cuGraphMemAllocNodeGetParams
@Cast(value="CUresult") public static int cuGraphMemFreeNodeGetParams(CUgraphNode_st hNode, @Cast(value="CUdeviceptr*") LongBuffer dptr_out)
@Cast(value="CUresult") public static int cuGraphMemFreeNodeGetParams(CUgraphNode_st hNode, @Cast(value="CUdeviceptr*") long[] dptr_out)
@Cast(value="CUresult") public static int cuDeviceGraphMemTrim(@Cast(value="CUdevice") int device)
device
- - The device for which cached memory should be freed.::cuGraphAddMemAllocNode,
::cuGraphAddMemFreeNode,
::cuDeviceSetGraphMemAttribute,
::cuDeviceGetGraphMemAttribute
@Cast(value="CUresult") public static int cuDeviceGetGraphMemAttribute(@Cast(value="CUdevice") int device, @Cast(value="CUgraphMem_attribute") int attr, Pointer value)
device
- - Specifies the scope of the queryattr
- - attribute to getvalue
- - retrieved value::cuDeviceSetGraphMemAttribute,
::cuGraphAddMemAllocNode,
::cuGraphAddMemFreeNode
@Cast(value="CUresult") public static int cuDeviceSetGraphMemAttribute(@Cast(value="CUdevice") int device, @Cast(value="CUgraphMem_attribute") int attr, Pointer value)
device
- - Specifies the scope of the queryattr
- - attribute to getvalue
- - pointer to value to set::cuDeviceGetGraphMemAttribute,
::cuGraphAddMemAllocNode,
::cuGraphAddMemFreeNode
@Cast(value="CUresult") public static int cuGraphClone(@ByPtrPtr CUgraph_st phGraphClone, CUgraph_st originalGraph)
phGraphClone
- - Returns newly created cloned graphoriginalGraph
- - Graph to clone::cuGraphCreate,
::cuGraphNodeFindInClone
@Cast(value="CUresult") public static int cuGraphNodeFindInClone(@ByPtrPtr CUgraphNode_st phNode, CUgraphNode_st hOriginalNode, CUgraph_st hClonedGraph)
phNode
- - Returns handle to the cloned nodehOriginalNode
- - Handle to the original nodehClonedGraph
- - Cloned graph to query::cuGraphClone
@Cast(value="CUresult") public static int cuGraphNodeGetType(CUgraphNode_st hNode, @Cast(value="CUgraphNodeType*") IntPointer type)
hNode
- - Node to querytype
- - Pointer to return the node type::cuGraphGetNodes,
::cuGraphGetRootNodes,
::cuGraphChildGraphNodeGetGraph,
::cuGraphKernelNodeGetParams,
::cuGraphKernelNodeSetParams,
::cuGraphHostNodeGetParams,
::cuGraphHostNodeSetParams,
::cuGraphMemcpyNodeGetParams,
::cuGraphMemcpyNodeSetParams,
::cuGraphMemsetNodeGetParams,
::cuGraphMemsetNodeSetParams
@Cast(value="CUresult") public static int cuGraphNodeGetType(CUgraphNode_st hNode, @Cast(value="CUgraphNodeType*") IntBuffer type)
@Cast(value="CUresult") public static int cuGraphNodeGetType(CUgraphNode_st hNode, @Cast(value="CUgraphNodeType*") int[] type)
@Cast(value="CUresult") public static int cuGraphGetNodes(CUgraph_st hGraph, @ByPtrPtr CUgraphNode_st nodes, @Cast(value="size_t*") SizeTPointer numNodes)
hGraph
- - Graph to querynodes
- - Pointer to return the nodesnumNodes
- - See description::cuGraphCreate,
::cuGraphGetRootNodes,
::cuGraphGetEdges,
::cuGraphNodeGetType,
::cuGraphNodeGetDependencies,
::cuGraphNodeGetDependentNodes
@Cast(value="CUresult") public static int cuGraphGetRootNodes(CUgraph_st hGraph, @ByPtrPtr CUgraphNode_st rootNodes, @Cast(value="size_t*") SizeTPointer numRootNodes)
hGraph
- - Graph to queryrootNodes
- - Pointer to return the root nodesnumRootNodes
- - See description::cuGraphCreate,
::cuGraphGetNodes,
::cuGraphGetEdges,
::cuGraphNodeGetType,
::cuGraphNodeGetDependencies,
::cuGraphNodeGetDependentNodes
@Cast(value="CUresult") public static int cuGraphGetEdges(CUgraph_st hGraph, @ByPtrPtr CUgraphNode_st from, @ByPtrPtr CUgraphNode_st to, @Cast(value="size_t*") SizeTPointer numEdges)
hGraph
- - Graph to get the edges fromfrom
- - Location to return edge endpointsto
- - Location to return edge endpointsnumEdges
- - See description::cuGraphGetNodes,
::cuGraphGetRootNodes,
::cuGraphAddDependencies,
::cuGraphRemoveDependencies,
::cuGraphNodeGetDependencies,
::cuGraphNodeGetDependentNodes
@Cast(value="CUresult") public static int cuGraphGetEdges_v2(CUgraph_st hGraph, @ByPtrPtr CUgraphNode_st from, @ByPtrPtr CUgraphNode_st to, CUgraphEdgeData edgeData, @Cast(value="size_t*") SizeTPointer numEdges)
hGraph
- - Graph to get the edges fromfrom
- - Location to return edge endpointsto
- - Location to return edge endpointsedgeData
- - Optional location to return edge datanumEdges
- - See description::cuGraphGetNodes,
::cuGraphGetRootNodes,
::cuGraphAddDependencies,
::cuGraphRemoveDependencies,
::cuGraphNodeGetDependencies,
::cuGraphNodeGetDependentNodes
@Cast(value="CUresult") public static int cuGraphNodeGetDependencies(CUgraphNode_st hNode, @ByPtrPtr CUgraphNode_st dependencies, @Cast(value="size_t*") SizeTPointer numDependencies)
hNode
- - Node to querydependencies
- - Pointer to return the dependenciesnumDependencies
- - See description::cuGraphNodeGetDependentNodes,
::cuGraphGetNodes,
::cuGraphGetRootNodes,
::cuGraphGetEdges,
::cuGraphAddDependencies,
::cuGraphRemoveDependencies
@Cast(value="CUresult") public static int cuGraphNodeGetDependencies_v2(CUgraphNode_st hNode, @ByPtrPtr CUgraphNode_st dependencies, CUgraphEdgeData edgeData, @Cast(value="size_t*") SizeTPointer numDependencies)
hNode
- - Node to querydependencies
- - Pointer to return the dependenciesedgeData
- - Optional array to return edge data for each dependencynumDependencies
- - See description::cuGraphNodeGetDependentNodes,
::cuGraphGetNodes,
::cuGraphGetRootNodes,
::cuGraphGetEdges,
::cuGraphAddDependencies,
::cuGraphRemoveDependencies
@Cast(value="CUresult") public static int cuGraphNodeGetDependentNodes(CUgraphNode_st hNode, @ByPtrPtr CUgraphNode_st dependentNodes, @Cast(value="size_t*") SizeTPointer numDependentNodes)
hNode
- - Node to querydependentNodes
- - Pointer to return the dependent nodesnumDependentNodes
- - See description::cuGraphNodeGetDependencies,
::cuGraphGetNodes,
::cuGraphGetRootNodes,
::cuGraphGetEdges,
::cuGraphAddDependencies,
::cuGraphRemoveDependencies
@Cast(value="CUresult") public static int cuGraphNodeGetDependentNodes_v2(CUgraphNode_st hNode, @ByPtrPtr CUgraphNode_st dependentNodes, CUgraphEdgeData edgeData, @Cast(value="size_t*") SizeTPointer numDependentNodes)
hNode
- - Node to querydependentNodes
- - Pointer to return the dependent nodesedgeData
- - Optional pointer to return edge data for dependent nodesnumDependentNodes
- - See description::cuGraphNodeGetDependencies,
::cuGraphGetNodes,
::cuGraphGetRootNodes,
::cuGraphGetEdges,
::cuGraphAddDependencies,
::cuGraphRemoveDependencies
@Cast(value="CUresult") public static int cuGraphAddDependencies(CUgraph_st hGraph, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st from, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st to, @Cast(value="size_t") long numDependencies)
hGraph
- - Graph to which dependencies are addedfrom
- - Array of nodes that provide the dependenciesto
- - Array of dependent nodesnumDependencies
- - Number of dependencies to be added::cuGraphRemoveDependencies,
::cuGraphGetEdges,
::cuGraphNodeGetDependencies,
::cuGraphNodeGetDependentNodes
@Cast(value="CUresult") public static int cuGraphAddDependencies_v2(CUgraph_st hGraph, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st from, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st to, @Const CUgraphEdgeData edgeData, @Cast(value="size_t") long numDependencies)
hGraph
- - Graph to which dependencies are addedfrom
- - Array of nodes that provide the dependenciesto
- - Array of dependent nodesedgeData
- - Optional array of edge data. If NULL, default (zeroed) edge data is assumed.numDependencies
- - Number of dependencies to be added::cuGraphRemoveDependencies,
::cuGraphGetEdges,
::cuGraphNodeGetDependencies,
::cuGraphNodeGetDependentNodes
@Cast(value="CUresult") public static int cuGraphRemoveDependencies(CUgraph_st hGraph, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st from, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st to, @Cast(value="size_t") long numDependencies)
hGraph
- - Graph from which to remove dependenciesfrom
- - Array of nodes that provide the dependenciesto
- - Array of dependent nodesnumDependencies
- - Number of dependencies to be removed::cuGraphAddDependencies,
::cuGraphGetEdges,
::cuGraphNodeGetDependencies,
::cuGraphNodeGetDependentNodes
@Cast(value="CUresult") public static int cuGraphRemoveDependencies_v2(CUgraph_st hGraph, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st from, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st to, @Const CUgraphEdgeData edgeData, @Cast(value="size_t") long numDependencies)
hGraph
- - Graph from which to remove dependenciesfrom
- - Array of nodes that provide the dependenciesto
- - Array of dependent nodesedgeData
- - Optional array of edge data. If NULL, edge data is assumed to
be default (zeroed).numDependencies
- - Number of dependencies to be removed::cuGraphAddDependencies,
::cuGraphGetEdges,
::cuGraphNodeGetDependencies,
::cuGraphNodeGetDependentNodes
@Cast(value="CUresult") public static int cuGraphDestroyNode(CUgraphNode_st hNode)
hNode
- - Node to remove::cuGraphAddChildGraphNode,
::cuGraphAddEmptyNode,
::cuGraphAddKernelNode,
::cuGraphAddHostNode,
::cuGraphAddMemcpyNode,
::cuGraphAddMemsetNode
@Cast(value="CUresult") public static int cuGraphInstantiate(@ByPtrPtr CUgraphExec_st phGraphExec, CUgraph_st hGraph, @Cast(value="unsigned long long") long flags)
phGraphExec
- - Returns instantiated graphhGraph
- - Graph to instantiateflags
- - Flags to control instantiation. See ::CUgraphInstantiate_flags.::cuGraphInstantiate,
::cuGraphCreate,
::cuGraphUpload,
::cuGraphLaunch,
::cuGraphExecDestroy
@Cast(value="CUresult") public static int cuGraphInstantiateWithParams(@ByPtrPtr CUgraphExec_st phGraphExec, CUgraph_st hGraph, CUDA_GRAPH_INSTANTIATE_PARAMS instantiateParams)
typedef struct {
cuuint64_t flags;
CUstream hUploadStream;
CUgraphNode hErrNode_out;
CUgraphInstantiateResult result_out;
} CUDA_GRAPH_INSTANTIATE_PARAMS;
The \p flags field controls the behavior of instantiation and subsequent
graph launches. Valid flags are:
- ::CUDA_GRAPH_INSTANTIATE_FLAG_AUTO_FREE_ON_LAUNCH, which configures a
graph containing memory allocation nodes to automatically free any
unfreed memory allocations before the graph is relaunched.
- ::CUDA_GRAPH_INSTANTIATE_FLAG_UPLOAD, which will perform an upload of the graph
into \p hUploadStream once the graph has been instantiated.
- ::CUDA_GRAPH_INSTANTIATE_FLAG_DEVICE_LAUNCH, which configures the graph for launch
from the device. If this flag is passed, the executable graph handle returned can be
used to launch the graph from both the host and device. This flag can only be used
on platforms which support unified addressing. This flag cannot be used in
conjunction with ::CUDA_GRAPH_INSTANTIATE_FLAG_AUTO_FREE_ON_LAUNCH.
- ::CUDA_GRAPH_INSTANTIATE_FLAG_USE_NODE_PRIORITY, which causes the graph
to use the priorities from the per-node attributes rather than the priority
of the launch stream during execution. Note that priorities are only available
on kernel nodes, and are copied from stream priority during stream capture.
If \p hGraph contains any allocation or free nodes, there can be at most one
executable graph in existence for that graph at a time. An attempt to instantiate a
second executable graph before destroying the first with ::cuGraphExecDestroy will
result in an error.
If \p hGraph contains kernels which call device-side cudaGraphLaunch() from multiple
contexts, this will result in an error.
Graphs instantiated for launch on the device have additional restrictions which do not
apply to host graphs:
- The graph's nodes must reside on a single context.
- The graph can only contain kernel nodes, memcpy nodes, memset nodes, and child graph nodes.
- The graph cannot be empty and must contain at least one kernel, memcpy, or memset node.
Operation-specific restrictions are outlined below.
- Kernel nodes:
- Use of CUDA Dynamic Parallelism is not permitted.
- Cooperative launches are permitted as long as MPS is not in use.
- Memcpy nodes:
- Only copies involving device memory and/or pinned device-mapped host memory are permitted.
- Copies involving CUDA arrays are not permitted.
- Both operands must be accessible from the current context, and the current context must
match the context of other nodes in the graph.
In the event of an error, the \p result_out and \p hErrNode_out fields will contain more
information about the nature of the error. Possible error reporting includes:
- ::CUDA_GRAPH_INSTANTIATE_ERROR, if passed an invalid value or if an unexpected error occurred
which is described by the return value of the function. \p hErrNode_out will be set to NULL.
- ::CUDA_GRAPH_INSTANTIATE_INVALID_STRUCTURE, if the graph structure is invalid. \p hErrNode_out
will be set to one of the offending nodes.
- ::CUDA_GRAPH_INSTANTIATE_NODE_OPERATION_NOT_SUPPORTED, if the graph is instantiated for device
launch but contains a node of an unsupported node type, or a node which performs unsupported
operations, such as use of CUDA dynamic parallelism within a kernel node. \p hErrNode_out will
be set to this node.
- ::CUDA_GRAPH_INSTANTIATE_MULTIPLE_CTXS_NOT_SUPPORTED, if the graph is instantiated for device
launch but a node’s context differs from that of another node. This error can also be returned
if a graph is not instantiated for device launch and it contains kernels which call device-side
cudaGraphLaunch() from multiple contexts. \p hErrNode_out will be set to this node.
If instantiation is successful, \p result_out will be set to ::CUDA_GRAPH_INSTANTIATE_SUCCESS,
and \p hErrNode_out will be set to NULL.phGraphExec
- - Returns instantiated graphhGraph
- - Graph to instantiateinstantiateParams
- - Instantiation parameters::cuGraphCreate,
::cuGraphInstantiate,
::cuGraphExecDestroy
@Cast(value="CUresult") public static int cuGraphExecGetFlags(CUgraphExec_st hGraphExec, @Cast(value="cuuint64_t*") LongPointer flags)
hGraphExec
- - The executable graph to queryflags
- - Returns the instantiation flags::cuGraphInstantiate,
::cuGraphInstantiateWithParams
@Cast(value="CUresult") public static int cuGraphExecGetFlags(CUgraphExec_st hGraphExec, @Cast(value="cuuint64_t*") LongBuffer flags)
@Cast(value="CUresult") public static int cuGraphExecGetFlags(CUgraphExec_st hGraphExec, @Cast(value="cuuint64_t*") long[] flags)
@Cast(value="CUresult") public static int cuGraphExecKernelNodeSetParams(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, @Cast(value="const CUDA_KERNEL_NODE_PARAMS*") CUDA_KERNEL_NODE_PARAMS_v2 nodeParams)
hGraphExec
- - The executable graph in which to set the specified nodehNode
- - kernel node from the graph from which graphExec was instantiatednodeParams
- - Updated Parameters to set::cuGraphExecNodeSetParams,
::cuGraphAddKernelNode,
::cuGraphKernelNodeSetParams,
::cuGraphExecMemcpyNodeSetParams,
::cuGraphExecMemsetNodeSetParams,
::cuGraphExecHostNodeSetParams,
::cuGraphExecChildGraphNodeSetParams,
::cuGraphExecEventRecordNodeSetEvent,
::cuGraphExecEventWaitNodeSetEvent,
::cuGraphExecExternalSemaphoresSignalNodeSetParams,
::cuGraphExecExternalSemaphoresWaitNodeSetParams,
::cuGraphExecUpdate,
::cuGraphInstantiate
@Cast(value="CUresult") public static int cuGraphExecMemcpyNodeSetParams(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, @Cast(value="const CUDA_MEMCPY3D*") CUDA_MEMCPY3D_v2 copyParams, CUctx_st ctx)
hGraphExec
- - The executable graph in which to set the specified nodehNode
- - Memcpy node from the graph which was used to instantiate graphExeccopyParams
- - The updated parameters to setctx
- - Context on which to run the node::cuGraphExecNodeSetParams,
::cuGraphAddMemcpyNode,
::cuGraphMemcpyNodeSetParams,
::cuGraphExecKernelNodeSetParams,
::cuGraphExecMemsetNodeSetParams,
::cuGraphExecHostNodeSetParams,
::cuGraphExecChildGraphNodeSetParams,
::cuGraphExecEventRecordNodeSetEvent,
::cuGraphExecEventWaitNodeSetEvent,
::cuGraphExecExternalSemaphoresSignalNodeSetParams,
::cuGraphExecExternalSemaphoresWaitNodeSetParams,
::cuGraphExecUpdate,
::cuGraphInstantiate
@Cast(value="CUresult") public static int cuGraphExecMemsetNodeSetParams(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, @Cast(value="const CUDA_MEMSET_NODE_PARAMS*") CUDA_MEMSET_NODE_PARAMS_v1 memsetParams, CUctx_st ctx)
hGraphExec
- - The executable graph in which to set the specified nodehNode
- - Memset node from the graph which was used to instantiate graphExecmemsetParams
- - The updated parameters to setctx
- - Context on which to run the node::cuGraphExecNodeSetParams,
::cuGraphAddMemsetNode,
::cuGraphMemsetNodeSetParams,
::cuGraphExecKernelNodeSetParams,
::cuGraphExecMemcpyNodeSetParams,
::cuGraphExecHostNodeSetParams,
::cuGraphExecChildGraphNodeSetParams,
::cuGraphExecEventRecordNodeSetEvent,
::cuGraphExecEventWaitNodeSetEvent,
::cuGraphExecExternalSemaphoresSignalNodeSetParams,
::cuGraphExecExternalSemaphoresWaitNodeSetParams,
::cuGraphExecUpdate,
::cuGraphInstantiate
@Cast(value="CUresult") public static int cuGraphExecHostNodeSetParams(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, @Cast(value="const CUDA_HOST_NODE_PARAMS*") CUDA_HOST_NODE_PARAMS_v1 nodeParams)
hGraphExec
- - The executable graph in which to set the specified nodehNode
- - Host node from the graph which was used to instantiate graphExecnodeParams
- - The updated parameters to set::cuGraphExecNodeSetParams,
::cuGraphAddHostNode,
::cuGraphHostNodeSetParams,
::cuGraphExecKernelNodeSetParams,
::cuGraphExecMemcpyNodeSetParams,
::cuGraphExecMemsetNodeSetParams,
::cuGraphExecChildGraphNodeSetParams,
::cuGraphExecEventRecordNodeSetEvent,
::cuGraphExecEventWaitNodeSetEvent,
::cuGraphExecExternalSemaphoresSignalNodeSetParams,
::cuGraphExecExternalSemaphoresWaitNodeSetParams,
::cuGraphExecUpdate,
::cuGraphInstantiate
@Cast(value="CUresult") public static int cuGraphExecChildGraphNodeSetParams(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, CUgraph_st childGraph)
hGraphExec
- - The executable graph in which to set the specified nodehNode
- - Host node from the graph which was used to instantiate graphExecchildGraph
- - The graph supplying the updated parameters::cuGraphExecNodeSetParams,
::cuGraphAddChildGraphNode,
::cuGraphChildGraphNodeGetGraph,
::cuGraphExecKernelNodeSetParams,
::cuGraphExecMemcpyNodeSetParams,
::cuGraphExecMemsetNodeSetParams,
::cuGraphExecHostNodeSetParams,
::cuGraphExecEventRecordNodeSetEvent,
::cuGraphExecEventWaitNodeSetEvent,
::cuGraphExecExternalSemaphoresSignalNodeSetParams,
::cuGraphExecExternalSemaphoresWaitNodeSetParams,
::cuGraphExecUpdate,
::cuGraphInstantiate
@Cast(value="CUresult") public static int cuGraphExecEventRecordNodeSetEvent(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, CUevent_st event)
hGraphExec
- - The executable graph in which to set the specified nodehNode
- - event record node from the graph from which graphExec was instantiatedevent
- - Updated event to use::cuGraphExecNodeSetParams,
::cuGraphAddEventRecordNode,
::cuGraphEventRecordNodeGetEvent,
::cuGraphEventWaitNodeSetEvent,
::cuEventRecordWithFlags,
::cuStreamWaitEvent,
::cuGraphExecKernelNodeSetParams,
::cuGraphExecMemcpyNodeSetParams,
::cuGraphExecMemsetNodeSetParams,
::cuGraphExecHostNodeSetParams,
::cuGraphExecChildGraphNodeSetParams,
::cuGraphExecEventWaitNodeSetEvent,
::cuGraphExecExternalSemaphoresSignalNodeSetParams,
::cuGraphExecExternalSemaphoresWaitNodeSetParams,
::cuGraphExecUpdate,
::cuGraphInstantiate
@Cast(value="CUresult") public static int cuGraphExecEventWaitNodeSetEvent(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, CUevent_st event)
hGraphExec
- - The executable graph in which to set the specified nodehNode
- - event wait node from the graph from which graphExec was instantiatedevent
- - Updated event to use::cuGraphExecNodeSetParams,
::cuGraphAddEventWaitNode,
::cuGraphEventWaitNodeGetEvent,
::cuGraphEventRecordNodeSetEvent,
::cuEventRecordWithFlags,
::cuStreamWaitEvent,
::cuGraphExecKernelNodeSetParams,
::cuGraphExecMemcpyNodeSetParams,
::cuGraphExecMemsetNodeSetParams,
::cuGraphExecHostNodeSetParams,
::cuGraphExecChildGraphNodeSetParams,
::cuGraphExecEventRecordNodeSetEvent,
::cuGraphExecExternalSemaphoresSignalNodeSetParams,
::cuGraphExecExternalSemaphoresWaitNodeSetParams,
::cuGraphExecUpdate,
::cuGraphInstantiate
@Cast(value="CUresult") public static int cuGraphExecExternalSemaphoresSignalNodeSetParams(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, @Cast(value="const CUDA_EXT_SEM_SIGNAL_NODE_PARAMS*") CUDA_EXT_SEM_SIGNAL_NODE_PARAMS_v1 nodeParams)
hGraphExec
- - The executable graph in which to set the specified nodehNode
- - semaphore signal node from the graph from which graphExec was instantiatednodeParams
- - Updated Parameters to set::cuGraphExecNodeSetParams,
::cuGraphAddExternalSemaphoresSignalNode,
::cuImportExternalSemaphore,
::cuSignalExternalSemaphoresAsync,
::cuWaitExternalSemaphoresAsync,
::cuGraphExecKernelNodeSetParams,
::cuGraphExecMemcpyNodeSetParams,
::cuGraphExecMemsetNodeSetParams,
::cuGraphExecHostNodeSetParams,
::cuGraphExecChildGraphNodeSetParams,
::cuGraphExecEventRecordNodeSetEvent,
::cuGraphExecEventWaitNodeSetEvent,
::cuGraphExecExternalSemaphoresWaitNodeSetParams,
::cuGraphExecUpdate,
::cuGraphInstantiate
@Cast(value="CUresult") public static int cuGraphExecExternalSemaphoresWaitNodeSetParams(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, @Cast(value="const CUDA_EXT_SEM_WAIT_NODE_PARAMS*") CUDA_EXT_SEM_WAIT_NODE_PARAMS_v1 nodeParams)
hGraphExec
- - The executable graph in which to set the specified nodehNode
- - semaphore wait node from the graph from which graphExec was instantiatednodeParams
- - Updated Parameters to set::cuGraphExecNodeSetParams,
::cuGraphAddExternalSemaphoresWaitNode,
::cuImportExternalSemaphore,
::cuSignalExternalSemaphoresAsync,
::cuWaitExternalSemaphoresAsync,
::cuGraphExecKernelNodeSetParams,
::cuGraphExecMemcpyNodeSetParams,
::cuGraphExecMemsetNodeSetParams,
::cuGraphExecHostNodeSetParams,
::cuGraphExecChildGraphNodeSetParams,
::cuGraphExecEventRecordNodeSetEvent,
::cuGraphExecEventWaitNodeSetEvent,
::cuGraphExecExternalSemaphoresSignalNodeSetParams,
::cuGraphExecUpdate,
::cuGraphInstantiate
@Cast(value="CUresult") public static int cuGraphNodeSetEnabled(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, @Cast(value="unsigned int") int isEnabled)
hGraphExec
- - The executable graph in which to set the specified nodehNode
- - Node from the graph from which graphExec was instantiatedisEnabled
- - Node is enabled if != 0, otherwise the node is disabled::cuGraphNodeGetEnabled,
::cuGraphExecUpdate,
::cuGraphInstantiate
::cuGraphLaunch
@Cast(value="CUresult") public static int cuGraphNodeGetEnabled(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, @Cast(value="unsigned int*") IntPointer isEnabled)
hGraphExec
- - The executable graph in which to set the specified nodehNode
- - Node from the graph from which graphExec was instantiatedisEnabled
- - Location to return the enabled status of the node::cuGraphNodeSetEnabled,
::cuGraphExecUpdate,
::cuGraphInstantiate
::cuGraphLaunch
@Cast(value="CUresult") public static int cuGraphNodeGetEnabled(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, @Cast(value="unsigned int*") IntBuffer isEnabled)
@Cast(value="CUresult") public static int cuGraphNodeGetEnabled(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, @Cast(value="unsigned int*") int[] isEnabled)
@Cast(value="CUresult") public static int cuGraphUpload(CUgraphExec_st hGraphExec, CUstream_st hStream)
hGraphExec
- - Executable graph to uploadhStream
- - Stream in which to upload the graph::cuGraphInstantiate,
::cuGraphLaunch,
::cuGraphExecDestroy
@Cast(value="CUresult") public static int cuGraphLaunch(CUgraphExec_st hGraphExec, CUstream_st hStream)
hGraphExec
- - Executable graph to launchhStream
- - Stream in which to launch the graph::cuGraphInstantiate,
::cuGraphUpload,
::cuGraphExecDestroy
@Cast(value="CUresult") public static int cuGraphExecDestroy(CUgraphExec_st hGraphExec)
hGraphExec
- - Executable graph to destroy::cuGraphInstantiate,
::cuGraphUpload,
::cuGraphLaunch
@Cast(value="CUresult") public static int cuGraphDestroy(CUgraph_st hGraph)
hGraph
- - Graph to destroy::cuGraphCreate
@Cast(value="CUresult") public static int cuGraphExecUpdate(CUgraphExec_st hGraphExec, CUgraph_st hGraph, @Cast(value="CUgraphExecUpdateResultInfo*") CUgraphExecUpdateResultInfo_v1 resultInfo)
hGraphExec
- The instantiated graph to be updatedhGraph
- The graph containing the updated parametersresultInfo
- the error info structure::cuGraphInstantiate
@Cast(value="CUresult") public static int cuGraphKernelNodeCopyAttributes(CUgraphNode_st dst, CUgraphNode_st src)
dst
- [out] Destination nodesrc
- [in] Source node
For list of attributes see ::CUkernelNodeAttrID::CUaccessPolicyWindow
@Cast(value="CUresult") public static int cuGraphKernelNodeGetAttribute(CUgraphNode_st hNode, @Cast(value="CUkernelNodeAttrID") int attr, @Cast(value="CUkernelNodeAttrValue*") CUlaunchAttributeValue value_out)
hNode
- [in]attr
- [in]value_out
- [out]::CUaccessPolicyWindow
@Cast(value="CUresult") public static int cuGraphKernelNodeSetAttribute(CUgraphNode_st hNode, @Cast(value="CUkernelNodeAttrID") int attr, @Cast(value="const CUkernelNodeAttrValue*") CUlaunchAttributeValue value)
hNode
- [out]attr
- [in]value
- [out]::CUaccessPolicyWindow
@Cast(value="CUresult") public static int cuGraphDebugDotPrint(CUgraph_st hGraph, @Cast(value="const char*") BytePointer path, @Cast(value="unsigned int") int flags)
hGraph
- - The graph to create a DOT file frompath
- - The path to write the DOT file toflags
- - Flags from CUgraphDebugDot_flags for specifying which additional node information to write@Cast(value="CUresult") public static int cuGraphDebugDotPrint(CUgraph_st hGraph, String path, @Cast(value="unsigned int") int flags)
@Cast(value="CUresult") public static int cuUserObjectCreate(@ByPtrPtr CUuserObject_st object_out, Pointer ptr, CUhostFn destroy, @Cast(value="unsigned int") int initialRefcount, @Cast(value="unsigned int") int flags)
object_out
- - Location to return the user object handleptr
- - The pointer to pass to the destroy functiondestroy
- - Callback to free the user object when it is no longer in useinitialRefcount
- - The initial refcount to create the object with, typically 1. The
initial references are owned by the calling thread.flags
- - Currently it is required to pass ::CU_USER_OBJECT_NO_DESTRUCTOR_SYNC,
which is the only defined flag. This indicates that the destroy
callback cannot be waited on by any CUDA API. Users requiring
synchronization of the callback should signal its completion
manually.::cuUserObjectRetain,
::cuUserObjectRelease,
::cuGraphRetainUserObject,
::cuGraphReleaseUserObject,
::cuGraphCreate
@Cast(value="CUresult") public static int cuUserObjectRetain(CUuserObject_st object, @Cast(value="unsigned int") int count)
object
- - The object to retaincount
- - The number of references to retain, typically 1. Must be nonzero
and not larger than INT_MAX.::cuUserObjectCreate,
::cuUserObjectRelease,
::cuGraphRetainUserObject,
::cuGraphReleaseUserObject,
::cuGraphCreate
@Cast(value="CUresult") public static int cuUserObjectRelease(CUuserObject_st object, @Cast(value="unsigned int") int count)
object
- - The object to releasecount
- - The number of references to release, typically 1. Must be nonzero
and not larger than INT_MAX.::cuUserObjectCreate,
::cuUserObjectRetain,
::cuGraphRetainUserObject,
::cuGraphReleaseUserObject,
::cuGraphCreate
@Cast(value="CUresult") public static int cuGraphRetainUserObject(CUgraph_st graph, CUuserObject_st object, @Cast(value="unsigned int") int count, @Cast(value="unsigned int") int flags)
graph
- - The graph to associate the reference withobject
- - The user object to retain a reference forcount
- - The number of references to add to the graph, typically 1. Must be
nonzero and not larger than INT_MAX.flags
- - The optional flag ::CU_GRAPH_USER_OBJECT_MOVE transfers references
from the calling thread, rather than create new references. Pass 0
to create new references.::cuUserObjectCreate,
::cuUserObjectRetain,
::cuUserObjectRelease,
::cuGraphReleaseUserObject,
::cuGraphCreate
@Cast(value="CUresult") public static int cuGraphReleaseUserObject(CUgraph_st graph, CUuserObject_st object, @Cast(value="unsigned int") int count)
graph
- - The graph that will release the referenceobject
- - The user object to release a reference forcount
- - The number of references to release, typically 1. Must be nonzero
and not larger than INT_MAX.::cuUserObjectCreate,
::cuUserObjectRetain,
::cuUserObjectRelease,
::cuGraphRetainUserObject,
::cuGraphCreate
@Cast(value="CUresult") public static int cuGraphAddNode(@ByPtrPtr CUgraphNode_st phGraphNode, CUgraph_st hGraph, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st dependencies, @Cast(value="size_t") long numDependencies, CUgraphNodeParams nodeParams)
phGraphNode
- - Returns newly created nodehGraph
- - Graph to which to add the nodedependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciesnodeParams
- - Specification of the node::cuGraphCreate,
::cuGraphNodeSetParams,
::cuGraphExecNodeSetParams
@Cast(value="CUresult") public static int cuGraphAddNode_v2(@ByPtrPtr CUgraphNode_st phGraphNode, CUgraph_st hGraph, @Cast(value="const CUgraphNode*") @ByPtrPtr CUgraphNode_st dependencies, @Const CUgraphEdgeData dependencyData, @Cast(value="size_t") long numDependencies, CUgraphNodeParams nodeParams)
phGraphNode
- - Returns newly created nodehGraph
- - Graph to which to add the nodedependencies
- - Dependencies of the nodedependencyData
- - Optional edge data for the dependencies. If NULL, the data is
assumed to be default (zeroed) for all dependencies.numDependencies
- - Number of dependenciesnodeParams
- - Specification of the node::cuGraphCreate,
::cuGraphNodeSetParams,
::cuGraphExecNodeSetParams
@Cast(value="CUresult") public static int cuGraphNodeSetParams(CUgraphNode_st hNode, CUgraphNodeParams nodeParams)
hNode
- - Node to set the parameters fornodeParams
- - Parameters to copy::cuGraphAddNode,
::cuGraphExecNodeSetParams
@Cast(value="CUresult") public static int cuGraphExecNodeSetParams(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, CUgraphNodeParams nodeParams)
Node type | Allowed changes |
---|---|
kernel | See ::cuGraphExecKernelNodeSetParams |
memcpy | Addresses for 1-dimensional copies if allocated in same context; see ::cuGraphExecMemcpyNodeSetParams |
memset | Addresses for 1-dimensional memsets if allocated in same context; see ::cuGraphExecMemsetNodeSetParams |
host | Unrestricted |
child graph | Topology must match and restrictions apply recursively; see ::cuGraphExecUpdate |
event wait | Unrestricted |
event record | Unrestricted |
external semaphore signal | Number of semaphore operations cannot change |
external semaphore wait | Number of semaphore operations cannot change |
memory allocation | API unsupported |
memory free | API unsupported |
batch memops | Addresses, values, and operation type for wait operations; see ::cuGraphExecBatchMemOpNodeSetParams |
hGraphExec
- - The executable graph in which to update the specified nodehNode
- - Corresponding node from the graph from which graphExec was instantiatednodeParams
- - Updated Parameters to set::cuGraphAddNode,
::cuGraphNodeSetParams
::cuGraphExecUpdate,
::cuGraphInstantiate
@Cast(value="CUresult") public static int cuGraphConditionalHandleCreate(@Cast(value="CUgraphConditionalHandle*") LongPointer pHandle_out, CUgraph_st hGraph, CUctx_st ctx, @Cast(value="unsigned int") int defaultLaunchValue, @Cast(value="unsigned int") int flags)
pHandle_out
- - Pointer used to return the handle to the caller.hGraph
- - Graph which will contain the conditional node using this handle.ctx
- - Context for the handle and associated conditional node.defaultLaunchValue
- - Optional initial value for the conditional variable.flags
- - Currently must be CU_GRAPH_COND_ASSIGN_DEFAULT or 0.::cuGraphAddNode
@Cast(value="CUresult") public static int cuGraphConditionalHandleCreate(@Cast(value="CUgraphConditionalHandle*") LongBuffer pHandle_out, CUgraph_st hGraph, CUctx_st ctx, @Cast(value="unsigned int") int defaultLaunchValue, @Cast(value="unsigned int") int flags)
@Cast(value="CUresult") public static int cuGraphConditionalHandleCreate(@Cast(value="CUgraphConditionalHandle*") long[] pHandle_out, CUgraph_st hGraph, CUctx_st ctx, @Cast(value="unsigned int") int defaultLaunchValue, @Cast(value="unsigned int") int flags)
@Cast(value="CUresult") public static int cuOccupancyMaxActiveBlocksPerMultiprocessor(IntPointer numBlocks, CUfunc_st func, int blockSize, @Cast(value="size_t") long dynamicSMemSize)
numBlocks
- - Returned occupancyfunc
- - Kernel for which occupancy is calculatedblockSize
- - Block size the kernel is intended to be launched withdynamicSMemSize
- - Per-block dynamic shared memory usage intended, in bytes::cudaOccupancyMaxActiveBlocksPerMultiprocessor
@Cast(value="CUresult") public static int cuOccupancyMaxActiveBlocksPerMultiprocessor(IntBuffer numBlocks, CUfunc_st func, int blockSize, @Cast(value="size_t") long dynamicSMemSize)
@Cast(value="CUresult") public static int cuOccupancyMaxActiveBlocksPerMultiprocessor(int[] numBlocks, CUfunc_st func, int blockSize, @Cast(value="size_t") long dynamicSMemSize)
@Cast(value="CUresult") public static int cuOccupancyMaxActiveBlocksPerMultiprocessorWithFlags(IntPointer numBlocks, CUfunc_st func, int blockSize, @Cast(value="size_t") long dynamicSMemSize, @Cast(value="unsigned int") int flags)
numBlocks
- - Returned occupancyfunc
- - Kernel for which occupancy is calculatedblockSize
- - Block size the kernel is intended to be launched withdynamicSMemSize
- - Per-block dynamic shared memory usage intended, in bytesflags
- - Requested behavior for the occupancy calculator::cudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlags
@Cast(value="CUresult") public static int cuOccupancyMaxActiveBlocksPerMultiprocessorWithFlags(IntBuffer numBlocks, CUfunc_st func, int blockSize, @Cast(value="size_t") long dynamicSMemSize, @Cast(value="unsigned int") int flags)
@Cast(value="CUresult") public static int cuOccupancyMaxActiveBlocksPerMultiprocessorWithFlags(int[] numBlocks, CUfunc_st func, int blockSize, @Cast(value="size_t") long dynamicSMemSize, @Cast(value="unsigned int") int flags)
@Cast(value="CUresult") public static int cuOccupancyMaxPotentialBlockSize(IntPointer minGridSize, IntPointer blockSize, CUfunc_st func, CUoccupancyB2DSize blockSizeToDynamicSMemSize, @Cast(value="size_t") long dynamicSMemSize, int blockSizeLimit)
// Take block size, returns dynamic shared memory needed
size_t blockToSmem(int blockSize);
minGridSize
- - Returned minimum grid size needed to achieve the maximum occupancyblockSize
- - Returned maximum block size that can achieve the maximum occupancyfunc
- - Kernel for which launch configuration is calculatedblockSizeToDynamicSMemSize
- - A function that calculates how much per-block dynamic shared memory \p func uses based on the block sizedynamicSMemSize
- - Dynamic shared memory usage intended, in bytesblockSizeLimit
- - The maximum block size \p func is designed to handle::cudaOccupancyMaxPotentialBlockSize
@Cast(value="CUresult") public static int cuOccupancyMaxPotentialBlockSize(IntBuffer minGridSize, IntBuffer blockSize, CUfunc_st func, CUoccupancyB2DSize blockSizeToDynamicSMemSize, @Cast(value="size_t") long dynamicSMemSize, int blockSizeLimit)
@Cast(value="CUresult") public static int cuOccupancyMaxPotentialBlockSize(int[] minGridSize, int[] blockSize, CUfunc_st func, CUoccupancyB2DSize blockSizeToDynamicSMemSize, @Cast(value="size_t") long dynamicSMemSize, int blockSizeLimit)
@Cast(value="CUresult") public static int cuOccupancyMaxPotentialBlockSizeWithFlags(IntPointer minGridSize, IntPointer blockSize, CUfunc_st func, CUoccupancyB2DSize blockSizeToDynamicSMemSize, @Cast(value="size_t") long dynamicSMemSize, int blockSizeLimit, @Cast(value="unsigned int") int flags)
minGridSize
- - Returned minimum grid size needed to achieve the maximum occupancyblockSize
- - Returned maximum block size that can achieve the maximum occupancyfunc
- - Kernel for which launch configuration is calculatedblockSizeToDynamicSMemSize
- - A function that calculates how much per-block dynamic shared memory \p func uses based on the block sizedynamicSMemSize
- - Dynamic shared memory usage intended, in bytesblockSizeLimit
- - The maximum block size \p func is designed to handleflags
- - Options::cudaOccupancyMaxPotentialBlockSizeWithFlags
@Cast(value="CUresult") public static int cuOccupancyMaxPotentialBlockSizeWithFlags(IntBuffer minGridSize, IntBuffer blockSize, CUfunc_st func, CUoccupancyB2DSize blockSizeToDynamicSMemSize, @Cast(value="size_t") long dynamicSMemSize, int blockSizeLimit, @Cast(value="unsigned int") int flags)
@Cast(value="CUresult") public static int cuOccupancyMaxPotentialBlockSizeWithFlags(int[] minGridSize, int[] blockSize, CUfunc_st func, CUoccupancyB2DSize blockSizeToDynamicSMemSize, @Cast(value="size_t") long dynamicSMemSize, int blockSizeLimit, @Cast(value="unsigned int") int flags)
@Cast(value="CUresult") public static int cuOccupancyAvailableDynamicSMemPerBlock(@Cast(value="size_t*") SizeTPointer dynamicSmemSize, CUfunc_st func, int numBlocks, int blockSize)
dynamicSmemSize
- - Returned maximum dynamic shared memoryfunc
- - Kernel function for which occupancy is calculatednumBlocks
- - Number of blocks to fit on SMblockSize
- - Size of the blocks@Cast(value="CUresult") public static int cuOccupancyMaxPotentialClusterSize(IntPointer clusterSize, CUfunc_st func, @Const CUlaunchConfig config)
clusterSize
- - Returned maximum cluster size that can be launched
for the given kernel function and launch configurationfunc
- - Kernel function for which maximum cluster
size is calculatedconfig
- - Launch configuration for the given kernel function::cudaFuncGetAttributes,
::cuFuncGetAttribute
@Cast(value="CUresult") public static int cuOccupancyMaxPotentialClusterSize(IntBuffer clusterSize, CUfunc_st func, @Const CUlaunchConfig config)
@Cast(value="CUresult") public static int cuOccupancyMaxPotentialClusterSize(int[] clusterSize, CUfunc_st func, @Const CUlaunchConfig config)
@Cast(value="CUresult") public static int cuOccupancyMaxActiveClusters(IntPointer numClusters, CUfunc_st func, @Const CUlaunchConfig config)
numClusters
- - Returned maximum number of clusters that
could co-exist on the target devicefunc
- - Kernel function for which maximum number
of clusters are calculatedconfig
- - Launch configuration for the given kernel function::cudaFuncGetAttributes,
::cuFuncGetAttribute
@Cast(value="CUresult") public static int cuOccupancyMaxActiveClusters(IntBuffer numClusters, CUfunc_st func, @Const CUlaunchConfig config)
@Cast(value="CUresult") public static int cuOccupancyMaxActiveClusters(int[] numClusters, CUfunc_st func, @Const CUlaunchConfig config)
@Cast(value="CUresult") @Deprecated public static int cuTexRefSetArray(CUtexref_st hTexRef, CUarray_st hArray, @Cast(value="unsigned int") int Flags)
hTexRef
- - Texture reference to bindhArray
- - Array to bindFlags
- - Options (must be ::CU_TRSA_OVERRIDE_FORMAT)::cuTexRefSetAddress,
::cuTexRefSetAddress2D, ::cuTexRefSetAddressMode,
::cuTexRefSetFilterMode, ::cuTexRefSetFlags, ::cuTexRefSetFormat,
::cuTexRefGetAddress, ::cuTexRefGetAddressMode, ::cuTexRefGetArray,
::cuTexRefGetFilterMode, ::cuTexRefGetFlags, ::cuTexRefGetFormat
@Cast(value="CUresult") @Deprecated public static int cuTexRefSetMipmappedArray(CUtexref_st hTexRef, CUmipmappedArray_st hMipmappedArray, @Cast(value="unsigned int") int Flags)
hTexRef
- - Texture reference to bindhMipmappedArray
- - Mipmapped array to bindFlags
- - Options (must be ::CU_TRSA_OVERRIDE_FORMAT)::cuTexRefSetAddress,
::cuTexRefSetAddress2D, ::cuTexRefSetAddressMode,
::cuTexRefSetFilterMode, ::cuTexRefSetFlags, ::cuTexRefSetFormat,
::cuTexRefGetAddress, ::cuTexRefGetAddressMode, ::cuTexRefGetArray,
::cuTexRefGetFilterMode, ::cuTexRefGetFlags, ::cuTexRefGetFormat
@Cast(value="CUresult") @Deprecated public static int cuTexRefSetAddress(@Cast(value="size_t*") SizeTPointer ByteOffset, CUtexref_st hTexRef, @Cast(value="CUdeviceptr") long dptr, @Cast(value="size_t") long bytes)
ByteOffset
- - Returned byte offsethTexRef
- - Texture reference to binddptr
- - Device pointer to bindbytes
- - Size of memory to bind in bytes::cuTexRefSetAddress2D, ::cuTexRefSetAddressMode, ::cuTexRefSetArray,
::cuTexRefSetFilterMode, ::cuTexRefSetFlags, ::cuTexRefSetFormat,
::cuTexRefGetAddress, ::cuTexRefGetAddressMode, ::cuTexRefGetArray,
::cuTexRefGetFilterMode, ::cuTexRefGetFlags, ::cuTexRefGetFormat
@Cast(value="CUresult") @Deprecated public static int cuTexRefSetAddress2D(CUtexref_st hTexRef, @Cast(value="const CUDA_ARRAY_DESCRIPTOR*") CUDA_ARRAY_DESCRIPTOR_v2 desc, @Cast(value="CUdeviceptr") long dptr, @Cast(value="size_t") long Pitch)
hTexRef
- - Texture reference to binddesc
- - Descriptor of CUDA arraydptr
- - Device pointer to bindPitch
- - Line pitch in bytes::cuTexRefSetAddress,
::cuTexRefSetAddressMode, ::cuTexRefSetArray,
::cuTexRefSetFilterMode, ::cuTexRefSetFlags, ::cuTexRefSetFormat,
::cuTexRefGetAddress, ::cuTexRefGetAddressMode, ::cuTexRefGetArray,
::cuTexRefGetFilterMode, ::cuTexRefGetFlags, ::cuTexRefGetFormat
@Cast(value="CUresult") @Deprecated public static int cuTexRefSetFormat(CUtexref_st hTexRef, @Cast(value="CUarray_format") int fmt, int NumPackedComponents)
hTexRef
- - Texture referencefmt
- - Format to setNumPackedComponents
- - Number of components per array element::cuTexRefSetAddress,
::cuTexRefSetAddress2D, ::cuTexRefSetAddressMode, ::cuTexRefSetArray,
::cuTexRefSetFilterMode, ::cuTexRefSetFlags,
::cuTexRefGetAddress, ::cuTexRefGetAddressMode, ::cuTexRefGetArray,
::cuTexRefGetFilterMode, ::cuTexRefGetFlags, ::cuTexRefGetFormat,
::cudaCreateChannelDesc
@Cast(value="CUresult") @Deprecated public static int cuTexRefSetAddressMode(CUtexref_st hTexRef, int dim, @Cast(value="CUaddress_mode") int am)
typedef enum CUaddress_mode_enum {
CU_TR_ADDRESS_MODE_WRAP = 0,
CU_TR_ADDRESS_MODE_CLAMP = 1,
CU_TR_ADDRESS_MODE_MIRROR = 2,
CU_TR_ADDRESS_MODE_BORDER = 3
} CUaddress_mode;
Note that this call has no effect if \p hTexRef is bound to linear memory.
Also, if the flag, ::CU_TRSF_NORMALIZED_COORDINATES, is not set, the only
supported address mode is ::CU_TR_ADDRESS_MODE_CLAMP.hTexRef
- - Texture referencedim
- - Dimensionam
- - Addressing mode to set::cuTexRefSetAddress,
::cuTexRefSetAddress2D, ::cuTexRefSetArray,
::cuTexRefSetFilterMode, ::cuTexRefSetFlags, ::cuTexRefSetFormat,
::cuTexRefGetAddress, ::cuTexRefGetAddressMode, ::cuTexRefGetArray,
::cuTexRefGetFilterMode, ::cuTexRefGetFlags, ::cuTexRefGetFormat
@Cast(value="CUresult") @Deprecated public static int cuTexRefSetFilterMode(CUtexref_st hTexRef, @Cast(value="CUfilter_mode") int fm)
typedef enum CUfilter_mode_enum {
CU_TR_FILTER_MODE_POINT = 0,
CU_TR_FILTER_MODE_LINEAR = 1
} CUfilter_mode;
Note that this call has no effect if \p hTexRef is bound to linear memory.hTexRef
- - Texture referencefm
- - Filtering mode to set::cuTexRefSetAddress,
::cuTexRefSetAddress2D, ::cuTexRefSetAddressMode, ::cuTexRefSetArray,
::cuTexRefSetFlags, ::cuTexRefSetFormat,
::cuTexRefGetAddress, ::cuTexRefGetAddressMode, ::cuTexRefGetArray,
::cuTexRefGetFilterMode, ::cuTexRefGetFlags, ::cuTexRefGetFormat
@Cast(value="CUresult") @Deprecated public static int cuTexRefSetMipmapFilterMode(CUtexref_st hTexRef, @Cast(value="CUfilter_mode") int fm)
typedef enum CUfilter_mode_enum {
CU_TR_FILTER_MODE_POINT = 0,
CU_TR_FILTER_MODE_LINEAR = 1
} CUfilter_mode;
Note that this call has no effect if \p hTexRef is not bound to a mipmapped array.hTexRef
- - Texture referencefm
- - Filtering mode to set::cuTexRefSetAddress,
::cuTexRefSetAddress2D, ::cuTexRefSetAddressMode, ::cuTexRefSetArray,
::cuTexRefSetFlags, ::cuTexRefSetFormat,
::cuTexRefGetAddress, ::cuTexRefGetAddressMode, ::cuTexRefGetArray,
::cuTexRefGetFilterMode, ::cuTexRefGetFlags, ::cuTexRefGetFormat
@Cast(value="CUresult") @Deprecated public static int cuTexRefSetMipmapLevelBias(CUtexref_st hTexRef, float bias)
hTexRef
- - Texture referencebias
- - Mipmap level bias::cuTexRefSetAddress,
::cuTexRefSetAddress2D, ::cuTexRefSetAddressMode, ::cuTexRefSetArray,
::cuTexRefSetFlags, ::cuTexRefSetFormat,
::cuTexRefGetAddress, ::cuTexRefGetAddressMode, ::cuTexRefGetArray,
::cuTexRefGetFilterMode, ::cuTexRefGetFlags, ::cuTexRefGetFormat
@Cast(value="CUresult") @Deprecated public static int cuTexRefSetMipmapLevelClamp(CUtexref_st hTexRef, float minMipmapLevelClamp, float maxMipmapLevelClamp)
hTexRef
- - Texture referenceminMipmapLevelClamp
- - Mipmap min level clampmaxMipmapLevelClamp
- - Mipmap max level clamp::cuTexRefSetAddress,
::cuTexRefSetAddress2D, ::cuTexRefSetAddressMode, ::cuTexRefSetArray,
::cuTexRefSetFlags, ::cuTexRefSetFormat,
::cuTexRefGetAddress, ::cuTexRefGetAddressMode, ::cuTexRefGetArray,
::cuTexRefGetFilterMode, ::cuTexRefGetFlags, ::cuTexRefGetFormat
@Cast(value="CUresult") @Deprecated public static int cuTexRefSetMaxAnisotropy(CUtexref_st hTexRef, @Cast(value="unsigned int") int maxAniso)
hTexRef
- - Texture referencemaxAniso
- - Maximum anisotropy::cuTexRefSetAddress,
::cuTexRefSetAddress2D, ::cuTexRefSetAddressMode, ::cuTexRefSetArray,
::cuTexRefSetFlags, ::cuTexRefSetFormat,
::cuTexRefGetAddress, ::cuTexRefGetAddressMode, ::cuTexRefGetArray,
::cuTexRefGetFilterMode, ::cuTexRefGetFlags, ::cuTexRefGetFormat
@Cast(value="CUresult") @Deprecated public static int cuTexRefSetBorderColor(CUtexref_st hTexRef, FloatPointer pBorderColor)
hTexRef
- - Texture referencepBorderColor
- - RGBA color::cuTexRefSetAddressMode,
::cuTexRefGetAddressMode, ::cuTexRefGetBorderColor
@Cast(value="CUresult") @Deprecated public static int cuTexRefSetBorderColor(CUtexref_st hTexRef, FloatBuffer pBorderColor)
@Cast(value="CUresult") @Deprecated public static int cuTexRefSetBorderColor(CUtexref_st hTexRef, float[] pBorderColor)
@Cast(value="CUresult") @Deprecated public static int cuTexRefSetFlags(CUtexref_st hTexRef, @Cast(value="unsigned int") int Flags)
hTexRef
- - Texture referenceFlags
- - Optional flags to set::cuTexRefSetAddress,
::cuTexRefSetAddress2D, ::cuTexRefSetAddressMode, ::cuTexRefSetArray,
::cuTexRefSetFilterMode, ::cuTexRefSetFormat,
::cuTexRefGetAddress, ::cuTexRefGetAddressMode, ::cuTexRefGetArray,
::cuTexRefGetFilterMode, ::cuTexRefGetFlags, ::cuTexRefGetFormat
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetAddress(@Cast(value="CUdeviceptr*") LongPointer pdptr, CUtexref_st hTexRef)
pdptr
- - Returned device addresshTexRef
- - Texture reference::cuTexRefSetAddress,
::cuTexRefSetAddress2D, ::cuTexRefSetAddressMode, ::cuTexRefSetArray,
::cuTexRefSetFilterMode, ::cuTexRefSetFlags, ::cuTexRefSetFormat,
::cuTexRefGetAddressMode, ::cuTexRefGetArray,
::cuTexRefGetFilterMode, ::cuTexRefGetFlags, ::cuTexRefGetFormat
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetAddress(@Cast(value="CUdeviceptr*") LongBuffer pdptr, CUtexref_st hTexRef)
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetAddress(@Cast(value="CUdeviceptr*") long[] pdptr, CUtexref_st hTexRef)
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetArray(@ByPtrPtr CUarray_st phArray, CUtexref_st hTexRef)
phArray
- - Returned arrayhTexRef
- - Texture reference::cuTexRefSetAddress,
::cuTexRefSetAddress2D, ::cuTexRefSetAddressMode, ::cuTexRefSetArray,
::cuTexRefSetFilterMode, ::cuTexRefSetFlags, ::cuTexRefSetFormat,
::cuTexRefGetAddress, ::cuTexRefGetAddressMode,
::cuTexRefGetFilterMode, ::cuTexRefGetFlags, ::cuTexRefGetFormat
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetMipmappedArray(@ByPtrPtr CUmipmappedArray_st phMipmappedArray, CUtexref_st hTexRef)
phMipmappedArray
- - Returned mipmapped arrayhTexRef
- - Texture reference::cuTexRefSetAddress,
::cuTexRefSetAddress2D, ::cuTexRefSetAddressMode, ::cuTexRefSetArray,
::cuTexRefSetFilterMode, ::cuTexRefSetFlags, ::cuTexRefSetFormat,
::cuTexRefGetAddress, ::cuTexRefGetAddressMode,
::cuTexRefGetFilterMode, ::cuTexRefGetFlags, ::cuTexRefGetFormat
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetAddressMode(@Cast(value="CUaddress_mode*") IntPointer pam, CUtexref_st hTexRef, int dim)
pam
- - Returned addressing modehTexRef
- - Texture referencedim
- - Dimension::cuTexRefSetAddress,
::cuTexRefSetAddress2D, ::cuTexRefSetAddressMode, ::cuTexRefSetArray,
::cuTexRefSetFilterMode, ::cuTexRefSetFlags, ::cuTexRefSetFormat,
::cuTexRefGetAddress, ::cuTexRefGetArray,
::cuTexRefGetFilterMode, ::cuTexRefGetFlags, ::cuTexRefGetFormat
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetAddressMode(@Cast(value="CUaddress_mode*") IntBuffer pam, CUtexref_st hTexRef, int dim)
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetAddressMode(@Cast(value="CUaddress_mode*") int[] pam, CUtexref_st hTexRef, int dim)
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetFilterMode(@Cast(value="CUfilter_mode*") IntPointer pfm, CUtexref_st hTexRef)
pfm
- - Returned filtering modehTexRef
- - Texture reference::cuTexRefSetAddress,
::cuTexRefSetAddress2D, ::cuTexRefSetAddressMode, ::cuTexRefSetArray,
::cuTexRefSetFilterMode, ::cuTexRefSetFlags, ::cuTexRefSetFormat,
::cuTexRefGetAddress, ::cuTexRefGetAddressMode, ::cuTexRefGetArray,
::cuTexRefGetFlags, ::cuTexRefGetFormat
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetFilterMode(@Cast(value="CUfilter_mode*") IntBuffer pfm, CUtexref_st hTexRef)
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetFilterMode(@Cast(value="CUfilter_mode*") int[] pfm, CUtexref_st hTexRef)
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetFormat(@Cast(value="CUarray_format*") IntPointer pFormat, IntPointer pNumChannels, CUtexref_st hTexRef)
pFormat
- - Returned formatpNumChannels
- - Returned number of componentshTexRef
- - Texture reference::cuTexRefSetAddress,
::cuTexRefSetAddress2D, ::cuTexRefSetAddressMode, ::cuTexRefSetArray,
::cuTexRefSetFilterMode, ::cuTexRefSetFlags, ::cuTexRefSetFormat,
::cuTexRefGetAddress, ::cuTexRefGetAddressMode, ::cuTexRefGetArray,
::cuTexRefGetFilterMode, ::cuTexRefGetFlags
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetFormat(@Cast(value="CUarray_format*") IntBuffer pFormat, IntBuffer pNumChannels, CUtexref_st hTexRef)
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetFormat(@Cast(value="CUarray_format*") int[] pFormat, int[] pNumChannels, CUtexref_st hTexRef)
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetMipmapFilterMode(@Cast(value="CUfilter_mode*") IntPointer pfm, CUtexref_st hTexRef)
pfm
- - Returned mipmap filtering modehTexRef
- - Texture reference::cuTexRefSetAddress,
::cuTexRefSetAddress2D, ::cuTexRefSetAddressMode, ::cuTexRefSetArray,
::cuTexRefSetFlags, ::cuTexRefSetFormat,
::cuTexRefGetAddress, ::cuTexRefGetAddressMode, ::cuTexRefGetArray,
::cuTexRefGetFilterMode, ::cuTexRefGetFlags, ::cuTexRefGetFormat
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetMipmapFilterMode(@Cast(value="CUfilter_mode*") IntBuffer pfm, CUtexref_st hTexRef)
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetMipmapFilterMode(@Cast(value="CUfilter_mode*") int[] pfm, CUtexref_st hTexRef)
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetMipmapLevelBias(FloatPointer pbias, CUtexref_st hTexRef)
pbias
- - Returned mipmap level biashTexRef
- - Texture reference::cuTexRefSetAddress,
::cuTexRefSetAddress2D, ::cuTexRefSetAddressMode, ::cuTexRefSetArray,
::cuTexRefSetFlags, ::cuTexRefSetFormat,
::cuTexRefGetAddress, ::cuTexRefGetAddressMode, ::cuTexRefGetArray,
::cuTexRefGetFilterMode, ::cuTexRefGetFlags, ::cuTexRefGetFormat
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetMipmapLevelBias(FloatBuffer pbias, CUtexref_st hTexRef)
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetMipmapLevelBias(float[] pbias, CUtexref_st hTexRef)
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetMipmapLevelClamp(FloatPointer pminMipmapLevelClamp, FloatPointer pmaxMipmapLevelClamp, CUtexref_st hTexRef)
pminMipmapLevelClamp
- - Returned mipmap min level clamppmaxMipmapLevelClamp
- - Returned mipmap max level clamphTexRef
- - Texture reference::cuTexRefSetAddress,
::cuTexRefSetAddress2D, ::cuTexRefSetAddressMode, ::cuTexRefSetArray,
::cuTexRefSetFlags, ::cuTexRefSetFormat,
::cuTexRefGetAddress, ::cuTexRefGetAddressMode, ::cuTexRefGetArray,
::cuTexRefGetFilterMode, ::cuTexRefGetFlags, ::cuTexRefGetFormat
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetMipmapLevelClamp(FloatBuffer pminMipmapLevelClamp, FloatBuffer pmaxMipmapLevelClamp, CUtexref_st hTexRef)
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetMipmapLevelClamp(float[] pminMipmapLevelClamp, float[] pmaxMipmapLevelClamp, CUtexref_st hTexRef)
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetMaxAnisotropy(IntPointer pmaxAniso, CUtexref_st hTexRef)
pmaxAniso
- - Returned maximum anisotropyhTexRef
- - Texture reference::cuTexRefSetAddress,
::cuTexRefSetAddress2D, ::cuTexRefSetAddressMode, ::cuTexRefSetArray,
::cuTexRefSetFlags, ::cuTexRefSetFormat,
::cuTexRefGetAddress, ::cuTexRefGetAddressMode, ::cuTexRefGetArray,
::cuTexRefGetFilterMode, ::cuTexRefGetFlags, ::cuTexRefGetFormat
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetMaxAnisotropy(IntBuffer pmaxAniso, CUtexref_st hTexRef)
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetMaxAnisotropy(int[] pmaxAniso, CUtexref_st hTexRef)
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetBorderColor(FloatPointer pBorderColor, CUtexref_st hTexRef)
hTexRef
- - Texture referencepBorderColor
- - Returned Type and Value of RGBA color::cuTexRefSetAddressMode,
::cuTexRefSetAddressMode, ::cuTexRefSetBorderColor
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetBorderColor(FloatBuffer pBorderColor, CUtexref_st hTexRef)
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetBorderColor(float[] pBorderColor, CUtexref_st hTexRef)
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetFlags(@Cast(value="unsigned int*") IntPointer pFlags, CUtexref_st hTexRef)
pFlags
- - Returned flagshTexRef
- - Texture reference::cuTexRefSetAddress,
::cuTexRefSetAddress2D, ::cuTexRefSetAddressMode, ::cuTexRefSetArray,
::cuTexRefSetFilterMode, ::cuTexRefSetFlags, ::cuTexRefSetFormat,
::cuTexRefGetAddress, ::cuTexRefGetAddressMode, ::cuTexRefGetArray,
::cuTexRefGetFilterMode, ::cuTexRefGetFormat
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetFlags(@Cast(value="unsigned int*") IntBuffer pFlags, CUtexref_st hTexRef)
@Cast(value="CUresult") @Deprecated public static int cuTexRefGetFlags(@Cast(value="unsigned int*") int[] pFlags, CUtexref_st hTexRef)
@Cast(value="CUresult") @Deprecated public static int cuTexRefCreate(@ByPtrPtr CUtexref_st pTexRef)
pTexRef
- - Returned texture reference::cuTexRefDestroy
@Cast(value="CUresult") @Deprecated public static int cuTexRefDestroy(CUtexref_st hTexRef)
hTexRef
- - Texture reference to destroy::cuTexRefCreate
@Cast(value="CUresult") @Deprecated public static int cuSurfRefSetArray(CUsurfref_st hSurfRef, CUarray_st hArray, @Cast(value="unsigned int") int Flags)
hSurfRef
- - Surface reference handlehArray
- - CUDA array handleFlags
- - set to 0::cuModuleGetSurfRef,
::cuSurfRefGetArray
@Cast(value="CUresult") @Deprecated public static int cuSurfRefGetArray(@ByPtrPtr CUarray_st phArray, CUsurfref_st hSurfRef)
phArray
- - Surface reference handlehSurfRef
- - Surface reference handle::cuModuleGetSurfRef, ::cuSurfRefSetArray
@Cast(value="CUresult") public static int cuTexObjectCreate(@Cast(value="CUtexObject*") LongPointer pTexObject, @Cast(value="const CUDA_RESOURCE_DESC*") CUDA_RESOURCE_DESC_v1 pResDesc, @Cast(value="const CUDA_TEXTURE_DESC*") CUDA_TEXTURE_DESC_v1 pTexDesc, @Cast(value="const CUDA_RESOURCE_VIEW_DESC*") CUDA_RESOURCE_VIEW_DESC_v1 pResViewDesc)
typedef struct CUDA_RESOURCE_DESC_st
{
CUresourcetype resType;
union {
struct {
CUarray hArray;
} array;
struct {
CUmipmappedArray hMipmappedArray;
} mipmap;
struct {
CUdeviceptr devPtr;
CUarray_format format;
unsigned int numChannels;
size_t sizeInBytes;
} linear;
struct {
CUdeviceptr devPtr;
CUarray_format format;
unsigned int numChannels;
size_t width;
size_t height;
size_t pitchInBytes;
} pitch2D;
} res;
unsigned int flags;
} CUDA_RESOURCE_DESC;
where:
- ::CUDA_RESOURCE_DESC::resType specifies the type of resource to texture from.
CUresourceType is defined as:
typedef enum CUresourcetype_enum {
CU_RESOURCE_TYPE_ARRAY = 0x00,
CU_RESOURCE_TYPE_MIPMAPPED_ARRAY = 0x01,
CU_RESOURCE_TYPE_LINEAR = 0x02,
CU_RESOURCE_TYPE_PITCH2D = 0x03
} CUresourcetype;
\par
If ::CUDA_RESOURCE_DESC::resType is set to ::CU_RESOURCE_TYPE_ARRAY, ::CUDA_RESOURCE_DESC::res::array::hArray
must be set to a valid CUDA array handle.
\par
If ::CUDA_RESOURCE_DESC::resType is set to ::CU_RESOURCE_TYPE_MIPMAPPED_ARRAY, ::CUDA_RESOURCE_DESC::res::mipmap::hMipmappedArray
must be set to a valid CUDA mipmapped array handle.
\par
If ::CUDA_RESOURCE_DESC::resType is set to ::CU_RESOURCE_TYPE_LINEAR, ::CUDA_RESOURCE_DESC::res::linear::devPtr
must be set to a valid device pointer, that is aligned to ::CU_DEVICE_ATTRIBUTE_TEXTURE_ALIGNMENT.
::CUDA_RESOURCE_DESC::res::linear::format and ::CUDA_RESOURCE_DESC::res::linear::numChannels
describe the format of each component and the number of components per array element. ::CUDA_RESOURCE_DESC::res::linear::sizeInBytes
specifies the size of the array in bytes. The total number of elements in the linear address range cannot exceed
::CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE1D_LINEAR_WIDTH. The number of elements is computed as (sizeInBytes / (sizeof(format) * numChannels)).
\par
If ::CUDA_RESOURCE_DESC::resType is set to ::CU_RESOURCE_TYPE_PITCH2D, ::CUDA_RESOURCE_DESC::res::pitch2D::devPtr
must be set to a valid device pointer, that is aligned to ::CU_DEVICE_ATTRIBUTE_TEXTURE_ALIGNMENT.
::CUDA_RESOURCE_DESC::res::pitch2D::format and ::CUDA_RESOURCE_DESC::res::pitch2D::numChannels
describe the format of each component and the number of components per array element. ::CUDA_RESOURCE_DESC::res::pitch2D::width
and ::CUDA_RESOURCE_DESC::res::pitch2D::height specify the width and height of the array in elements, and cannot exceed
::CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_LINEAR_WIDTH and ::CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_LINEAR_HEIGHT respectively.
::CUDA_RESOURCE_DESC::res::pitch2D::pitchInBytes specifies the pitch between two rows in bytes and has to be aligned to
::CU_DEVICE_ATTRIBUTE_TEXTURE_PITCH_ALIGNMENT. Pitch cannot exceed ::CU_DEVICE_ATTRIBUTE_MAXIMUM_TEXTURE2D_LINEAR_PITCH.
- ::flags must be set to zero.
The ::CUDA_TEXTURE_DESC struct is defined as
typedef struct CUDA_TEXTURE_DESC_st {
CUaddress_mode addressMode[3];
CUfilter_mode filterMode;
unsigned int flags;
unsigned int maxAnisotropy;
CUfilter_mode mipmapFilterMode;
float mipmapLevelBias;
float minMipmapLevelClamp;
float maxMipmapLevelClamp;
} CUDA_TEXTURE_DESC;
where
- ::CUDA_TEXTURE_DESC::addressMode specifies the addressing mode for each dimension of the texture data. ::CUaddress_mode is defined as:
typedef enum CUaddress_mode_enum {
CU_TR_ADDRESS_MODE_WRAP = 0,
CU_TR_ADDRESS_MODE_CLAMP = 1,
CU_TR_ADDRESS_MODE_MIRROR = 2,
CU_TR_ADDRESS_MODE_BORDER = 3
} CUaddress_mode;
This is ignored if ::CUDA_RESOURCE_DESC::resType is ::CU_RESOURCE_TYPE_LINEAR. Also, if the flag, ::CU_TRSF_NORMALIZED_COORDINATES
is not set, the only supported address mode is ::CU_TR_ADDRESS_MODE_CLAMP.
- ::CUDA_TEXTURE_DESC::filterMode specifies the filtering mode to be used when fetching from the texture. CUfilter_mode is defined as:
typedef enum CUfilter_mode_enum {
CU_TR_FILTER_MODE_POINT = 0,
CU_TR_FILTER_MODE_LINEAR = 1
} CUfilter_mode;
This is ignored if ::CUDA_RESOURCE_DESC::resType is ::CU_RESOURCE_TYPE_LINEAR.
- ::CUDA_TEXTURE_DESC::flags can be any combination of the following:
- ::CU_TRSF_READ_AS_INTEGER, which suppresses the default behavior of
having the texture promote integer data to floating point data in the
range [0, 1]. Note that texture with 32-bit integer format would not be
promoted, regardless of whether or not this flag is specified.
- ::CU_TRSF_NORMALIZED_COORDINATES, which suppresses the default behavior
of having the texture coordinates range from [0, Dim) where Dim is the
width or height of the CUDA array. Instead, the texture coordinates
[0, 1.0) reference the entire breadth of the array dimension; Note that
for CUDA mipmapped arrays, this flag has to be set.
- ::CU_TRSF_DISABLE_TRILINEAR_OPTIMIZATION, which disables any trilinear
filtering optimizations. Trilinear optimizations improve texture filtering
performance by allowing bilinear filtering on textures in scenarios where
it can closely approximate the expected results.
- ::CU_TRSF_SEAMLESS_CUBEMAP, which enables seamless cube map filtering.
This flag can only be specified if the underlying resource is a CUDA array
or a CUDA mipmapped array that was created with the flag ::CUDA_ARRAY3D_CUBEMAP.
When seamless cube map filtering is enabled, texture address modes specified
by ::CUDA_TEXTURE_DESC::addressMode are ignored. Instead, if the ::CUDA_TEXTURE_DESC::filterMode
is set to ::CU_TR_FILTER_MODE_POINT the address mode ::CU_TR_ADDRESS_MODE_CLAMP
will be applied for all dimensions. If the ::CUDA_TEXTURE_DESC::filterMode is
set to ::CU_TR_FILTER_MODE_LINEAR seamless cube map filtering will be performed
when sampling along the cube face borders.
- ::CUDA_TEXTURE_DESC::maxAnisotropy specifies the maximum anisotropy ratio to be used when doing anisotropic filtering. This value will be
clamped to the range [1,16].
- ::CUDA_TEXTURE_DESC::mipmapFilterMode specifies the filter mode when the calculated mipmap level lies between two defined mipmap levels.
- ::CUDA_TEXTURE_DESC::mipmapLevelBias specifies the offset to be applied to the calculated mipmap level.
- ::CUDA_TEXTURE_DESC::minMipmapLevelClamp specifies the lower end of the mipmap level range to clamp access to.
- ::CUDA_TEXTURE_DESC::maxMipmapLevelClamp specifies the upper end of the mipmap level range to clamp access to.
The ::CUDA_RESOURCE_VIEW_DESC struct is defined as
typedef struct CUDA_RESOURCE_VIEW_DESC_st
{
CUresourceViewFormat format;
size_t width;
size_t height;
size_t depth;
unsigned int firstMipmapLevel;
unsigned int lastMipmapLevel;
unsigned int firstLayer;
unsigned int lastLayer;
} CUDA_RESOURCE_VIEW_DESC;
where:
- ::CUDA_RESOURCE_VIEW_DESC::format specifies how the data contained in the CUDA array or CUDA mipmapped array should
be interpreted. Note that this can incur a change in size of the texture data. If the resource view format is a block
compressed format, then the underlying CUDA array or CUDA mipmapped array has to have a base of format ::CU_AD_FORMAT_UNSIGNED_INT32.
with 2 or 4 channels, depending on the block compressed format. For ex., BC1 and BC4 require the underlying CUDA array to have
a format of ::CU_AD_FORMAT_UNSIGNED_INT32 with 2 channels. The other BC formats require the underlying resource to have the same base
format but with 4 channels.
- ::CUDA_RESOURCE_VIEW_DESC::width specifies the new width of the texture data. If the resource view format is a block
compressed format, this value has to be 4 times the original width of the resource. For non block compressed formats,
this value has to be equal to that of the original resource.
- ::CUDA_RESOURCE_VIEW_DESC::height specifies the new height of the texture data. If the resource view format is a block
compressed format, this value has to be 4 times the original height of the resource. For non block compressed formats,
this value has to be equal to that of the original resource.
- ::CUDA_RESOURCE_VIEW_DESC::depth specifies the new depth of the texture data. This value has to be equal to that of the
original resource.
- ::CUDA_RESOURCE_VIEW_DESC::firstMipmapLevel specifies the most detailed mipmap level. This will be the new mipmap level zero.
For non-mipmapped resources, this value has to be zero.::CUDA_TEXTURE_DESC::minMipmapLevelClamp and ::CUDA_TEXTURE_DESC::maxMipmapLevelClamp
will be relative to this value. For ex., if the firstMipmapLevel is set to 2, and a minMipmapLevelClamp of 1.2 is specified,
then the actual minimum mipmap level clamp will be 3.2.
- ::CUDA_RESOURCE_VIEW_DESC::lastMipmapLevel specifies the least detailed mipmap level. For non-mipmapped resources, this value
has to be zero.
- ::CUDA_RESOURCE_VIEW_DESC::firstLayer specifies the first layer index for layered textures. This will be the new layer zero.
For non-layered resources, this value has to be zero.
- ::CUDA_RESOURCE_VIEW_DESC::lastLayer specifies the last layer index for layered textures. For non-layered resources,
this value has to be zero.pTexObject
- - Texture object to createpResDesc
- - Resource descriptorpTexDesc
- - Texture descriptorpResViewDesc
- - Resource view descriptor::cuTexObjectDestroy,
::cudaCreateTextureObject
@Cast(value="CUresult") public static int cuTexObjectCreate(@Cast(value="CUtexObject*") LongBuffer pTexObject, @Cast(value="const CUDA_RESOURCE_DESC*") CUDA_RESOURCE_DESC_v1 pResDesc, @Cast(value="const CUDA_TEXTURE_DESC*") CUDA_TEXTURE_DESC_v1 pTexDesc, @Cast(value="const CUDA_RESOURCE_VIEW_DESC*") CUDA_RESOURCE_VIEW_DESC_v1 pResViewDesc)
@Cast(value="CUresult") public static int cuTexObjectCreate(@Cast(value="CUtexObject*") long[] pTexObject, @Cast(value="const CUDA_RESOURCE_DESC*") CUDA_RESOURCE_DESC_v1 pResDesc, @Cast(value="const CUDA_TEXTURE_DESC*") CUDA_TEXTURE_DESC_v1 pTexDesc, @Cast(value="const CUDA_RESOURCE_VIEW_DESC*") CUDA_RESOURCE_VIEW_DESC_v1 pResViewDesc)
@Cast(value="CUresult") public static int cuTexObjectDestroy(@Cast(value="CUtexObject") long texObject)
texObject
- - Texture object to destroy::cuTexObjectCreate,
::cudaDestroyTextureObject
@Cast(value="CUresult") public static int cuTexObjectGetResourceDesc(@Cast(value="CUDA_RESOURCE_DESC*") CUDA_RESOURCE_DESC_v1 pResDesc, @Cast(value="CUtexObject") long texObject)
pResDesc
- - Resource descriptortexObject
- - Texture object::cuTexObjectCreate,
::cudaGetTextureObjectResourceDesc,
@Cast(value="CUresult") public static int cuTexObjectGetTextureDesc(@Cast(value="CUDA_TEXTURE_DESC*") CUDA_TEXTURE_DESC_v1 pTexDesc, @Cast(value="CUtexObject") long texObject)
pTexDesc
- - Texture descriptortexObject
- - Texture object::cuTexObjectCreate,
::cudaGetTextureObjectTextureDesc
@Cast(value="CUresult") public static int cuTexObjectGetResourceViewDesc(@Cast(value="CUDA_RESOURCE_VIEW_DESC*") CUDA_RESOURCE_VIEW_DESC_v1 pResViewDesc, @Cast(value="CUtexObject") long texObject)
pResViewDesc
- - Resource view descriptortexObject
- - Texture object::cuTexObjectCreate,
::cudaGetTextureObjectResourceViewDesc
@Cast(value="CUresult") public static int cuSurfObjectCreate(@Cast(value="CUsurfObject*") LongPointer pSurfObject, @Cast(value="const CUDA_RESOURCE_DESC*") CUDA_RESOURCE_DESC_v1 pResDesc)
pSurfObject
- - Surface object to createpResDesc
- - Resource descriptor::cuSurfObjectDestroy,
::cudaCreateSurfaceObject
@Cast(value="CUresult") public static int cuSurfObjectCreate(@Cast(value="CUsurfObject*") LongBuffer pSurfObject, @Cast(value="const CUDA_RESOURCE_DESC*") CUDA_RESOURCE_DESC_v1 pResDesc)
@Cast(value="CUresult") public static int cuSurfObjectCreate(@Cast(value="CUsurfObject*") long[] pSurfObject, @Cast(value="const CUDA_RESOURCE_DESC*") CUDA_RESOURCE_DESC_v1 pResDesc)
@Cast(value="CUresult") public static int cuSurfObjectDestroy(@Cast(value="CUsurfObject") long surfObject)
surfObject
- - Surface object to destroy::cuSurfObjectCreate,
::cudaDestroySurfaceObject
@Cast(value="CUresult") public static int cuSurfObjectGetResourceDesc(@Cast(value="CUDA_RESOURCE_DESC*") CUDA_RESOURCE_DESC_v1 pResDesc, @Cast(value="CUsurfObject") long surfObject)
pResDesc
- - Resource descriptorsurfObject
- - Surface object::cuSurfObjectCreate,
::cudaGetSurfaceObjectResourceDesc
@Cast(value="CUresult") public static int cuTensorMapEncodeTiled(CUtensorMap tensorMap, @Cast(value="CUtensorMapDataType") int tensorDataType, @Cast(value="cuuint32_t") int tensorRank, Pointer globalAddress, @Cast(value="const cuuint64_t*") LongPointer globalDim, @Cast(value="const cuuint64_t*") LongPointer globalStrides, @Cast(value="const cuuint32_t*") IntPointer boxDim, @Cast(value="const cuuint32_t*") IntPointer elementStrides, @Cast(value="CUtensorMapInterleave") int interleave, @Cast(value="CUtensorMapSwizzle") int swizzle, @Cast(value="CUtensorMapL2promotion") int l2Promotion, @Cast(value="CUtensorMapFloatOOBfill") int oobFill)
typedef enum CUtensorMapDataType_enum {
CU_TENSOR_MAP_DATA_TYPE_UINT8 = 0, // 1 byte
CU_TENSOR_MAP_DATA_TYPE_UINT16, // 2 bytes
CU_TENSOR_MAP_DATA_TYPE_UINT32, // 4 bytes
CU_TENSOR_MAP_DATA_TYPE_INT32, // 4 bytes
CU_TENSOR_MAP_DATA_TYPE_UINT64, // 8 bytes
CU_TENSOR_MAP_DATA_TYPE_INT64, // 8 bytes
CU_TENSOR_MAP_DATA_TYPE_FLOAT16, // 2 bytes
CU_TENSOR_MAP_DATA_TYPE_FLOAT32, // 4 bytes
CU_TENSOR_MAP_DATA_TYPE_FLOAT64, // 8 bytes
CU_TENSOR_MAP_DATA_TYPE_BFLOAT16, // 2 bytes
CU_TENSOR_MAP_DATA_TYPE_FLOAT32_FTZ, // 4 bytes
CU_TENSOR_MAP_DATA_TYPE_TFLOAT32, // 4 bytes
CU_TENSOR_MAP_DATA_TYPE_TFLOAT32_FTZ // 4 bytes
} CUtensorMapDataType;
- \p tensorRank must be non-zero and less than or equal to the maximum supported dimensionality of 5. If \p interleave is not
::CU_TENSOR_MAP_INTERLEAVE_NONE, then \p tensorRank must additionally be greater than or equal to 3.
- \p globalAddress, which specifies the starting address of the memory region described, must be 32 byte aligned when \p interleave is
::CU_TENSOR_MAP_INTERLEAVE_32B and 16 byte aligned otherwise.
- \p globalDim array, which specifies tensor size of each of the \p tensorRank dimensions, must be non-zero and less than or
equal to 2^32.
- \p globalStrides array, which specifies tensor stride of each of the lower \p tensorRank - 1 dimensions in bytes, must be a
multiple of 16 and less than 2^40. Additionally, the stride must be a multiple of 32 when \p interleave is ::CU_TENSOR_MAP_INTERLEAVE_32B.
Each following dimension specified includes previous dimension stride:
globalStrides[0] = globalDim[0] * elementSizeInBytes(tensorDataType) + padding[0];
for (i = 1; i < tensorRank - 1; i++)
globalStrides[i] = globalStrides[i – 1] * (globalDim[i] + padding[i]);
assert(globalStrides[i] >= globalDim[i]);
- \p boxDim array, which specifies number of elements to be traversed along each of the \p tensorRank dimensions, must be non-zero
and less than or equal to 256.
When \p interleave is ::CU_TENSOR_MAP_INTERLEAVE_NONE, { \p boxDim[0] * elementSizeInBytes( \p tensorDataType ) } must be a multiple
of 16 bytes.
- \p elementStrides array, which specifies the iteration step along each of the \p tensorRank dimensions, must be non-zero and less
than or equal to 8. Note that when \p interleave is ::CU_TENSOR_MAP_INTERLEAVE_NONE, the first element of this array is ignored since
TMA doesn’t support the stride for dimension zero.
When all elements of \p elementStrides array is one, \p boxDim specifies the number of elements to load. However, if the \p elementStrides[i]
is not equal to one, then TMA loads ceil( \p boxDim[i] / \p elementStrides[i]) number of elements along i-th dimension. To load N elements along
i-th dimension, \p boxDim[i] must be set to N * \p elementStrides[i].
- \p interleave specifies the interleaved layout of type ::CUtensorMapInterleave, which is defined as:
typedef enum CUtensorMapInterleave_enum {
CU_TENSOR_MAP_INTERLEAVE_NONE = 0,
CU_TENSOR_MAP_INTERLEAVE_16B,
CU_TENSOR_MAP_INTERLEAVE_32B
} CUtensorMapInterleave;
TMA supports interleaved layouts like NC/8HWC8 where C8 utilizes 16 bytes in memory assuming 2 byte per channel or NC/16HWC16 where C16
uses 32 bytes.
When \p interleave is ::CU_TENSOR_MAP_INTERLEAVE_NONE and \p swizzle is not ::CU_TENSOR_MAP_SWIZZLE_NONE, the bounding box inner dimension
(computed as \p boxDim[0] multiplied by element size derived from \p tensorDataType) must be less than or equal to the swizzle size.
- CU_TENSOR_MAP_SWIZZLE_32B implies the bounding box inner dimension will be <= 32.
- CU_TENSOR_MAP_SWIZZLE_64B implies the bounding box inner dimension will be <= 64.
- CU_TENSOR_MAP_SWIZZLE_128B implies the bounding box inner dimension will be <= 128.
- \p swizzle, which specifies the shared memory bank swizzling pattern, has to be of type ::CUtensorMapSwizzle which is defined as:
typedef enum CUtensorMapSwizzle_enum {
CU_TENSOR_MAP_SWIZZLE_NONE = 0,
CU_TENSOR_MAP_SWIZZLE_32B,
CU_TENSOR_MAP_SWIZZLE_64B,
CU_TENSOR_MAP_SWIZZLE_128B
} CUtensorMapSwizzle;
Data are organized in a specific order in global memory; however, this may not match the order in which the application accesses data
in shared memory. This difference in data organization may cause bank conflicts when shared memory is accessed. In order to avoid this
problem, data can be loaded to shared memory with shuffling across shared memory banks.
When \p interleave is ::CU_TENSOR_MAP_INTERLEAVE_32B, \p swizzle must be ::CU_TENSOR_MAP_SWIZZLE_32B.
Other interleave modes can have any swizzling pattern.
- \p l2Promotion specifies L2 fetch size which indicates the byte granurality at which L2 requests is filled from DRAM. It must be of
type ::CUtensorMapL2promotion, which is defined as:
typedef enum CUtensorMapL2promotion_enum {
CU_TENSOR_MAP_L2_PROMOTION_NONE = 0,
CU_TENSOR_MAP_L2_PROMOTION_L2_64B,
CU_TENSOR_MAP_L2_PROMOTION_L2_128B,
CU_TENSOR_MAP_L2_PROMOTION_L2_256B
} CUtensorMapL2promotion;
- \p oobFill, which indicates whether zero or a special NaN constant should be used to fill out-of-bound elements, must be of type
::CUtensorMapFloatOOBfill which is defined as:
typedef enum CUtensorMapFloatOOBfill_enum {
CU_TENSOR_MAP_FLOAT_OOB_FILL_NONE = 0,
CU_TENSOR_MAP_FLOAT_OOB_FILL_NAN_REQUEST_ZERO_FMA
} CUtensorMapFloatOOBfill;
Note that ::CU_TENSOR_MAP_FLOAT_OOB_FILL_NAN_REQUEST_ZERO_FMA can only be used when \p tensorDataType represents a floating-point data type.tensorMap
- - Tensor map object to createtensorDataType
- - Tensor data typetensorRank
- - Dimensionality of tensorglobalAddress
- - Starting address of memory region described by tensorglobalDim
- - Array containing tensor size (number of elements) along each of the \p tensorRank dimensionsglobalStrides
- - Array containing stride size (in bytes) along each of the \p tensorRank - 1 dimensionsboxDim
- - Array containing traversal box size (number of elments) along each of the \p tensorRank dimensions. Specifies how many elements to be traversed along each tensor dimension.elementStrides
- - Array containing traversal stride in each of the \p tensorRank dimensionsinterleave
- - Type of interleaved layout the tensor addressesswizzle
- - Bank swizzling pattern inside shared memoryl2Promotion
- - L2 promotion sizeoobFill
- - Indicate whether zero or special NaN constant must be used to fill out-of-bound elements::cuTensorMapEncodeIm2col,
::cuTensorMapReplaceAddress
@Cast(value="CUresult") public static int cuTensorMapEncodeTiled(CUtensorMap tensorMap, @Cast(value="CUtensorMapDataType") int tensorDataType, @Cast(value="cuuint32_t") int tensorRank, Pointer globalAddress, @Cast(value="const cuuint64_t*") LongBuffer globalDim, @Cast(value="const cuuint64_t*") LongBuffer globalStrides, @Cast(value="const cuuint32_t*") IntBuffer boxDim, @Cast(value="const cuuint32_t*") IntBuffer elementStrides, @Cast(value="CUtensorMapInterleave") int interleave, @Cast(value="CUtensorMapSwizzle") int swizzle, @Cast(value="CUtensorMapL2promotion") int l2Promotion, @Cast(value="CUtensorMapFloatOOBfill") int oobFill)
@Cast(value="CUresult") public static int cuTensorMapEncodeTiled(CUtensorMap tensorMap, @Cast(value="CUtensorMapDataType") int tensorDataType, @Cast(value="cuuint32_t") int tensorRank, Pointer globalAddress, @Cast(value="const cuuint64_t*") long[] globalDim, @Cast(value="const cuuint64_t*") long[] globalStrides, @Cast(value="const cuuint32_t*") int[] boxDim, @Cast(value="const cuuint32_t*") int[] elementStrides, @Cast(value="CUtensorMapInterleave") int interleave, @Cast(value="CUtensorMapSwizzle") int swizzle, @Cast(value="CUtensorMapL2promotion") int l2Promotion, @Cast(value="CUtensorMapFloatOOBfill") int oobFill)
@Cast(value="CUresult") public static int cuTensorMapEncodeIm2col(CUtensorMap tensorMap, @Cast(value="CUtensorMapDataType") int tensorDataType, @Cast(value="cuuint32_t") int tensorRank, Pointer globalAddress, @Cast(value="const cuuint64_t*") LongPointer globalDim, @Cast(value="const cuuint64_t*") LongPointer globalStrides, @Const IntPointer pixelBoxLowerCorner, @Const IntPointer pixelBoxUpperCorner, @Cast(value="cuuint32_t") int channelsPerPixel, @Cast(value="cuuint32_t") int pixelsPerColumn, @Cast(value="const cuuint32_t*") IntPointer elementStrides, @Cast(value="CUtensorMapInterleave") int interleave, @Cast(value="CUtensorMapSwizzle") int swizzle, @Cast(value="CUtensorMapL2promotion") int l2Promotion, @Cast(value="CUtensorMapFloatOOBfill") int oobFill)
typedef enum CUtensorMapDataType_enum {
CU_TENSOR_MAP_DATA_TYPE_UINT8 = 0, // 1 byte
CU_TENSOR_MAP_DATA_TYPE_UINT16, // 2 bytes
CU_TENSOR_MAP_DATA_TYPE_UINT32, // 4 bytes
CU_TENSOR_MAP_DATA_TYPE_INT32, // 4 bytes
CU_TENSOR_MAP_DATA_TYPE_UINT64, // 8 bytes
CU_TENSOR_MAP_DATA_TYPE_INT64, // 8 bytes
CU_TENSOR_MAP_DATA_TYPE_FLOAT16, // 2 bytes
CU_TENSOR_MAP_DATA_TYPE_FLOAT32, // 4 bytes
CU_TENSOR_MAP_DATA_TYPE_FLOAT64, // 8 bytes
CU_TENSOR_MAP_DATA_TYPE_BFLOAT16, // 2 bytes
CU_TENSOR_MAP_DATA_TYPE_FLOAT32_FTZ, // 4 bytes
CU_TENSOR_MAP_DATA_TYPE_TFLOAT32, // 4 bytes
CU_TENSOR_MAP_DATA_TYPE_TFLOAT32_FTZ // 4 bytes
} CUtensorMapDataType;
- \p tensorRank, which specifies the number of tensor dimensions, must be 3, 4, or 5.
- \p globalAddress, which specifies the starting address of the memory region described, must be 32 byte aligned when \p interleave is
::CU_TENSOR_MAP_INTERLEAVE_32B and 16 byte aligned otherwise.
- \p globalDim array, which specifies tensor size of each of the \p tensorRank dimensions, must be non-zero and less than or
equal to 2^32.
- \p globalStrides array, which specifies tensor stride of each of the lower \p tensorRank - 1 dimensions in bytes, must be a
multiple of 16 and less than 2^40. Additionally, the stride must be a multiple of 32 when \p interleave is ::CU_TENSOR_MAP_INTERLEAVE_32B.
Each following dimension specified includes previous dimension stride:
globalStrides[0] = globalDim[0] * elementSizeInBytes(tensorDataType) + padding[0];
for (i = 1; i < tensorRank - 1; i++)
globalStrides[i] = globalStrides[i – 1] * (globalDim[i] + padding[i]);
assert(globalStrides[i] >= globalDim[i]);
- \p pixelBoxLowerCorner array specifies the coordinate offsets {D, H, W} of the bounding box from top/left/front corner. The number of
offsets and their precision depend on the tensor dimensionality:
- When \p tensorRank is 3, one signed offset within range [-32768, 32767] is supported.
- When \p tensorRank is 4, two signed offsets each within range [-128, 127] are supported.
- When \p tensorRank is 5, three offsets each within range [-16, 15] are supported.
- \p pixelBoxUpperCorner array specifies the coordinate offsets {D, H, W} of the bounding box from bottom/right/back corner. The number of
offsets and their precision depend on the tensor dimensionality:
- When \p tensorRank is 3, one signed offset within range [-32768, 32767] is supported.
- When \p tensorRank is 4, two signed offsets each within range [-128, 127] are supported.
- When \p tensorRank is 5, three offsets each within range [-16, 15] are supported.
The bounding box specified by \p pixelBoxLowerCorner and \p pixelBoxUpperCorner must have non-zero area.
- \p channelsPerPixel, which specifies the number of elements which must be accessed along C dimension, must be less than or equal to 256.
- \p pixelsPerColumn, which specifies the number of elements that must be accessed along the {N, D, H, W} dimensions, must be less than or
equal to 1024.
- \p elementStrides array, which specifies the iteration step along each of the \p tensorRank dimensions, must be non-zero and less
than or equal to 8. Note that when \p interleave is ::CU_TENSOR_MAP_INTERLEAVE_NONE, the first element of this array is ignored since
TMA doesn’t support the stride for dimension zero.
When all elements of the \p elementStrides array are one, \p boxDim specifies the number of elements to load. However, if \p elementStrides[i]
is not equal to one for some \p i, then TMA loads ceil( \p boxDim[i] / \p elementStrides[i]) number of elements along i-th dimension.
To load N elements along i-th dimension, \p boxDim[i] must be set to N * \p elementStrides[i].
- \p interleave specifies the interleaved layout of type ::CUtensorMapInterleave, which is defined as:
typedef enum CUtensorMapInterleave_enum {
CU_TENSOR_MAP_INTERLEAVE_NONE = 0,
CU_TENSOR_MAP_INTERLEAVE_16B,
CU_TENSOR_MAP_INTERLEAVE_32B
} CUtensorMapInterleave;
TMA supports interleaved layouts like NC/8HWC8 where C8 utilizes 16 bytes in memory assuming 2 byte per channel or NC/16HWC16 where C16
uses 32 bytes.
When \p interleave is ::CU_TENSOR_MAP_INTERLEAVE_NONE and \p swizzle is not ::CU_TENSOR_MAP_SWIZZLE_NONE, the bounding box inner dimension
(computed as \p boxDim[0] multiplied by element size derived from \p tensorDataType) must be less than or equal to the swizzle size.
- CU_TENSOR_MAP_SWIZZLE_32B implies the bounding box inner dimension will be <= 32.
- CU_TENSOR_MAP_SWIZZLE_64B implies the bounding box inner dimension will be <= 64.
- CU_TENSOR_MAP_SWIZZLE_128B implies the bounding box inner dimension will be <= 128.
- \p swizzle, which specifies the shared memory bank swizzling pattern, has to be of type ::CUtensorMapSwizzle which is defined as:
typedef enum CUtensorMapSwizzle_enum {
CU_TENSOR_MAP_SWIZZLE_NONE = 0,
CU_TENSOR_MAP_SWIZZLE_32B,
CU_TENSOR_MAP_SWIZZLE_64B,
CU_TENSOR_MAP_SWIZZLE_128B
} CUtensorMapSwizzle;
Data are organized in a specific order in global memory; however, this may not match the order in which the application accesses data
in shared memory. This difference in data organization may cause bank conflicts when shared memory is accessed. In order to avoid this
problem, data can be loaded to shared memory with shuffling across shared memory banks.
When \p interleave is ::CU_TENSOR_MAP_INTERLEAVE_32B, \p swizzle must be ::CU_TENSOR_MAP_SWIZZLE_32B.
Other interleave modes can have any swizzling pattern.
- \p l2Promotion specifies L2 fetch size which indicates the byte granularity at which L2 requests are filled from DRAM. It must be of
type ::CUtensorMapL2promotion, which is defined as:
typedef enum CUtensorMapL2promotion_enum {
CU_TENSOR_MAP_L2_PROMOTION_NONE = 0,
CU_TENSOR_MAP_L2_PROMOTION_L2_64B,
CU_TENSOR_MAP_L2_PROMOTION_L2_128B,
CU_TENSOR_MAP_L2_PROMOTION_L2_256B
} CUtensorMapL2promotion;
- \p oobFill, which indicates whether zero or a special NaN constant should be used to fill out-of-bound elements, must be of type
::CUtensorMapFloatOOBfill which is defined as:
typedef enum CUtensorMapFloatOOBfill_enum {
CU_TENSOR_MAP_FLOAT_OOB_FILL_NONE = 0,
CU_TENSOR_MAP_FLOAT_OOB_FILL_NAN_REQUEST_ZERO_FMA
} CUtensorMapFloatOOBfill;
Note that ::CU_TENSOR_MAP_FLOAT_OOB_FILL_NAN_REQUEST_ZERO_FMA can only be used when \p tensorDataType represents a floating-point data type.tensorMap
- - Tensor map object to createtensorDataType
- - Tensor data typetensorRank
- - Dimensionality of tensor; must be at least 3globalAddress
- - Starting address of memory region described by tensorglobalDim
- - Array containing tensor size (number of elements) along each of the \p tensorRank dimensionsglobalStrides
- - Array containing stride size (in bytes) along each of the \p tensorRank - 1 dimensionspixelBoxLowerCorner
- - Array containing DHW dimensions of lower box cornerpixelBoxUpperCorner
- - Array containing DHW dimensions of upper box cornerchannelsPerPixel
- - Number of channels per pixelpixelsPerColumn
- - Number of pixels per columnelementStrides
- - Array containing traversal stride in each of the \p tensorRank dimensionsinterleave
- - Type of interleaved layout the tensor addressesswizzle
- - Bank swizzling pattern inside shared memoryl2Promotion
- - L2 promotion sizeoobFill
- - Indicate whether zero or special NaN constant will be used to fill out-of-bound elements::cuTensorMapEncodeTiled,
::cuTensorMapReplaceAddress
@Cast(value="CUresult") public static int cuTensorMapEncodeIm2col(CUtensorMap tensorMap, @Cast(value="CUtensorMapDataType") int tensorDataType, @Cast(value="cuuint32_t") int tensorRank, Pointer globalAddress, @Cast(value="const cuuint64_t*") LongBuffer globalDim, @Cast(value="const cuuint64_t*") LongBuffer globalStrides, @Const IntBuffer pixelBoxLowerCorner, @Const IntBuffer pixelBoxUpperCorner, @Cast(value="cuuint32_t") int channelsPerPixel, @Cast(value="cuuint32_t") int pixelsPerColumn, @Cast(value="const cuuint32_t*") IntBuffer elementStrides, @Cast(value="CUtensorMapInterleave") int interleave, @Cast(value="CUtensorMapSwizzle") int swizzle, @Cast(value="CUtensorMapL2promotion") int l2Promotion, @Cast(value="CUtensorMapFloatOOBfill") int oobFill)
@Cast(value="CUresult") public static int cuTensorMapEncodeIm2col(CUtensorMap tensorMap, @Cast(value="CUtensorMapDataType") int tensorDataType, @Cast(value="cuuint32_t") int tensorRank, Pointer globalAddress, @Cast(value="const cuuint64_t*") long[] globalDim, @Cast(value="const cuuint64_t*") long[] globalStrides, @Const int[] pixelBoxLowerCorner, @Const int[] pixelBoxUpperCorner, @Cast(value="cuuint32_t") int channelsPerPixel, @Cast(value="cuuint32_t") int pixelsPerColumn, @Cast(value="const cuuint32_t*") int[] elementStrides, @Cast(value="CUtensorMapInterleave") int interleave, @Cast(value="CUtensorMapSwizzle") int swizzle, @Cast(value="CUtensorMapL2promotion") int l2Promotion, @Cast(value="CUtensorMapFloatOOBfill") int oobFill)
@Cast(value="CUresult") public static int cuTensorMapReplaceAddress(CUtensorMap tensorMap, Pointer globalAddress)
tensorMap
- - Tensor map object to modifyglobalAddress
- - Starting address of memory region described by tensor, must follow previous alignment requirements::cuTensorMapEncodeTiled,
::cuTensorMapEncodeIm2col
@Cast(value="CUresult") public static int cuDeviceCanAccessPeer(IntPointer canAccessPeer, @Cast(value="CUdevice") int dev, @Cast(value="CUdevice") int peerDev)
canAccessPeer
- - Returned access capabilitydev
- - Device from which allocations on \p peerDev are to
be directly accessed.peerDev
- - Device on which the allocations to be directly accessed
by \p dev reside.::cuCtxEnablePeerAccess,
::cuCtxDisablePeerAccess,
::cudaDeviceCanAccessPeer
@Cast(value="CUresult") public static int cuDeviceCanAccessPeer(IntBuffer canAccessPeer, @Cast(value="CUdevice") int dev, @Cast(value="CUdevice") int peerDev)
@Cast(value="CUresult") public static int cuDeviceCanAccessPeer(int[] canAccessPeer, @Cast(value="CUdevice") int dev, @Cast(value="CUdevice") int peerDev)
@Cast(value="CUresult") public static int cuCtxEnablePeerAccess(CUctx_st peerContext, @Cast(value="unsigned int") int Flags)
peerContext
- - Peer context to enable direct access to from the current contextFlags
- - Reserved for future use and must be set to 0::cuDeviceCanAccessPeer,
::cuCtxDisablePeerAccess,
::cudaDeviceEnablePeerAccess
@Cast(value="CUresult") public static int cuCtxDisablePeerAccess(CUctx_st peerContext)
peerContext
- - Peer context to disable direct access to::cuDeviceCanAccessPeer,
::cuCtxEnablePeerAccess,
::cudaDeviceDisablePeerAccess
@Cast(value="CUresult") public static int cuGraphicsUnregisterResource(CUgraphicsResource_st resource)
resource
- - Resource to unregister::cuGraphicsD3D9RegisterResource,
::cuGraphicsD3D10RegisterResource,
::cuGraphicsD3D11RegisterResource,
::cuGraphicsGLRegisterBuffer,
::cuGraphicsGLRegisterImage,
::cudaGraphicsUnregisterResource
@Cast(value="CUresult") public static int cuGraphicsSubResourceGetMappedArray(@ByPtrPtr CUarray_st pArray, CUgraphicsResource_st resource, @Cast(value="unsigned int") int arrayIndex, @Cast(value="unsigned int") int mipLevel)
pArray
- - Returned array through which a subresource of \p resource may be accessedresource
- - Mapped resource to accessarrayIndex
- - Array index for array textures or cubemap face
index as defined by ::CUarray_cubemap_face for
cubemap textures for the subresource to accessmipLevel
- - Mipmap level for the subresource to access::cuGraphicsResourceGetMappedPointer,
::cudaGraphicsSubResourceGetMappedArray
@Cast(value="CUresult") public static int cuGraphicsResourceGetMappedMipmappedArray(@ByPtrPtr CUmipmappedArray_st pMipmappedArray, CUgraphicsResource_st resource)
pMipmappedArray
- - Returned mipmapped array through which \p resource may be accessedresource
- - Mapped resource to access::cuGraphicsResourceGetMappedPointer,
::cudaGraphicsResourceGetMappedMipmappedArray
@Cast(value="CUresult") public static int cuGraphicsResourceGetMappedPointer(@Cast(value="CUdeviceptr*") LongPointer pDevPtr, @Cast(value="size_t*") SizeTPointer pSize, CUgraphicsResource_st resource)
pDevPtr
- - Returned pointer through which \p resource may be accessedpSize
- - Returned size of the buffer accessible starting at \p *pPointerresource
- - Mapped resource to access::cuGraphicsMapResources,
::cuGraphicsSubResourceGetMappedArray,
::cudaGraphicsResourceGetMappedPointer
@Cast(value="CUresult") public static int cuGraphicsResourceGetMappedPointer(@Cast(value="CUdeviceptr*") LongBuffer pDevPtr, @Cast(value="size_t*") SizeTPointer pSize, CUgraphicsResource_st resource)
@Cast(value="CUresult") public static int cuGraphicsResourceGetMappedPointer(@Cast(value="CUdeviceptr*") long[] pDevPtr, @Cast(value="size_t*") SizeTPointer pSize, CUgraphicsResource_st resource)
@Cast(value="CUresult") public static int cuGraphicsResourceSetMapFlags(CUgraphicsResource_st resource, @Cast(value="unsigned int") int flags)
- ::CU_GRAPHICS_MAP_RESOURCE_FLAGS_NONE: Specifies no hints about how this resource will be used. It is therefore assumed that this resource will be read from and written to by CUDA kernels. This is the default value. - ::CU_GRAPHICS_MAP_RESOURCE_FLAGS_READONLY: Specifies that CUDA kernels which access this resource will not write to this resource. - ::CU_GRAPHICS_MAP_RESOURCE_FLAGS_WRITEDISCARD: Specifies that CUDA kernels which access this resource will not read from this resource and will write over the entire contents of the resource, so none of the data previously stored in the resource will be preserved. If \p resource is presently mapped for access by CUDA then ::CUDA_ERROR_ALREADY_MAPPED is returned. If \p flags is not one of the above values then ::CUDA_ERROR_INVALID_VALUE is returned.
resource
- - Registered resource to set flags forflags
- - Parameters for resource mapping::cuGraphicsMapResources,
::cudaGraphicsResourceSetMapFlags
@Cast(value="CUresult") public static int cuGraphicsMapResources(@Cast(value="unsigned int") int count, @ByPtrPtr CUgraphicsResource_st resources, CUstream_st hStream)
count
- - Number of resources to mapresources
- - Resources to map for CUDA usagehStream
- - Stream with which to synchronize::cuGraphicsResourceGetMappedPointer,
::cuGraphicsSubResourceGetMappedArray,
::cuGraphicsUnmapResources,
::cudaGraphicsMapResources
@Cast(value="CUresult") public static int cuGraphicsUnmapResources(@Cast(value="unsigned int") int count, @ByPtrPtr CUgraphicsResource_st resources, CUstream_st hStream)
count
- - Number of resources to unmapresources
- - Resources to unmaphStream
- - Stream with which to synchronize::cuGraphicsMapResources,
::cudaGraphicsUnmapResources
@Cast(value="CUresult") public static int cuGetProcAddress(@Cast(value="const char*") BytePointer symbol, @Cast(value="void**") PointerPointer pfn, int cudaVersion, @Cast(value="cuuint64_t") long flags, @Cast(value="CUdriverProcAddressQueryResult*") IntPointer symbolStatus)
symbol
- - The base name of the driver API function to look for. As an example,
for the driver API ::cuMemAlloc_v2, \p symbol would be cuMemAlloc and
\p cudaVersion would be the ABI compatible CUDA version for the _v2 variant.pfn
- - Location to return the function pointer to the requested driver functioncudaVersion
- - The CUDA version to look for the requested driver symbolflags
- - Flags to specify search options.symbolStatus
- - Optional location to store the status of the search for
\p symbol based on \p cudaVersion. See ::CUdriverProcAddressQueryResult
for possible values.::cudaGetDriverEntryPoint
@Cast(value="CUresult") public static int cuGetProcAddress(@Cast(value="const char*") BytePointer symbol, @Cast(value="void**") @ByPtrPtr Pointer pfn, int cudaVersion, @Cast(value="cuuint64_t") long flags, @Cast(value="CUdriverProcAddressQueryResult*") IntPointer symbolStatus)
@Cast(value="CUresult") public static int cuGetProcAddress(String symbol, @Cast(value="void**") @ByPtrPtr Pointer pfn, int cudaVersion, @Cast(value="cuuint64_t") long flags, @Cast(value="CUdriverProcAddressQueryResult*") IntBuffer symbolStatus)
@Cast(value="CUresult") public static int cuGetProcAddress(@Cast(value="const char*") BytePointer symbol, @Cast(value="void**") @ByPtrPtr Pointer pfn, int cudaVersion, @Cast(value="cuuint64_t") long flags, @Cast(value="CUdriverProcAddressQueryResult*") int[] symbolStatus)
@Cast(value="CUresult") public static int cuGetProcAddress(String symbol, @Cast(value="void**") @ByPtrPtr Pointer pfn, int cudaVersion, @Cast(value="cuuint64_t") long flags, @Cast(value="CUdriverProcAddressQueryResult*") IntPointer symbolStatus)
@Cast(value="CUresult") public static int cuGetProcAddress(@Cast(value="const char*") BytePointer symbol, @Cast(value="void**") @ByPtrPtr Pointer pfn, int cudaVersion, @Cast(value="cuuint64_t") long flags, @Cast(value="CUdriverProcAddressQueryResult*") IntBuffer symbolStatus)
@Cast(value="CUresult") public static int cuGetProcAddress(String symbol, @Cast(value="void**") @ByPtrPtr Pointer pfn, int cudaVersion, @Cast(value="cuuint64_t") long flags, @Cast(value="CUdriverProcAddressQueryResult*") int[] symbolStatus)
@Cast(value="CUresult") public static int cuCoredumpGetAttribute(@Cast(value="CUcoredumpSettings") int attrib, Pointer value, @Cast(value="size_t*") SizeTPointer size)
attrib
- - The enum defining which value to fetch.value
- - void* containing the requested data.size
- - The size of the memory region \p value points to.::cuCoredumpGetAttributeGlobal,
::cuCoredumpSetAttribute,
::cuCoredumpSetAttributeGlobal
@Cast(value="CUresult") public static int cuCoredumpGetAttributeGlobal(@Cast(value="CUcoredumpSettings") int attrib, Pointer value, @Cast(value="size_t*") SizeTPointer size)
attrib
- - The enum defining which value to fetch.value
- - void* containing the requested data.size
- - The size of the memory region \p value points to.::cuCoredumpGetAttribute,
::cuCoredumpSetAttribute,
::cuCoredumpSetAttributeGlobal
@Cast(value="CUresult") public static int cuCoredumpSetAttribute(@Cast(value="CUcoredumpSettings") int attrib, Pointer value, @Cast(value="size_t*") SizeTPointer size)
attrib
- - The enum defining which value to set.value
- - void* containing the requested data.size
- - The size of the memory region \p value points to.::cuCoredumpGetAttributeGlobal,
::cuCoredumpGetAttribute,
::cuCoredumpSetAttributeGlobal
@Cast(value="CUresult") public static int cuCoredumpSetAttributeGlobal(@Cast(value="CUcoredumpSettings") int attrib, Pointer value, @Cast(value="size_t*") SizeTPointer size)
attrib
- - The enum defining which value to set.value
- - void* containing the requested data.size
- - The size of the memory region \p value points to.::cuCoredumpGetAttribute,
::cuCoredumpGetAttributeGlobal,
::cuCoredumpSetAttribute
@Cast(value="CUresult") public static int cuGetExportTable(@Cast(value="const void**") PointerPointer ppExportTable, @Const CUuuid pExportTableId)
@Cast(value="CUresult") public static int cuGetExportTable(@Cast(value="const void**") @ByPtrPtr Pointer ppExportTable, @Const CUuuid pExportTableId)
@MemberGetter public static CUstream_st cudaStreamLegacy()
@MemberGetter public static CUstream_st cudaStreamPerThread()
@MemberGetter public static int __CUDART_API_VERSION()
@Cast(value="cudaError_t") public static int cudaDeviceReset()
::cudaDeviceSynchronize
@Cast(value="cudaError_t") public static int cudaDeviceSynchronize()
::cudaDeviceReset,
::cuCtxSynchronize
@Cast(value="cudaError_t") public static int cudaDeviceSetLimit(@Cast(value="cudaLimit") int _limit, @Cast(value="size_t") long value)
limit
- - Limit to setvalue
- - Size of limit::cudaDeviceGetLimit,
::cuCtxSetLimit
@Cast(value="cudaError_t") public static int cudaDeviceGetLimit(@Cast(value="size_t*") SizeTPointer pValue, @Cast(value="cudaLimit") int _limit)
limit
- - Limit to querypValue
- - Returned size of the limit::cudaDeviceSetLimit,
::cuCtxGetLimit
@Cast(value="cudaError_t") public static int cudaDeviceGetTexture1DLinearMaxWidth(@Cast(value="size_t*") SizeTPointer maxWidthInElements, @Const cudaChannelFormatDesc fmtDesc, int device)
maxWidthInElements
- - Returns maximum number of texture elements allocatable for given \p fmtDesc.fmtDesc
- - Texture format description.::cuDeviceGetTexture1DLinearMaxWidth
@Cast(value="cudaError_t") public static int cudaDeviceGetCacheConfig(@Cast(value="cudaFuncCache*") IntPointer pCacheConfig)
pCacheConfig
- - Returned cache configuration::cudaDeviceSetCacheConfig,
\ref ::cudaFuncSetCacheConfig(const void*, enum cudaFuncCache) "cudaFuncSetCacheConfig (C API)",
\ref ::cudaFuncSetCacheConfig(T*, enum cudaFuncCache) "cudaFuncSetCacheConfig (C++ API)",
::cuCtxGetCacheConfig
@Cast(value="cudaError_t") public static int cudaDeviceGetCacheConfig(@Cast(value="cudaFuncCache*") IntBuffer pCacheConfig)
@Cast(value="cudaError_t") public static int cudaDeviceGetCacheConfig(@Cast(value="cudaFuncCache*") int[] pCacheConfig)
@Cast(value="cudaError_t") public static int cudaDeviceGetStreamPriorityRange(IntPointer leastPriority, IntPointer greatestPriority)
leastPriority
- - Pointer to an int in which the numerical value for least
stream priority is returnedgreatestPriority
- - Pointer to an int in which the numerical value for greatest
stream priority is returned::cudaStreamCreateWithPriority,
::cudaStreamGetPriority,
::cuCtxGetStreamPriorityRange
@Cast(value="cudaError_t") public static int cudaDeviceGetStreamPriorityRange(IntBuffer leastPriority, IntBuffer greatestPriority)
@Cast(value="cudaError_t") public static int cudaDeviceGetStreamPriorityRange(int[] leastPriority, int[] greatestPriority)
@Cast(value="cudaError_t") public static int cudaDeviceSetCacheConfig(@Cast(value="cudaFuncCache") int cacheConfig)
cacheConfig
- - Requested cache configuration::cudaDeviceGetCacheConfig,
\ref ::cudaFuncSetCacheConfig(const void*, enum cudaFuncCache) "cudaFuncSetCacheConfig (C API)",
\ref ::cudaFuncSetCacheConfig(T*, enum cudaFuncCache) "cudaFuncSetCacheConfig (C++ API)",
::cuCtxSetCacheConfig
@Cast(value="cudaError_t") public static int cudaDeviceGetSharedMemConfig(@Cast(value="cudaSharedMemConfig*") IntPointer pConfig)
pConfig
- - Returned cache configuration::cudaDeviceSetCacheConfig,
::cudaDeviceGetCacheConfig,
::cudaDeviceSetSharedMemConfig,
::cudaFuncSetCacheConfig,
::cuCtxGetSharedMemConfig
@Cast(value="cudaError_t") public static int cudaDeviceGetSharedMemConfig(@Cast(value="cudaSharedMemConfig*") IntBuffer pConfig)
@Cast(value="cudaError_t") public static int cudaDeviceGetSharedMemConfig(@Cast(value="cudaSharedMemConfig*") int[] pConfig)
@Cast(value="cudaError_t") public static int cudaDeviceSetSharedMemConfig(@Cast(value="cudaSharedMemConfig") int config)
config
- - Requested cache configuration::cudaDeviceSetCacheConfig,
::cudaDeviceGetCacheConfig,
::cudaDeviceGetSharedMemConfig,
::cudaFuncSetCacheConfig,
::cuCtxSetSharedMemConfig
@Cast(value="cudaError_t") public static int cudaDeviceGetByPCIBusId(IntPointer device, @Cast(value="const char*") BytePointer pciBusId)
device
- - Returned device ordinalpciBusId
- - String in one of the following forms:
[domain]:[bus]:[device].[function]
[domain]:[bus]:[device]
[bus]:[device].[function]
where \p domain, \p bus, \p device, and \p function are all hexadecimal values::cudaDeviceGetPCIBusId,
::cuDeviceGetByPCIBusId
@Cast(value="cudaError_t") public static int cudaDeviceGetByPCIBusId(IntBuffer device, String pciBusId)
@Cast(value="cudaError_t") public static int cudaDeviceGetByPCIBusId(int[] device, @Cast(value="const char*") BytePointer pciBusId)
@Cast(value="cudaError_t") public static int cudaDeviceGetByPCIBusId(IntPointer device, String pciBusId)
@Cast(value="cudaError_t") public static int cudaDeviceGetByPCIBusId(IntBuffer device, @Cast(value="const char*") BytePointer pciBusId)
@Cast(value="cudaError_t") public static int cudaDeviceGetByPCIBusId(int[] device, String pciBusId)
@Cast(value="cudaError_t") public static int cudaDeviceGetPCIBusId(@Cast(value="char*") BytePointer pciBusId, int len, int device)
pciBusId
- - Returned identifier string for the device in the following format
[domain]:[bus]:[device].[function]
where \p domain, \p bus, \p device, and \p function are all hexadecimal values.
pciBusId should be large enough to store 13 characters including the NULL-terminator.len
- - Maximum length of string to store in \p namedevice
- - Device to get identifier string for::cudaDeviceGetByPCIBusId,
::cuDeviceGetPCIBusId
@Cast(value="cudaError_t") public static int cudaDeviceGetPCIBusId(@Cast(value="char*") ByteBuffer pciBusId, int len, int device)
@Cast(value="cudaError_t") public static int cudaDeviceGetPCIBusId(@Cast(value="char*") byte[] pciBusId, int len, int device)
@Cast(value="cudaError_t") public static int cudaIpcGetEventHandle(cudaIpcEventHandle_t handle, CUevent_st event)
handle
- - Pointer to a user allocated cudaIpcEventHandle
in which to return the opaque event handleevent
- - Event allocated with ::cudaEventInterprocess and
::cudaEventDisableTiming flags.::cudaEventCreate,
::cudaEventDestroy,
::cudaEventSynchronize,
::cudaEventQuery,
::cudaStreamWaitEvent,
::cudaIpcOpenEventHandle,
::cudaIpcGetMemHandle,
::cudaIpcOpenMemHandle,
::cudaIpcCloseMemHandle,
::cuIpcGetEventHandle
@Cast(value="cudaError_t") public static int cudaIpcOpenEventHandle(@ByPtrPtr CUevent_st event, @ByVal cudaIpcEventHandle_t handle)
event
- - Returns the imported eventhandle
- - Interprocess handle to open::cudaEventCreate,
::cudaEventDestroy,
::cudaEventSynchronize,
::cudaEventQuery,
::cudaStreamWaitEvent,
::cudaIpcGetEventHandle,
::cudaIpcGetMemHandle,
::cudaIpcOpenMemHandle,
::cudaIpcCloseMemHandle,
::cuIpcOpenEventHandle
@Cast(value="cudaError_t") public static int cudaIpcGetMemHandle(cudaIpcMemHandle_t handle, Pointer devPtr)
handle
- - Pointer to user allocated ::cudaIpcMemHandle to return
the handle in.devPtr
- - Base pointer to previously allocated device memory::cudaMalloc,
::cudaFree,
::cudaIpcGetEventHandle,
::cudaIpcOpenEventHandle,
::cudaIpcOpenMemHandle,
::cudaIpcCloseMemHandle,
::cuIpcGetMemHandle
@Cast(value="cudaError_t") public static int cudaIpcOpenMemHandle(@Cast(value="void**") PointerPointer devPtr, @ByVal cudaIpcMemHandle_t handle, @Cast(value="unsigned int") int flags)
devPtr
- - Returned device pointerhandle
- - ::cudaIpcMemHandle to openflags
- - Flags for this operation. Must be specified as ::cudaIpcMemLazyEnablePeerAccess::cudaMalloc,
::cudaFree,
::cudaIpcGetEventHandle,
::cudaIpcOpenEventHandle,
::cudaIpcGetMemHandle,
::cudaIpcCloseMemHandle,
::cudaDeviceEnablePeerAccess,
::cudaDeviceCanAccessPeer,
::cuIpcOpenMemHandle
@Cast(value="cudaError_t") public static int cudaIpcOpenMemHandle(@Cast(value="void**") @ByPtrPtr Pointer devPtr, @ByVal cudaIpcMemHandle_t handle, @Cast(value="unsigned int") int flags)
@Cast(value="cudaError_t") public static int cudaIpcCloseMemHandle(Pointer devPtr)
devPtr
- - Device pointer returned by ::cudaIpcOpenMemHandle::cudaMalloc,
::cudaFree,
::cudaIpcGetEventHandle,
::cudaIpcOpenEventHandle,
::cudaIpcGetMemHandle,
::cudaIpcOpenMemHandle,
::cuIpcCloseMemHandle
@Cast(value="cudaError_t") public static int cudaDeviceFlushGPUDirectRDMAWrites(@Cast(value="cudaFlushGPUDirectRDMAWritesTarget") int target, @Cast(value="cudaFlushGPUDirectRDMAWritesScope") int scope)
target
- - The target of the operation, see cudaFlushGPUDirectRDMAWritesTargetscope
- - The scope of the operation, see cudaFlushGPUDirectRDMAWritesScope::cuFlushGPUDirectRDMAWrites
@Cast(value="cudaError_t") @Deprecated public static int cudaThreadExit()
::cudaDeviceReset
@Cast(value="cudaError_t") @Deprecated public static int cudaThreadSynchronize()
::cudaDeviceSynchronize
@Cast(value="cudaError_t") @Deprecated public static int cudaThreadSetLimit(@Cast(value="cudaLimit") int _limit, @Cast(value="size_t") long value)
limit
- - Limit to setvalue
- - Size in bytes of limit::cudaDeviceSetLimit
@Cast(value="cudaError_t") @Deprecated public static int cudaThreadGetLimit(@Cast(value="size_t*") SizeTPointer pValue, @Cast(value="cudaLimit") int _limit)
limit
- - Limit to querypValue
- - Returned size in bytes of limit::cudaDeviceGetLimit
@Cast(value="cudaError_t") @Deprecated public static int cudaThreadGetCacheConfig(@Cast(value="cudaFuncCache*") IntPointer pCacheConfig)
pCacheConfig
- - Returned cache configuration::cudaDeviceGetCacheConfig
@Cast(value="cudaError_t") @Deprecated public static int cudaThreadGetCacheConfig(@Cast(value="cudaFuncCache*") IntBuffer pCacheConfig)
@Cast(value="cudaError_t") @Deprecated public static int cudaThreadGetCacheConfig(@Cast(value="cudaFuncCache*") int[] pCacheConfig)
@Cast(value="cudaError_t") @Deprecated public static int cudaThreadSetCacheConfig(@Cast(value="cudaFuncCache") int cacheConfig)
cacheConfig
- - Requested cache configuration::cudaDeviceSetCacheConfig
@Cast(value="cudaError_t") public static int cudaGetLastError()
::cudaPeekAtLastError, ::cudaGetErrorName, ::cudaGetErrorString, ::cudaError
@Cast(value="cudaError_t") public static int cudaPeekAtLastError()
::cudaGetLastError, ::cudaGetErrorName, ::cudaGetErrorString, ::cudaError
@Cast(value="const char*") public static BytePointer cudaGetErrorName(@Cast(value="cudaError_t") int error)
error
- - Error code to convert to string::cudaGetErrorString, ::cudaGetLastError, ::cudaPeekAtLastError, ::cudaError,
::cuGetErrorName
@Cast(value="const char*") public static BytePointer cudaGetErrorString(@Cast(value="cudaError_t") int error)
error
- - Error code to convert to string::cudaGetErrorName, ::cudaGetLastError, ::cudaPeekAtLastError, ::cudaError,
::cuGetErrorString
@Cast(value="cudaError_t") public static int cudaGetDeviceCount(IntPointer count)
count
- - Returns the number of devices with compute capability
greater or equal to 2.0::cudaGetDevice, ::cudaSetDevice, ::cudaGetDeviceProperties,
::cudaChooseDevice,
::cudaInitDevice,
::cuDeviceGetCount
@Cast(value="cudaError_t") public static int cudaGetDeviceProperties(cudaDeviceProp prop, int device)
struct cudaDeviceProp {
char name[256];
cudaUUID_t uuid;
size_t totalGlobalMem;
size_t sharedMemPerBlock;
int regsPerBlock;
int warpSize;
size_t memPitch;
int maxThreadsPerBlock;
int maxThreadsDim[3];
int maxGridSize[3];
int clockRate;
size_t totalConstMem;
int major;
int minor;
size_t textureAlignment;
size_t texturePitchAlignment;
int deviceOverlap;
int multiProcessorCount;
int kernelExecTimeoutEnabled;
int integrated;
int canMapHostMemory;
int computeMode;
int maxTexture1D;
int maxTexture1DMipmap;
int maxTexture1DLinear;
int maxTexture2D[2];
int maxTexture2DMipmap[2];
int maxTexture2DLinear[3];
int maxTexture2DGather[2];
int maxTexture3D[3];
int maxTexture3DAlt[3];
int maxTextureCubemap;
int maxTexture1DLayered[2];
int maxTexture2DLayered[3];
int maxTextureCubemapLayered[2];
int maxSurface1D;
int maxSurface2D[2];
int maxSurface3D[3];
int maxSurface1DLayered[2];
int maxSurface2DLayered[3];
int maxSurfaceCubemap;
int maxSurfaceCubemapLayered[2];
size_t surfaceAlignment;
int concurrentKernels;
int ECCEnabled;
int pciBusID;
int pciDeviceID;
int pciDomainID;
int tccDriver;
int asyncEngineCount;
int unifiedAddressing;
int memoryClockRate;
int memoryBusWidth;
int l2CacheSize;
int persistingL2CacheMaxSize;
int maxThreadsPerMultiProcessor;
int streamPrioritiesSupported;
int globalL1CacheSupported;
int localL1CacheSupported;
size_t sharedMemPerMultiprocessor;
int regsPerMultiprocessor;
int managedMemory;
int isMultiGpuBoard;
int multiGpuBoardGroupID;
int singleToDoublePrecisionPerfRatio;
int pageableMemoryAccess;
int concurrentManagedAccess;
int computePreemptionSupported;
int canUseHostPointerForRegisteredMem;
int cooperativeLaunch;
int cooperativeMultiDeviceLaunch;
int pageableMemoryAccessUsesHostPageTables;
int directManagedMemAccessFromHost;
int accessPolicyMaxWindowSize;
}
where:
- \ref ::cudaDeviceProp::name "name[256]" is an ASCII string identifying
the device.
- \ref ::cudaDeviceProp::uuid "uuid" is a 16-byte unique identifier.
- \ref ::cudaDeviceProp::totalGlobalMem "totalGlobalMem" is the total
amount of global memory available on the device in bytes.
- \ref ::cudaDeviceProp::sharedMemPerBlock "sharedMemPerBlock" is the
maximum amount of shared memory available to a thread block in bytes.
- \ref ::cudaDeviceProp::regsPerBlock "regsPerBlock" is the maximum number
of 32-bit registers available to a thread block.
- \ref ::cudaDeviceProp::warpSize "warpSize" is the warp size in threads.
- \ref ::cudaDeviceProp::memPitch "memPitch" is the maximum pitch in
bytes allowed by the memory copy functions that involve memory regions
allocated through ::cudaMallocPitch().
- \ref ::cudaDeviceProp::maxThreadsPerBlock "maxThreadsPerBlock" is the
maximum number of threads per block.
- \ref ::cudaDeviceProp::maxThreadsDim "maxThreadsDim[3]" contains the
maximum size of each dimension of a block.
- \ref ::cudaDeviceProp::maxGridSize "maxGridSize[3]" contains the
maximum size of each dimension of a grid.
- \ref ::cudaDeviceProp::clockRate "clockRate" is the clock frequency in
kilohertz.
- \ref ::cudaDeviceProp::totalConstMem "totalConstMem" is the total amount
of constant memory available on the device in bytes.
- \ref ::cudaDeviceProp::major "major",
\ref ::cudaDeviceProp::minor "minor" are the major and minor revision
numbers defining the device's compute capability.
- \ref ::cudaDeviceProp::textureAlignment "textureAlignment" is the
alignment requirement; texture base addresses that are aligned to
\ref ::cudaDeviceProp::textureAlignment "textureAlignment" bytes do not
need an offset applied to texture fetches.
- \ref ::cudaDeviceProp::texturePitchAlignment "texturePitchAlignment" is the
pitch alignment requirement for 2D texture references that are bound to
pitched memory.
- \ref ::cudaDeviceProp::deviceOverlap "deviceOverlap" is 1 if the device
can concurrently copy memory between host and device while executing a
kernel, or 0 if not. Deprecated, use instead asyncEngineCount.
- \ref ::cudaDeviceProp::multiProcessorCount "multiProcessorCount" is the
number of multiprocessors on the device.
- \ref ::cudaDeviceProp::kernelExecTimeoutEnabled "kernelExecTimeoutEnabled"
is 1 if there is a run time limit for kernels executed on the device, or
0 if not.
- \ref ::cudaDeviceProp::integrated "integrated" is 1 if the device is an
integrated (motherboard) GPU and 0 if it is a discrete (card) component.
- \ref ::cudaDeviceProp::canMapHostMemory "canMapHostMemory" is 1 if the
device can map host memory into the CUDA address space for use with
::cudaHostAlloc()/::cudaHostGetDevicePointer(), or 0 if not.
- \ref ::cudaDeviceProp::computeMode "computeMode" is the compute mode
that the device is currently in. Available modes are as follows:
- cudaComputeModeDefault: Default mode - Device is not restricted and
multiple threads can use ::cudaSetDevice() with this device.
- cudaComputeModeProhibited: Compute-prohibited mode - No threads can use
::cudaSetDevice() with this device.
- cudaComputeModeExclusiveProcess: Compute-exclusive-process mode - Many
threads in one process will be able to use ::cudaSetDevice() with this device.
prop
- - Properties for the specified devicedevice
- - Device number to get properties for::cudaGetDeviceCount, ::cudaGetDevice, ::cudaSetDevice, ::cudaChooseDevice,
::cudaDeviceGetAttribute,
::cudaInitDevice,
::cuDeviceGetAttribute,
::cuDeviceGetName
@Cast(value="cudaError_t") public static int cudaDeviceGetAttribute(IntPointer value, @Cast(value="cudaDeviceAttr") int attr, int device)
value
- - Returned device attribute valueattr
- - Device attribute to querydevice
- - Device number to query::cudaGetDeviceCount, ::cudaGetDevice, ::cudaSetDevice, ::cudaChooseDevice,
::cudaGetDeviceProperties,
::cudaInitDevice,
::cuDeviceGetAttribute
@Cast(value="cudaError_t") public static int cudaDeviceGetAttribute(IntBuffer value, @Cast(value="cudaDeviceAttr") int attr, int device)
@Cast(value="cudaError_t") public static int cudaDeviceGetAttribute(int[] value, @Cast(value="cudaDeviceAttr") int attr, int device)
@Cast(value="cudaError_t") public static int cudaDeviceGetDefaultMemPool(@ByPtrPtr CUmemPoolHandle_st memPool, int device)
::cuDeviceGetDefaultMemPool, ::cudaMallocAsync, ::cudaMemPoolTrimTo, ::cudaMemPoolGetAttribute, ::cudaDeviceSetMemPool, ::cudaMemPoolSetAttribute, ::cudaMemPoolSetAccess
@Cast(value="cudaError_t") public static int cudaDeviceSetMemPool(int device, CUmemPoolHandle_st memPool)
::cuDeviceSetMemPool, ::cudaDeviceGetMemPool, ::cudaDeviceGetDefaultMemPool, ::cudaMemPoolCreate, ::cudaMemPoolDestroy, ::cudaMallocFromPoolAsync
@Cast(value="cudaError_t") public static int cudaDeviceGetMemPool(@ByPtrPtr CUmemPoolHandle_st memPool, int device)
::cuDeviceGetMemPool, ::cudaDeviceGetDefaultMemPool, ::cudaDeviceSetMemPool
@Cast(value="cudaError_t") public static int cudaDeviceGetP2PAttribute(IntPointer value, @Cast(value="cudaDeviceP2PAttr") int attr, int srcDevice, int dstDevice)
value
- - Returned value of the requested attributeattrib
- - The requested attribute of the link between \p srcDevice and \p dstDevice.srcDevice
- - The source device of the target link.dstDevice
- - The destination device of the target link.::cudaDeviceEnablePeerAccess,
::cudaDeviceDisablePeerAccess,
::cudaDeviceCanAccessPeer,
::cuDeviceGetP2PAttribute
@Cast(value="cudaError_t") public static int cudaDeviceGetP2PAttribute(IntBuffer value, @Cast(value="cudaDeviceP2PAttr") int attr, int srcDevice, int dstDevice)
@Cast(value="cudaError_t") public static int cudaDeviceGetP2PAttribute(int[] value, @Cast(value="cudaDeviceP2PAttr") int attr, int srcDevice, int dstDevice)
@Cast(value="cudaError_t") public static int cudaChooseDevice(IntPointer device, @Const cudaDeviceProp prop)
device
- - Device with best matchprop
- - Desired device properties::cudaGetDeviceCount, ::cudaGetDevice, ::cudaSetDevice,
::cudaGetDeviceProperties,
::cudaInitDevice
@Cast(value="cudaError_t") public static int cudaChooseDevice(IntBuffer device, @Const cudaDeviceProp prop)
@Cast(value="cudaError_t") public static int cudaChooseDevice(int[] device, @Const cudaDeviceProp prop)
@Cast(value="cudaError_t") public static int cudaInitDevice(int device, @Cast(value="unsigned int") int deviceFlags, @Cast(value="unsigned int") int flags)
device
- - Device on which the runtime will initialize itself.deviceFlags
- - Parameters for device operation.flags
- - Flags for controlling the device initialization.::cudaGetDeviceCount, ::cudaGetDevice, ::cudaGetDeviceProperties,
::cudaChooseDevice, ::cudaSetDevice
::cuCtxSetCurrent
@Cast(value="cudaError_t") public static int cudaSetDevice(int device)
device
- - Device on which the active host thread should execute the
device code.::cudaGetDeviceCount, ::cudaGetDevice, ::cudaGetDeviceProperties,
::cudaChooseDevice,
::cudaInitDevice,
::cuCtxSetCurrent
@Cast(value="cudaError_t") public static int cudaGetDevice(IntPointer device)
device
- - Returns the device on which the active host thread
executes the device code.::cudaGetDeviceCount, ::cudaSetDevice, ::cudaGetDeviceProperties,
::cudaChooseDevice,
::cuCtxGetCurrent
@Cast(value="cudaError_t") public static int cudaSetValidDevices(IntPointer device_arr, int len)
device_arr
- - List of devices to trylen
- - Number of devices in specified list::cudaGetDeviceCount, ::cudaSetDevice, ::cudaGetDeviceProperties,
::cudaSetDeviceFlags,
::cudaChooseDevice
@Cast(value="cudaError_t") public static int cudaSetValidDevices(IntBuffer device_arr, int len)
@Cast(value="cudaError_t") public static int cudaSetValidDevices(int[] device_arr, int len)
@Cast(value="cudaError_t") public static int cudaSetDeviceFlags(@Cast(value="unsigned int") int flags)
flags
- - Parameters for device operation::cudaGetDeviceFlags, ::cudaGetDeviceCount, ::cudaGetDevice, ::cudaGetDeviceProperties,
::cudaSetDevice, ::cudaSetValidDevices,
::cudaInitDevice,
::cudaChooseDevice,
::cuDevicePrimaryCtxSetFlags
@Cast(value="cudaError_t") public static int cudaGetDeviceFlags(@Cast(value="unsigned int*") IntPointer flags)
flags
- - Pointer to store the device flags::cudaGetDevice, ::cudaGetDeviceProperties,
::cudaSetDevice, ::cudaSetDeviceFlags,
::cudaInitDevice,
::cuCtxGetFlags,
::cuDevicePrimaryCtxGetState
@Cast(value="cudaError_t") public static int cudaGetDeviceFlags(@Cast(value="unsigned int*") IntBuffer flags)
@Cast(value="cudaError_t") public static int cudaGetDeviceFlags(@Cast(value="unsigned int*") int[] flags)
@Cast(value="cudaError_t") public static int cudaStreamCreate(@ByPtrPtr CUstream_st pStream)
pStream
- - Pointer to new stream identifier::cudaStreamCreateWithPriority,
::cudaStreamCreateWithFlags,
::cudaStreamGetPriority,
::cudaStreamGetFlags,
::cudaStreamQuery,
::cudaStreamSynchronize,
::cudaStreamWaitEvent,
::cudaStreamAddCallback,
::cudaStreamDestroy,
::cuStreamCreate
@Cast(value="cudaError_t") public static int cudaStreamCreateWithFlags(@ByPtrPtr CUstream_st pStream, @Cast(value="unsigned int") int flags)
pStream
- - Pointer to new stream identifierflags
- - Parameters for stream creation::cudaStreamCreate,
::cudaStreamCreateWithPriority,
::cudaStreamGetFlags,
::cudaStreamQuery,
::cudaStreamSynchronize,
::cudaStreamWaitEvent,
::cudaStreamAddCallback,
::cudaStreamDestroy,
::cuStreamCreate
@Cast(value="cudaError_t") public static int cudaStreamCreateWithPriority(@ByPtrPtr CUstream_st pStream, @Cast(value="unsigned int") int flags, int priority)
pStream
- - Pointer to new stream identifierflags
- - Flags for stream creation. See ::cudaStreamCreateWithFlags for a list of valid flags that can be passedpriority
- - Priority of the stream. Lower numbers represent higher priorities.
See ::cudaDeviceGetStreamPriorityRange for more information about
the meaningful stream priorities that can be passed.::cudaStreamCreate,
::cudaStreamCreateWithFlags,
::cudaDeviceGetStreamPriorityRange,
::cudaStreamGetPriority,
::cudaStreamQuery,
::cudaStreamWaitEvent,
::cudaStreamAddCallback,
::cudaStreamSynchronize,
::cudaStreamDestroy,
::cuStreamCreateWithPriority
@Cast(value="cudaError_t") public static int cudaStreamGetPriority(CUstream_st hStream, IntPointer priority)
hStream
- - Handle to the stream to be queriedpriority
- - Pointer to a signed integer in which the stream's priority is returned::cudaStreamCreateWithPriority,
::cudaDeviceGetStreamPriorityRange,
::cudaStreamGetFlags,
::cuStreamGetPriority
@Cast(value="cudaError_t") public static int cudaStreamGetPriority(CUstream_st hStream, IntBuffer priority)
@Cast(value="cudaError_t") public static int cudaStreamGetPriority(CUstream_st hStream, int[] priority)
@Cast(value="cudaError_t") public static int cudaStreamGetFlags(CUstream_st hStream, @Cast(value="unsigned int*") IntPointer flags)
hStream
- - Handle to the stream to be queriedflags
- - Pointer to an unsigned integer in which the stream's flags are returned::cudaStreamCreateWithPriority,
::cudaStreamCreateWithFlags,
::cudaStreamGetPriority,
::cuStreamGetFlags
@Cast(value="cudaError_t") public static int cudaStreamGetFlags(CUstream_st hStream, @Cast(value="unsigned int*") IntBuffer flags)
@Cast(value="cudaError_t") public static int cudaStreamGetFlags(CUstream_st hStream, @Cast(value="unsigned int*") int[] flags)
@Cast(value="cudaError_t") public static int cudaStreamGetId(CUstream_st hStream, @Cast(value="unsigned long long*") LongPointer streamId)
hStream
- - Handle to the stream to be queriedstreamId
- - Pointer to an unsigned long long in which the stream Id is returned::cudaStreamCreateWithPriority,
::cudaStreamCreateWithFlags,
::cudaStreamGetPriority,
::cudaStreamGetFlags,
::cuStreamGetId
@Cast(value="cudaError_t") public static int cudaStreamGetId(CUstream_st hStream, @Cast(value="unsigned long long*") LongBuffer streamId)
@Cast(value="cudaError_t") public static int cudaStreamGetId(CUstream_st hStream, @Cast(value="unsigned long long*") long[] streamId)
@Cast(value="cudaError_t") public static int cudaCtxResetPersistingL2Cache()
::cudaAccessPolicyWindow
@Cast(value="cudaError_t") public static int cudaStreamCopyAttributes(CUstream_st dst, CUstream_st src)
dst
- [out] Destination streamsrc
- [in] Source stream
For attributes see ::cudaStreamAttrID::cudaAccessPolicyWindow
@Cast(value="cudaError_t") public static int cudaStreamGetAttribute(CUstream_st hStream, @Cast(value="cudaLaunchAttributeID") int attr, cudaLaunchAttributeValue value_out)
hStream
- [in]attr
- [in]value_out
- [out]::cudaAccessPolicyWindow
@Cast(value="cudaError_t") public static int cudaStreamSetAttribute(CUstream_st hStream, @Cast(value="cudaLaunchAttributeID") int attr, @Const cudaLaunchAttributeValue value)
hStream
- [out]attr
- [in]value
- [in]::cudaAccessPolicyWindow
@Cast(value="cudaError_t") public static int cudaStreamDestroy(CUstream_st stream)
stream
- - Stream identifier::cudaStreamCreate,
::cudaStreamCreateWithFlags,
::cudaStreamQuery,
::cudaStreamWaitEvent,
::cudaStreamSynchronize,
::cudaStreamAddCallback,
::cuStreamDestroy
@Cast(value="cudaError_t") public static int cudaStreamWaitEvent(CUstream_st stream, CUevent_st event, @Cast(value="unsigned int") int flags)
stream
- - Stream to waitevent
- - Event to wait onflags
- - Parameters for the operation(See above)::cudaStreamCreate, ::cudaStreamCreateWithFlags, ::cudaStreamQuery, ::cudaStreamSynchronize, ::cudaStreamAddCallback, ::cudaStreamDestroy,
::cuStreamWaitEvent
@Cast(value="cudaError_t") public static int cudaStreamWaitEvent(CUstream_st stream, CUevent_st event)
@Cast(value="cudaError_t") public static int cudaStreamAddCallback(CUstream_st stream, cudaStreamCallback_t callback, Pointer userData, @Cast(value="unsigned int") int flags)
stream
- - Stream to add callback tocallback
- - The function to call once preceding stream operations are completeuserData
- - User specified data to be passed to the callback functionflags
- - Reserved for future use, must be 0::cudaStreamCreate, ::cudaStreamCreateWithFlags, ::cudaStreamQuery, ::cudaStreamSynchronize, ::cudaStreamWaitEvent, ::cudaStreamDestroy, ::cudaMallocManaged, ::cudaStreamAttachMemAsync,
::cudaLaunchHostFunc, ::cuStreamAddCallback
@Cast(value="cudaError_t") public static int cudaStreamSynchronize(CUstream_st stream)
stream
- - Stream identifier::cudaStreamCreate, ::cudaStreamCreateWithFlags, ::cudaStreamQuery, ::cudaStreamWaitEvent, ::cudaStreamAddCallback, ::cudaStreamDestroy,
::cuStreamSynchronize
@Cast(value="cudaError_t") public static int cudaStreamQuery(CUstream_st stream)
stream
- - Stream identifier::cudaStreamCreate, ::cudaStreamCreateWithFlags, ::cudaStreamWaitEvent, ::cudaStreamSynchronize, ::cudaStreamAddCallback, ::cudaStreamDestroy,
::cuStreamQuery
@Cast(value="cudaError_t") public static int cudaStreamAttachMemAsync(CUstream_st stream, Pointer devPtr, @Cast(value="size_t") long length, @Cast(value="unsigned int") int flags)
stream
- - Stream in which to enqueue the attach operationdevPtr
- - Pointer to memory (must be a pointer to managed memory or
to a valid host-accessible region of system-allocated
memory)length
- - Length of memory (defaults to zero)flags
- - Must be one of ::cudaMemAttachGlobal, ::cudaMemAttachHost or ::cudaMemAttachSingle (defaults to ::cudaMemAttachSingle)::cudaStreamCreate, ::cudaStreamCreateWithFlags, ::cudaStreamWaitEvent, ::cudaStreamSynchronize, ::cudaStreamAddCallback, ::cudaStreamDestroy, ::cudaMallocManaged,
::cuStreamAttachMemAsync
@Cast(value="cudaError_t") public static int cudaStreamAttachMemAsync(CUstream_st stream, Pointer devPtr)
@Cast(value="cudaError_t") public static int cudaStreamAttachMemAsync(CUstream_st stream, Pointer devPtr, @Cast(value="unsigned int") int flags)
@Cast(value="cudaError_t") public static int cudaStreamBeginCapture(CUstream_st stream, @Cast(value="cudaStreamCaptureMode") int mode)
stream
- - Stream in which to initiate capturemode
- - Controls the interaction of this capture sequence with other API
calls that are potentially unsafe. For more details see
::cudaThreadExchangeStreamCaptureMode.::cudaStreamCreate,
::cudaStreamIsCapturing,
::cudaStreamEndCapture,
::cudaThreadExchangeStreamCaptureMode
@Cast(value="cudaError_t") public static int cudaStreamBeginCaptureToGraph(CUstream_st stream, CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st dependencies, @Const cudaGraphEdgeData dependencyData, @Cast(value="size_t") long numDependencies, @Cast(value="cudaStreamCaptureMode") int mode)
stream
- - Stream in which to initiate capture.graph
- - Graph to capture into.dependencies
- - Dependencies of the first node captured in the stream. Can be NULL if numDependencies is 0.dependencyData
- - Optional array of data associated with each dependency.numDependencies
- - Number of dependencies.mode
- - Controls the interaction of this capture sequence with other API
calls that are potentially unsafe. For more details see
::cudaThreadExchangeStreamCaptureMode.::cudaStreamCreate,
::cudaStreamIsCapturing,
::cudaStreamEndCapture,
::cudaThreadExchangeStreamCaptureMode
@Cast(value="cudaError_t") public static int cudaThreadExchangeStreamCaptureMode(@Cast(value="cudaStreamCaptureMode*") IntPointer mode)
cudaStreamCaptureMode mode = desiredMode;
cudaThreadExchangeStreamCaptureMode(&mode);
...
cudaThreadExchangeStreamCaptureMode(&mode); // restore previous mode
During stream capture (see ::cudaStreamBeginCapture), some actions, such as a call
to ::cudaMalloc, may be unsafe. In the case of ::cudaMalloc, the operation is
not enqueued asynchronously to a stream, and is not observed by stream capture.
Therefore, if the sequence of operations captured via ::cudaStreamBeginCapture
depended on the allocation being replayed whenever the graph is launched, the
captured graph would be invalid.
Therefore, stream capture places restrictions on API calls that can be made within
or concurrently to a ::cudaStreamBeginCapture-::cudaStreamEndCapture sequence. This
behavior can be controlled via this API and flags to ::cudaStreamBeginCapture.
A thread's mode is one of the following:
- \p cudaStreamCaptureModeGlobal: This is the default mode. If the local thread has
an ongoing capture sequence that was not initiated with
\p cudaStreamCaptureModeRelaxed at \p cuStreamBeginCapture, or if any other thread
has a concurrent capture sequence initiated with \p cudaStreamCaptureModeGlobal,
this thread is prohibited from potentially unsafe API calls.
- \p cudaStreamCaptureModeThreadLocal: If the local thread has an ongoing capture
sequence not initiated with \p cudaStreamCaptureModeRelaxed, it is prohibited
from potentially unsafe API calls. Concurrent capture sequences in other threads
are ignored.
- \p cudaStreamCaptureModeRelaxed: The local thread is not prohibited from potentially
unsafe API calls. Note that the thread is still prohibited from API calls which
necessarily conflict with stream capture, for example, attempting ::cudaEventQuery
on an event that was last recorded inside a capture sequence.mode
- - Pointer to mode value to swap with the current mode::cudaStreamBeginCapture
@Cast(value="cudaError_t") public static int cudaThreadExchangeStreamCaptureMode(@Cast(value="cudaStreamCaptureMode*") IntBuffer mode)
@Cast(value="cudaError_t") public static int cudaThreadExchangeStreamCaptureMode(@Cast(value="cudaStreamCaptureMode*") int[] mode)
@Cast(value="cudaError_t") public static int cudaStreamEndCapture(CUstream_st stream, @ByPtrPtr CUgraph_st pGraph)
stream
- - Stream to querypGraph
- - The captured graph::cudaStreamCreate,
::cudaStreamBeginCapture,
::cudaStreamIsCapturing,
::cudaGraphDestroy
@Cast(value="cudaError_t") public static int cudaStreamIsCapturing(CUstream_st stream, @Cast(value="cudaStreamCaptureStatus*") IntPointer pCaptureStatus)
stream
- - Stream to querypCaptureStatus
- - Returns the stream's capture status::cudaStreamCreate,
::cudaStreamBeginCapture,
::cudaStreamEndCapture
@Cast(value="cudaError_t") public static int cudaStreamIsCapturing(CUstream_st stream, @Cast(value="cudaStreamCaptureStatus*") IntBuffer pCaptureStatus)
@Cast(value="cudaError_t") public static int cudaStreamIsCapturing(CUstream_st stream, @Cast(value="cudaStreamCaptureStatus*") int[] pCaptureStatus)
@Cast(value="cudaError_t") public static int cudaStreamGetCaptureInfo(CUstream_st stream, @Cast(value="cudaStreamCaptureStatus*") IntPointer captureStatus_out, @Cast(value="unsigned long long*") LongPointer id_out, @ByPtrPtr CUgraph_st graph_out, @Cast(value="const cudaGraphNode_t**") @ByPtrPtr PointerPointer dependencies_out, @Cast(value="size_t*") SizeTPointer numDependencies_out)
stream
- - The stream to querycaptureStatus_out
- - Location to return the capture status of the stream; requiredid_out
- - Optional location to return an id for the capture sequence, which is
unique over the lifetime of the processgraph_out
- - Optional location to return the graph being captured into. All
operations other than destroy and node removal are permitted on the graph
while the capture sequence is in progress. This API does not transfer
ownership of the graph, which is transferred or destroyed at
::cudaStreamEndCapture. Note that the graph handle may be invalidated before
end of capture for certain errors. Nodes that are or become
unreachable from the original stream at ::cudaStreamEndCapture due to direct
actions on the graph do not trigger ::cudaErrorStreamCaptureUnjoined.dependencies_out
- - Optional location to store a pointer to an array of nodes.
The next node to be captured in the stream will depend on this set of nodes,
absent operations such as event wait which modify this set. The array pointer
is valid until the next API call which operates on the stream or until the
capture is terminated. The node handles may be copied out and are valid until
they or the graph is destroyed. The driver-owned array may also be passed
directly to APIs that operate on the graph (not the stream) without copying.numDependencies_out
- - Optional location to store the size of the array
returned in dependencies_out.::cudaStreamGetCaptureInfo_v3,
::cudaStreamBeginCapture,
::cudaStreamIsCapturing,
::cudaStreamUpdateCaptureDependencies
@Cast(value="cudaError_t") public static int cudaStreamGetCaptureInfo(CUstream_st stream, @Cast(value="cudaStreamCaptureStatus*") IntPointer captureStatus_out)
@Cast(value="cudaError_t") public static int cudaStreamGetCaptureInfo(CUstream_st stream, @Cast(value="cudaStreamCaptureStatus*") IntBuffer captureStatus_out, @Cast(value="unsigned long long*") LongBuffer id_out, @ByPtrPtr CUgraph_st graph_out, @Cast(value="const cudaGraphNode_t**") @ByPtrPtr PointerPointer dependencies_out, @Cast(value="size_t*") SizeTPointer numDependencies_out)
@Cast(value="cudaError_t") public static int cudaStreamGetCaptureInfo(CUstream_st stream, @Cast(value="cudaStreamCaptureStatus*") IntBuffer captureStatus_out)
@Cast(value="cudaError_t") public static int cudaStreamGetCaptureInfo(CUstream_st stream, @Cast(value="cudaStreamCaptureStatus*") int[] captureStatus_out, @Cast(value="unsigned long long*") long[] id_out, @ByPtrPtr CUgraph_st graph_out, @Cast(value="const cudaGraphNode_t**") @ByPtrPtr PointerPointer dependencies_out, @Cast(value="size_t*") SizeTPointer numDependencies_out)
@Cast(value="cudaError_t") public static int cudaStreamGetCaptureInfo(CUstream_st stream, @Cast(value="cudaStreamCaptureStatus*") int[] captureStatus_out)
@Cast(value="cudaError_t") public static int cudaStreamGetCaptureInfo_v3(CUstream_st stream, @Cast(value="cudaStreamCaptureStatus*") IntPointer captureStatus_out, @Cast(value="unsigned long long*") LongPointer id_out, @ByPtrPtr CUgraph_st graph_out, @Cast(value="const cudaGraphNode_t**") @ByPtrPtr PointerPointer dependencies_out, @Cast(value="const cudaGraphEdgeData**") PointerPointer edgeData_out, @Cast(value="size_t*") SizeTPointer numDependencies_out)
stream
- - The stream to querycaptureStatus_out
- - Location to return the capture status of the stream; requiredid_out
- - Optional location to return an id for the capture sequence, which is
unique over the lifetime of the processgraph_out
- - Optional location to return the graph being captured into. All
operations other than destroy and node removal are permitted on the graph
while the capture sequence is in progress. This API does not transfer
ownership of the graph, which is transferred or destroyed at
::cudaStreamEndCapture. Note that the graph handle may be invalidated before
end of capture for certain errors. Nodes that are or become
unreachable from the original stream at ::cudaStreamEndCapture due to direct
actions on the graph do not trigger ::cudaErrorStreamCaptureUnjoined.dependencies_out
- - Optional location to store a pointer to an array of nodes.
The next node to be captured in the stream will depend on this set of nodes,
absent operations such as event wait which modify this set. The array pointer
is valid until the next API call which operates on the stream or until the
capture is terminated. The node handles may be copied out and are valid until
they or the graph is destroyed. The driver-owned array may also be passed
directly to APIs that operate on the graph (not the stream) without copying.edgeData_out
- - Optional location to store a pointer to an array of graph edge
data. This array parallels \c dependencies_out; the next node to be added
has an edge to \c dependencies_out[i] with annotation \c edgeData_out[i] for
each \c i. The array pointer is valid until the next API call which operates
on the stream or until the capture is terminated.numDependencies_out
- - Optional location to store the size of the array
returned in dependencies_out.::cudaStreamBeginCapture,
::cudaStreamIsCapturing,
::cudaStreamUpdateCaptureDependencies
@Cast(value="cudaError_t") public static int cudaStreamGetCaptureInfo_v3(CUstream_st stream, @Cast(value="cudaStreamCaptureStatus*") IntPointer captureStatus_out)
@Cast(value="cudaError_t") public static int cudaStreamGetCaptureInfo_v3(CUstream_st stream, @Cast(value="cudaStreamCaptureStatus*") IntPointer captureStatus_out, @Cast(value="unsigned long long*") LongPointer id_out, @ByPtrPtr CUgraph_st graph_out, @Cast(value="const cudaGraphNode_t**") @ByPtrPtr PointerPointer dependencies_out, @Const @ByPtrPtr cudaGraphEdgeData edgeData_out, @Cast(value="size_t*") SizeTPointer numDependencies_out)
@Cast(value="cudaError_t") public static int cudaStreamGetCaptureInfo_v3(CUstream_st stream, @Cast(value="cudaStreamCaptureStatus*") IntBuffer captureStatus_out, @Cast(value="unsigned long long*") LongBuffer id_out, @ByPtrPtr CUgraph_st graph_out, @Cast(value="const cudaGraphNode_t**") @ByPtrPtr PointerPointer dependencies_out, @Const @ByPtrPtr cudaGraphEdgeData edgeData_out, @Cast(value="size_t*") SizeTPointer numDependencies_out)
@Cast(value="cudaError_t") public static int cudaStreamGetCaptureInfo_v3(CUstream_st stream, @Cast(value="cudaStreamCaptureStatus*") IntBuffer captureStatus_out)
@Cast(value="cudaError_t") public static int cudaStreamGetCaptureInfo_v3(CUstream_st stream, @Cast(value="cudaStreamCaptureStatus*") int[] captureStatus_out, @Cast(value="unsigned long long*") long[] id_out, @ByPtrPtr CUgraph_st graph_out, @Cast(value="const cudaGraphNode_t**") @ByPtrPtr PointerPointer dependencies_out, @Const @ByPtrPtr cudaGraphEdgeData edgeData_out, @Cast(value="size_t*") SizeTPointer numDependencies_out)
@Cast(value="cudaError_t") public static int cudaStreamGetCaptureInfo_v3(CUstream_st stream, @Cast(value="cudaStreamCaptureStatus*") int[] captureStatus_out)
@Cast(value="cudaError_t") public static int cudaStreamUpdateCaptureDependencies(CUstream_st stream, @ByPtrPtr CUgraphNode_st dependencies, @Cast(value="size_t") long numDependencies, @Cast(value="unsigned int") int flags)
stream
- - The stream to updatedependencies
- - The set of dependencies to addnumDependencies
- - The size of the dependencies arrayflags
- - See above::cudaStreamBeginCapture,
::cudaStreamGetCaptureInfo,
@Cast(value="cudaError_t") public static int cudaStreamUpdateCaptureDependencies(CUstream_st stream, @ByPtrPtr CUgraphNode_st dependencies, @Cast(value="size_t") long numDependencies)
@Cast(value="cudaError_t") public static int cudaStreamUpdateCaptureDependencies_v2(CUstream_st stream, @ByPtrPtr CUgraphNode_st dependencies, @Const cudaGraphEdgeData dependencyData, @Cast(value="size_t") long numDependencies, @Cast(value="unsigned int") int flags)
stream
- - The stream to updatedependencies
- - The set of dependencies to adddependencyData
- - Optional array of data associated with each dependency.numDependencies
- - The size of the dependencies arrayflags
- - See above::cudaStreamBeginCapture,
::cudaStreamGetCaptureInfo,
@Cast(value="cudaError_t") public static int cudaStreamUpdateCaptureDependencies_v2(CUstream_st stream, @ByPtrPtr CUgraphNode_st dependencies, @Const cudaGraphEdgeData dependencyData, @Cast(value="size_t") long numDependencies)
@Cast(value="cudaError_t") public static int cudaEventCreate(@ByPtrPtr CUevent_st event)
event
- - Newly created event::cudaEventCreate(cudaEvent_t*, unsigned int) "cudaEventCreate (C++ API)",
::cudaEventCreateWithFlags, ::cudaEventRecord, ::cudaEventQuery,
::cudaEventSynchronize, ::cudaEventDestroy, ::cudaEventElapsedTime,
::cudaStreamWaitEvent,
::cuEventCreate
@Cast(value="cudaError_t") public static int cudaEventCreateWithFlags(@ByPtrPtr CUevent_st event, @Cast(value="unsigned int") int flags)
event
- - Newly created eventflags
- - Flags for new event::cudaEventCreate(cudaEvent_t*) "cudaEventCreate (C API)",
::cudaEventSynchronize, ::cudaEventDestroy, ::cudaEventElapsedTime,
::cudaStreamWaitEvent,
::cuEventCreate
@Cast(value="cudaError_t") public static int cudaEventRecord(CUevent_st event, CUstream_st stream)
event
- - Event to recordstream
- - Stream in which to record event::cudaEventCreate(cudaEvent_t*) "cudaEventCreate (C API)",
::cudaEventCreateWithFlags, ::cudaEventQuery,
::cudaEventSynchronize, ::cudaEventDestroy, ::cudaEventElapsedTime,
::cudaStreamWaitEvent,
::cudaEventRecordWithFlags,
::cuEventRecord
@Cast(value="cudaError_t") public static int cudaEventRecord(CUevent_st event)
@Cast(value="cudaError_t") public static int cudaEventRecordWithFlags(CUevent_st event, CUstream_st stream, @Cast(value="unsigned int") int flags)
event
- - Event to recordstream
- - Stream in which to record eventflags
- - Parameters for the operation(See above)::cudaEventCreate(cudaEvent_t*) "cudaEventCreate (C API)",
::cudaEventCreateWithFlags, ::cudaEventQuery,
::cudaEventSynchronize, ::cudaEventDestroy, ::cudaEventElapsedTime,
::cudaStreamWaitEvent,
::cudaEventRecord,
::cuEventRecord,
@Cast(value="cudaError_t") public static int cudaEventRecordWithFlags(CUevent_st event)
@Cast(value="cudaError_t") public static int cudaEventQuery(CUevent_st event)
event
- - Event to query::cudaEventCreate(cudaEvent_t*) "cudaEventCreate (C API)",
::cudaEventCreateWithFlags, ::cudaEventRecord,
::cudaEventSynchronize, ::cudaEventDestroy, ::cudaEventElapsedTime,
::cuEventQuery
@Cast(value="cudaError_t") public static int cudaEventSynchronize(CUevent_st event)
event
- - Event to wait for::cudaEventCreate(cudaEvent_t*) "cudaEventCreate (C API)",
::cudaEventCreateWithFlags, ::cudaEventRecord,
::cudaEventQuery, ::cudaEventDestroy, ::cudaEventElapsedTime,
::cuEventSynchronize
@Cast(value="cudaError_t") public static int cudaEventDestroy(CUevent_st event)
event
- - Event to destroy::cudaEventCreate(cudaEvent_t*) "cudaEventCreate (C API)",
::cudaEventCreateWithFlags, ::cudaEventQuery,
::cudaEventSynchronize, ::cudaEventRecord, ::cudaEventElapsedTime,
::cuEventDestroy
@Cast(value="cudaError_t") public static int cudaEventElapsedTime(FloatPointer ms, CUevent_st start, CUevent_st end)
ms
- - Time between \p start and \p end in msstart
- - Starting eventend
- - Ending event::cudaEventCreate(cudaEvent_t*) "cudaEventCreate (C API)",
::cudaEventCreateWithFlags, ::cudaEventQuery,
::cudaEventSynchronize, ::cudaEventDestroy, ::cudaEventRecord,
::cuEventElapsedTime
@Cast(value="cudaError_t") public static int cudaEventElapsedTime(FloatBuffer ms, CUevent_st start, CUevent_st end)
@Cast(value="cudaError_t") public static int cudaEventElapsedTime(float[] ms, CUevent_st start, CUevent_st end)
@Cast(value="cudaError_t") public static int cudaImportExternalMemory(@ByPtrPtr CUexternalMemory_st extMem_out, @Const cudaExternalMemoryHandleDesc memHandleDesc)
typedef struct cudaExternalMemoryHandleDesc_st {
cudaExternalMemoryHandleType type;
union {
int fd;
struct {
void *handle;
const void *name;
} win32;
const void *nvSciBufObject;
} handle;
unsigned long long size;
unsigned int flags;
} cudaExternalMemoryHandleDesc;
where ::cudaExternalMemoryHandleDesc::type specifies the type
of handle being imported. ::cudaExternalMemoryHandleType is
defined as:
typedef enum cudaExternalMemoryHandleType_enum {
cudaExternalMemoryHandleTypeOpaqueFd = 1,
cudaExternalMemoryHandleTypeOpaqueWin32 = 2,
cudaExternalMemoryHandleTypeOpaqueWin32Kmt = 3,
cudaExternalMemoryHandleTypeD3D12Heap = 4,
cudaExternalMemoryHandleTypeD3D12Resource = 5,
cudaExternalMemoryHandleTypeD3D11Resource = 6,
cudaExternalMemoryHandleTypeD3D11ResourceKmt = 7,
cudaExternalMemoryHandleTypeNvSciBuf = 8
} cudaExternalMemoryHandleType;
If ::cudaExternalMemoryHandleDesc::type is
::cudaExternalMemoryHandleTypeOpaqueFd, then
::cudaExternalMemoryHandleDesc::handle::fd must be a valid
file descriptor referencing a memory object. Ownership of
the file descriptor is transferred to the CUDA driver when the
handle is imported successfully. Performing any operations on the
file descriptor after it is imported results in undefined behavior.
If ::cudaExternalMemoryHandleDesc::type is
::cudaExternalMemoryHandleTypeOpaqueWin32, then exactly one
of ::cudaExternalMemoryHandleDesc::handle::win32::handle and
::cudaExternalMemoryHandleDesc::handle::win32::name must not be
NULL. If ::cudaExternalMemoryHandleDesc::handle::win32::handle
is not NULL, then it must represent a valid shared NT handle that
references a memory object. Ownership of this handle is
not transferred to CUDA after the import operation, so the
application must release the handle using the appropriate system
call. If ::cudaExternalMemoryHandleDesc::handle::win32::name
is not NULL, then it must point to a NULL-terminated array of
UTF-16 characters that refers to a memory object.
If ::cudaExternalMemoryHandleDesc::type is
::cudaExternalMemoryHandleTypeOpaqueWin32Kmt, then
::cudaExternalMemoryHandleDesc::handle::win32::handle must
be non-NULL and
::cudaExternalMemoryHandleDesc::handle::win32::name
must be NULL. The handle specified must be a globally shared KMT
handle. This handle does not hold a reference to the underlying
object, and thus will be invalid when all references to the
memory object are destroyed.
If ::cudaExternalMemoryHandleDesc::type is
::cudaExternalMemoryHandleTypeD3D12Heap, then exactly one
of ::cudaExternalMemoryHandleDesc::handle::win32::handle and
::cudaExternalMemoryHandleDesc::handle::win32::name must not be
NULL. If ::cudaExternalMemoryHandleDesc::handle::win32::handle
is not NULL, then it must represent a valid shared NT handle that
is returned by ID3D12Device::CreateSharedHandle when referring to a
ID3D12Heap object. This handle holds a reference to the underlying
object. If ::cudaExternalMemoryHandleDesc::handle::win32::name
is not NULL, then it must point to a NULL-terminated array of
UTF-16 characters that refers to a ID3D12Heap object.
If ::cudaExternalMemoryHandleDesc::type is
::cudaExternalMemoryHandleTypeD3D12Resource, then exactly one
of ::cudaExternalMemoryHandleDesc::handle::win32::handle and
::cudaExternalMemoryHandleDesc::handle::win32::name must not be
NULL. If ::cudaExternalMemoryHandleDesc::handle::win32::handle
is not NULL, then it must represent a valid shared NT handle that
is returned by ID3D12Device::CreateSharedHandle when referring to a
ID3D12Resource object. This handle holds a reference to the
underlying object. If
::cudaExternalMemoryHandleDesc::handle::win32::name
is not NULL, then it must point to a NULL-terminated array of
UTF-16 characters that refers to a ID3D12Resource object.
If ::cudaExternalMemoryHandleDesc::type is
::cudaExternalMemoryHandleTypeD3D11Resource,then exactly one
of ::cudaExternalMemoryHandleDesc::handle::win32::handle and
::cudaExternalMemoryHandleDesc::handle::win32::name must not be
NULL. If ::cudaExternalMemoryHandleDesc::handle::win32::handle is
not NULL, then it must represent a valid shared NT handle that is
returned by IDXGIResource1::CreateSharedHandle when referring to a
ID3D11Resource object. If
::cudaExternalMemoryHandleDesc::handle::win32::name
is not NULL, then it must point to a NULL-terminated array of
UTF-16 characters that refers to a ID3D11Resource object.
If ::cudaExternalMemoryHandleDesc::type is
::cudaExternalMemoryHandleTypeD3D11ResourceKmt, then
::cudaExternalMemoryHandleDesc::handle::win32::handle must
be non-NULL and ::cudaExternalMemoryHandleDesc::handle::win32::name
must be NULL. The handle specified must be a valid shared KMT
handle that is returned by IDXGIResource::GetSharedHandle when
referring to a ID3D11Resource object.
If ::cudaExternalMemoryHandleDesc::type is
::cudaExternalMemoryHandleTypeNvSciBuf, then
::cudaExternalMemoryHandleDesc::handle::nvSciBufObject must be NON-NULL
and reference a valid NvSciBuf object.
If the NvSciBuf object imported into CUDA is also mapped by other drivers, then the
application must use ::cudaWaitExternalSemaphoresAsync or ::cudaSignalExternalSemaphoresAsync
as approprriate barriers to maintain coherence between CUDA and the other drivers.
See ::cudaExternalSemaphoreWaitSkipNvSciBufMemSync and ::cudaExternalSemaphoreSignalSkipNvSciBufMemSync
for memory synchronization.
The size of the memory object must be specified in
::cudaExternalMemoryHandleDesc::size.
Specifying the flag ::cudaExternalMemoryDedicated in
::cudaExternalMemoryHandleDesc::flags indicates that the
resource is a dedicated resource. The definition of what a
dedicated resource is outside the scope of this extension.
This flag must be set if ::cudaExternalMemoryHandleDesc::type
is one of the following:
::cudaExternalMemoryHandleTypeD3D12Resource
::cudaExternalMemoryHandleTypeD3D11Resource
::cudaExternalMemoryHandleTypeD3D11ResourceKmtextMem_out
- - Returned handle to an external memory objectmemHandleDesc
- - Memory import handle descriptor::cudaDestroyExternalMemory,
::cudaExternalMemoryGetMappedBuffer,
::cudaExternalMemoryGetMappedMipmappedArray
@Cast(value="cudaError_t") public static int cudaExternalMemoryGetMappedBuffer(@Cast(value="void**") PointerPointer devPtr, CUexternalMemory_st extMem, @Const cudaExternalMemoryBufferDesc bufferDesc)
typedef struct cudaExternalMemoryBufferDesc_st {
unsigned long long offset;
unsigned long long size;
unsigned int flags;
} cudaExternalMemoryBufferDesc;
where ::cudaExternalMemoryBufferDesc::offset is the offset in
the memory object where the buffer's base address is.
::cudaExternalMemoryBufferDesc::size is the size of the buffer.
::cudaExternalMemoryBufferDesc::flags must be zero.
The offset and size have to be suitably aligned to match the
requirements of the external API. Mapping two buffers whose ranges
overlap may or may not result in the same virtual address being
returned for the overlapped portion. In such cases, the application
must ensure that all accesses to that region from the GPU are
volatile. Otherwise writes made via one address are not guaranteed
to be visible via the other address, even if they're issued by the
same thread. It is recommended that applications map the combined
range instead of mapping separate buffers and then apply the
appropriate offsets to the returned pointer to derive the
individual buffers.
The returned pointer \p devPtr must be freed using ::cudaFree.devPtr
- - Returned device pointer to bufferextMem
- - Handle to external memory objectbufferDesc
- - Buffer descriptor::cudaImportExternalMemory,
::cudaDestroyExternalMemory,
::cudaExternalMemoryGetMappedMipmappedArray
@Cast(value="cudaError_t") public static int cudaExternalMemoryGetMappedBuffer(@Cast(value="void**") @ByPtrPtr Pointer devPtr, CUexternalMemory_st extMem, @Const cudaExternalMemoryBufferDesc bufferDesc)
@Cast(value="cudaError_t") public static int cudaExternalMemoryGetMappedMipmappedArray(@ByPtrPtr cudaMipmappedArray mipmap, CUexternalMemory_st extMem, @Const cudaExternalMemoryMipmappedArrayDesc mipmapDesc)
typedef struct cudaExternalMemoryMipmappedArrayDesc_st {
unsigned long long offset;
cudaChannelFormatDesc formatDesc;
cudaExtent extent;
unsigned int flags;
unsigned int numLevels;
} cudaExternalMemoryMipmappedArrayDesc;
where ::cudaExternalMemoryMipmappedArrayDesc::offset is the
offset in the memory object where the base level of the mipmap
chain is.
::cudaExternalMemoryMipmappedArrayDesc::formatDesc describes the
format of the data.
::cudaExternalMemoryMipmappedArrayDesc::extent specifies the
dimensions of the base level of the mipmap chain.
::cudaExternalMemoryMipmappedArrayDesc::flags are flags associated
with CUDA mipmapped arrays. For further details, please refer to
the documentation for ::cudaMalloc3DArray. Note that if the mipmapped
array is bound as a color target in the graphics API, then the flag
::cudaArrayColorAttachment must be specified in
::cudaExternalMemoryMipmappedArrayDesc::flags.
::cudaExternalMemoryMipmappedArrayDesc::numLevels specifies
the total number of levels in the mipmap chain.
The returned CUDA mipmapped array must be freed using ::cudaFreeMipmappedArray.mipmap
- - Returned CUDA mipmapped arrayextMem
- - Handle to external memory objectmipmapDesc
- - CUDA array descriptor::cudaImportExternalMemory,
::cudaDestroyExternalMemory,
::cudaExternalMemoryGetMappedBuffer
\note If ::cudaExternalMemoryHandleDesc::type is
::cudaExternalMemoryHandleTypeNvSciBuf, then
::cudaExternalMemoryMipmappedArrayDesc::numLevels must not be greater than 1.
@Cast(value="cudaError_t") public static int cudaDestroyExternalMemory(CUexternalMemory_st extMem)
extMem
- - External memory object to be destroyed::cudaImportExternalMemory,
::cudaExternalMemoryGetMappedBuffer,
::cudaExternalMemoryGetMappedMipmappedArray
@Cast(value="cudaError_t") public static int cudaImportExternalSemaphore(@ByPtrPtr CUexternalSemaphore_st extSem_out, @Const cudaExternalSemaphoreHandleDesc semHandleDesc)
typedef struct cudaExternalSemaphoreHandleDesc_st {
cudaExternalSemaphoreHandleType type;
union {
int fd;
struct {
void *handle;
const void *name;
} win32;
const void* NvSciSyncObj;
} handle;
unsigned int flags;
} cudaExternalSemaphoreHandleDesc;
where ::cudaExternalSemaphoreHandleDesc::type specifies the type of
handle being imported. ::cudaExternalSemaphoreHandleType is defined
as:
typedef enum cudaExternalSemaphoreHandleType_enum {
cudaExternalSemaphoreHandleTypeOpaqueFd = 1,
cudaExternalSemaphoreHandleTypeOpaqueWin32 = 2,
cudaExternalSemaphoreHandleTypeOpaqueWin32Kmt = 3,
cudaExternalSemaphoreHandleTypeD3D12Fence = 4,
cudaExternalSemaphoreHandleTypeD3D11Fence = 5,
cudaExternalSemaphoreHandleTypeNvSciSync = 6,
cudaExternalSemaphoreHandleTypeKeyedMutex = 7,
cudaExternalSemaphoreHandleTypeKeyedMutexKmt = 8,
cudaExternalSemaphoreHandleTypeTimelineSemaphoreFd = 9,
cudaExternalSemaphoreHandleTypeTimelineSemaphoreWin32 = 10
} cudaExternalSemaphoreHandleType;
If ::cudaExternalSemaphoreHandleDesc::type is
::cudaExternalSemaphoreHandleTypeOpaqueFd, then
::cudaExternalSemaphoreHandleDesc::handle::fd must be a valid file
descriptor referencing a synchronization object. Ownership of the
file descriptor is transferred to the CUDA driver when the handle
is imported successfully. Performing any operations on the file
descriptor after it is imported results in undefined behavior.
If ::cudaExternalSemaphoreHandleDesc::type is
::cudaExternalSemaphoreHandleTypeOpaqueWin32, then exactly one of
::cudaExternalSemaphoreHandleDesc::handle::win32::handle and
::cudaExternalSemaphoreHandleDesc::handle::win32::name must not be
NULL. If ::cudaExternalSemaphoreHandleDesc::handle::win32::handle
is not NULL, then it must represent a valid shared NT handle that
references a synchronization object. Ownership of this handle is
not transferred to CUDA after the import operation, so the
application must release the handle using the appropriate system
call. If ::cudaExternalSemaphoreHandleDesc::handle::win32::name is
not NULL, then it must name a valid synchronization object.
If ::cudaExternalSemaphoreHandleDesc::type is
::cudaExternalSemaphoreHandleTypeOpaqueWin32Kmt, then
::cudaExternalSemaphoreHandleDesc::handle::win32::handle must be
non-NULL and ::cudaExternalSemaphoreHandleDesc::handle::win32::name
must be NULL. The handle specified must be a globally shared KMT
handle. This handle does not hold a reference to the underlying
object, and thus will be invalid when all references to the
synchronization object are destroyed.
If ::cudaExternalSemaphoreHandleDesc::type is
::cudaExternalSemaphoreHandleTypeD3D12Fence, then exactly one of
::cudaExternalSemaphoreHandleDesc::handle::win32::handle and
::cudaExternalSemaphoreHandleDesc::handle::win32::name must not be
NULL. If ::cudaExternalSemaphoreHandleDesc::handle::win32::handle
is not NULL, then it must represent a valid shared NT handle that
is returned by ID3D12Device::CreateSharedHandle when referring to a
ID3D12Fence object. This handle holds a reference to the underlying
object. If ::cudaExternalSemaphoreHandleDesc::handle::win32::name
is not NULL, then it must name a valid synchronization object that
refers to a valid ID3D12Fence object.
If ::cudaExternalSemaphoreHandleDesc::type is
::cudaExternalSemaphoreHandleTypeD3D11Fence, then exactly one of
::cudaExternalSemaphoreHandleDesc::handle::win32::handle and
::cudaExternalSemaphoreHandleDesc::handle::win32::name must not be
NULL. If ::cudaExternalSemaphoreHandleDesc::handle::win32::handle
is not NULL, then it must represent a valid shared NT handle that
is returned by ID3D11Fence::CreateSharedHandle. If
::cudaExternalSemaphoreHandleDesc::handle::win32::name
is not NULL, then it must name a valid synchronization object that
refers to a valid ID3D11Fence object.
If ::cudaExternalSemaphoreHandleDesc::type is
::cudaExternalSemaphoreHandleTypeNvSciSync, then
::cudaExternalSemaphoreHandleDesc::handle::nvSciSyncObj
represents a valid NvSciSyncObj.
::cudaExternalSemaphoreHandleTypeKeyedMutex, then exactly one of
::cudaExternalSemaphoreHandleDesc::handle::win32::handle and
::cudaExternalSemaphoreHandleDesc::handle::win32::name must not be
NULL. If ::cudaExternalSemaphoreHandleDesc::handle::win32::handle
is not NULL, then it represent a valid shared NT handle that
is returned by IDXGIResource1::CreateSharedHandle when referring to
a IDXGIKeyedMutex object.
If ::cudaExternalSemaphoreHandleDesc::type is
::cudaExternalSemaphoreHandleTypeKeyedMutexKmt, then
::cudaExternalSemaphoreHandleDesc::handle::win32::handle must be
non-NULL and ::cudaExternalSemaphoreHandleDesc::handle::win32::name
must be NULL. The handle specified must represent a valid KMT
handle that is returned by IDXGIResource::GetSharedHandle when
referring to a IDXGIKeyedMutex object.
If ::cudaExternalSemaphoreHandleDesc::type is
::cudaExternalSemaphoreHandleTypeTimelineSemaphoreFd, then
::cudaExternalSemaphoreHandleDesc::handle::fd must be a valid file
descriptor referencing a synchronization object. Ownership of the
file descriptor is transferred to the CUDA driver when the handle
is imported successfully. Performing any operations on the file
descriptor after it is imported results in undefined behavior.
If ::cudaExternalSemaphoreHandleDesc::type is
::cudaExternalSemaphoreHandleTypeTimelineSemaphoreWin32, then exactly one of
::cudaExternalSemaphoreHandleDesc::handle::win32::handle and
::cudaExternalSemaphoreHandleDesc::handle::win32::name must not be
NULL. If ::cudaExternalSemaphoreHandleDesc::handle::win32::handle
is not NULL, then it must represent a valid shared NT handle that
references a synchronization object. Ownership of this handle is
not transferred to CUDA after the import operation, so the
application must release the handle using the appropriate system
call. If ::cudaExternalSemaphoreHandleDesc::handle::win32::name is
not NULL, then it must name a valid synchronization object.extSem_out
- - Returned handle to an external semaphoresemHandleDesc
- - Semaphore import handle descriptor::cudaDestroyExternalSemaphore,
::cudaSignalExternalSemaphoresAsync,
::cudaWaitExternalSemaphoresAsync
@Cast(value="cudaError_t") public static int cudaSignalExternalSemaphoresAsync(@Cast(value="const cudaExternalSemaphore_t*") @ByPtrPtr CUexternalSemaphore_st extSemArray, @Const cudaExternalSemaphoreSignalParams paramsArray, @Cast(value="unsigned int") int numExtSems, CUstream_st stream)
extSemArray
- - Set of external semaphores to be signaledparamsArray
- - Array of semaphore parametersnumExtSems
- - Number of semaphores to signalstream
- - Stream to enqueue the signal operations in::cudaImportExternalSemaphore,
::cudaDestroyExternalSemaphore,
::cudaWaitExternalSemaphoresAsync
@Cast(value="cudaError_t") public static int cudaSignalExternalSemaphoresAsync(@Cast(value="const cudaExternalSemaphore_t*") @ByPtrPtr CUexternalSemaphore_st extSemArray, @Const cudaExternalSemaphoreSignalParams paramsArray, @Cast(value="unsigned int") int numExtSems)
@Cast(value="cudaError_t") public static int cudaWaitExternalSemaphoresAsync(@Cast(value="const cudaExternalSemaphore_t*") @ByPtrPtr CUexternalSemaphore_st extSemArray, @Const cudaExternalSemaphoreWaitParams paramsArray, @Cast(value="unsigned int") int numExtSems, CUstream_st stream)
extSemArray
- - External semaphores to be waited onparamsArray
- - Array of semaphore parametersnumExtSems
- - Number of semaphores to wait onstream
- - Stream to enqueue the wait operations in::cudaImportExternalSemaphore,
::cudaDestroyExternalSemaphore,
::cudaSignalExternalSemaphoresAsync
@Cast(value="cudaError_t") public static int cudaWaitExternalSemaphoresAsync(@Cast(value="const cudaExternalSemaphore_t*") @ByPtrPtr CUexternalSemaphore_st extSemArray, @Const cudaExternalSemaphoreWaitParams paramsArray, @Cast(value="unsigned int") int numExtSems)
@Cast(value="cudaError_t") public static int cudaDestroyExternalSemaphore(CUexternalSemaphore_st extSem)
extSem
- - External semaphore to be destroyed::cudaImportExternalSemaphore,
::cudaSignalExternalSemaphoresAsync,
::cudaWaitExternalSemaphoresAsync
@Cast(value="cudaError_t") public static int cudaLaunchKernel(@Const Pointer func, @ByVal dim3 gridDim, @ByVal dim3 blockDim, @Cast(value="void**") PointerPointer args, @Cast(value="size_t") long sharedMem, CUstream_st stream)
func
- - Device function symbolgridDim
- - Grid dimentionsblockDim
- - Block dimentionsargs
- - ArgumentssharedMem
- - Shared memorystream
- - Stream identifier::cudaLaunchKernel(const T *func, dim3 gridDim, dim3 blockDim, void **args, size_t sharedMem, cudaStream_t stream) "cudaLaunchKernel (C++ API)",
::cuLaunchKernel
@Cast(value="cudaError_t") public static int cudaLaunchKernel(@Const Pointer func, @ByVal dim3 gridDim, @ByVal dim3 blockDim, @Cast(value="void**") @ByPtrPtr Pointer args, @Cast(value="size_t") long sharedMem, CUstream_st stream)
@Cast(value="cudaError_t") public static int cudaLaunchKernelExC(@Const cudaLaunchConfig_t config, @Const Pointer func, @Cast(value="void**") PointerPointer args)
config
- - Launch configurationfunc
- - Kernel to launchargs
- - Array of pointers to kernel parameters::cudaLaunchKernelEx(const cudaLaunchConfig_t *config, void (*kernel)(ExpTypes...), ActTypes &&... args) "cudaLaunchKernelEx (C++ API)",
::cuLaunchKernelEx
@Cast(value="cudaError_t") public static int cudaLaunchKernelExC(@Const cudaLaunchConfig_t config, @Const Pointer func, @Cast(value="void**") @ByPtrPtr Pointer args)
@Cast(value="cudaError_t") public static int cudaLaunchCooperativeKernel(@Const Pointer func, @ByVal dim3 gridDim, @ByVal dim3 blockDim, @Cast(value="void**") PointerPointer args, @Cast(value="size_t") long sharedMem, CUstream_st stream)
func
- - Device function symbolgridDim
- - Grid dimentionsblockDim
- - Block dimentionsargs
- - ArgumentssharedMem
- - Shared memorystream
- - Stream identifier::cudaLaunchCooperativeKernel(const T *func, dim3 gridDim, dim3 blockDim, void **args, size_t sharedMem, cudaStream_t stream) "cudaLaunchCooperativeKernel (C++ API)",
::cudaLaunchCooperativeKernelMultiDevice,
::cuLaunchCooperativeKernel
@Cast(value="cudaError_t") public static int cudaLaunchCooperativeKernel(@Const Pointer func, @ByVal dim3 gridDim, @ByVal dim3 blockDim, @Cast(value="void**") @ByPtrPtr Pointer args, @Cast(value="size_t") long sharedMem, CUstream_st stream)
@Cast(value="cudaError_t") @Deprecated public static int cudaLaunchCooperativeKernelMultiDevice(cudaLaunchParams launchParamsList, @Cast(value="unsigned int") int numDevices, @Cast(value="unsigned int") int flags)
struct cudaLaunchParams
{
void *func;
dim3 gridDim;
dim3 blockDim;
void **args;
size_t sharedMem;
cudaStream_t stream;
};
where:
- ::cudaLaunchParams::func specifies the kernel to be launched. This same functions must
be launched on all devices. For templated functions, pass the function symbol as follows:
func_namelaunchParamsList
- - List of launch parameters, one per devicenumDevices
- - Size of the \p launchParamsList arrayflags
- - Flags to control launch behavior::cudaLaunchCooperativeKernel(const T *func, dim3 gridDim, dim3 blockDim, void **args, size_t sharedMem, cudaStream_t stream) "cudaLaunchCooperativeKernel (C++ API)",
::cudaLaunchCooperativeKernel,
::cuLaunchCooperativeKernelMultiDevice
@Cast(value="cudaError_t") @Deprecated public static int cudaLaunchCooperativeKernelMultiDevice(cudaLaunchParams launchParamsList, @Cast(value="unsigned int") int numDevices)
@Cast(value="cudaError_t") public static int cudaFuncSetCacheConfig(@Const Pointer func, @Cast(value="cudaFuncCache") int cacheConfig)
func
- - Device function symbolcacheConfig
- - Requested cache configuration::cudaFuncSetCacheConfig(T*, enum cudaFuncCache) "cudaFuncSetCacheConfig (C++ API)",
\ref ::cudaFuncGetAttributes(struct cudaFuncAttributes*, const void*) "cudaFuncGetAttributes (C API)",
\ref ::cudaLaunchKernel(const void *func, dim3 gridDim, dim3 blockDim, void **args, size_t sharedMem, cudaStream_t stream) "cudaLaunchKernel (C API)",
::cuFuncSetCacheConfig
@Cast(value="cudaError_t") public static int cudaFuncSetSharedMemConfig(@Const Pointer func, @Cast(value="cudaSharedMemConfig") int config)
func
- - Device function symbolconfig
- - Requested shared memory configuration::cudaDeviceSetSharedMemConfig,
::cudaDeviceGetSharedMemConfig,
::cudaDeviceSetCacheConfig,
::cudaDeviceGetCacheConfig,
::cudaFuncSetCacheConfig,
::cuFuncSetSharedMemConfig
@Cast(value="cudaError_t") public static int cudaFuncGetAttributes(cudaFuncAttributes attr, @Const Pointer func)
attr
- - Return pointer to function's attributesfunc
- - Device function symbol::cudaFuncSetCacheConfig(const void*, enum cudaFuncCache) "cudaFuncSetCacheConfig (C API)",
\ref ::cudaFuncGetAttributes(struct cudaFuncAttributes*, T*) "cudaFuncGetAttributes (C++ API)",
\ref ::cudaLaunchKernel(const void *func, dim3 gridDim, dim3 blockDim, void **args, size_t sharedMem, cudaStream_t stream) "cudaLaunchKernel (C API)",
::cuFuncGetAttribute
@Cast(value="cudaError_t") public static int cudaFuncSetAttribute(@Const Pointer func, @Cast(value="cudaFuncAttribute") int attr, int value)
func
- - Function to get attributes ofattr
- - Attribute to setvalue
- - Value to set@Cast(value="cudaError_t") public static int cudaFuncGetName(@Cast(value="const char**") PointerPointer name, @Const Pointer func)
name
- - The returned name of the functionfunc
- - The function pointer to retrieve name for@Cast(value="cudaError_t") public static int cudaFuncGetName(@Cast(value="const char**") @ByPtrPtr BytePointer name, @Const Pointer func)
@Cast(value="cudaError_t") public static int cudaFuncGetName(@Cast(value="const char**") @ByPtrPtr ByteBuffer name, @Const Pointer func)
@Cast(value="cudaError_t") public static int cudaFuncGetName(@Cast(value="const char**") @ByPtrPtr byte[] name, @Const Pointer func)
@Cast(value="cudaError_t") @Deprecated public static int cudaSetDoubleForDevice(DoublePointer d)
d
- - Double to convert::cudaFuncSetCacheConfig(const void*, enum cudaFuncCache) "cudaFuncSetCacheConfig (C API)",
\ref ::cudaFuncGetAttributes(struct cudaFuncAttributes*, const void*) "cudaFuncGetAttributes (C API)",
::cudaSetDoubleForHost
@Cast(value="cudaError_t") @Deprecated public static int cudaSetDoubleForDevice(DoubleBuffer d)
@Cast(value="cudaError_t") @Deprecated public static int cudaSetDoubleForDevice(double[] d)
@Cast(value="cudaError_t") @Deprecated public static int cudaSetDoubleForHost(DoublePointer d)
d
- - Double to convert::cudaFuncSetCacheConfig(const void*, enum cudaFuncCache) "cudaFuncSetCacheConfig (C API)",
\ref ::cudaFuncGetAttributes(struct cudaFuncAttributes*, const void*) "cudaFuncGetAttributes (C API)",
::cudaSetDoubleForDevice
@Cast(value="cudaError_t") @Deprecated public static int cudaSetDoubleForHost(DoubleBuffer d)
@Cast(value="cudaError_t") @Deprecated public static int cudaSetDoubleForHost(double[] d)
@Cast(value="cudaError_t") public static int cudaLaunchHostFunc(CUstream_st stream, cudaHostFn_t fn, Pointer userData)
hStream
- - Stream to enqueue function call infn
- - The function to call once preceding stream operations are completeuserData
- - User-specified data to be passed to the function::cudaStreamCreate,
::cudaStreamQuery,
::cudaStreamSynchronize,
::cudaStreamWaitEvent,
::cudaStreamDestroy,
::cudaMallocManaged,
::cudaStreamAttachMemAsync,
::cudaStreamAddCallback,
::cuLaunchHostFunc
@Cast(value="cudaError_t") public static int cudaOccupancyMaxActiveBlocksPerMultiprocessor(IntPointer numBlocks, @Const Pointer func, int blockSize, @Cast(value="size_t") long dynamicSMemSize)
numBlocks
- - Returned occupancyfunc
- - Kernel function for which occupancy is calculatedblockSize
- - Block size the kernel is intended to be launched withdynamicSMemSize
- - Per-block dynamic shared memory usage intended, in bytes::cudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlags,
\ref ::cudaOccupancyMaxPotentialBlockSize(int*, int*, T, size_t, int) "cudaOccupancyMaxPotentialBlockSize (C++ API)",
\ref ::cudaOccupancyMaxPotentialBlockSizeWithFlags(int*, int*, T, size_t, int, unsigned int) "cudaOccupancyMaxPotentialBlockSizeWithFlags (C++ API)",
\ref ::cudaOccupancyMaxPotentialBlockSizeVariableSMem(int*, int*, T, UnaryFunction, int) "cudaOccupancyMaxPotentialBlockSizeVariableSMem (C++ API)",
\ref ::cudaOccupancyMaxPotentialBlockSizeVariableSMemWithFlags(int*, int*, T, UnaryFunction, int, unsigned int) "cudaOccupancyMaxPotentialBlockSizeVariableSMemWithFlags (C++ API)",
\ref ::cudaOccupancyAvailableDynamicSMemPerBlock(size_t*, T, int, int) "cudaOccupancyAvailableDynamicSMemPerBlock (C++ API)",
::cuOccupancyMaxActiveBlocksPerMultiprocessor
@Cast(value="cudaError_t") public static int cudaOccupancyMaxActiveBlocksPerMultiprocessor(IntBuffer numBlocks, @Const Pointer func, int blockSize, @Cast(value="size_t") long dynamicSMemSize)
@Cast(value="cudaError_t") public static int cudaOccupancyMaxActiveBlocksPerMultiprocessor(int[] numBlocks, @Const Pointer func, int blockSize, @Cast(value="size_t") long dynamicSMemSize)
@Cast(value="cudaError_t") public static int cudaOccupancyAvailableDynamicSMemPerBlock(@Cast(value="size_t*") SizeTPointer dynamicSmemSize, @Const Pointer func, int numBlocks, int blockSize)
dynamicSmemSize
- - Returned maximum dynamic shared memoryfunc
- - Kernel function for which occupancy is calculatednumBlocks
- - Number of blocks to fit on SMblockSize
- - Size of the block::cudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlags,
\ref ::cudaOccupancyMaxPotentialBlockSize(int*, int*, T, size_t, int) "cudaOccupancyMaxPotentialBlockSize (C++ API)",
\ref ::cudaOccupancyMaxPotentialBlockSizeWithFlags(int*, int*, T, size_t, int, unsigned int) "cudaOccupancyMaxPotentialBlockSizeWithFlags (C++ API)",
\ref ::cudaOccupancyMaxPotentialBlockSizeVariableSMem(int*, int*, T, UnaryFunction, int) "cudaOccupancyMaxPotentialBlockSizeVariableSMem (C++ API)",
\ref ::cudaOccupancyMaxPotentialBlockSizeVariableSMemWithFlags(int*, int*, T, UnaryFunction, int, unsigned int) "cudaOccupancyMaxPotentialBlockSizeVariableSMemWithFlags (C++ API)",
::cudaOccupancyAvailableDynamicSMemPerBlock
@Cast(value="cudaError_t") public static int cudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlags(IntPointer numBlocks, @Const Pointer func, int blockSize, @Cast(value="size_t") long dynamicSMemSize, @Cast(value="unsigned int") int flags)
numBlocks
- - Returned occupancyfunc
- - Kernel function for which occupancy is calculatedblockSize
- - Block size the kernel is intended to be launched withdynamicSMemSize
- - Per-block dynamic shared memory usage intended, in bytesflags
- - Requested behavior for the occupancy calculator::cudaOccupancyMaxActiveBlocksPerMultiprocessor,
\ref ::cudaOccupancyMaxPotentialBlockSize(int*, int*, T, size_t, int) "cudaOccupancyMaxPotentialBlockSize (C++ API)",
\ref ::cudaOccupancyMaxPotentialBlockSizeWithFlags(int*, int*, T, size_t, int, unsigned int) "cudaOccupancyMaxPotentialBlockSizeWithFlags (C++ API)",
\ref ::cudaOccupancyMaxPotentialBlockSizeVariableSMem(int*, int*, T, UnaryFunction, int) "cudaOccupancyMaxPotentialBlockSizeVariableSMem (C++ API)",
\ref ::cudaOccupancyMaxPotentialBlockSizeVariableSMemWithFlags(int*, int*, T, UnaryFunction, int, unsigned int) "cudaOccupancyMaxPotentialBlockSizeVariableSMemWithFlags (C++ API)",
\ref ::cudaOccupancyAvailableDynamicSMemPerBlock(size_t*, T, int, int) "cudaOccupancyAvailableDynamicSMemPerBlock (C++ API)",
::cuOccupancyMaxActiveBlocksPerMultiprocessorWithFlags
@Cast(value="cudaError_t") public static int cudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlags(IntBuffer numBlocks, @Const Pointer func, int blockSize, @Cast(value="size_t") long dynamicSMemSize, @Cast(value="unsigned int") int flags)
@Cast(value="cudaError_t") public static int cudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlags(int[] numBlocks, @Const Pointer func, int blockSize, @Cast(value="size_t") long dynamicSMemSize, @Cast(value="unsigned int") int flags)
@Cast(value="cudaError_t") public static int cudaOccupancyMaxPotentialClusterSize(IntPointer clusterSize, @Const Pointer func, @Const cudaLaunchConfig_t launchConfig)
clusterSize
- - Returned maximum cluster size that can be launched
for the given kernel function and launch configurationfunc
- - Kernel function for which maximum cluster
size is calculatedconfig
- - Launch configuration for the given kernel function\ref ::cudaOccupancyMaxPotentialClusterSize(int*, T, const cudaLaunchConfig_t*) "cudaOccupancyMaxPotentialClusterSize (C++ API)",
::cuOccupancyMaxPotentialClusterSize
@Cast(value="cudaError_t") public static int cudaOccupancyMaxPotentialClusterSize(IntBuffer clusterSize, @Const Pointer func, @Const cudaLaunchConfig_t launchConfig)
@Cast(value="cudaError_t") public static int cudaOccupancyMaxPotentialClusterSize(int[] clusterSize, @Const Pointer func, @Const cudaLaunchConfig_t launchConfig)
@Cast(value="cudaError_t") public static int cudaOccupancyMaxActiveClusters(IntPointer numClusters, @Const Pointer func, @Const cudaLaunchConfig_t launchConfig)
numClusters
- - Returned maximum number of clusters that
could co-exist on the target devicefunc
- - Kernel function for which maximum number
of clusters are calculatedconfig
- - Launch configuration for the given kernel function\ref ::cudaOccupancyMaxActiveClusters(int*, T, const cudaLaunchConfig_t*) "cudaOccupancyMaxActiveClusters (C++ API)",
::cuOccupancyMaxActiveClusters
@Cast(value="cudaError_t") public static int cudaOccupancyMaxActiveClusters(IntBuffer numClusters, @Const Pointer func, @Const cudaLaunchConfig_t launchConfig)
@Cast(value="cudaError_t") public static int cudaOccupancyMaxActiveClusters(int[] numClusters, @Const Pointer func, @Const cudaLaunchConfig_t launchConfig)
@Cast(value="cudaError_t") public static int cudaMallocManaged(@Cast(value="void**") PointerPointer devPtr, @Cast(value="size_t") long size, @Cast(value="unsigned int") int flags)
devPtr
- - Pointer to allocated device memorysize
- - Requested allocation size in bytesflags
- - Must be either ::cudaMemAttachGlobal or ::cudaMemAttachHost (defaults to ::cudaMemAttachGlobal)::cudaMallocPitch, ::cudaFree, ::cudaMallocArray, ::cudaFreeArray,
::cudaMalloc3D, ::cudaMalloc3DArray,
\ref ::cudaMallocHost(void**, size_t) "cudaMallocHost (C API)",
::cudaFreeHost, ::cudaHostAlloc, ::cudaDeviceGetAttribute, ::cudaStreamAttachMemAsync,
::cuMemAllocManaged
@Cast(value="cudaError_t") public static int cudaMallocManaged(@Cast(value="void**") @ByPtrPtr Pointer devPtr, @Cast(value="size_t") long size)
@Cast(value="cudaError_t") public static int cudaMallocManaged(@Cast(value="void**") @ByPtrPtr Pointer devPtr, @Cast(value="size_t") long size, @Cast(value="unsigned int") int flags)
@Cast(value="cudaError_t") public static int cudaMalloc(@Cast(value="void**") PointerPointer devPtr, @Cast(value="size_t") long size)
devPtr
- - Pointer to allocated device memorysize
- - Requested allocation size in bytes::cudaMallocPitch, ::cudaFree, ::cudaMallocArray, ::cudaFreeArray,
::cudaMalloc3D, ::cudaMalloc3DArray,
\ref ::cudaMallocHost(void**, size_t) "cudaMallocHost (C API)",
::cudaFreeHost, ::cudaHostAlloc,
::cuMemAlloc
@Cast(value="cudaError_t") public static int cudaMalloc(@Cast(value="void**") @ByPtrPtr Pointer devPtr, @Cast(value="size_t") long size)
@Cast(value="cudaError_t") public static int cudaMallocHost(@Cast(value="void**") PointerPointer ptr, @Cast(value="size_t") long size)
On systems where ::pageableMemoryAccessUsesHostPageTables is true, ::cudaMallocHost may not page-lock the allocated memory.
Page-locking excessive amounts of memory with ::cudaMallocHost() may degrade system performance, since it reduces the amount of memory available to the system for paging. As a result, this function is best used sparingly to allocate staging areas for data exchange between host and device.
ptr
- - Pointer to allocated host memorysize
- - Requested allocation size in bytes::cudaMalloc, ::cudaMallocPitch, ::cudaMallocArray, ::cudaMalloc3D,
::cudaMalloc3DArray, ::cudaHostAlloc, ::cudaFree, ::cudaFreeArray,
\ref ::cudaMallocHost(void**, size_t, unsigned int) "cudaMallocHost (C++ API)",
::cudaFreeHost, ::cudaHostAlloc,
::cuMemAllocHost
@Cast(value="cudaError_t") public static int cudaMallocHost(@Cast(value="void**") @ByPtrPtr Pointer ptr, @Cast(value="size_t") long size)
@Cast(value="cudaError_t") public static int cudaMallocPitch(@Cast(value="void**") PointerPointer devPtr, @Cast(value="size_t*") SizeTPointer pitch, @Cast(value="size_t") long width, @Cast(value="size_t") long height)
T* pElement = (T*)((char*)BaseAddress + Row * pitch) + Column;
For allocations of 2D arrays, it is recommended that programmers consider
performing pitch allocations using ::cudaMallocPitch(). Due to pitch
alignment restrictions in the hardware, this is especially true if the
application will be performing 2D memory copies between different regions
of device memory (whether linear memory or CUDA arrays).devPtr
- - Pointer to allocated pitched device memorypitch
- - Pitch for allocationwidth
- - Requested pitched allocation width (in bytes)height
- - Requested pitched allocation height::cudaMalloc, ::cudaFree, ::cudaMallocArray, ::cudaFreeArray,
\ref ::cudaMallocHost(void**, size_t) "cudaMallocHost (C API)",
::cudaFreeHost, ::cudaMalloc3D, ::cudaMalloc3DArray,
::cudaHostAlloc,
::cuMemAllocPitch
@Cast(value="cudaError_t") public static int cudaMallocPitch(@Cast(value="void**") @ByPtrPtr Pointer devPtr, @Cast(value="size_t*") SizeTPointer pitch, @Cast(value="size_t") long width, @Cast(value="size_t") long height)
@Cast(value="cudaError_t") public static int cudaMallocArray(@ByPtrPtr cudaArray array, @Const cudaChannelFormatDesc desc, @Cast(value="size_t") long width, @Cast(value="size_t") long height, @Cast(value="unsigned int") int flags)
struct cudaChannelFormatDesc {
int x, y, z, w;
enum cudaChannelFormatKind f;
};
where ::cudaChannelFormatKind is one of ::cudaChannelFormatKindSigned,
::cudaChannelFormatKindUnsigned, or ::cudaChannelFormatKindFloat.
The \p flags parameter enables different options to be specified that affect
the allocation, as follows.
- ::cudaArrayDefault: This flag's value is defined to be 0 and provides default array allocation
- ::cudaArraySurfaceLoadStore: Allocates an array that can be read from or written to using a surface reference
- ::cudaArrayTextureGather: This flag indicates that texture gather operations will be performed on the array.
- ::cudaArraySparse: Allocates a CUDA array without physical backing memory. The subregions within this sparse array
can later be mapped onto a physical memory allocation by calling ::cuMemMapArrayAsync.
The physical backing memory must be allocated via ::cuMemCreate.
- ::cudaArrayDeferredMapping: Allocates a CUDA array without physical backing memory. The entire array can
later be mapped onto a physical memory allocation by calling ::cuMemMapArrayAsync.
The physical backing memory must be allocated via ::cuMemCreate.
\p width and \p height must meet certain size requirements. See ::cudaMalloc3DArray() for more details.array
- - Pointer to allocated array in device memorydesc
- - Requested channel formatwidth
- - Requested array allocation widthheight
- - Requested array allocation heightflags
- - Requested properties of allocated array::cudaMalloc, ::cudaMallocPitch, ::cudaFree, ::cudaFreeArray,
\ref ::cudaMallocHost(void**, size_t) "cudaMallocHost (C API)",
::cudaFreeHost, ::cudaMalloc3D, ::cudaMalloc3DArray,
::cudaHostAlloc,
::cuArrayCreate
@Cast(value="cudaError_t") public static int cudaMallocArray(@ByPtrPtr cudaArray array, @Const cudaChannelFormatDesc desc, @Cast(value="size_t") long width)
@Cast(value="cudaError_t") public static int cudaFree(Pointer devPtr)
devPtr
- - Device pointer to memory to free::cudaMalloc, ::cudaMallocPitch, ::cudaMallocManaged, ::cudaMallocArray, ::cudaFreeArray, ::cudaMallocAsync, ::cudaMallocFromPoolAsync
\ref ::cudaMallocHost(void**, size_t) "cudaMallocHost (C API)",
::cudaFreeHost, ::cudaMalloc3D, ::cudaMalloc3DArray, ::cudaFreeAsync
::cudaHostAlloc,
::cuMemFree
@Cast(value="cudaError_t") public static int cudaFreeHost(Pointer ptr)
ptr
- - Pointer to memory to free::cudaMalloc, ::cudaMallocPitch, ::cudaFree, ::cudaMallocArray,
::cudaFreeArray,
\ref ::cudaMallocHost(void**, size_t) "cudaMallocHost (C API)",
::cudaMalloc3D, ::cudaMalloc3DArray, ::cudaHostAlloc,
::cuMemFreeHost
@Cast(value="cudaError_t") public static int cudaFreeArray(cudaArray array)
array
- - Pointer to array to free::cudaMalloc, ::cudaMallocPitch, ::cudaFree, ::cudaMallocArray,
\ref ::cudaMallocHost(void**, size_t) "cudaMallocHost (C API)",
::cudaFreeHost, ::cudaHostAlloc,
::cuArrayDestroy
@Cast(value="cudaError_t") public static int cudaFreeMipmappedArray(cudaMipmappedArray mipmappedArray)
mipmappedArray
- - Pointer to mipmapped array to free::cudaMalloc, ::cudaMallocPitch, ::cudaFree, ::cudaMallocArray,
\ref ::cudaMallocHost(void**, size_t) "cudaMallocHost (C API)",
::cudaFreeHost, ::cudaHostAlloc,
::cuMipmappedArrayDestroy
@Cast(value="cudaError_t") public static int cudaHostAlloc(@Cast(value="void**") PointerPointer pHost, @Cast(value="size_t") long size, @Cast(value="unsigned int") int flags)
pHost
- - Device pointer to allocated memorysize
- - Requested allocation size in bytesflags
- - Requested properties of allocated memory::cudaSetDeviceFlags,
\ref ::cudaMallocHost(void**, size_t) "cudaMallocHost (C API)",
::cudaFreeHost,
::cudaGetDeviceFlags,
::cuMemHostAlloc
@Cast(value="cudaError_t") public static int cudaHostAlloc(@Cast(value="void**") @ByPtrPtr Pointer pHost, @Cast(value="size_t") long size, @Cast(value="unsigned int") int flags)
@Cast(value="cudaError_t") public static int cudaHostRegister(Pointer ptr, @Cast(value="size_t") long size, @Cast(value="unsigned int") int flags)
ptr
- - Host pointer to memory to page-locksize
- - Size in bytes of the address range to page-lock in bytesflags
- - Flags for allocation request::cudaHostUnregister, ::cudaHostGetFlags, ::cudaHostGetDevicePointer,
::cuMemHostRegister
@Cast(value="cudaError_t") public static int cudaHostUnregister(Pointer ptr)
ptr
- - Host pointer to memory to unregister::cudaHostUnregister,
::cuMemHostUnregister
@Cast(value="cudaError_t") public static int cudaHostGetDevicePointer(@Cast(value="void**") PointerPointer pDevice, Pointer pHost, @Cast(value="unsigned int") int flags)
pDevice
- - Returned device pointer for mapped memorypHost
- - Requested host pointer mappingflags
- - Flags for extensions (must be 0 for now)::cudaSetDeviceFlags, ::cudaHostAlloc,
::cuMemHostGetDevicePointer
@Cast(value="cudaError_t") public static int cudaHostGetDevicePointer(@Cast(value="void**") @ByPtrPtr Pointer pDevice, Pointer pHost, @Cast(value="unsigned int") int flags)
@Cast(value="cudaError_t") public static int cudaHostGetFlags(@Cast(value="unsigned int*") IntPointer pFlags, Pointer pHost)
pFlags
- - Returned flags wordpHost
- - Host pointer::cudaHostAlloc,
::cuMemHostGetFlags
@Cast(value="cudaError_t") public static int cudaHostGetFlags(@Cast(value="unsigned int*") IntBuffer pFlags, Pointer pHost)
@Cast(value="cudaError_t") public static int cudaHostGetFlags(@Cast(value="unsigned int*") int[] pFlags, Pointer pHost)
@Cast(value="cudaError_t") public static int cudaMalloc3D(cudaPitchedPtr pitchedDevPtr, @ByVal cudaExtent extent)
pitchedDevPtr
- - Pointer to allocated pitched device memoryextent
- - Requested allocation size (\p width field in bytes)::cudaMallocPitch, ::cudaFree, ::cudaMemcpy3D, ::cudaMemset3D,
::cudaMalloc3DArray, ::cudaMallocArray, ::cudaFreeArray,
\ref ::cudaMallocHost(void**, size_t) "cudaMallocHost (C API)",
::cudaFreeHost, ::cudaHostAlloc, ::make_cudaPitchedPtr, ::make_cudaExtent,
::cuMemAllocPitch
@Cast(value="cudaError_t") public static int cudaMalloc3DArray(@ByPtrPtr cudaArray array, @Const cudaChannelFormatDesc desc, @ByVal cudaExtent extent, @Cast(value="unsigned int") int flags)
struct cudaChannelFormatDesc {
int x, y, z, w;
enum cudaChannelFormatKind f;
};
where ::cudaChannelFormatKind is one of ::cudaChannelFormatKindSigned,
::cudaChannelFormatKindUnsigned, or ::cudaChannelFormatKindFloat.
::cudaMalloc3DArray() can allocate the following:
- A 1D array is allocated if the height and depth extents are both zero.
- A 2D array is allocated if only the depth extent is zero.
- A 3D array is allocated if all three extents are non-zero.
- A 1D layered CUDA array is allocated if only the height extent is zero and
the cudaArrayLayered flag is set. Each layer is a 1D array. The number of layers is
determined by the depth extent.
- A 2D layered CUDA array is allocated if all three extents are non-zero and
the cudaArrayLayered flag is set. Each layer is a 2D array. The number of layers is
determined by the depth extent.
- A cubemap CUDA array is allocated if all three extents are non-zero and the
cudaArrayCubemap flag is set. Width must be equal to height, and depth must be six. A cubemap is
a special type of 2D layered CUDA array, where the six layers represent the six faces of a cube.
The order of the six layers in memory is the same as that listed in ::cudaGraphicsCubeFace.
- A cubemap layered CUDA array is allocated if all three extents are non-zero, and both,
cudaArrayCubemap and cudaArrayLayered flags are set. Width must be equal to height, and depth must be
a multiple of six. A cubemap layered CUDA array is a special type of 2D layered CUDA array that consists
of a collection of cubemaps. The first six layers represent the first cubemap, the next six layers form
the second cubemap, and so on.
The \p flags parameter enables different options to be specified that affect
the allocation, as follows.
- ::cudaArrayDefault: This flag's value is defined to be 0 and provides default array allocation
- ::cudaArrayLayered: Allocates a layered CUDA array, with the depth extent indicating the number of layers
- ::cudaArrayCubemap: Allocates a cubemap CUDA array. Width must be equal to height, and depth must be six.
If the cudaArrayLayered flag is also set, depth must be a multiple of six.
- ::cudaArraySurfaceLoadStore: Allocates a CUDA array that could be read from or written to using a surface
reference.
- ::cudaArrayTextureGather: This flag indicates that texture gather operations will be performed on the CUDA
array. Texture gather can only be performed on 2D CUDA arrays.
- ::cudaArraySparse: Allocates a CUDA array without physical backing memory. The subregions within this sparse array
can later be mapped onto a physical memory allocation by calling ::cuMemMapArrayAsync. This flag can only be used for
creating 2D, 3D or 2D layered sparse CUDA arrays. The physical backing memory must be allocated via ::cuMemCreate.
- ::cudaArrayDeferredMapping: Allocates a CUDA array without physical backing memory. The entire array can
later be mapped onto a physical memory allocation by calling ::cuMemMapArrayAsync. The physical backing memory must be allocated
via ::cuMemCreate.
The width, height and depth extents must meet certain size requirements as listed in the following table.
All values are specified in elements.
Note that 2D CUDA arrays have different size requirements if the ::cudaArrayTextureGather flag is set. In that
case, the valid range for (width, height, depth) is ((1,maxTexture2DGather[0]), (1,maxTexture2DGather[1]), 0).
\xmlonly
array
- - Pointer to allocated array in device memorydesc
- - Requested channel formatextent
- - Requested allocation size (\p width field in elements)flags
- - Flags for extensions::cudaMalloc3D, ::cudaMalloc, ::cudaMallocPitch, ::cudaFree,
::cudaFreeArray,
\ref ::cudaMallocHost(void**, size_t) "cudaMallocHost (C API)",
::cudaFreeHost, ::cudaHostAlloc,
::make_cudaExtent,
::cuArray3DCreate
@Cast(value="cudaError_t") public static int cudaMalloc3DArray(@ByPtrPtr cudaArray array, @Const cudaChannelFormatDesc desc, @ByVal cudaExtent extent)
@Cast(value="cudaError_t") public static int cudaMallocMipmappedArray(@ByPtrPtr cudaMipmappedArray mipmappedArray, @Const cudaChannelFormatDesc desc, @ByVal cudaExtent extent, @Cast(value="unsigned int") int numLevels, @Cast(value="unsigned int") int flags)
struct cudaChannelFormatDesc {
int x, y, z, w;
enum cudaChannelFormatKind f;
};
where ::cudaChannelFormatKind is one of ::cudaChannelFormatKindSigned,
::cudaChannelFormatKindUnsigned, or ::cudaChannelFormatKindFloat.
::cudaMallocMipmappedArray() can allocate the following:
- A 1D mipmapped array is allocated if the height and depth extents are both zero.
- A 2D mipmapped array is allocated if only the depth extent is zero.
- A 3D mipmapped array is allocated if all three extents are non-zero.
- A 1D layered CUDA mipmapped array is allocated if only the height extent is zero and
the cudaArrayLayered flag is set. Each layer is a 1D mipmapped array. The number of layers is
determined by the depth extent.
- A 2D layered CUDA mipmapped array is allocated if all three extents are non-zero and
the cudaArrayLayered flag is set. Each layer is a 2D mipmapped array. The number of layers is
determined by the depth extent.
- A cubemap CUDA mipmapped array is allocated if all three extents are non-zero and the
cudaArrayCubemap flag is set. Width must be equal to height, and depth must be six.
The order of the six layers in memory is the same as that listed in ::cudaGraphicsCubeFace.
- A cubemap layered CUDA mipmapped array is allocated if all three extents are non-zero, and both,
cudaArrayCubemap and cudaArrayLayered flags are set. Width must be equal to height, and depth must be
a multiple of six. A cubemap layered CUDA mipmapped array is a special type of 2D layered CUDA mipmapped
array that consists of a collection of cubemap mipmapped arrays. The first six layers represent the
first cubemap mipmapped array, the next six layers form the second cubemap mipmapped array, and so on.
The \p flags parameter enables different options to be specified that affect
the allocation, as follows.
- ::cudaArrayDefault: This flag's value is defined to be 0 and provides default mipmapped array allocation
- ::cudaArrayLayered: Allocates a layered CUDA mipmapped array, with the depth extent indicating the number of layers
- ::cudaArrayCubemap: Allocates a cubemap CUDA mipmapped array. Width must be equal to height, and depth must be six.
If the cudaArrayLayered flag is also set, depth must be a multiple of six.
- ::cudaArraySurfaceLoadStore: This flag indicates that individual mipmap levels of the CUDA mipmapped array
will be read from or written to using a surface reference.
- ::cudaArrayTextureGather: This flag indicates that texture gather operations will be performed on the CUDA
array. Texture gather can only be performed on 2D CUDA mipmapped arrays, and the gather operations are
performed only on the most detailed mipmap level.
- ::cudaArraySparse: Allocates a CUDA mipmapped array without physical backing memory. The subregions within this sparse array
can later be mapped onto a physical memory allocation by calling ::cuMemMapArrayAsync. This flag can only be used for creating
2D, 3D or 2D layered sparse CUDA mipmapped arrays. The physical backing memory must be allocated via ::cuMemCreate.
- ::cudaArrayDeferredMapping: Allocates a CUDA mipmapped array without physical backing memory. The entire array can
later be mapped onto a physical memory allocation by calling ::cuMemMapArrayAsync. The physical backing memory must be allocated
via ::cuMemCreate.
The width, height and depth extents must meet certain size requirements as listed in the following table.
All values are specified in elements.
\xmlonly
mipmappedArray
- - Pointer to allocated mipmapped array in device memorydesc
- - Requested channel formatextent
- - Requested allocation size (\p width field in elements)numLevels
- - Number of mipmap levels to allocateflags
- - Flags for extensions::cudaMalloc3D, ::cudaMalloc, ::cudaMallocPitch, ::cudaFree,
::cudaFreeArray,
\ref ::cudaMallocHost(void**, size_t) "cudaMallocHost (C API)",
::cudaFreeHost, ::cudaHostAlloc,
::make_cudaExtent,
::cuMipmappedArrayCreate
@Cast(value="cudaError_t") public static int cudaMallocMipmappedArray(@ByPtrPtr cudaMipmappedArray mipmappedArray, @Const cudaChannelFormatDesc desc, @ByVal cudaExtent extent, @Cast(value="unsigned int") int numLevels)
@Cast(value="cudaError_t") public static int cudaGetMipmappedArrayLevel(@ByPtrPtr cudaArray levelArray, cudaMipmappedArray mipmappedArray, @Cast(value="unsigned int") int level)
levelArray
- - Returned mipmap level CUDA arraymipmappedArray
- - CUDA mipmapped arraylevel
- - Mipmap level::cudaMalloc3D, ::cudaMalloc, ::cudaMallocPitch, ::cudaFree,
::cudaFreeArray,
\ref ::cudaMallocHost(void**, size_t) "cudaMallocHost (C API)",
::cudaFreeHost, ::cudaHostAlloc,
::make_cudaExtent,
::cuMipmappedArrayGetLevel
@Cast(value="cudaError_t") public static int cudaMemcpy3D(@Const cudaMemcpy3DParms p)
struct cudaExtent {
size_t width;
size_t height;
size_t depth;
};
struct cudaExtent make_cudaExtent(size_t w, size_t h, size_t d);
struct cudaPos {
size_t x;
size_t y;
size_t z;
};
struct cudaPos make_cudaPos(size_t x, size_t y, size_t z);
struct cudaMemcpy3DParms {
cudaArray_t srcArray;
struct cudaPos srcPos;
struct cudaPitchedPtr srcPtr;
cudaArray_t dstArray;
struct cudaPos dstPos;
struct cudaPitchedPtr dstPtr;
struct cudaExtent extent;
enum cudaMemcpyKind kind;
};
::cudaMemcpy3D() copies data betwen two 3D objects. The source and
destination objects may be in either host memory, device memory, or a CUDA
array. The source, destination, extent, and kind of copy performed is
specified by the ::cudaMemcpy3DParms struct which should be initialized to
zero before use:
cudaMemcpy3DParms myParms = {0};
The struct passed to ::cudaMemcpy3D() must specify one of \p srcArray or
\p srcPtr and one of \p dstArray or \p dstPtr. Passing more than one
non-zero source or destination will cause ::cudaMemcpy3D() to return an
error.
The \p srcPos and \p dstPos fields are optional offsets into the source and
destination objects and are defined in units of each object's elements. The
element for a host or device pointer is assumed to be unsigned char.
The \p extent field defines the dimensions of the transferred area in
elements. If a CUDA array is participating in the copy, the extent is
defined in terms of that array's elements. If no CUDA array is
participating in the copy then the extents are defined in elements of
unsigned char.
The \p kind field defines the direction of the copy. It must be one of
::cudaMemcpyHostToHost, ::cudaMemcpyHostToDevice, ::cudaMemcpyDeviceToHost,
::cudaMemcpyDeviceToDevice, or ::cudaMemcpyDefault. Passing
::cudaMemcpyDefault is recommended, in which case the type of transfer is
inferred from the pointer values. However, ::cudaMemcpyDefault is only
allowed on systems that support unified virtual addressing.
For ::cudaMemcpyHostToHost or ::cudaMemcpyHostToDevice or ::cudaMemcpyDeviceToHost
passed as kind and cudaArray type passed as source or destination, if the kind
implies cudaArray type to be present on the host, ::cudaMemcpy3D() will
disregard that implication and silently correct the kind based on the fact that
cudaArray type can only be present on the device.
If the source and destination are both arrays, ::cudaMemcpy3D() will return
an error if they do not have the same element size.
The source and destination object may not overlap. If overlapping source
and destination objects are specified, undefined behavior will result.
The source object must entirely contain the region defined by \p srcPos
and \p extent. The destination object must entirely contain the region
defined by \p dstPos and \p extent.
::cudaMemcpy3D() returns an error if the pitch of \p srcPtr or \p dstPtr
exceeds the maximum allowed. The pitch of a ::cudaPitchedPtr allocated
with ::cudaMalloc3D() will always be valid.p
- - 3D memory copy parameters::cudaMalloc3D, ::cudaMalloc3DArray, ::cudaMemset3D, ::cudaMemcpy3DAsync,
::cudaMemcpy, ::cudaMemcpy2D,
::cudaMemcpy2DToArray, ::cudaMemcpy2DFromArray,
::cudaMemcpy2DArrayToArray, ::cudaMemcpyToSymbol,
::cudaMemcpyFromSymbol, ::cudaMemcpyAsync, ::cudaMemcpy2DAsync,
::cudaMemcpy2DToArrayAsync,
::cudaMemcpy2DFromArrayAsync,
::cudaMemcpyToSymbolAsync, ::cudaMemcpyFromSymbolAsync,
::make_cudaExtent, ::make_cudaPos,
::cuMemcpy3D
@Cast(value="cudaError_t") public static int cudaMemcpy3DPeer(@Const cudaMemcpy3DPeerParms p)
p
- - Parameters for the memory copy::cudaMemcpy, ::cudaMemcpyPeer, ::cudaMemcpyAsync, ::cudaMemcpyPeerAsync,
::cudaMemcpy3DPeerAsync,
::cuMemcpy3DPeer
@Cast(value="cudaError_t") public static int cudaMemcpy3DAsync(@Const cudaMemcpy3DParms p, CUstream_st stream)
struct cudaExtent {
size_t width;
size_t height;
size_t depth;
};
struct cudaExtent make_cudaExtent(size_t w, size_t h, size_t d);
struct cudaPos {
size_t x;
size_t y;
size_t z;
};
struct cudaPos make_cudaPos(size_t x, size_t y, size_t z);
struct cudaMemcpy3DParms {
cudaArray_t srcArray;
struct cudaPos srcPos;
struct cudaPitchedPtr srcPtr;
cudaArray_t dstArray;
struct cudaPos dstPos;
struct cudaPitchedPtr dstPtr;
struct cudaExtent extent;
enum cudaMemcpyKind kind;
};
::cudaMemcpy3DAsync() copies data betwen two 3D objects. The source and
destination objects may be in either host memory, device memory, or a CUDA
array. The source, destination, extent, and kind of copy performed is
specified by the ::cudaMemcpy3DParms struct which should be initialized to
zero before use:
cudaMemcpy3DParms myParms = {0};
The struct passed to ::cudaMemcpy3DAsync() must specify one of \p srcArray
or \p srcPtr and one of \p dstArray or \p dstPtr. Passing more than one
non-zero source or destination will cause ::cudaMemcpy3DAsync() to return an
error.
The \p srcPos and \p dstPos fields are optional offsets into the source and
destination objects and are defined in units of each object's elements. The
element for a host or device pointer is assumed to be unsigned char.
For CUDA arrays, positions must be in the range [0, 2048) for any
dimension.
The \p extent field defines the dimensions of the transferred area in
elements. If a CUDA array is participating in the copy, the extent is
defined in terms of that array's elements. If no CUDA array is
participating in the copy then the extents are defined in elements of
unsigned char.
The \p kind field defines the direction of the copy. It must be one of
::cudaMemcpyHostToHost, ::cudaMemcpyHostToDevice, ::cudaMemcpyDeviceToHost,
::cudaMemcpyDeviceToDevice, or ::cudaMemcpyDefault. Passing
::cudaMemcpyDefault is recommended, in which case the type of transfer is
inferred from the pointer values. However, ::cudaMemcpyDefault is only
allowed on systems that support unified virtual addressing.
For ::cudaMemcpyHostToHost or ::cudaMemcpyHostToDevice or ::cudaMemcpyDeviceToHost
passed as kind and cudaArray type passed as source or destination, if the kind
implies cudaArray type to be present on the host, ::cudaMemcpy3DAsync() will
disregard that implication and silently correct the kind based on the fact that
cudaArray type can only be present on the device.
If the source and destination are both arrays, ::cudaMemcpy3DAsync() will
return an error if they do not have the same element size.
The source and destination object may not overlap. If overlapping source
and destination objects are specified, undefined behavior will result.
The source object must lie entirely within the region defined by \p srcPos
and \p extent. The destination object must lie entirely within the region
defined by \p dstPos and \p extent.
::cudaMemcpy3DAsync() returns an error if the pitch of \p srcPtr or
\p dstPtr exceeds the maximum allowed. The pitch of a
::cudaPitchedPtr allocated with ::cudaMalloc3D() will always be valid.
::cudaMemcpy3DAsync() is asynchronous with respect to the host, so
the call may return before the copy is complete. The copy can optionally
be associated to a stream by passing a non-zero \p stream argument. If
\p kind is ::cudaMemcpyHostToDevice or ::cudaMemcpyDeviceToHost and \p stream
is non-zero, the copy may overlap with operations in other streams.
The device version of this function only handles device to device copies and
cannot be given local or shared pointers.p
- - 3D memory copy parametersstream
- - Stream identifier::cudaMalloc3D, ::cudaMalloc3DArray, ::cudaMemset3D, ::cudaMemcpy3D,
::cudaMemcpy, ::cudaMemcpy2D,
::cudaMemcpy2DToArray, :::cudaMemcpy2DFromArray,
::cudaMemcpy2DArrayToArray, ::cudaMemcpyToSymbol,
::cudaMemcpyFromSymbol, ::cudaMemcpyAsync, ::cudaMemcpy2DAsync,
::cudaMemcpy2DToArrayAsync,
::cudaMemcpy2DFromArrayAsync,
::cudaMemcpyToSymbolAsync, ::cudaMemcpyFromSymbolAsync,
::make_cudaExtent, ::make_cudaPos,
::cuMemcpy3DAsync
@Cast(value="cudaError_t") public static int cudaMemcpy3DAsync(@Const cudaMemcpy3DParms p)
@Cast(value="cudaError_t") public static int cudaMemcpy3DPeerAsync(@Const cudaMemcpy3DPeerParms p, CUstream_st stream)
p
- - Parameters for the memory copystream
- - Stream identifier::cudaMemcpy, ::cudaMemcpyPeer, ::cudaMemcpyAsync, ::cudaMemcpyPeerAsync,
::cudaMemcpy3DPeerAsync,
::cuMemcpy3DPeerAsync
@Cast(value="cudaError_t") public static int cudaMemcpy3DPeerAsync(@Const cudaMemcpy3DPeerParms p)
@Cast(value="cudaError_t") public static int cudaMemGetInfo(@Cast(value="size_t*") SizeTPointer _free, @Cast(value="size_t*") SizeTPointer total)
free
- - Returned free memory in bytestotal
- - Returned total memory in bytes::cuMemGetInfo
@Cast(value="cudaError_t") public static int cudaArrayGetInfo(cudaChannelFormatDesc desc, cudaExtent extent, @Cast(value="unsigned int*") IntPointer flags, cudaArray array)
desc
- - Returned array typeextent
- - Returned array shape. 2D arrays will have depth of zeroflags
- - Returned array flagsarray
- - The ::cudaArray to get info for::cuArrayGetDescriptor,
::cuArray3DGetDescriptor
@Cast(value="cudaError_t") public static int cudaArrayGetInfo(cudaChannelFormatDesc desc, cudaExtent extent, @Cast(value="unsigned int*") IntBuffer flags, cudaArray array)
@Cast(value="cudaError_t") public static int cudaArrayGetInfo(cudaChannelFormatDesc desc, cudaExtent extent, @Cast(value="unsigned int*") int[] flags, cudaArray array)
@Cast(value="cudaError_t") public static int cudaArrayGetPlane(@ByPtrPtr cudaArray pPlaneArray, cudaArray hArray, @Cast(value="unsigned int") int planeIdx)
pPlaneArray
- - Returned CUDA array referenced by the \p planeIdxhArray
- - CUDA arrayplaneIdx
- - Plane index::cuArrayGetPlane
@Cast(value="cudaError_t") public static int cudaArrayGetMemoryRequirements(cudaArrayMemoryRequirements memoryRequirements, cudaArray array, int device)
memoryRequirements
- [out] - Pointer to ::cudaArrayMemoryRequirementsarray
- [in] - CUDA array to get the memory requirements ofdevice
- [in] - Device to get the memory requirements for::cudaMipmappedArrayGetMemoryRequirements
@Cast(value="cudaError_t") public static int cudaMipmappedArrayGetMemoryRequirements(cudaArrayMemoryRequirements memoryRequirements, cudaMipmappedArray mipmap, int device)
memoryRequirements
- [out] - Pointer to ::cudaArrayMemoryRequirementsmipmap
- [in] - CUDA mipmapped array to get the memory requirements ofdevice
- [in] - Device to get the memory requirements for::cudaArrayGetMemoryRequirements
@Cast(value="cudaError_t") public static int cudaArrayGetSparseProperties(cudaArraySparseProperties sparseProperties, cudaArray array)
sparseProperties
- [out] - Pointer to return the ::cudaArraySparsePropertiesarray
- [in] - The CUDA array to get the sparse properties of::cudaMipmappedArrayGetSparseProperties,
::cuMemMapArrayAsync
@Cast(value="cudaError_t") public static int cudaMipmappedArrayGetSparseProperties(cudaArraySparseProperties sparseProperties, cudaMipmappedArray mipmap)
sparseProperties
- [out] - Pointer to return ::cudaArraySparsePropertiesmipmap
- [in] - The CUDA mipmapped array to get the sparse properties of::cudaArrayGetSparseProperties,
::cuMemMapArrayAsync
@Cast(value="cudaError_t") public static int cudaMemcpy(Pointer dst, @Const Pointer src, @Cast(value="size_t") long count, @Cast(value="cudaMemcpyKind") int kind)
dst
- - Destination memory addresssrc
- - Source memory addresscount
- - Size in bytes to copykind
- - Type of transfer::cudaMemcpy2D,
::cudaMemcpy2DToArray, ::cudaMemcpy2DFromArray,
::cudaMemcpy2DArrayToArray, ::cudaMemcpyToSymbol,
::cudaMemcpyFromSymbol, ::cudaMemcpyAsync, ::cudaMemcpy2DAsync,
::cudaMemcpy2DToArrayAsync,
::cudaMemcpy2DFromArrayAsync,
::cudaMemcpyToSymbolAsync, ::cudaMemcpyFromSymbolAsync,
::cuMemcpyDtoH,
::cuMemcpyHtoD,
::cuMemcpyDtoD,
::cuMemcpy
@Cast(value="cudaError_t") public static int cudaMemcpyPeer(Pointer dst, int dstDevice, @Const Pointer src, int srcDevice, @Cast(value="size_t") long count)
dst
- - Destination device pointerdstDevice
- - Destination devicesrc
- - Source device pointersrcDevice
- - Source devicecount
- - Size of memory copy in bytes::cudaMemcpy, ::cudaMemcpyAsync, ::cudaMemcpyPeerAsync,
::cudaMemcpy3DPeerAsync,
::cuMemcpyPeer
@Cast(value="cudaError_t") public static int cudaMemcpy2D(Pointer dst, @Cast(value="size_t") long dpitch, @Const Pointer src, @Cast(value="size_t") long spitch, @Cast(value="size_t") long width, @Cast(value="size_t") long height, @Cast(value="cudaMemcpyKind") int kind)
dst
- - Destination memory addressdpitch
- - Pitch of destination memorysrc
- - Source memory addressspitch
- - Pitch of source memorywidth
- - Width of matrix transfer (columns in bytes)height
- - Height of matrix transfer (rows)kind
- - Type of transfer::cudaMemcpy,
::cudaMemcpy2DToArray, ::cudaMemcpy2DFromArray,
::cudaMemcpy2DArrayToArray, ::cudaMemcpyToSymbol,
::cudaMemcpyFromSymbol, ::cudaMemcpyAsync, ::cudaMemcpy2DAsync,
::cudaMemcpy2DToArrayAsync,
::cudaMemcpy2DFromArrayAsync,
::cudaMemcpyToSymbolAsync, ::cudaMemcpyFromSymbolAsync,
::cuMemcpy2D,
::cuMemcpy2DUnaligned
@Cast(value="cudaError_t") public static int cudaMemcpy2DToArray(cudaArray dst, @Cast(value="size_t") long wOffset, @Cast(value="size_t") long hOffset, @Const Pointer src, @Cast(value="size_t") long spitch, @Cast(value="size_t") long width, @Cast(value="size_t") long height, @Cast(value="cudaMemcpyKind") int kind)
dst
- - Destination memory addresswOffset
- - Destination starting X offset (columns in bytes)hOffset
- - Destination starting Y offset (rows)src
- - Source memory addressspitch
- - Pitch of source memorywidth
- - Width of matrix transfer (columns in bytes)height
- - Height of matrix transfer (rows)kind
- - Type of transfer::cudaMemcpy, ::cudaMemcpy2D,
::cudaMemcpy2DFromArray,
::cudaMemcpy2DArrayToArray, ::cudaMemcpyToSymbol,
::cudaMemcpyFromSymbol, ::cudaMemcpyAsync, ::cudaMemcpy2DAsync,
::cudaMemcpy2DToArrayAsync,
::cudaMemcpy2DFromArrayAsync,
::cudaMemcpyToSymbolAsync, ::cudaMemcpyFromSymbolAsync,
::cuMemcpy2D,
::cuMemcpy2DUnaligned
@Cast(value="cudaError_t") public static int cudaMemcpy2DFromArray(Pointer dst, @Cast(value="size_t") long dpitch, cudaArray src, @Cast(value="size_t") long wOffset, @Cast(value="size_t") long hOffset, @Cast(value="size_t") long width, @Cast(value="size_t") long height, @Cast(value="cudaMemcpyKind") int kind)
dst
- - Destination memory addressdpitch
- - Pitch of destination memorysrc
- - Source memory addresswOffset
- - Source starting X offset (columns in bytes)hOffset
- - Source starting Y offset (rows)width
- - Width of matrix transfer (columns in bytes)height
- - Height of matrix transfer (rows)kind
- - Type of transfer::cudaMemcpy, ::cudaMemcpy2D,
::cudaMemcpy2DToArray,
::cudaMemcpy2DArrayToArray, ::cudaMemcpyToSymbol,
::cudaMemcpyFromSymbol, ::cudaMemcpyAsync, ::cudaMemcpy2DAsync,
::cudaMemcpy2DToArrayAsync,
::cudaMemcpy2DFromArrayAsync,
::cudaMemcpyToSymbolAsync, ::cudaMemcpyFromSymbolAsync,
::cuMemcpy2D,
::cuMemcpy2DUnaligned
@Cast(value="cudaError_t") public static int cudaMemcpy2DArrayToArray(cudaArray dst, @Cast(value="size_t") long wOffsetDst, @Cast(value="size_t") long hOffsetDst, cudaArray src, @Cast(value="size_t") long wOffsetSrc, @Cast(value="size_t") long hOffsetSrc, @Cast(value="size_t") long width, @Cast(value="size_t") long height, @Cast(value="cudaMemcpyKind") int kind)
dst
- - Destination memory addresswOffsetDst
- - Destination starting X offset (columns in bytes)hOffsetDst
- - Destination starting Y offset (rows)src
- - Source memory addresswOffsetSrc
- - Source starting X offset (columns in bytes)hOffsetSrc
- - Source starting Y offset (rows)width
- - Width of matrix transfer (columns in bytes)height
- - Height of matrix transfer (rows)kind
- - Type of transfer::cudaMemcpy, ::cudaMemcpy2D,
::cudaMemcpy2DToArray, ::cudaMemcpy2DFromArray,
::cudaMemcpyToSymbol,
::cudaMemcpyFromSymbol, ::cudaMemcpyAsync, ::cudaMemcpy2DAsync,
::cudaMemcpy2DToArrayAsync,
::cudaMemcpy2DFromArrayAsync,
::cudaMemcpyToSymbolAsync, ::cudaMemcpyFromSymbolAsync,
::cuMemcpy2D,
::cuMemcpy2DUnaligned
@Cast(value="cudaError_t") public static int cudaMemcpy2DArrayToArray(cudaArray dst, @Cast(value="size_t") long wOffsetDst, @Cast(value="size_t") long hOffsetDst, cudaArray src, @Cast(value="size_t") long wOffsetSrc, @Cast(value="size_t") long hOffsetSrc, @Cast(value="size_t") long width, @Cast(value="size_t") long height)
@Cast(value="cudaError_t") public static int cudaMemcpyToSymbol(@Const Pointer symbol, @Const Pointer src, @Cast(value="size_t") long count, @Cast(value="size_t") long offset, @Cast(value="cudaMemcpyKind") int kind)
symbol
- - Device symbol addresssrc
- - Source memory addresscount
- - Size in bytes to copyoffset
- - Offset from start of symbol in byteskind
- - Type of transfer::cudaMemcpy, ::cudaMemcpy2D,
::cudaMemcpy2DToArray, ::cudaMemcpy2DFromArray,
::cudaMemcpy2DArrayToArray,
::cudaMemcpyFromSymbol, ::cudaMemcpyAsync, ::cudaMemcpy2DAsync,
::cudaMemcpy2DToArrayAsync,
::cudaMemcpy2DFromArrayAsync,
::cudaMemcpyToSymbolAsync, ::cudaMemcpyFromSymbolAsync,
::cuMemcpy,
::cuMemcpyHtoD,
::cuMemcpyDtoD
@Cast(value="cudaError_t") public static int cudaMemcpyToSymbol(@Const Pointer symbol, @Const Pointer src, @Cast(value="size_t") long count)
@Cast(value="cudaError_t") public static int cudaMemcpyFromSymbol(Pointer dst, @Const Pointer symbol, @Cast(value="size_t") long count, @Cast(value="size_t") long offset, @Cast(value="cudaMemcpyKind") int kind)
dst
- - Destination memory addresssymbol
- - Device symbol addresscount
- - Size in bytes to copyoffset
- - Offset from start of symbol in byteskind
- - Type of transfer::cudaMemcpy, ::cudaMemcpy2D,
::cudaMemcpy2DToArray, ::cudaMemcpy2DFromArray,
::cudaMemcpy2DArrayToArray, ::cudaMemcpyToSymbol,
::cudaMemcpyAsync, ::cudaMemcpy2DAsync,
::cudaMemcpy2DToArrayAsync,
::cudaMemcpy2DFromArrayAsync,
::cudaMemcpyToSymbolAsync, ::cudaMemcpyFromSymbolAsync,
::cuMemcpy,
::cuMemcpyDtoH,
::cuMemcpyDtoD
@Cast(value="cudaError_t") public static int cudaMemcpyFromSymbol(Pointer dst, @Const Pointer symbol, @Cast(value="size_t") long count)
@Cast(value="cudaError_t") public static int cudaMemcpyAsync(Pointer dst, @Const Pointer src, @Cast(value="size_t") long count, @Cast(value="cudaMemcpyKind") int kind, CUstream_st stream)
dst
- - Destination memory addresssrc
- - Source memory addresscount
- - Size in bytes to copykind
- - Type of transferstream
- - Stream identifier::cudaMemcpy, ::cudaMemcpy2D,
::cudaMemcpy2DToArray, ::cudaMemcpy2DFromArray,
::cudaMemcpy2DArrayToArray, ::cudaMemcpyToSymbol,
::cudaMemcpyFromSymbol, ::cudaMemcpy2DAsync,
::cudaMemcpy2DToArrayAsync,
::cudaMemcpy2DFromArrayAsync,
::cudaMemcpyToSymbolAsync, ::cudaMemcpyFromSymbolAsync,
::cuMemcpyAsync,
::cuMemcpyDtoHAsync,
::cuMemcpyHtoDAsync,
::cuMemcpyDtoDAsync
@Cast(value="cudaError_t") public static int cudaMemcpyAsync(Pointer dst, @Const Pointer src, @Cast(value="size_t") long count, @Cast(value="cudaMemcpyKind") int kind)
@Cast(value="cudaError_t") public static int cudaMemcpyPeerAsync(Pointer dst, int dstDevice, @Const Pointer src, int srcDevice, @Cast(value="size_t") long count, CUstream_st stream)
dst
- - Destination device pointerdstDevice
- - Destination devicesrc
- - Source device pointersrcDevice
- - Source devicecount
- - Size of memory copy in bytesstream
- - Stream identifier::cudaMemcpy, ::cudaMemcpyPeer, ::cudaMemcpyAsync,
::cudaMemcpy3DPeerAsync,
::cuMemcpyPeerAsync
@Cast(value="cudaError_t") public static int cudaMemcpyPeerAsync(Pointer dst, int dstDevice, @Const Pointer src, int srcDevice, @Cast(value="size_t") long count)
@Cast(value="cudaError_t") public static int cudaMemcpy2DAsync(Pointer dst, @Cast(value="size_t") long dpitch, @Const Pointer src, @Cast(value="size_t") long spitch, @Cast(value="size_t") long width, @Cast(value="size_t") long height, @Cast(value="cudaMemcpyKind") int kind, CUstream_st stream)
dst
- - Destination memory addressdpitch
- - Pitch of destination memorysrc
- - Source memory addressspitch
- - Pitch of source memorywidth
- - Width of matrix transfer (columns in bytes)height
- - Height of matrix transfer (rows)kind
- - Type of transferstream
- - Stream identifier::cudaMemcpy, ::cudaMemcpy2D,
::cudaMemcpy2DToArray, ::cudaMemcpy2DFromArray,
::cudaMemcpy2DArrayToArray, ::cudaMemcpyToSymbol,
::cudaMemcpyFromSymbol, ::cudaMemcpyAsync,
::cudaMemcpy2DToArrayAsync,
::cudaMemcpy2DFromArrayAsync,
::cudaMemcpyToSymbolAsync, ::cudaMemcpyFromSymbolAsync,
::cuMemcpy2DAsync
@Cast(value="cudaError_t") public static int cudaMemcpy2DAsync(Pointer dst, @Cast(value="size_t") long dpitch, @Const Pointer src, @Cast(value="size_t") long spitch, @Cast(value="size_t") long width, @Cast(value="size_t") long height, @Cast(value="cudaMemcpyKind") int kind)
@Cast(value="cudaError_t") public static int cudaMemcpy2DToArrayAsync(cudaArray dst, @Cast(value="size_t") long wOffset, @Cast(value="size_t") long hOffset, @Const Pointer src, @Cast(value="size_t") long spitch, @Cast(value="size_t") long width, @Cast(value="size_t") long height, @Cast(value="cudaMemcpyKind") int kind, CUstream_st stream)
dst
- - Destination memory addresswOffset
- - Destination starting X offset (columns in bytes)hOffset
- - Destination starting Y offset (rows)src
- - Source memory addressspitch
- - Pitch of source memorywidth
- - Width of matrix transfer (columns in bytes)height
- - Height of matrix transfer (rows)kind
- - Type of transferstream
- - Stream identifier::cudaMemcpy, ::cudaMemcpy2D,
::cudaMemcpy2DToArray, ::cudaMemcpy2DFromArray,
::cudaMemcpy2DArrayToArray, ::cudaMemcpyToSymbol,
::cudaMemcpyFromSymbol, ::cudaMemcpyAsync, ::cudaMemcpy2DAsync,
::cudaMemcpy2DFromArrayAsync,
::cudaMemcpyToSymbolAsync, ::cudaMemcpyFromSymbolAsync,
::cuMemcpy2DAsync
@Cast(value="cudaError_t") public static int cudaMemcpy2DToArrayAsync(cudaArray dst, @Cast(value="size_t") long wOffset, @Cast(value="size_t") long hOffset, @Const Pointer src, @Cast(value="size_t") long spitch, @Cast(value="size_t") long width, @Cast(value="size_t") long height, @Cast(value="cudaMemcpyKind") int kind)
@Cast(value="cudaError_t") public static int cudaMemcpy2DFromArrayAsync(Pointer dst, @Cast(value="size_t") long dpitch, cudaArray src, @Cast(value="size_t") long wOffset, @Cast(value="size_t") long hOffset, @Cast(value="size_t") long width, @Cast(value="size_t") long height, @Cast(value="cudaMemcpyKind") int kind, CUstream_st stream)
dst
- - Destination memory addressdpitch
- - Pitch of destination memorysrc
- - Source memory addresswOffset
- - Source starting X offset (columns in bytes)hOffset
- - Source starting Y offset (rows)width
- - Width of matrix transfer (columns in bytes)height
- - Height of matrix transfer (rows)kind
- - Type of transferstream
- - Stream identifier::cudaMemcpy, ::cudaMemcpy2D,
::cudaMemcpy2DToArray, ::cudaMemcpy2DFromArray,
::cudaMemcpy2DArrayToArray, ::cudaMemcpyToSymbol,
::cudaMemcpyFromSymbol, ::cudaMemcpyAsync, ::cudaMemcpy2DAsync,
::cudaMemcpy2DToArrayAsync,
::cudaMemcpyToSymbolAsync, ::cudaMemcpyFromSymbolAsync,
::cuMemcpy2DAsync
@Cast(value="cudaError_t") public static int cudaMemcpy2DFromArrayAsync(Pointer dst, @Cast(value="size_t") long dpitch, cudaArray src, @Cast(value="size_t") long wOffset, @Cast(value="size_t") long hOffset, @Cast(value="size_t") long width, @Cast(value="size_t") long height, @Cast(value="cudaMemcpyKind") int kind)
@Cast(value="cudaError_t") public static int cudaMemcpyToSymbolAsync(@Const Pointer symbol, @Const Pointer src, @Cast(value="size_t") long count, @Cast(value="size_t") long offset, @Cast(value="cudaMemcpyKind") int kind, CUstream_st stream)
symbol
- - Device symbol addresssrc
- - Source memory addresscount
- - Size in bytes to copyoffset
- - Offset from start of symbol in byteskind
- - Type of transferstream
- - Stream identifier::cudaMemcpy, ::cudaMemcpy2D,
::cudaMemcpy2DToArray, ::cudaMemcpy2DFromArray,
::cudaMemcpy2DArrayToArray, ::cudaMemcpyToSymbol,
::cudaMemcpyFromSymbol, ::cudaMemcpyAsync, ::cudaMemcpy2DAsync,
::cudaMemcpy2DToArrayAsync,
::cudaMemcpy2DFromArrayAsync,
::cudaMemcpyFromSymbolAsync,
::cuMemcpyAsync,
::cuMemcpyHtoDAsync,
::cuMemcpyDtoDAsync
@Cast(value="cudaError_t") public static int cudaMemcpyToSymbolAsync(@Const Pointer symbol, @Const Pointer src, @Cast(value="size_t") long count, @Cast(value="size_t") long offset, @Cast(value="cudaMemcpyKind") int kind)
@Cast(value="cudaError_t") public static int cudaMemcpyFromSymbolAsync(Pointer dst, @Const Pointer symbol, @Cast(value="size_t") long count, @Cast(value="size_t") long offset, @Cast(value="cudaMemcpyKind") int kind, CUstream_st stream)
dst
- - Destination memory addresssymbol
- - Device symbol addresscount
- - Size in bytes to copyoffset
- - Offset from start of symbol in byteskind
- - Type of transferstream
- - Stream identifier::cudaMemcpy, ::cudaMemcpy2D,
::cudaMemcpy2DToArray, ::cudaMemcpy2DFromArray,
::cudaMemcpy2DArrayToArray, ::cudaMemcpyToSymbol,
::cudaMemcpyFromSymbol, ::cudaMemcpyAsync, ::cudaMemcpy2DAsync,
::cudaMemcpy2DToArrayAsync,
::cudaMemcpy2DFromArrayAsync,
::cudaMemcpyToSymbolAsync,
::cuMemcpyAsync,
::cuMemcpyDtoHAsync,
::cuMemcpyDtoDAsync
@Cast(value="cudaError_t") public static int cudaMemcpyFromSymbolAsync(Pointer dst, @Const Pointer symbol, @Cast(value="size_t") long count, @Cast(value="size_t") long offset, @Cast(value="cudaMemcpyKind") int kind)
@Cast(value="cudaError_t") public static int cudaMemset(Pointer devPtr, int value, @Cast(value="size_t") long count)
devPtr
- - Pointer to device memoryvalue
- - Value to set for each byte of specified memorycount
- - Size in bytes to set::cuMemsetD8,
::cuMemsetD16,
::cuMemsetD32
@Cast(value="cudaError_t") public static int cudaMemset2D(Pointer devPtr, @Cast(value="size_t") long pitch, int value, @Cast(value="size_t") long width, @Cast(value="size_t") long height)
devPtr
- - Pointer to 2D device memorypitch
- - Pitch in bytes of 2D device memory(Unused if \p height is 1)value
- - Value to set for each byte of specified memorywidth
- - Width of matrix set (columns in bytes)height
- - Height of matrix set (rows)::cudaMemset, ::cudaMemset3D, ::cudaMemsetAsync,
::cudaMemset2DAsync, ::cudaMemset3DAsync,
::cuMemsetD2D8,
::cuMemsetD2D16,
::cuMemsetD2D32
@Cast(value="cudaError_t") public static int cudaMemset3D(@ByVal cudaPitchedPtr pitchedDevPtr, int value, @ByVal cudaExtent extent)
pitchedDevPtr
- - Pointer to pitched device memoryvalue
- - Value to set for each byte of specified memoryextent
- - Size parameters for where to set device memory (\p width field in bytes)::cudaMemset, ::cudaMemset2D,
::cudaMemsetAsync, ::cudaMemset2DAsync, ::cudaMemset3DAsync,
::cudaMalloc3D, ::make_cudaPitchedPtr,
::make_cudaExtent
@Cast(value="cudaError_t") public static int cudaMemsetAsync(Pointer devPtr, int value, @Cast(value="size_t") long count, CUstream_st stream)
devPtr
- - Pointer to device memoryvalue
- - Value to set for each byte of specified memorycount
- - Size in bytes to setstream
- - Stream identifier::cudaMemset, ::cudaMemset2D, ::cudaMemset3D,
::cudaMemset2DAsync, ::cudaMemset3DAsync,
::cuMemsetD8Async,
::cuMemsetD16Async,
::cuMemsetD32Async
@Cast(value="cudaError_t") public static int cudaMemsetAsync(Pointer devPtr, int value, @Cast(value="size_t") long count)
@Cast(value="cudaError_t") public static int cudaMemset2DAsync(Pointer devPtr, @Cast(value="size_t") long pitch, int value, @Cast(value="size_t") long width, @Cast(value="size_t") long height, CUstream_st stream)
devPtr
- - Pointer to 2D device memorypitch
- - Pitch in bytes of 2D device memory(Unused if \p height is 1)value
- - Value to set for each byte of specified memorywidth
- - Width of matrix set (columns in bytes)height
- - Height of matrix set (rows)stream
- - Stream identifier::cudaMemset, ::cudaMemset2D, ::cudaMemset3D,
::cudaMemsetAsync, ::cudaMemset3DAsync,
::cuMemsetD2D8Async,
::cuMemsetD2D16Async,
::cuMemsetD2D32Async
@Cast(value="cudaError_t") public static int cudaMemset2DAsync(Pointer devPtr, @Cast(value="size_t") long pitch, int value, @Cast(value="size_t") long width, @Cast(value="size_t") long height)
@Cast(value="cudaError_t") public static int cudaMemset3DAsync(@ByVal cudaPitchedPtr pitchedDevPtr, int value, @ByVal cudaExtent extent, CUstream_st stream)
pitchedDevPtr
- - Pointer to pitched device memoryvalue
- - Value to set for each byte of specified memoryextent
- - Size parameters for where to set device memory (\p width field in bytes)stream
- - Stream identifier::cudaMemset, ::cudaMemset2D, ::cudaMemset3D,
::cudaMemsetAsync, ::cudaMemset2DAsync,
::cudaMalloc3D, ::make_cudaPitchedPtr,
::make_cudaExtent
@Cast(value="cudaError_t") public static int cudaMemset3DAsync(@ByVal cudaPitchedPtr pitchedDevPtr, int value, @ByVal cudaExtent extent)
@Cast(value="cudaError_t") public static int cudaGetSymbolAddress(@Cast(value="void**") PointerPointer devPtr, @Const Pointer symbol)
devPtr
- - Return device pointer associated with symbolsymbol
- - Device symbol address::cudaGetSymbolAddress(void**, const T&) "cudaGetSymbolAddress (C++ API)",
\ref ::cudaGetSymbolSize(size_t*, const void*) "cudaGetSymbolSize (C API)",
::cuModuleGetGlobal
@Cast(value="cudaError_t") public static int cudaGetSymbolAddress(@Cast(value="void**") @ByPtrPtr Pointer devPtr, @Const Pointer symbol)
@Cast(value="cudaError_t") public static int cudaGetSymbolSize(@Cast(value="size_t*") SizeTPointer size, @Const Pointer symbol)
size
- - Size of object associated with symbolsymbol
- - Device symbol address::cudaGetSymbolAddress(void**, const void*) "cudaGetSymbolAddress (C API)",
\ref ::cudaGetSymbolSize(size_t*, const T&) "cudaGetSymbolSize (C++ API)",
::cuModuleGetGlobal
@Cast(value="cudaError_t") public static int cudaMemPrefetchAsync(@Const Pointer devPtr, @Cast(value="size_t") long count, int dstDevice, CUstream_st stream)
devPtr
- - Pointer to be prefetchedcount
- - Size in bytesdstDevice
- - Destination device to prefetch tostream
- - Stream to enqueue prefetch operation::cudaMemcpy, ::cudaMemcpyPeer, ::cudaMemcpyAsync,
::cudaMemcpy3DPeerAsync, ::cudaMemAdvise, ::cudaMemAdvise_v2
::cuMemPrefetchAsync
@Cast(value="cudaError_t") public static int cudaMemPrefetchAsync(@Const Pointer devPtr, @Cast(value="size_t") long count, int dstDevice)
@Cast(value="cudaError_t") public static int cudaMemPrefetchAsync_v2(@Const Pointer devPtr, @Cast(value="size_t") long count, @ByVal cudaMemLocation location, @Cast(value="unsigned int") int flags, CUstream_st stream)
@Cast(value="cudaError_t") public static int cudaMemPrefetchAsync_v2(@Const Pointer devPtr, @Cast(value="size_t") long count, @ByVal cudaMemLocation location, @Cast(value="unsigned int") int flags)
@Cast(value="cudaError_t") public static int cudaMemAdvise(@Const Pointer devPtr, @Cast(value="size_t") long count, @Cast(value="cudaMemoryAdvise") int advice, int device)
devPtr
- - Pointer to memory to set the advice forcount
- - Size in bytes of the memory rangeadvice
- - Advice to be applied for the specified memory rangedevice
- - Device to apply the advice for::cudaMemcpy, ::cudaMemcpyPeer, ::cudaMemcpyAsync,
::cudaMemcpy3DPeerAsync, ::cudaMemPrefetchAsync,
::cuMemAdvise
@Cast(value="cudaError_t") public static int cudaMemAdvise_v2(@Const Pointer devPtr, @Cast(value="size_t") long count, @Cast(value="cudaMemoryAdvise") int advice, @ByVal cudaMemLocation location)
devPtr
- - Pointer to memory to set the advice forcount
- - Size in bytes of the memory rangeadvice
- - Advice to be applied for the specified memory rangelocation
- - location to apply the advice for::cudaMemcpy, ::cudaMemcpyPeer, ::cudaMemcpyAsync,
::cudaMemcpy3DPeerAsync, ::cudaMemPrefetchAsync,
::cuMemAdvise, ::cuMemAdvise_v2
@Cast(value="cudaError_t") public static int cudaMemRangeGetAttribute(Pointer data, @Cast(value="size_t") long dataSize, @Cast(value="cudaMemRangeAttribute") int attribute, @Const Pointer devPtr, @Cast(value="size_t") long count)
data
- - A pointers to a memory location where the result
of each attribute query will be written to.dataSize
- - Array containing the size of dataattribute
- - The attribute to querydevPtr
- - Start of the range to querycount
- - Size of the range to query::cudaMemRangeGetAttributes, ::cudaMemPrefetchAsync,
::cudaMemAdvise,
::cuMemRangeGetAttribute
@Cast(value="cudaError_t") public static int cudaMemRangeGetAttributes(@Cast(value="void**") PointerPointer data, @Cast(value="size_t*") SizeTPointer dataSizes, @Cast(value="cudaMemRangeAttribute*") IntPointer attributes, @Cast(value="size_t") long numAttributes, @Const Pointer devPtr, @Cast(value="size_t") long count)
data
- - A two-dimensional array containing pointers to memory
locations where the result of each attribute query will be written to.dataSizes
- - Array containing the sizes of each resultattributes
- - An array of attributes to query
(numAttributes and the number of attributes in this array should match)numAttributes
- - Number of attributes to querydevPtr
- - Start of the range to querycount
- - Size of the range to query::cudaMemRangeGetAttribute, ::cudaMemAdvise,
::cudaMemPrefetchAsync,
::cuMemRangeGetAttributes
@Cast(value="cudaError_t") public static int cudaMemRangeGetAttributes(@Cast(value="void**") @ByPtrPtr Pointer data, @Cast(value="size_t*") SizeTPointer dataSizes, @Cast(value="cudaMemRangeAttribute*") IntPointer attributes, @Cast(value="size_t") long numAttributes, @Const Pointer devPtr, @Cast(value="size_t") long count)
@Cast(value="cudaError_t") public static int cudaMemRangeGetAttributes(@Cast(value="void**") @ByPtrPtr Pointer data, @Cast(value="size_t*") SizeTPointer dataSizes, @Cast(value="cudaMemRangeAttribute*") IntBuffer attributes, @Cast(value="size_t") long numAttributes, @Const Pointer devPtr, @Cast(value="size_t") long count)
@Cast(value="cudaError_t") public static int cudaMemRangeGetAttributes(@Cast(value="void**") @ByPtrPtr Pointer data, @Cast(value="size_t*") SizeTPointer dataSizes, @Cast(value="cudaMemRangeAttribute*") int[] attributes, @Cast(value="size_t") long numAttributes, @Const Pointer devPtr, @Cast(value="size_t") long count)
@Cast(value="cudaError_t") @Deprecated public static int cudaMemcpyToArray(cudaArray dst, @Cast(value="size_t") long wOffset, @Cast(value="size_t") long hOffset, @Const Pointer src, @Cast(value="size_t") long count, @Cast(value="cudaMemcpyKind") int kind)
dst
- - Destination memory addresswOffset
- - Destination starting X offset (columns in bytes)hOffset
- - Destination starting Y offset (rows)src
- - Source memory addresscount
- - Size in bytes to copykind
- - Type of transfer::cudaMemcpy, ::cudaMemcpy2D,
::cudaMemcpy2DToArray, ::cudaMemcpyFromArray, ::cudaMemcpy2DFromArray,
::cudaMemcpyArrayToArray, ::cudaMemcpy2DArrayToArray, ::cudaMemcpyToSymbol,
::cudaMemcpyFromSymbol, ::cudaMemcpyAsync, ::cudaMemcpy2DAsync,
::cudaMemcpyToArrayAsync, ::cudaMemcpy2DToArrayAsync,
::cudaMemcpyFromArrayAsync, ::cudaMemcpy2DFromArrayAsync,
::cudaMemcpyToSymbolAsync, ::cudaMemcpyFromSymbolAsync,
::cuMemcpyHtoA,
::cuMemcpyDtoA
@Cast(value="cudaError_t") @Deprecated public static int cudaMemcpyFromArray(Pointer dst, cudaArray src, @Cast(value="size_t") long wOffset, @Cast(value="size_t") long hOffset, @Cast(value="size_t") long count, @Cast(value="cudaMemcpyKind") int kind)
dst
- - Destination memory addresssrc
- - Source memory addresswOffset
- - Source starting X offset (columns in bytes)hOffset
- - Source starting Y offset (rows)count
- - Size in bytes to copykind
- - Type of transfer::cudaMemcpy, ::cudaMemcpy2D, ::cudaMemcpyToArray,
::cudaMemcpy2DToArray, ::cudaMemcpy2DFromArray,
::cudaMemcpyArrayToArray, ::cudaMemcpy2DArrayToArray, ::cudaMemcpyToSymbol,
::cudaMemcpyFromSymbol, ::cudaMemcpyAsync, ::cudaMemcpy2DAsync,
::cudaMemcpyToArrayAsync, ::cudaMemcpy2DToArrayAsync,
::cudaMemcpyFromArrayAsync, ::cudaMemcpy2DFromArrayAsync,
::cudaMemcpyToSymbolAsync, ::cudaMemcpyFromSymbolAsync,
::cuMemcpyAtoH,
::cuMemcpyAtoD
@Cast(value="cudaError_t") @Deprecated public static int cudaMemcpyArrayToArray(cudaArray dst, @Cast(value="size_t") long wOffsetDst, @Cast(value="size_t") long hOffsetDst, cudaArray src, @Cast(value="size_t") long wOffsetSrc, @Cast(value="size_t") long hOffsetSrc, @Cast(value="size_t") long count, @Cast(value="cudaMemcpyKind") int kind)
dst
- - Destination memory addresswOffsetDst
- - Destination starting X offset (columns in bytes)hOffsetDst
- - Destination starting Y offset (rows)src
- - Source memory addresswOffsetSrc
- - Source starting X offset (columns in bytes)hOffsetSrc
- - Source starting Y offset (rows)count
- - Size in bytes to copykind
- - Type of transfer::cudaMemcpy, ::cudaMemcpy2D, ::cudaMemcpyToArray,
::cudaMemcpy2DToArray, ::cudaMemcpyFromArray, ::cudaMemcpy2DFromArray,
::cudaMemcpy2DArrayToArray, ::cudaMemcpyToSymbol,
::cudaMemcpyFromSymbol, ::cudaMemcpyAsync, ::cudaMemcpy2DAsync,
::cudaMemcpyToArrayAsync, ::cudaMemcpy2DToArrayAsync,
::cudaMemcpyFromArrayAsync, ::cudaMemcpy2DFromArrayAsync,
::cudaMemcpyToSymbolAsync, ::cudaMemcpyFromSymbolAsync,
::cuMemcpyAtoA
@Cast(value="cudaError_t") @Deprecated public static int cudaMemcpyArrayToArray(cudaArray dst, @Cast(value="size_t") long wOffsetDst, @Cast(value="size_t") long hOffsetDst, cudaArray src, @Cast(value="size_t") long wOffsetSrc, @Cast(value="size_t") long hOffsetSrc, @Cast(value="size_t") long count)
@Cast(value="cudaError_t") @Deprecated public static int cudaMemcpyToArrayAsync(cudaArray dst, @Cast(value="size_t") long wOffset, @Cast(value="size_t") long hOffset, @Const Pointer src, @Cast(value="size_t") long count, @Cast(value="cudaMemcpyKind") int kind, CUstream_st stream)
dst
- - Destination memory addresswOffset
- - Destination starting X offset (columns in bytes)hOffset
- - Destination starting Y offset (rows)src
- - Source memory addresscount
- - Size in bytes to copykind
- - Type of transferstream
- - Stream identifier::cudaMemcpy, ::cudaMemcpy2D, ::cudaMemcpyToArray,
::cudaMemcpy2DToArray, ::cudaMemcpyFromArray, ::cudaMemcpy2DFromArray,
::cudaMemcpyArrayToArray, ::cudaMemcpy2DArrayToArray, ::cudaMemcpyToSymbol,
::cudaMemcpyFromSymbol, ::cudaMemcpyAsync, ::cudaMemcpy2DAsync,
::cudaMemcpy2DToArrayAsync,
::cudaMemcpyFromArrayAsync, ::cudaMemcpy2DFromArrayAsync,
::cudaMemcpyToSymbolAsync, ::cudaMemcpyFromSymbolAsync,
::cuMemcpyHtoAAsync,
::cuMemcpy2DAsync
@Cast(value="cudaError_t") @Deprecated public static int cudaMemcpyToArrayAsync(cudaArray dst, @Cast(value="size_t") long wOffset, @Cast(value="size_t") long hOffset, @Const Pointer src, @Cast(value="size_t") long count, @Cast(value="cudaMemcpyKind") int kind)
@Cast(value="cudaError_t") @Deprecated public static int cudaMemcpyFromArrayAsync(Pointer dst, cudaArray src, @Cast(value="size_t") long wOffset, @Cast(value="size_t") long hOffset, @Cast(value="size_t") long count, @Cast(value="cudaMemcpyKind") int kind, CUstream_st stream)
dst
- - Destination memory addresssrc
- - Source memory addresswOffset
- - Source starting X offset (columns in bytes)hOffset
- - Source starting Y offset (rows)count
- - Size in bytes to copykind
- - Type of transferstream
- - Stream identifier::cudaMemcpy, ::cudaMemcpy2D, ::cudaMemcpyToArray,
::cudaMemcpy2DToArray, ::cudaMemcpyFromArray, ::cudaMemcpy2DFromArray,
::cudaMemcpyArrayToArray, ::cudaMemcpy2DArrayToArray, ::cudaMemcpyToSymbol,
::cudaMemcpyFromSymbol, ::cudaMemcpyAsync, ::cudaMemcpy2DAsync,
::cudaMemcpyToArrayAsync, ::cudaMemcpy2DToArrayAsync,
::cudaMemcpy2DFromArrayAsync,
::cudaMemcpyToSymbolAsync, ::cudaMemcpyFromSymbolAsync,
::cuMemcpyAtoHAsync,
::cuMemcpy2DAsync
@Cast(value="cudaError_t") @Deprecated public static int cudaMemcpyFromArrayAsync(Pointer dst, cudaArray src, @Cast(value="size_t") long wOffset, @Cast(value="size_t") long hOffset, @Cast(value="size_t") long count, @Cast(value="cudaMemcpyKind") int kind)
@Cast(value="cudaError_t") public static int cudaMallocAsync(@Cast(value="void**") PointerPointer devPtr, @Cast(value="size_t") long size, CUstream_st hStream)
devPtr
- [out] - Returned device pointersize
- [in] - Number of bytes to allocatehStream
- [in] - The stream establishing the stream ordering contract and the memory pool to allocate from::cuMemAllocAsync,
\ref ::cudaMallocAsync(void** ptr, size_t size, cudaMemPool_t memPool, cudaStream_t stream) "cudaMallocAsync (C++ API)",
::cudaMallocFromPoolAsync, ::cudaFreeAsync, ::cudaDeviceSetMemPool, ::cudaDeviceGetDefaultMemPool, ::cudaDeviceGetMemPool, ::cudaMemPoolSetAccess, ::cudaMemPoolSetAttribute, ::cudaMemPoolGetAttribute
@Cast(value="cudaError_t") public static int cudaMallocAsync(@Cast(value="void**") @ByPtrPtr Pointer devPtr, @Cast(value="size_t") long size, CUstream_st hStream)
@Cast(value="cudaError_t") public static int cudaFreeAsync(Pointer devPtr, CUstream_st hStream)
dptr
- - memory to freehStream
- - The stream establishing the stream ordering promise::cuMemFreeAsync, ::cudaMallocAsync
@Cast(value="cudaError_t") public static int cudaMemPoolTrimTo(CUmemPoolHandle_st memPool, @Cast(value="size_t") long minBytesToKeep)
pool
- [in] - The memory pool to trimminBytesToKeep
- [in] - If the pool has less than minBytesToKeep reserved,
the TrimTo operation is a no-op. Otherwise the pool will be guaranteed to have
at least minBytesToKeep bytes reserved after the operation.::cuMemPoolTrimTo, ::cudaMallocAsync, ::cudaFreeAsync, ::cudaDeviceGetDefaultMemPool, ::cudaDeviceGetMemPool, ::cudaMemPoolCreate
@Cast(value="cudaError_t") public static int cudaMemPoolSetAttribute(CUmemPoolHandle_st memPool, @Cast(value="cudaMemPoolAttr") int attr, Pointer value)
pool
- [in] - The memory pool to modifyattr
- [in] - The attribute to modifyvalue
- [in] - Pointer to the value to assign::cuMemPoolSetAttribute, ::cudaMallocAsync, ::cudaFreeAsync, ::cudaDeviceGetDefaultMemPool, ::cudaDeviceGetMemPool, ::cudaMemPoolCreate
@Cast(value="cudaError_t") public static int cudaMemPoolGetAttribute(CUmemPoolHandle_st memPool, @Cast(value="cudaMemPoolAttr") int attr, Pointer value)
pool
- [in] - The memory pool to get attributes ofattr
- [in] - The attribute to getvalue
- [in] - Retrieved value::cuMemPoolGetAttribute, ::cudaMallocAsync, ::cudaFreeAsync, ::cudaDeviceGetDefaultMemPool, ::cudaDeviceGetMemPool, ::cudaMemPoolCreate
@Cast(value="cudaError_t") public static int cudaMemPoolSetAccess(CUmemPoolHandle_st memPool, @Const cudaMemAccessDesc descList, @Cast(value="size_t") long count)
pool
- [in] - The pool being modifiedmap
- [in] - Array of access descriptors. Each descriptor instructs the access to enable for a single gpucount
- [in] - Number of descriptors in the map array.::cuMemPoolSetAccess, ::cudaMemPoolGetAccess, ::cudaMallocAsync, cudaFreeAsync
@Cast(value="cudaError_t") public static int cudaMemPoolGetAccess(@Cast(value="cudaMemAccessFlags*") IntPointer flags, CUmemPoolHandle_st memPool, cudaMemLocation location)
flags
- [out] - the accessibility of the pool from the specified locationmemPool
- [in] - the pool being queriedlocation
- [in] - the location accessing the pool::cuMemPoolGetAccess, ::cudaMemPoolSetAccess
@Cast(value="cudaError_t") public static int cudaMemPoolGetAccess(@Cast(value="cudaMemAccessFlags*") IntBuffer flags, CUmemPoolHandle_st memPool, cudaMemLocation location)
@Cast(value="cudaError_t") public static int cudaMemPoolGetAccess(@Cast(value="cudaMemAccessFlags*") int[] flags, CUmemPoolHandle_st memPool, cudaMemLocation location)
@Cast(value="cudaError_t") public static int cudaMemPoolCreate(@ByPtrPtr CUmemPoolHandle_st memPool, @Const cudaMemPoolProps poolProps)
::cuMemPoolCreate, ::cudaDeviceSetMemPool, ::cudaMallocFromPoolAsync, ::cudaMemPoolExportToShareableHandle, ::cudaDeviceGetDefaultMemPool, ::cudaDeviceGetMemPool
@Cast(value="cudaError_t") public static int cudaMemPoolDestroy(CUmemPoolHandle_st memPool)
cuMemPoolDestroy, ::cudaFreeAsync, ::cudaDeviceSetMemPool, ::cudaDeviceGetDefaultMemPool, ::cudaDeviceGetMemPool, ::cudaMemPoolCreate
@Cast(value="cudaError_t") public static int cudaMallocFromPoolAsync(@Cast(value="void**") PointerPointer ptr, @Cast(value="size_t") long size, CUmemPoolHandle_st memPool, CUstream_st stream)
ptr
- [out] - Returned device pointerbytesize
- [in] - Number of bytes to allocatememPool
- [in] - The pool to allocate fromstream
- [in] - The stream establishing the stream ordering semantic::cuMemAllocFromPoolAsync,
\ref ::cudaMallocAsync(void** ptr, size_t size, cudaMemPool_t memPool, cudaStream_t stream) "cudaMallocAsync (C++ API)",
::cudaMallocAsync, ::cudaFreeAsync, ::cudaDeviceGetDefaultMemPool, ::cudaMemPoolCreate, ::cudaMemPoolSetAccess, ::cudaMemPoolSetAttribute
@Cast(value="cudaError_t") public static int cudaMallocFromPoolAsync(@Cast(value="void**") @ByPtrPtr Pointer ptr, @Cast(value="size_t") long size, CUmemPoolHandle_st memPool, CUstream_st stream)
@Cast(value="cudaError_t") public static int cudaMemPoolExportToShareableHandle(Pointer shareableHandle, CUmemPoolHandle_st memPool, @Cast(value="cudaMemAllocationHandleType") int handleType, @Cast(value="unsigned int") int flags)
handle_out
- [out] - pointer to the location in which to store the requested handlepool
- [in] - pool to exporthandleType
- [in] - the type of handle to createflags
- [in] - must be 0::cuMemPoolExportToShareableHandle, ::cudaMemPoolImportFromShareableHandle, ::cudaMemPoolExportPointer, ::cudaMemPoolImportPointer
@Cast(value="cudaError_t") public static int cudaMemPoolImportFromShareableHandle(@ByPtrPtr CUmemPoolHandle_st memPool, Pointer shareableHandle, @Cast(value="cudaMemAllocationHandleType") int handleType, @Cast(value="unsigned int") int flags)
pool_out
- [out] - Returned memory poolhandle
- [in] - OS handle of the pool to openhandleType
- [in] - The type of handle being importedflags
- [in] - must be 0::cuMemPoolImportFromShareableHandle, ::cudaMemPoolExportToShareableHandle, ::cudaMemPoolExportPointer, ::cudaMemPoolImportPointer
@Cast(value="cudaError_t") public static int cudaMemPoolExportPointer(cudaMemPoolPtrExportData exportData, Pointer ptr)
shareData_out
- [out] - Returned export dataptr
- [in] - pointer to memory being exported::cuMemPoolExportPointer, ::cudaMemPoolExportToShareableHandle, ::cudaMemPoolImportFromShareableHandle, ::cudaMemPoolImportPointer
@Cast(value="cudaError_t") public static int cudaMemPoolImportPointer(@Cast(value="void**") PointerPointer ptr, CUmemPoolHandle_st memPool, cudaMemPoolPtrExportData exportData)
ptr_out
- [out] - pointer to imported memorypool
- [in] - pool from which to importshareData
- [in] - data specifying the memory to import::cuMemPoolImportPointer, ::cudaMemPoolExportToShareableHandle, ::cudaMemPoolImportFromShareableHandle, ::cudaMemPoolExportPointer
@Cast(value="cudaError_t") public static int cudaMemPoolImportPointer(@Cast(value="void**") @ByPtrPtr Pointer ptr, CUmemPoolHandle_st memPool, cudaMemPoolPtrExportData exportData)
@Cast(value="cudaError_t") public static int cudaPointerGetAttributes(cudaPointerAttributes attributes, @Const Pointer ptr)
struct cudaPointerAttributes {
enum cudaMemoryType type;
int device;
void *devicePointer;
void *hostPointer;
}
In this structure, the individual fields mean
- \ref ::cudaPointerAttributes::type identifies type of memory. It can be
::cudaMemoryTypeUnregistered for unregistered host memory,
::cudaMemoryTypeHost for registered host memory, ::cudaMemoryTypeDevice for device
memory or ::cudaMemoryTypeManaged for managed memory.
- \ref ::cudaPointerAttributes::device "device" is the device against which
\p ptr was allocated. If \p ptr has memory type ::cudaMemoryTypeDevice
then this identifies the device on which the memory referred to by \p ptr
physically resides. If \p ptr has memory type ::cudaMemoryTypeHost then this
identifies the device which was current when the allocation was made
(and if that device is deinitialized then this allocation will vanish
with that device's state).
- \ref ::cudaPointerAttributes::devicePointer "devicePointer" is
the device pointer alias through which the memory referred to by \p ptr
may be accessed on the current device.
If the memory referred to by \p ptr cannot be accessed directly by the
current device then this is NULL.
- \ref ::cudaPointerAttributes::hostPointer "hostPointer" is
the host pointer alias through which the memory referred to by \p ptr
may be accessed on the host.
If the memory referred to by \p ptr cannot be accessed directly by the
host then this is NULL.attributes
- - Attributes for the specified pointerptr
- - Pointer to get attributes for::cudaGetDeviceCount, ::cudaGetDevice, ::cudaSetDevice,
::cudaChooseDevice,
::cudaInitDevice,
::cuPointerGetAttributes
@Cast(value="cudaError_t") public static int cudaDeviceCanAccessPeer(IntPointer canAccessPeer, int device, int peerDevice)
canAccessPeer
- - Returned access capabilitydevice
- - Device from which allocations on \p peerDevice are to
be directly accessed.peerDevice
- - Device on which the allocations to be directly accessed
by \p device reside.::cudaDeviceEnablePeerAccess,
::cudaDeviceDisablePeerAccess,
::cuDeviceCanAccessPeer
@Cast(value="cudaError_t") public static int cudaDeviceCanAccessPeer(IntBuffer canAccessPeer, int device, int peerDevice)
@Cast(value="cudaError_t") public static int cudaDeviceCanAccessPeer(int[] canAccessPeer, int device, int peerDevice)
@Cast(value="cudaError_t") public static int cudaDeviceEnablePeerAccess(int peerDevice, @Cast(value="unsigned int") int flags)
peerDevice
- - Peer device to enable direct access to from the current deviceflags
- - Reserved for future use and must be set to 0::cudaDeviceCanAccessPeer,
::cudaDeviceDisablePeerAccess,
::cuCtxEnablePeerAccess
@Cast(value="cudaError_t") public static int cudaDeviceDisablePeerAccess(int peerDevice)
peerDevice
- - Peer device to disable direct access to::cudaDeviceCanAccessPeer,
::cudaDeviceEnablePeerAccess,
::cuCtxDisablePeerAccess
@Cast(value="cudaError_t") public static int cudaGraphicsUnregisterResource(cudaGraphicsResource resource)
resource
- - Resource to unregister::cudaGraphicsD3D9RegisterResource,
::cudaGraphicsD3D10RegisterResource,
::cudaGraphicsD3D11RegisterResource,
::cudaGraphicsGLRegisterBuffer,
::cudaGraphicsGLRegisterImage,
::cuGraphicsUnregisterResource
@Cast(value="cudaError_t") public static int cudaGraphicsResourceSetMapFlags(cudaGraphicsResource resource, @Cast(value="unsigned int") int flags)
resource
- - Registered resource to set flags forflags
- - Parameters for resource mapping::cudaGraphicsMapResources,
::cuGraphicsResourceSetMapFlags
@Cast(value="cudaError_t") public static int cudaGraphicsMapResources(int count, @ByPtrPtr cudaGraphicsResource resources, CUstream_st stream)
count
- - Number of resources to mapresources
- - Resources to map for CUDAstream
- - Stream for synchronization::cudaGraphicsResourceGetMappedPointer,
::cudaGraphicsSubResourceGetMappedArray,
::cudaGraphicsUnmapResources,
::cuGraphicsMapResources
@Cast(value="cudaError_t") public static int cudaGraphicsMapResources(int count, @ByPtrPtr cudaGraphicsResource resources)
@Cast(value="cudaError_t") public static int cudaGraphicsUnmapResources(int count, @ByPtrPtr cudaGraphicsResource resources, CUstream_st stream)
count
- - Number of resources to unmapresources
- - Resources to unmapstream
- - Stream for synchronization::cudaGraphicsMapResources,
::cuGraphicsUnmapResources
@Cast(value="cudaError_t") public static int cudaGraphicsUnmapResources(int count, @ByPtrPtr cudaGraphicsResource resources)
@Cast(value="cudaError_t") public static int cudaGraphicsResourceGetMappedPointer(@Cast(value="void**") PointerPointer devPtr, @Cast(value="size_t*") SizeTPointer size, cudaGraphicsResource resource)
devPtr
- - Returned pointer through which \p resource may be accessedsize
- - Returned size of the buffer accessible starting at \p *devPtrresource
- - Mapped resource to access::cudaGraphicsMapResources,
::cudaGraphicsSubResourceGetMappedArray,
::cuGraphicsResourceGetMappedPointer
@Cast(value="cudaError_t") public static int cudaGraphicsResourceGetMappedPointer(@Cast(value="void**") @ByPtrPtr Pointer devPtr, @Cast(value="size_t*") SizeTPointer size, cudaGraphicsResource resource)
@Cast(value="cudaError_t") public static int cudaGraphicsSubResourceGetMappedArray(@ByPtrPtr cudaArray array, cudaGraphicsResource resource, @Cast(value="unsigned int") int arrayIndex, @Cast(value="unsigned int") int mipLevel)
array
- - Returned array through which a subresource of \p resource may be accessedresource
- - Mapped resource to accessarrayIndex
- - Array index for array textures or cubemap face
index as defined by ::cudaGraphicsCubeFace for
cubemap textures for the subresource to accessmipLevel
- - Mipmap level for the subresource to access::cudaGraphicsResourceGetMappedPointer,
::cuGraphicsSubResourceGetMappedArray
@Cast(value="cudaError_t") public static int cudaGraphicsResourceGetMappedMipmappedArray(@ByPtrPtr cudaMipmappedArray mipmappedArray, cudaGraphicsResource resource)
mipmappedArray
- - Returned mipmapped array through which \p resource may be accessedresource
- - Mapped resource to access::cudaGraphicsResourceGetMappedPointer,
::cuGraphicsResourceGetMappedMipmappedArray
@Cast(value="cudaError_t") public static int cudaGetChannelDesc(cudaChannelFormatDesc desc, cudaArray array)
desc
- - Channel formatarray
- - Memory array on device::cudaCreateChannelDesc(int, int, int, int, cudaChannelFormatKind) "cudaCreateChannelDesc (C API)",
::cudaCreateTextureObject, ::cudaCreateSurfaceObject
@ByVal public static cudaChannelFormatDesc cudaCreateChannelDesc(int x, int y, int z, int w, @Cast(value="cudaChannelFormatKind") int f)
struct cudaChannelFormatDesc {
int x, y, z, w;
enum cudaChannelFormatKind f;
};
where ::cudaChannelFormatKind is one of ::cudaChannelFormatKindSigned,
::cudaChannelFormatKindUnsigned, or ::cudaChannelFormatKindFloat.x
- - X componenty
- - Y componentz
- - Z componentw
- - W componentf
- - Channel format::cudaCreateChannelDesc(void) "cudaCreateChannelDesc (C++ API)",
::cudaGetChannelDesc, ::cudaCreateTextureObject, ::cudaCreateSurfaceObject
@Cast(value="cudaError_t") public static int cudaCreateTextureObject(@Cast(value="cudaTextureObject_t*") LongPointer pTexObject, @Const cudaResourceDesc pResDesc, @Const cudaTextureDesc pTexDesc, @Const cudaResourceViewDesc pResViewDesc)
struct cudaResourceDesc {
enum cudaResourceType resType;
union {
struct {
cudaArray_t array;
} array;
struct {
cudaMipmappedArray_t mipmap;
} mipmap;
struct {
void *devPtr;
struct cudaChannelFormatDesc desc;
size_t sizeInBytes;
} linear;
struct {
void *devPtr;
struct cudaChannelFormatDesc desc;
size_t width;
size_t height;
size_t pitchInBytes;
} pitch2D;
} res;
};
where:
- ::cudaResourceDesc::resType specifies the type of resource to texture from.
CUresourceType is defined as:
enum cudaResourceType {
cudaResourceTypeArray = 0x00,
cudaResourceTypeMipmappedArray = 0x01,
cudaResourceTypeLinear = 0x02,
cudaResourceTypePitch2D = 0x03
};
\par
If ::cudaResourceDesc::resType is set to ::cudaResourceTypeArray, ::cudaResourceDesc::res::array::array
must be set to a valid CUDA array handle.
\par
If ::cudaResourceDesc::resType is set to ::cudaResourceTypeMipmappedArray, ::cudaResourceDesc::res::mipmap::mipmap
must be set to a valid CUDA mipmapped array handle and ::cudaTextureDesc::normalizedCoords must be set to true.
\par
If ::cudaResourceDesc::resType is set to ::cudaResourceTypeLinear, ::cudaResourceDesc::res::linear::devPtr
must be set to a valid device pointer, that is aligned to ::cudaDeviceProp::textureAlignment.
::cudaResourceDesc::res::linear::desc describes the format and the number of components per array element. ::cudaResourceDesc::res::linear::sizeInBytes
specifies the size of the array in bytes. The total number of elements in the linear address range cannot exceed
::cudaDeviceProp::maxTexture1DLinear. The number of elements is computed as (sizeInBytes / sizeof(desc)).
\par
If ::cudaResourceDesc::resType is set to ::cudaResourceTypePitch2D, ::cudaResourceDesc::res::pitch2D::devPtr
must be set to a valid device pointer, that is aligned to ::cudaDeviceProp::textureAlignment.
::cudaResourceDesc::res::pitch2D::desc describes the format and the number of components per array element. ::cudaResourceDesc::res::pitch2D::width
and ::cudaResourceDesc::res::pitch2D::height specify the width and height of the array in elements, and cannot exceed
::cudaDeviceProp::maxTexture2DLinear[0] and ::cudaDeviceProp::maxTexture2DLinear[1] respectively.
::cudaResourceDesc::res::pitch2D::pitchInBytes specifies the pitch between two rows in bytes and has to be aligned to
::cudaDeviceProp::texturePitchAlignment. Pitch cannot exceed ::cudaDeviceProp::maxTexture2DLinear[2].
The ::cudaTextureDesc struct is defined as
struct cudaTextureDesc {
enum cudaTextureAddressMode addressMode[3];
enum cudaTextureFilterMode filterMode;
enum cudaTextureReadMode readMode;
int sRGB;
float borderColor[4];
int normalizedCoords;
unsigned int maxAnisotropy;
enum cudaTextureFilterMode mipmapFilterMode;
float mipmapLevelBias;
float minMipmapLevelClamp;
float maxMipmapLevelClamp;
int disableTrilinearOptimization;
int seamlessCubemap;
};
where
- ::cudaTextureDesc::addressMode specifies the addressing mode for each dimension of the texture data. ::cudaTextureAddressMode is defined as:
enum cudaTextureAddressMode {
cudaAddressModeWrap = 0,
cudaAddressModeClamp = 1,
cudaAddressModeMirror = 2,
cudaAddressModeBorder = 3
};
This is ignored if ::cudaResourceDesc::resType is ::cudaResourceTypeLinear. Also, if ::cudaTextureDesc::normalizedCoords
is set to zero, ::cudaAddressModeWrap and ::cudaAddressModeMirror won't be supported and will be switched to ::cudaAddressModeClamp.
- ::cudaTextureDesc::filterMode specifies the filtering mode to be used when fetching from the texture. ::cudaTextureFilterMode is defined as:
enum cudaTextureFilterMode {
cudaFilterModePoint = 0,
cudaFilterModeLinear = 1
};
This is ignored if ::cudaResourceDesc::resType is ::cudaResourceTypeLinear.
- ::cudaTextureDesc::readMode specifies whether integer data should be converted to floating point or not. ::cudaTextureReadMode is defined as:
enum cudaTextureReadMode {
cudaReadModeElementType = 0,
cudaReadModeNormalizedFloat = 1
};
Note that this applies only to 8-bit and 16-bit integer formats. 32-bit integer format would not be promoted, regardless of
whether or not this ::cudaTextureDesc::readMode is set ::cudaReadModeNormalizedFloat is specified.
- ::cudaTextureDesc::sRGB specifies whether sRGB to linear conversion should be performed during texture fetch.
- ::cudaTextureDesc::borderColor specifies the float values of color. where:
::cudaTextureDesc::borderColor[0] contains value of 'R',
::cudaTextureDesc::borderColor[1] contains value of 'G',
::cudaTextureDesc::borderColor[2] contains value of 'B',
::cudaTextureDesc::borderColor[3] contains value of 'A'
Note that application using integer border color values will need to
struct cudaResourceViewDesc {
enum cudaResourceViewFormat format;
size_t width;
size_t height;
size_t depth;
unsigned int firstMipmapLevel;
unsigned int lastMipmapLevel;
unsigned int firstLayer;
unsigned int lastLayer;
};
where:
- ::cudaResourceViewDesc::format specifies how the data contained in the CUDA array or CUDA mipmapped array should
be interpreted. Note that this can incur a change in size of the texture data. If the resource view format is a block
compressed format, then the underlying CUDA array or CUDA mipmapped array has to have a 32-bit unsigned integer format
with 2 or 4 channels, depending on the block compressed format. For ex., BC1 and BC4 require the underlying CUDA array to have
a 32-bit unsigned int with 2 channels. The other BC formats require the underlying resource to have the same 32-bit unsigned int
format but with 4 channels.
- ::cudaResourceViewDesc::width specifies the new width of the texture data. If the resource view format is a block
compressed format, this value has to be 4 times the original width of the resource. For non block compressed formats,
this value has to be equal to that of the original resource.
- ::cudaResourceViewDesc::height specifies the new height of the texture data. If the resource view format is a block
compressed format, this value has to be 4 times the original height of the resource. For non block compressed formats,
this value has to be equal to that of the original resource.
- ::cudaResourceViewDesc::depth specifies the new depth of the texture data. This value has to be equal to that of the
original resource.
- ::cudaResourceViewDesc::firstMipmapLevel specifies the most detailed mipmap level. This will be the new mipmap level zero.
For non-mipmapped resources, this value has to be zero.::cudaTextureDesc::minMipmapLevelClamp and ::cudaTextureDesc::maxMipmapLevelClamp
will be relative to this value. For ex., if the firstMipmapLevel is set to 2, and a minMipmapLevelClamp of 1.2 is specified,
then the actual minimum mipmap level clamp will be 3.2.
- ::cudaResourceViewDesc::lastMipmapLevel specifies the least detailed mipmap level. For non-mipmapped resources, this value
has to be zero.
- ::cudaResourceViewDesc::firstLayer specifies the first layer index for layered textures. This will be the new layer zero.
For non-layered resources, this value has to be zero.
- ::cudaResourceViewDesc::lastLayer specifies the last layer index for layered textures. For non-layered resources,
this value has to be zero.pTexObject
- - Texture object to createpResDesc
- - Resource descriptorpTexDesc
- - Texture descriptorpResViewDesc
- - Resource view descriptor::cudaDestroyTextureObject,
::cuTexObjectCreate
@Cast(value="cudaError_t") public static int cudaCreateTextureObject(@Cast(value="cudaTextureObject_t*") LongBuffer pTexObject, @Const cudaResourceDesc pResDesc, @Const cudaTextureDesc pTexDesc, @Const cudaResourceViewDesc pResViewDesc)
@Cast(value="cudaError_t") public static int cudaCreateTextureObject(@Cast(value="cudaTextureObject_t*") long[] pTexObject, @Const cudaResourceDesc pResDesc, @Const cudaTextureDesc pTexDesc, @Const cudaResourceViewDesc pResViewDesc)
@Cast(value="cudaError_t") public static int cudaDestroyTextureObject(@Cast(value="cudaTextureObject_t") long texObject)
texObject
- - Texture object to destroy::cudaCreateTextureObject,
::cuTexObjectDestroy
@Cast(value="cudaError_t") public static int cudaGetTextureObjectResourceDesc(cudaResourceDesc pResDesc, @Cast(value="cudaTextureObject_t") long texObject)
pResDesc
- - Resource descriptortexObject
- - Texture object::cudaCreateTextureObject,
::cuTexObjectGetResourceDesc
@Cast(value="cudaError_t") public static int cudaGetTextureObjectTextureDesc(cudaTextureDesc pTexDesc, @Cast(value="cudaTextureObject_t") long texObject)
pTexDesc
- - Texture descriptortexObject
- - Texture object::cudaCreateTextureObject,
::cuTexObjectGetTextureDesc
@Cast(value="cudaError_t") public static int cudaGetTextureObjectResourceViewDesc(cudaResourceViewDesc pResViewDesc, @Cast(value="cudaTextureObject_t") long texObject)
pResViewDesc
- - Resource view descriptortexObject
- - Texture object::cudaCreateTextureObject,
::cuTexObjectGetResourceViewDesc
@Cast(value="cudaError_t") public static int cudaCreateSurfaceObject(@Cast(value="cudaSurfaceObject_t*") LongPointer pSurfObject, @Const cudaResourceDesc pResDesc)
pSurfObject
- - Surface object to createpResDesc
- - Resource descriptor::cudaDestroySurfaceObject,
::cuSurfObjectCreate
@Cast(value="cudaError_t") public static int cudaCreateSurfaceObject(@Cast(value="cudaSurfaceObject_t*") LongBuffer pSurfObject, @Const cudaResourceDesc pResDesc)
@Cast(value="cudaError_t") public static int cudaCreateSurfaceObject(@Cast(value="cudaSurfaceObject_t*") long[] pSurfObject, @Const cudaResourceDesc pResDesc)
@Cast(value="cudaError_t") public static int cudaDestroySurfaceObject(@Cast(value="cudaSurfaceObject_t") long surfObject)
surfObject
- - Surface object to destroy::cudaCreateSurfaceObject,
::cuSurfObjectDestroy
@Cast(value="cudaError_t") public static int cudaGetSurfaceObjectResourceDesc(cudaResourceDesc pResDesc, @Cast(value="cudaSurfaceObject_t") long surfObject)
pResDesc
- - Resource descriptorsurfObject
- - Surface object::cudaCreateSurfaceObject,
::cuSurfObjectGetResourceDesc
@Cast(value="cudaError_t") public static int cudaDriverGetVersion(IntPointer driverVersion)
driverVersion
- - Returns the CUDA driver version.::cudaRuntimeGetVersion,
::cuDriverGetVersion
@Cast(value="cudaError_t") public static int cudaDriverGetVersion(IntBuffer driverVersion)
@Cast(value="cudaError_t") public static int cudaDriverGetVersion(int[] driverVersion)
@Cast(value="cudaError_t") public static int cudaRuntimeGetVersion(IntPointer runtimeVersion)
runtimeVersion
- - Returns the CUDA Runtime version.::cudaDriverGetVersion,
::cuDriverGetVersion
@Cast(value="cudaError_t") public static int cudaRuntimeGetVersion(IntBuffer runtimeVersion)
@Cast(value="cudaError_t") public static int cudaRuntimeGetVersion(int[] runtimeVersion)
@Cast(value="cudaError_t") public static int cudaGraphCreate(@ByPtrPtr CUgraph_st pGraph, @Cast(value="unsigned int") int flags)
pGraph
- - Returns newly created graphflags
- - Graph creation flags, must be 0::cudaGraphAddChildGraphNode,
::cudaGraphAddEmptyNode,
::cudaGraphAddKernelNode,
::cudaGraphAddHostNode,
::cudaGraphAddMemcpyNode,
::cudaGraphAddMemsetNode,
::cudaGraphInstantiate,
::cudaGraphDestroy,
::cudaGraphGetNodes,
::cudaGraphGetRootNodes,
::cudaGraphGetEdges,
::cudaGraphClone
@Cast(value="cudaError_t") public static int cudaGraphAddKernelNode(@ByPtrPtr CUgraphNode_st pGraphNode, CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st pDependencies, @Cast(value="size_t") long numDependencies, @Const cudaKernelNodeParams pNodeParams)
struct cudaKernelNodeParams
{
void* func;
dim3 gridDim;
dim3 blockDim;
unsigned int sharedMemBytes;
void **kernelParams;
void **extra;
};
When the graph is launched, the node will invoke kernel \p func on a (\p gridDim.x x
\p gridDim.y x \p gridDim.z) grid of blocks. Each block contains
(\p blockDim.x x \p blockDim.y x \p blockDim.z) threads.
\p sharedMem sets the amount of dynamic shared memory that will be
available to each thread block.
Kernel parameters to \p func can be specified in one of two ways:
1) Kernel parameters can be specified via \p kernelParams. If the kernel has N
parameters, then \p kernelParams needs to be an array of N pointers. Each pointer,
from \p kernelParams[0] to \p kernelParams[N-1], points to the region of memory from which the actual
parameter will be copied. The number of kernel parameters and their offsets and sizes do not need
to be specified as that information is retrieved directly from the kernel's image.
2) Kernel parameters can also be packaged by the application into a single buffer that is passed in
via \p extra. This places the burden on the application of knowing each kernel
parameter's size and alignment/padding within the buffer. The \p extra parameter exists
to allow this function to take additional less commonly used arguments. \p extra specifies
a list of names of extra settings and their corresponding values. Each extra setting name is
immediately followed by the corresponding value. The list must be terminated with either NULL or
CU_LAUNCH_PARAM_END.
- ::CU_LAUNCH_PARAM_END, which indicates the end of the \p extra
array;
- ::CU_LAUNCH_PARAM_BUFFER_POINTER, which specifies that the next
value in \p extra will be a pointer to a buffer
containing all the kernel parameters for launching kernel
\p func;
- ::CU_LAUNCH_PARAM_BUFFER_SIZE, which specifies that the next
value in \p extra will be a pointer to a size_t
containing the size of the buffer specified with
::CU_LAUNCH_PARAM_BUFFER_POINTER;
The error ::cudaErrorInvalidValue will be returned if kernel parameters are specified with both
\p kernelParams and \p extra (i.e. both \p kernelParams and
\p extra are non-NULL).
The \p kernelParams or \p extra array, as well as the argument values it points to,
are copied during this call.
\note Kernels launched using graphs must not use texture and surface references. Reading or
writing through any texture or surface reference is undefined behavior.
This restriction does not apply to texture and surface objects.pGraphNode
- - Returns newly created nodegraph
- - Graph to which to add the nodepDependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciespNodeParams
- - Parameters for the GPU execution node::cudaGraphAddNode,
::cudaLaunchKernel,
::cudaGraphKernelNodeGetParams,
::cudaGraphKernelNodeSetParams,
::cudaGraphCreate,
::cudaGraphDestroyNode,
::cudaGraphAddChildGraphNode,
::cudaGraphAddEmptyNode,
::cudaGraphAddHostNode,
::cudaGraphAddMemcpyNode,
::cudaGraphAddMemsetNode
@Cast(value="cudaError_t") public static int cudaGraphKernelNodeGetParams(CUgraphNode_st node, cudaKernelNodeParams pNodeParams)
node
- - Node to get the parameters forpNodeParams
- - Pointer to return the parameters::cudaLaunchKernel,
::cudaGraphAddKernelNode,
::cudaGraphKernelNodeSetParams
@Cast(value="cudaError_t") public static int cudaGraphKernelNodeSetParams(CUgraphNode_st node, @Const cudaKernelNodeParams pNodeParams)
node
- - Node to set the parameters forpNodeParams
- - Parameters to copy::cudaGraphNodeSetParams,
::cudaLaunchKernel,
::cudaGraphAddKernelNode,
::cudaGraphKernelNodeGetParams
@Cast(value="cudaError_t") public static int cudaGraphKernelNodeCopyAttributes(CUgraphNode_st hSrc, CUgraphNode_st hDst)
dst
- [out] Destination nodesrc
- [in] Source node
For list of attributes see ::cudaKernelNodeAttrID::cudaAccessPolicyWindow
@Cast(value="cudaError_t") public static int cudaGraphKernelNodeGetAttribute(CUgraphNode_st hNode, @Cast(value="cudaLaunchAttributeID") int attr, cudaLaunchAttributeValue value_out)
hNode
- [in]attr
- [in]value_out
- [out]::cudaAccessPolicyWindow
@Cast(value="cudaError_t") public static int cudaGraphKernelNodeSetAttribute(CUgraphNode_st hNode, @Cast(value="cudaLaunchAttributeID") int attr, @Const cudaLaunchAttributeValue value)
hNode
- [out]attr
- [in]value
- [out]::cudaAccessPolicyWindow
@Cast(value="cudaError_t") public static int cudaGraphAddMemcpyNode(@ByPtrPtr CUgraphNode_st pGraphNode, CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st pDependencies, @Cast(value="size_t") long numDependencies, @Const cudaMemcpy3DParms pCopyParams)
pGraphNode
- - Returns newly created nodegraph
- - Graph to which to add the nodepDependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciespCopyParams
- - Parameters for the memory copy::cudaGraphAddNode,
::cudaMemcpy3D,
::cudaGraphAddMemcpyNodeToSymbol,
::cudaGraphAddMemcpyNodeFromSymbol,
::cudaGraphAddMemcpyNode1D,
::cudaGraphMemcpyNodeGetParams,
::cudaGraphMemcpyNodeSetParams,
::cudaGraphCreate,
::cudaGraphDestroyNode,
::cudaGraphAddChildGraphNode,
::cudaGraphAddEmptyNode,
::cudaGraphAddKernelNode,
::cudaGraphAddHostNode,
::cudaGraphAddMemsetNode
@Cast(value="cudaError_t") public static int cudaGraphAddMemcpyNodeToSymbol(@ByPtrPtr CUgraphNode_st pGraphNode, CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st pDependencies, @Cast(value="size_t") long numDependencies, @Const Pointer symbol, @Const Pointer src, @Cast(value="size_t") long count, @Cast(value="size_t") long offset, @Cast(value="cudaMemcpyKind") int kind)
pGraphNode
- - Returns newly created nodegraph
- - Graph to which to add the nodepDependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciessymbol
- - Device symbol addresssrc
- - Source memory addresscount
- - Size in bytes to copyoffset
- - Offset from start of symbol in byteskind
- - Type of transfer::cudaMemcpyToSymbol,
::cudaGraphAddMemcpyNode,
::cudaGraphAddMemcpyNodeFromSymbol,
::cudaGraphMemcpyNodeGetParams,
::cudaGraphMemcpyNodeSetParams,
::cudaGraphMemcpyNodeSetParamsToSymbol,
::cudaGraphMemcpyNodeSetParamsFromSymbol,
::cudaGraphCreate,
::cudaGraphDestroyNode,
::cudaGraphAddChildGraphNode,
::cudaGraphAddEmptyNode,
::cudaGraphAddKernelNode,
::cudaGraphAddHostNode,
::cudaGraphAddMemsetNode
@Cast(value="cudaError_t") public static int cudaGraphAddMemcpyNodeFromSymbol(@ByPtrPtr CUgraphNode_st pGraphNode, CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st pDependencies, @Cast(value="size_t") long numDependencies, Pointer dst, @Const Pointer symbol, @Cast(value="size_t") long count, @Cast(value="size_t") long offset, @Cast(value="cudaMemcpyKind") int kind)
pGraphNode
- - Returns newly created nodegraph
- - Graph to which to add the nodepDependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciesdst
- - Destination memory addresssymbol
- - Device symbol addresscount
- - Size in bytes to copyoffset
- - Offset from start of symbol in byteskind
- - Type of transfer::cudaMemcpyFromSymbol,
::cudaGraphAddMemcpyNode,
::cudaGraphAddMemcpyNodeToSymbol,
::cudaGraphMemcpyNodeGetParams,
::cudaGraphMemcpyNodeSetParams,
::cudaGraphMemcpyNodeSetParamsFromSymbol,
::cudaGraphMemcpyNodeSetParamsToSymbol,
::cudaGraphCreate,
::cudaGraphDestroyNode,
::cudaGraphAddChildGraphNode,
::cudaGraphAddEmptyNode,
::cudaGraphAddKernelNode,
::cudaGraphAddHostNode,
::cudaGraphAddMemsetNode
@Cast(value="cudaError_t") public static int cudaGraphAddMemcpyNode1D(@ByPtrPtr CUgraphNode_st pGraphNode, CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st pDependencies, @Cast(value="size_t") long numDependencies, Pointer dst, @Const Pointer src, @Cast(value="size_t") long count, @Cast(value="cudaMemcpyKind") int kind)
pGraphNode
- - Returns newly created nodegraph
- - Graph to which to add the nodepDependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciesdst
- - Destination memory addresssrc
- - Source memory addresscount
- - Size in bytes to copykind
- - Type of transfer::cudaMemcpy,
::cudaGraphAddMemcpyNode,
::cudaGraphMemcpyNodeGetParams,
::cudaGraphMemcpyNodeSetParams,
::cudaGraphMemcpyNodeSetParams1D,
::cudaGraphCreate,
::cudaGraphDestroyNode,
::cudaGraphAddChildGraphNode,
::cudaGraphAddEmptyNode,
::cudaGraphAddKernelNode,
::cudaGraphAddHostNode,
::cudaGraphAddMemsetNode
@Cast(value="cudaError_t") public static int cudaGraphMemcpyNodeGetParams(CUgraphNode_st node, cudaMemcpy3DParms pNodeParams)
node
- - Node to get the parameters forpNodeParams
- - Pointer to return the parameters::cudaMemcpy3D,
::cudaGraphAddMemcpyNode,
::cudaGraphMemcpyNodeSetParams
@Cast(value="cudaError_t") public static int cudaGraphMemcpyNodeSetParams(CUgraphNode_st node, @Const cudaMemcpy3DParms pNodeParams)
node
- - Node to set the parameters forpNodeParams
- - Parameters to copy::cudaGraphNodeSetParams,
::cudaMemcpy3D,
::cudaGraphMemcpyNodeSetParamsToSymbol,
::cudaGraphMemcpyNodeSetParamsFromSymbol,
::cudaGraphMemcpyNodeSetParams1D,
::cudaGraphAddMemcpyNode,
::cudaGraphMemcpyNodeGetParams
@Cast(value="cudaError_t") public static int cudaGraphMemcpyNodeSetParamsToSymbol(CUgraphNode_st node, @Const Pointer symbol, @Const Pointer src, @Cast(value="size_t") long count, @Cast(value="size_t") long offset, @Cast(value="cudaMemcpyKind") int kind)
node
- - Node to set the parameters forsymbol
- - Device symbol addresssrc
- - Source memory addresscount
- - Size in bytes to copyoffset
- - Offset from start of symbol in byteskind
- - Type of transfer::cudaMemcpyToSymbol,
::cudaGraphMemcpyNodeSetParams,
::cudaGraphMemcpyNodeSetParamsFromSymbol,
::cudaGraphAddMemcpyNode,
::cudaGraphMemcpyNodeGetParams
@Cast(value="cudaError_t") public static int cudaGraphMemcpyNodeSetParamsFromSymbol(CUgraphNode_st node, Pointer dst, @Const Pointer symbol, @Cast(value="size_t") long count, @Cast(value="size_t") long offset, @Cast(value="cudaMemcpyKind") int kind)
node
- - Node to set the parameters fordst
- - Destination memory addresssymbol
- - Device symbol addresscount
- - Size in bytes to copyoffset
- - Offset from start of symbol in byteskind
- - Type of transfer::cudaMemcpyFromSymbol,
::cudaGraphMemcpyNodeSetParams,
::cudaGraphMemcpyNodeSetParamsToSymbol,
::cudaGraphAddMemcpyNode,
::cudaGraphMemcpyNodeGetParams
@Cast(value="cudaError_t") public static int cudaGraphMemcpyNodeSetParams1D(CUgraphNode_st node, Pointer dst, @Const Pointer src, @Cast(value="size_t") long count, @Cast(value="cudaMemcpyKind") int kind)
node
- - Node to set the parameters fordst
- - Destination memory addresssrc
- - Source memory addresscount
- - Size in bytes to copykind
- - Type of transfer::cudaMemcpy,
::cudaGraphMemcpyNodeSetParams,
::cudaGraphAddMemcpyNode,
::cudaGraphMemcpyNodeGetParams
@Cast(value="cudaError_t") public static int cudaGraphAddMemsetNode(@ByPtrPtr CUgraphNode_st pGraphNode, CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st pDependencies, @Cast(value="size_t") long numDependencies, @Const cudaMemsetParams pMemsetParams)
pGraphNode
- - Returns newly created nodegraph
- - Graph to which to add the nodepDependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciespMemsetParams
- - Parameters for the memory set::cudaGraphAddNode,
::cudaMemset2D,
::cudaGraphMemsetNodeGetParams,
::cudaGraphMemsetNodeSetParams,
::cudaGraphCreate,
::cudaGraphDestroyNode,
::cudaGraphAddChildGraphNode,
::cudaGraphAddEmptyNode,
::cudaGraphAddKernelNode,
::cudaGraphAddHostNode,
::cudaGraphAddMemcpyNode
@Cast(value="cudaError_t") public static int cudaGraphMemsetNodeGetParams(CUgraphNode_st node, cudaMemsetParams pNodeParams)
node
- - Node to get the parameters forpNodeParams
- - Pointer to return the parameters::cudaMemset2D,
::cudaGraphAddMemsetNode,
::cudaGraphMemsetNodeSetParams
@Cast(value="cudaError_t") public static int cudaGraphMemsetNodeSetParams(CUgraphNode_st node, @Const cudaMemsetParams pNodeParams)
node
- - Node to set the parameters forpNodeParams
- - Parameters to copy::cudaGraphNodeSetParams,
::cudaMemset2D,
::cudaGraphAddMemsetNode,
::cudaGraphMemsetNodeGetParams
@Cast(value="cudaError_t") public static int cudaGraphAddHostNode(@ByPtrPtr CUgraphNode_st pGraphNode, CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st pDependencies, @Cast(value="size_t") long numDependencies, @Const cudaHostNodeParams pNodeParams)
pGraphNode
- - Returns newly created nodegraph
- - Graph to which to add the nodepDependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciespNodeParams
- - Parameters for the host node::cudaGraphAddNode,
::cudaLaunchHostFunc,
::cudaGraphHostNodeGetParams,
::cudaGraphHostNodeSetParams,
::cudaGraphCreate,
::cudaGraphDestroyNode,
::cudaGraphAddChildGraphNode,
::cudaGraphAddEmptyNode,
::cudaGraphAddKernelNode,
::cudaGraphAddMemcpyNode,
::cudaGraphAddMemsetNode
@Cast(value="cudaError_t") public static int cudaGraphHostNodeGetParams(CUgraphNode_st node, cudaHostNodeParams pNodeParams)
node
- - Node to get the parameters forpNodeParams
- - Pointer to return the parameters::cudaLaunchHostFunc,
::cudaGraphAddHostNode,
::cudaGraphHostNodeSetParams
@Cast(value="cudaError_t") public static int cudaGraphHostNodeSetParams(CUgraphNode_st node, @Const cudaHostNodeParams pNodeParams)
node
- - Node to set the parameters forpNodeParams
- - Parameters to copy::cudaGraphNodeSetParams,
::cudaLaunchHostFunc,
::cudaGraphAddHostNode,
::cudaGraphHostNodeGetParams
@Cast(value="cudaError_t") public static int cudaGraphAddChildGraphNode(@ByPtrPtr CUgraphNode_st pGraphNode, CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st pDependencies, @Cast(value="size_t") long numDependencies, CUgraph_st childGraph)
pGraphNode
- - Returns newly created nodegraph
- - Graph to which to add the nodepDependencies
- - Dependencies of the nodenumDependencies
- - Number of dependencieschildGraph
- - The graph to clone into this node::cudaGraphAddNode,
::cudaGraphChildGraphNodeGetGraph,
::cudaGraphCreate,
::cudaGraphDestroyNode,
::cudaGraphAddEmptyNode,
::cudaGraphAddKernelNode,
::cudaGraphAddHostNode,
::cudaGraphAddMemcpyNode,
::cudaGraphAddMemsetNode,
::cudaGraphClone
@Cast(value="cudaError_t") public static int cudaGraphChildGraphNodeGetGraph(CUgraphNode_st node, @ByPtrPtr CUgraph_st pGraph)
node
- - Node to get the embedded graph forpGraph
- - Location to store a handle to the graph::cudaGraphAddChildGraphNode,
::cudaGraphNodeFindInClone
@Cast(value="cudaError_t") public static int cudaGraphAddEmptyNode(@ByPtrPtr CUgraphNode_st pGraphNode, CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st pDependencies, @Cast(value="size_t") long numDependencies)
pGraphNode
- - Returns newly created nodegraph
- - Graph to which to add the nodepDependencies
- - Dependencies of the nodenumDependencies
- - Number of dependencies::cudaGraphAddNode,
::cudaGraphCreate,
::cudaGraphDestroyNode,
::cudaGraphAddChildGraphNode,
::cudaGraphAddKernelNode,
::cudaGraphAddHostNode,
::cudaGraphAddMemcpyNode,
::cudaGraphAddMemsetNode
@Cast(value="cudaError_t") public static int cudaGraphAddEventRecordNode(@ByPtrPtr CUgraphNode_st pGraphNode, CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st pDependencies, @Cast(value="size_t") long numDependencies, CUevent_st event)
phGraphNode
- - Returns newly created nodehGraph
- - Graph to which to add the nodedependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciesevent
- - Event for the node::cudaGraphAddNode,
::cudaGraphAddEventWaitNode,
::cudaEventRecordWithFlags,
::cudaStreamWaitEvent,
::cudaGraphCreate,
::cudaGraphDestroyNode,
::cudaGraphAddChildGraphNode,
::cudaGraphAddEmptyNode,
::cudaGraphAddKernelNode,
::cudaGraphAddMemcpyNode,
::cudaGraphAddMemsetNode
@Cast(value="cudaError_t") public static int cudaGraphEventRecordNodeGetEvent(CUgraphNode_st node, @ByPtrPtr CUevent_st event_out)
hNode
- - Node to get the event forevent_out
- - Pointer to return the event::cudaGraphAddEventRecordNode,
::cudaGraphEventRecordNodeSetEvent,
::cudaGraphEventWaitNodeGetEvent,
::cudaEventRecordWithFlags,
::cudaStreamWaitEvent
@Cast(value="cudaError_t") public static int cudaGraphEventRecordNodeSetEvent(CUgraphNode_st node, CUevent_st event)
hNode
- - Node to set the event forevent
- - Event to use::cudaGraphNodeSetParams,
::cudaGraphAddEventRecordNode,
::cudaGraphEventRecordNodeGetEvent,
::cudaGraphEventWaitNodeSetEvent,
::cudaEventRecordWithFlags,
::cudaStreamWaitEvent
@Cast(value="cudaError_t") public static int cudaGraphAddEventWaitNode(@ByPtrPtr CUgraphNode_st pGraphNode, CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st pDependencies, @Cast(value="size_t") long numDependencies, CUevent_st event)
phGraphNode
- - Returns newly created nodehGraph
- - Graph to which to add the nodedependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciesevent
- - Event for the node::cudaGraphAddNode,
::cudaGraphAddEventRecordNode,
::cudaEventRecordWithFlags,
::cudaStreamWaitEvent,
::cudaGraphCreate,
::cudaGraphDestroyNode,
::cudaGraphAddChildGraphNode,
::cudaGraphAddEmptyNode,
::cudaGraphAddKernelNode,
::cudaGraphAddMemcpyNode,
::cudaGraphAddMemsetNode
@Cast(value="cudaError_t") public static int cudaGraphEventWaitNodeGetEvent(CUgraphNode_st node, @ByPtrPtr CUevent_st event_out)
hNode
- - Node to get the event forevent_out
- - Pointer to return the event::cudaGraphAddEventWaitNode,
::cudaGraphEventWaitNodeSetEvent,
::cudaGraphEventRecordNodeGetEvent,
::cudaEventRecordWithFlags,
::cudaStreamWaitEvent
@Cast(value="cudaError_t") public static int cudaGraphEventWaitNodeSetEvent(CUgraphNode_st node, CUevent_st event)
hNode
- - Node to set the event forevent
- - Event to use::cudaGraphNodeSetParams,
::cudaGraphAddEventWaitNode,
::cudaGraphEventWaitNodeGetEvent,
::cudaGraphEventRecordNodeSetEvent,
::cudaEventRecordWithFlags,
::cudaStreamWaitEvent
@Cast(value="cudaError_t") public static int cudaGraphAddExternalSemaphoresSignalNode(@ByPtrPtr CUgraphNode_st pGraphNode, CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st pDependencies, @Cast(value="size_t") long numDependencies, @Const cudaExternalSemaphoreSignalNodeParams nodeParams)
pGraphNode
- - Returns newly created nodegraph
- - Graph to which to add the nodepDependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciesnodeParams
- - Parameters for the node::cudaGraphAddNode,
::cudaGraphExternalSemaphoresSignalNodeGetParams,
::cudaGraphExternalSemaphoresSignalNodeSetParams,
::cudaGraphExecExternalSemaphoresSignalNodeSetParams,
::cudaGraphAddExternalSemaphoresWaitNode,
::cudaImportExternalSemaphore,
::cudaSignalExternalSemaphoresAsync,
::cudaWaitExternalSemaphoresAsync,
::cudaGraphCreate,
::cudaGraphDestroyNode,
::cudaGraphAddEventRecordNode,
::cudaGraphAddEventWaitNode,
::cudaGraphAddChildGraphNode,
::cudaGraphAddEmptyNode,
::cudaGraphAddKernelNode,
::cudaGraphAddMemcpyNode,
::cudaGraphAddMemsetNode
@Cast(value="cudaError_t") public static int cudaGraphExternalSemaphoresSignalNodeGetParams(CUgraphNode_st hNode, cudaExternalSemaphoreSignalNodeParams params_out)
hNode
- - Node to get the parameters forparams_out
- - Pointer to return the parameters::cudaLaunchKernel,
::cudaGraphAddExternalSemaphoresSignalNode,
::cudaGraphExternalSemaphoresSignalNodeSetParams,
::cudaGraphAddExternalSemaphoresWaitNode,
::cudaSignalExternalSemaphoresAsync,
::cudaWaitExternalSemaphoresAsync
@Cast(value="cudaError_t") public static int cudaGraphExternalSemaphoresSignalNodeSetParams(CUgraphNode_st hNode, @Const cudaExternalSemaphoreSignalNodeParams nodeParams)
hNode
- - Node to set the parameters fornodeParams
- - Parameters to copy::cudaGraphNodeSetParams,
::cudaGraphAddExternalSemaphoresSignalNode,
::cudaGraphExternalSemaphoresSignalNodeSetParams,
::cudaGraphAddExternalSemaphoresWaitNode,
::cudaSignalExternalSemaphoresAsync,
::cudaWaitExternalSemaphoresAsync
@Cast(value="cudaError_t") public static int cudaGraphAddExternalSemaphoresWaitNode(@ByPtrPtr CUgraphNode_st pGraphNode, CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st pDependencies, @Cast(value="size_t") long numDependencies, @Const cudaExternalSemaphoreWaitNodeParams nodeParams)
pGraphNode
- - Returns newly created nodegraph
- - Graph to which to add the nodepDependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciesnodeParams
- - Parameters for the node::cudaGraphAddNode,
::cudaGraphExternalSemaphoresWaitNodeGetParams,
::cudaGraphExternalSemaphoresWaitNodeSetParams,
::cudaGraphExecExternalSemaphoresWaitNodeSetParams,
::cudaGraphAddExternalSemaphoresSignalNode,
::cudaImportExternalSemaphore,
::cudaSignalExternalSemaphoresAsync,
::cudaWaitExternalSemaphoresAsync,
::cudaGraphCreate,
::cudaGraphDestroyNode,
::cudaGraphAddEventRecordNode,
::cudaGraphAddEventWaitNode,
::cudaGraphAddChildGraphNode,
::cudaGraphAddEmptyNode,
::cudaGraphAddKernelNode,
::cudaGraphAddMemcpyNode,
::cudaGraphAddMemsetNode
@Cast(value="cudaError_t") public static int cudaGraphExternalSemaphoresWaitNodeGetParams(CUgraphNode_st hNode, cudaExternalSemaphoreWaitNodeParams params_out)
hNode
- - Node to get the parameters forparams_out
- - Pointer to return the parameters::cudaLaunchKernel,
::cudaGraphAddExternalSemaphoresWaitNode,
::cudaGraphExternalSemaphoresWaitNodeSetParams,
::cudaGraphAddExternalSemaphoresWaitNode,
::cudaSignalExternalSemaphoresAsync,
::cudaWaitExternalSemaphoresAsync
@Cast(value="cudaError_t") public static int cudaGraphExternalSemaphoresWaitNodeSetParams(CUgraphNode_st hNode, @Const cudaExternalSemaphoreWaitNodeParams nodeParams)
hNode
- - Node to set the parameters fornodeParams
- - Parameters to copy::cudaGraphNodeSetParams,
::cudaGraphAddExternalSemaphoresWaitNode,
::cudaGraphExternalSemaphoresWaitNodeSetParams,
::cudaGraphAddExternalSemaphoresWaitNode,
::cudaSignalExternalSemaphoresAsync,
::cudaWaitExternalSemaphoresAsync
@Cast(value="cudaError_t") public static int cudaGraphAddMemAllocNode(@ByPtrPtr CUgraphNode_st pGraphNode, CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st pDependencies, @Cast(value="size_t") long numDependencies, cudaMemAllocNodeParams nodeParams)
pGraphNode
- - Returns newly created nodegraph
- - Graph to which to add the nodepDependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciesnodeParams
- - Parameters for the node
When ::cudaGraphAddMemAllocNode creates an allocation node, it returns the address of the allocation in
\p nodeParams.dptr. The allocation's address remains fixed across instantiations and launches.
If the allocation is freed in the same graph, by creating a free node using ::cudaGraphAddMemFreeNode,
the allocation can be accessed by nodes ordered after the allocation node but before the free node.
These allocations cannot be freed outside the owning graph, and they can only be freed once in the
owning graph.
If the allocation is not freed in the same graph, then it can be accessed not only by nodes in the
graph which are ordered after the allocation node, but also by stream operations ordered after the
graph's execution but before the allocation is freed.
Allocations which are not freed in the same graph can be freed by:
- passing the allocation to ::cudaMemFreeAsync or ::cudaMemFree;
- launching a graph with a free node for that allocation; or
- specifying ::cudaGraphInstantiateFlagAutoFreeOnLaunch during instantiation, which makes
each launch behave as though it called ::cudaMemFreeAsync for every unfreed allocation.
It is not possible to free an allocation in both the owning graph and another graph. If the allocation
is freed in the same graph, a free node cannot be added to another graph. If the allocation is freed
in another graph, a free node can no longer be added to the owning graph.
The following restrictions apply to graphs which contain allocation and/or memory free nodes:
- Nodes and edges of the graph cannot be deleted.
- The graph cannot be used in a child node.
- Only one instantiation of the graph may exist at any point in time.
- The graph cannot be cloned.::cudaGraphAddNode,
::cudaGraphAddMemFreeNode,
::cudaGraphMemAllocNodeGetParams,
::cudaDeviceGraphMemTrim,
::cudaDeviceGetGraphMemAttribute,
::cudaDeviceSetGraphMemAttribute,
::cudaMallocAsync,
::cudaFreeAsync,
::cudaGraphCreate,
::cudaGraphDestroyNode,
::cudaGraphAddChildGraphNode,
::cudaGraphAddEmptyNode,
::cudaGraphAddEventRecordNode,
::cudaGraphAddEventWaitNode,
::cudaGraphAddExternalSemaphoresSignalNode,
::cudaGraphAddExternalSemaphoresWaitNode,
::cudaGraphAddKernelNode,
::cudaGraphAddMemcpyNode,
::cudaGraphAddMemsetNode
@Cast(value="cudaError_t") public static int cudaGraphMemAllocNodeGetParams(CUgraphNode_st node, cudaMemAllocNodeParams params_out)
node
- - Node to get the parameters forparams_out
- - Pointer to return the parameters::cudaGraphAddMemAllocNode,
::cudaGraphMemFreeNodeGetParams
@Cast(value="cudaError_t") public static int cudaGraphAddMemFreeNode(@ByPtrPtr CUgraphNode_st pGraphNode, CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st pDependencies, @Cast(value="size_t") long numDependencies, Pointer dptr)
pGraphNode
- - Returns newly created nodegraph
- - Graph to which to add the nodepDependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciesdptr
- - Address of memory to free
::cudaGraphAddMemFreeNode will return ::cudaErrorInvalidValue if the user attempts to free:
- an allocation twice in the same graph.
- an address that was not returned by an allocation node.
- an invalid address.
The following restrictions apply to graphs which contain allocation and/or memory free nodes:
- Nodes and edges of the graph cannot be deleted.
- The graph cannot be used in a child node.
- Only one instantiation of the graph may exist at any point in time.
- The graph cannot be cloned.::cudaGraphAddNode,
::cudaGraphAddMemAllocNode,
::cudaGraphMemFreeNodeGetParams,
::cudaDeviceGraphMemTrim,
::cudaDeviceGetGraphMemAttribute,
::cudaDeviceSetGraphMemAttribute,
::cudaMallocAsync,
::cudaFreeAsync,
::cudaGraphCreate,
::cudaGraphDestroyNode,
::cudaGraphAddChildGraphNode,
::cudaGraphAddEmptyNode,
::cudaGraphAddEventRecordNode,
::cudaGraphAddEventWaitNode,
::cudaGraphAddExternalSemaphoresSignalNode,
::cudaGraphAddExternalSemaphoresWaitNode,
::cudaGraphAddKernelNode,
::cudaGraphAddMemcpyNode,
::cudaGraphAddMemsetNode
@Cast(value="cudaError_t") public static int cudaGraphMemFreeNodeGetParams(CUgraphNode_st node, Pointer dptr_out)
node
- - Node to get the parameters fordptr_out
- - Pointer to return the device address::cudaGraphAddMemFreeNode,
::cudaGraphMemFreeNodeGetParams
@Cast(value="cudaError_t") public static int cudaDeviceGraphMemTrim(int device)
device
- - The device for which cached memory should be freed.::cudaGraphAddMemAllocNode,
::cudaGraphAddMemFreeNode,
::cudaDeviceGetGraphMemAttribute,
::cudaDeviceSetGraphMemAttribute,
::cudaMallocAsync,
::cudaFreeAsync
@Cast(value="cudaError_t") public static int cudaDeviceGetGraphMemAttribute(int device, @Cast(value="cudaGraphMemAttributeType") int attr, Pointer value)
device
- - Specifies the scope of the queryattr
- - attribute to getvalue
- - retrieved value::cudaDeviceSetGraphMemAttribute,
::cudaGraphAddMemAllocNode,
::cudaGraphAddMemFreeNode,
::cudaDeviceGraphMemTrim,
::cudaMallocAsync,
::cudaFreeAsync
@Cast(value="cudaError_t") public static int cudaDeviceSetGraphMemAttribute(int device, @Cast(value="cudaGraphMemAttributeType") int attr, Pointer value)
device
- - Specifies the scope of the queryattr
- - attribute to getvalue
- - pointer to value to set::cudaDeviceGetGraphMemAttribute,
::cudaGraphAddMemAllocNode,
::cudaGraphAddMemFreeNode,
::cudaDeviceGraphMemTrim,
::cudaMallocAsync,
::cudaFreeAsync
@Cast(value="cudaError_t") public static int cudaGraphClone(@ByPtrPtr CUgraph_st pGraphClone, CUgraph_st originalGraph)
pGraphClone
- - Returns newly created cloned graphoriginalGraph
- - Graph to clone::cudaGraphCreate,
::cudaGraphNodeFindInClone
@Cast(value="cudaError_t") public static int cudaGraphNodeFindInClone(@ByPtrPtr CUgraphNode_st pNode, CUgraphNode_st originalNode, CUgraph_st clonedGraph)
pNode
- - Returns handle to the cloned nodeoriginalNode
- - Handle to the original nodeclonedGraph
- - Cloned graph to query::cudaGraphClone
@Cast(value="cudaError_t") public static int cudaGraphNodeGetType(CUgraphNode_st node, @Cast(value="cudaGraphNodeType*") IntPointer pType)
node
- - Node to querypType
- - Pointer to return the node type::cudaGraphGetNodes,
::cudaGraphGetRootNodes,
::cudaGraphChildGraphNodeGetGraph,
::cudaGraphKernelNodeGetParams,
::cudaGraphKernelNodeSetParams,
::cudaGraphHostNodeGetParams,
::cudaGraphHostNodeSetParams,
::cudaGraphMemcpyNodeGetParams,
::cudaGraphMemcpyNodeSetParams,
::cudaGraphMemsetNodeGetParams,
::cudaGraphMemsetNodeSetParams
@Cast(value="cudaError_t") public static int cudaGraphNodeGetType(CUgraphNode_st node, @Cast(value="cudaGraphNodeType*") IntBuffer pType)
@Cast(value="cudaError_t") public static int cudaGraphNodeGetType(CUgraphNode_st node, @Cast(value="cudaGraphNodeType*") int[] pType)
@Cast(value="cudaError_t") public static int cudaGraphGetNodes(CUgraph_st graph, @ByPtrPtr CUgraphNode_st nodes, @Cast(value="size_t*") SizeTPointer numNodes)
graph
- - Graph to querynodes
- - Pointer to return the nodesnumNodes
- - See description::cudaGraphCreate,
::cudaGraphGetRootNodes,
::cudaGraphGetEdges,
::cudaGraphNodeGetType,
::cudaGraphNodeGetDependencies,
::cudaGraphNodeGetDependentNodes
@Cast(value="cudaError_t") public static int cudaGraphGetRootNodes(CUgraph_st graph, @ByPtrPtr CUgraphNode_st pRootNodes, @Cast(value="size_t*") SizeTPointer pNumRootNodes)
graph
- - Graph to querypRootNodes
- - Pointer to return the root nodespNumRootNodes
- - See description::cudaGraphCreate,
::cudaGraphGetNodes,
::cudaGraphGetEdges,
::cudaGraphNodeGetType,
::cudaGraphNodeGetDependencies,
::cudaGraphNodeGetDependentNodes
@Cast(value="cudaError_t") public static int cudaGraphGetEdges(CUgraph_st graph, @ByPtrPtr CUgraphNode_st from, @ByPtrPtr CUgraphNode_st to, @Cast(value="size_t*") SizeTPointer numEdges)
graph
- - Graph to get the edges fromfrom
- - Location to return edge endpointsto
- - Location to return edge endpointsnumEdges
- - See description::cudaGraphGetNodes,
::cudaGraphGetRootNodes,
::cudaGraphAddDependencies,
::cudaGraphRemoveDependencies,
::cudaGraphNodeGetDependencies,
::cudaGraphNodeGetDependentNodes
@Cast(value="cudaError_t") public static int cudaGraphGetEdges_v2(CUgraph_st graph, @ByPtrPtr CUgraphNode_st from, @ByPtrPtr CUgraphNode_st to, cudaGraphEdgeData edgeData, @Cast(value="size_t*") SizeTPointer numEdges)
graph
- - Graph to get the edges fromfrom
- - Location to return edge endpointsto
- - Location to return edge endpointsedgeData
- - Optional location to return edge datanumEdges
- - See description::cudaGraphGetNodes,
::cudaGraphGetRootNodes,
::cudaGraphAddDependencies,
::cudaGraphRemoveDependencies,
::cudaGraphNodeGetDependencies,
::cudaGraphNodeGetDependentNodes
@Cast(value="cudaError_t") public static int cudaGraphNodeGetDependencies(CUgraphNode_st node, @ByPtrPtr CUgraphNode_st pDependencies, @Cast(value="size_t*") SizeTPointer pNumDependencies)
node
- - Node to querypDependencies
- - Pointer to return the dependenciespNumDependencies
- - See description::cudaGraphNodeGetDependentNodes,
::cudaGraphGetNodes,
::cudaGraphGetRootNodes,
::cudaGraphGetEdges,
::cudaGraphAddDependencies,
::cudaGraphRemoveDependencies
@Cast(value="cudaError_t") public static int cudaGraphNodeGetDependencies_v2(CUgraphNode_st node, @ByPtrPtr CUgraphNode_st pDependencies, cudaGraphEdgeData edgeData, @Cast(value="size_t*") SizeTPointer pNumDependencies)
node
- - Node to querypDependencies
- - Pointer to return the dependenciesedgeData
- - Optional array to return edge data for each dependencypNumDependencies
- - See description::cudaGraphNodeGetDependentNodes,
::cudaGraphGetNodes,
::cudaGraphGetRootNodes,
::cudaGraphGetEdges,
::cudaGraphAddDependencies,
::cudaGraphRemoveDependencies
@Cast(value="cudaError_t") public static int cudaGraphNodeGetDependentNodes(CUgraphNode_st node, @ByPtrPtr CUgraphNode_st pDependentNodes, @Cast(value="size_t*") SizeTPointer pNumDependentNodes)
node
- - Node to querypDependentNodes
- - Pointer to return the dependent nodespNumDependentNodes
- - See description::cudaGraphNodeGetDependencies,
::cudaGraphGetNodes,
::cudaGraphGetRootNodes,
::cudaGraphGetEdges,
::cudaGraphAddDependencies,
::cudaGraphRemoveDependencies
@Cast(value="cudaError_t") public static int cudaGraphNodeGetDependentNodes_v2(CUgraphNode_st node, @ByPtrPtr CUgraphNode_st pDependentNodes, cudaGraphEdgeData edgeData, @Cast(value="size_t*") SizeTPointer pNumDependentNodes)
node
- - Node to querypDependentNodes
- - Pointer to return the dependent nodesedgeData
- - Optional pointer to return edge data for dependent nodespNumDependentNodes
- - See description::cudaGraphNodeGetDependencies,
::cudaGraphGetNodes,
::cudaGraphGetRootNodes,
::cudaGraphGetEdges,
::cudaGraphAddDependencies,
::cudaGraphRemoveDependencies
@Cast(value="cudaError_t") public static int cudaGraphAddDependencies(CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st from, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st to, @Cast(value="size_t") long numDependencies)
graph
- - Graph to which dependencies are addedfrom
- - Array of nodes that provide the dependenciesto
- - Array of dependent nodesnumDependencies
- - Number of dependencies to be added::cudaGraphRemoveDependencies,
::cudaGraphGetEdges,
::cudaGraphNodeGetDependencies,
::cudaGraphNodeGetDependentNodes
@Cast(value="cudaError_t") public static int cudaGraphAddDependencies_v2(CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st from, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st to, @Const cudaGraphEdgeData edgeData, @Cast(value="size_t") long numDependencies)
graph
- - Graph to which dependencies are addedfrom
- - Array of nodes that provide the dependenciesto
- - Array of dependent nodesedgeData
- - Optional array of edge data. If NULL, default (zeroed) edge data is assumed.numDependencies
- - Number of dependencies to be added::cudaGraphRemoveDependencies,
::cudaGraphGetEdges,
::cudaGraphNodeGetDependencies,
::cudaGraphNodeGetDependentNodes
@Cast(value="cudaError_t") public static int cudaGraphRemoveDependencies(CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st from, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st to, @Cast(value="size_t") long numDependencies)
graph
- - Graph from which to remove dependenciesfrom
- - Array of nodes that provide the dependenciesto
- - Array of dependent nodesnumDependencies
- - Number of dependencies to be removed::cudaGraphAddDependencies,
::cudaGraphGetEdges,
::cudaGraphNodeGetDependencies,
::cudaGraphNodeGetDependentNodes
@Cast(value="cudaError_t") public static int cudaGraphRemoveDependencies_v2(CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st from, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st to, @Const cudaGraphEdgeData edgeData, @Cast(value="size_t") long numDependencies)
graph
- - Graph from which to remove dependenciesfrom
- - Array of nodes that provide the dependenciesto
- - Array of dependent nodesedgeData
- - Optional array of edge data. If NULL, edge data is assumed to
be default (zeroed).numDependencies
- - Number of dependencies to be removed::cudaGraphAddDependencies,
::cudaGraphGetEdges,
::cudaGraphNodeGetDependencies,
::cudaGraphNodeGetDependentNodes
@Cast(value="cudaError_t") public static int cudaGraphDestroyNode(CUgraphNode_st node)
node
- - Node to remove::cudaGraphAddChildGraphNode,
::cudaGraphAddEmptyNode,
::cudaGraphAddKernelNode,
::cudaGraphAddHostNode,
::cudaGraphAddMemcpyNode,
::cudaGraphAddMemsetNode
@Cast(value="cudaError_t") public static int cudaGraphInstantiate(@ByPtrPtr CUgraphExec_st pGraphExec, CUgraph_st graph, @Cast(value="unsigned long long") long flags)
pGraphExec
- - Returns instantiated graphgraph
- - Graph to instantiateflags
- - Flags to control instantiation. See ::CUgraphInstantiate_flags.::cudaGraphInstantiateWithFlags,
::cudaGraphCreate,
::cudaGraphUpload,
::cudaGraphLaunch,
::cudaGraphExecDestroy
@Cast(value="cudaError_t") public static int cudaGraphInstantiate(@ByPtrPtr CUgraphExec_st pGraphExec, CUgraph_st graph)
@Cast(value="cudaError_t") public static int cudaGraphInstantiateWithFlags(@ByPtrPtr CUgraphExec_st pGraphExec, CUgraph_st graph, @Cast(value="unsigned long long") long flags)
pGraphExec
- - Returns instantiated graphgraph
- - Graph to instantiateflags
- - Flags to control instantiation. See ::CUgraphInstantiate_flags.::cudaGraphInstantiate,
::cudaGraphCreate,
::cudaGraphUpload,
::cudaGraphLaunch,
::cudaGraphExecDestroy
@Cast(value="cudaError_t") public static int cudaGraphInstantiateWithFlags(@ByPtrPtr CUgraphExec_st pGraphExec, CUgraph_st graph)
@Cast(value="cudaError_t") public static int cudaGraphInstantiateWithParams(@ByPtrPtr CUgraphExec_st pGraphExec, CUgraph_st graph, cudaGraphInstantiateParams instantiateParams)
typedef struct {
unsigned long long flags;
cudaStream_t uploadStream;
cudaGraphNode_t errNode_out;
cudaGraphInstantiateResult result_out;
} cudaGraphInstantiateParams;
The \p flags field controls the behavior of instantiation and subsequent
graph launches. Valid flags are:
- ::cudaGraphInstantiateFlagAutoFreeOnLaunch, which configures a
graph containing memory allocation nodes to automatically free any
unfreed memory allocations before the graph is relaunched.
- ::cudaGraphInstantiateFlagUpload, which will perform an upload of the graph
into \p uploadStream once the graph has been instantiated.
- ::cudaGraphInstantiateFlagDeviceLaunch, which configures the graph for launch
from the device. If this flag is passed, the executable graph handle returned can be
used to launch the graph from both the host and device. This flag can only be used
on platforms which support unified addressing. This flag cannot be used in
conjunction with ::cudaGraphInstantiateFlagAutoFreeOnLaunch.
- ::cudaGraphInstantiateFlagUseNodePriority, which causes the graph
to use the priorities from the per-node attributes rather than the priority
of the launch stream during execution. Note that priorities are only available
on kernel nodes, and are copied from stream priority during stream capture.
If \p graph contains any allocation or free nodes, there can be at most one
executable graph in existence for that graph at a time. An attempt to instantiate a
second executable graph before destroying the first with ::cudaGraphExecDestroy will
result in an error.
If \p graph contains kernels which call device-side cudaGraphLaunch() from multiple
devices, this will result in an error.
Graphs instantiated for launch on the device have additional restrictions which do not
apply to host graphs:
- The graph's nodes must reside on a single device.
- The graph can only contain kernel nodes, memcpy nodes, memset nodes, and child graph nodes.
- The graph cannot be empty and must contain at least one kernel, memcpy, or memset node.
Operation-specific restrictions are outlined below.
- Kernel nodes:
- Use of CUDA Dynamic Parallelism is not permitted.
- Cooperative launches are permitted as long as MPS is not in use.
- Memcpy nodes:
- Only copies involving device memory and/or pinned device-mapped host memory are permitted.
- Copies involving CUDA arrays are not permitted.
- Both operands must be accessible from the current device, and the current device must
match the device of other nodes in the graph.
In the event of an error, the \p result_out and \p errNode_out fields will contain more
information about the nature of the error. Possible error reporting includes:
- ::cudaGraphInstantiateError, if passed an invalid value or if an unexpected error occurred
which is described by the return value of the function. \p errNode_out will be set to NULL.
- ::cudaGraphInstantiateInvalidStructure, if the graph structure is invalid. \p errNode_out
will be set to one of the offending nodes.
- ::cudaGraphInstantiateNodeOperationNotSupported, if the graph is instantiated for device
launch but contains a node of an unsupported node type, or a node which performs unsupported
operations, such as use of CUDA dynamic parallelism within a kernel node. \p errNode_out will
be set to this node.
- ::cudaGraphInstantiateMultipleDevicesNotSupported, if the graph is instantiated for device
launch but a node’s device differs from that of another node. This error can also be returned
if a graph is not instantiated for device launch and it contains kernels which call device-side
cudaGraphLaunch() from multiple devices. \p errNode_out will be set to this node.
If instantiation is successful, \p result_out will be set to ::cudaGraphInstantiateSuccess,
and \p hErrNode_out will be set to NULL.pGraphExec
- - Returns instantiated graphgraph
- - Graph to instantiateinstantiateParams
- - Instantiation parameters::cudaGraphCreate,
::cudaGraphInstantiate,
::cudaGraphInstantiateWithFlags,
::cudaGraphExecDestroy
@Cast(value="cudaError_t") public static int cudaGraphExecGetFlags(CUgraphExec_st graphExec, @Cast(value="unsigned long long*") LongPointer flags)
graphExec
- - The executable graph to queryflags
- - Returns the instantiation flags::cudaGraphInstantiate,
::cudaGraphInstantiateWithFlags,
::cudaGraphInstantiateWithParams
@Cast(value="cudaError_t") public static int cudaGraphExecGetFlags(CUgraphExec_st graphExec, @Cast(value="unsigned long long*") LongBuffer flags)
@Cast(value="cudaError_t") public static int cudaGraphExecGetFlags(CUgraphExec_st graphExec, @Cast(value="unsigned long long*") long[] flags)
@Cast(value="cudaError_t") public static int cudaGraphExecKernelNodeSetParams(CUgraphExec_st hGraphExec, CUgraphNode_st node, @Const cudaKernelNodeParams pNodeParams)
hGraphExec
- - The executable graph in which to set the specified nodenode
- - kernel node from the graph from which graphExec was instantiatedpNodeParams
- - Updated Parameters to set::cudaGraphExecNodeSetParams,
::cudaGraphAddKernelNode,
::cudaGraphKernelNodeSetParams,
::cudaGraphExecMemcpyNodeSetParams,
::cudaGraphExecMemsetNodeSetParams,
::cudaGraphExecHostNodeSetParams,
::cudaGraphExecChildGraphNodeSetParams,
::cudaGraphExecEventRecordNodeSetEvent,
::cudaGraphExecEventWaitNodeSetEvent,
::cudaGraphExecExternalSemaphoresSignalNodeSetParams,
::cudaGraphExecExternalSemaphoresWaitNodeSetParams,
::cudaGraphExecUpdate,
::cudaGraphInstantiate
@Cast(value="cudaError_t") public static int cudaGraphExecMemcpyNodeSetParams(CUgraphExec_st hGraphExec, CUgraphNode_st node, @Const cudaMemcpy3DParms pNodeParams)
hGraphExec
- - The executable graph in which to set the specified nodenode
- - Memcpy node from the graph which was used to instantiate graphExecpNodeParams
- - Updated Parameters to set::cudaGraphExecNodeSetParams,
::cudaGraphAddMemcpyNode,
::cudaGraphMemcpyNodeSetParams,
::cudaGraphExecMemcpyNodeSetParamsToSymbol,
::cudaGraphExecMemcpyNodeSetParamsFromSymbol,
::cudaGraphExecMemcpyNodeSetParams1D,
::cudaGraphExecKernelNodeSetParams,
::cudaGraphExecMemsetNodeSetParams,
::cudaGraphExecHostNodeSetParams,
::cudaGraphExecChildGraphNodeSetParams,
::cudaGraphExecEventRecordNodeSetEvent,
::cudaGraphExecEventWaitNodeSetEvent,
::cudaGraphExecExternalSemaphoresSignalNodeSetParams,
::cudaGraphExecExternalSemaphoresWaitNodeSetParams,
::cudaGraphExecUpdate,
::cudaGraphInstantiate
@Cast(value="cudaError_t") public static int cudaGraphExecMemcpyNodeSetParamsToSymbol(CUgraphExec_st hGraphExec, CUgraphNode_st node, @Const Pointer symbol, @Const Pointer src, @Cast(value="size_t") long count, @Cast(value="size_t") long offset, @Cast(value="cudaMemcpyKind") int kind)
hGraphExec
- - The executable graph in which to set the specified nodenode
- - Memcpy node from the graph which was used to instantiate graphExecsymbol
- - Device symbol addresssrc
- - Source memory addresscount
- - Size in bytes to copyoffset
- - Offset from start of symbol in byteskind
- - Type of transfer::cudaGraphAddMemcpyNode,
::cudaGraphAddMemcpyNodeToSymbol,
::cudaGraphMemcpyNodeSetParams,
::cudaGraphMemcpyNodeSetParamsToSymbol,
::cudaGraphExecMemcpyNodeSetParams,
::cudaGraphExecMemcpyNodeSetParamsFromSymbol,
::cudaGraphExecKernelNodeSetParams,
::cudaGraphExecMemsetNodeSetParams,
::cudaGraphExecHostNodeSetParams,
::cudaGraphExecChildGraphNodeSetParams,
::cudaGraphExecEventRecordNodeSetEvent,
::cudaGraphExecEventWaitNodeSetEvent,
::cudaGraphExecExternalSemaphoresSignalNodeSetParams,
::cudaGraphExecExternalSemaphoresWaitNodeSetParams,
::cudaGraphExecUpdate,
::cudaGraphInstantiate
@Cast(value="cudaError_t") public static int cudaGraphExecMemcpyNodeSetParamsFromSymbol(CUgraphExec_st hGraphExec, CUgraphNode_st node, Pointer dst, @Const Pointer symbol, @Cast(value="size_t") long count, @Cast(value="size_t") long offset, @Cast(value="cudaMemcpyKind") int kind)
hGraphExec
- - The executable graph in which to set the specified nodenode
- - Memcpy node from the graph which was used to instantiate graphExecdst
- - Destination memory addresssymbol
- - Device symbol addresscount
- - Size in bytes to copyoffset
- - Offset from start of symbol in byteskind
- - Type of transfer::cudaGraphAddMemcpyNode,
::cudaGraphAddMemcpyNodeFromSymbol,
::cudaGraphMemcpyNodeSetParams,
::cudaGraphMemcpyNodeSetParamsFromSymbol,
::cudaGraphExecMemcpyNodeSetParams,
::cudaGraphExecMemcpyNodeSetParamsToSymbol,
::cudaGraphExecKernelNodeSetParams,
::cudaGraphExecMemsetNodeSetParams,
::cudaGraphExecHostNodeSetParams,
::cudaGraphExecChildGraphNodeSetParams,
::cudaGraphExecEventRecordNodeSetEvent,
::cudaGraphExecEventWaitNodeSetEvent,
::cudaGraphExecExternalSemaphoresSignalNodeSetParams,
::cudaGraphExecExternalSemaphoresWaitNodeSetParams,
::cudaGraphExecUpdate,
::cudaGraphInstantiate
@Cast(value="cudaError_t") public static int cudaGraphExecMemcpyNodeSetParams1D(CUgraphExec_st hGraphExec, CUgraphNode_st node, Pointer dst, @Const Pointer src, @Cast(value="size_t") long count, @Cast(value="cudaMemcpyKind") int kind)
hGraphExec
- - The executable graph in which to set the specified nodenode
- - Memcpy node from the graph which was used to instantiate graphExecdst
- - Destination memory addresssrc
- - Source memory addresscount
- - Size in bytes to copykind
- - Type of transfer::cudaGraphAddMemcpyNode,
::cudaGraphAddMemcpyNode1D,
::cudaGraphMemcpyNodeSetParams,
::cudaGraphMemcpyNodeSetParams1D,
::cudaGraphExecMemcpyNodeSetParams,
::cudaGraphExecKernelNodeSetParams,
::cudaGraphExecMemsetNodeSetParams,
::cudaGraphExecHostNodeSetParams,
::cudaGraphExecChildGraphNodeSetParams,
::cudaGraphExecEventRecordNodeSetEvent,
::cudaGraphExecEventWaitNodeSetEvent,
::cudaGraphExecExternalSemaphoresSignalNodeSetParams,
::cudaGraphExecExternalSemaphoresWaitNodeSetParams,
::cudaGraphExecUpdate,
::cudaGraphInstantiate
@Cast(value="cudaError_t") public static int cudaGraphExecMemsetNodeSetParams(CUgraphExec_st hGraphExec, CUgraphNode_st node, @Const cudaMemsetParams pNodeParams)
hGraphExec
- - The executable graph in which to set the specified nodenode
- - Memset node from the graph which was used to instantiate graphExecpNodeParams
- - Updated Parameters to set::cudaGraphExecNodeSetParams,
::cudaGraphAddMemsetNode,
::cudaGraphMemsetNodeSetParams,
::cudaGraphExecKernelNodeSetParams,
::cudaGraphExecMemcpyNodeSetParams,
::cudaGraphExecHostNodeSetParams,
::cudaGraphExecChildGraphNodeSetParams,
::cudaGraphExecEventRecordNodeSetEvent,
::cudaGraphExecEventWaitNodeSetEvent,
::cudaGraphExecExternalSemaphoresSignalNodeSetParams,
::cudaGraphExecExternalSemaphoresWaitNodeSetParams,
::cudaGraphExecUpdate,
::cudaGraphInstantiate
@Cast(value="cudaError_t") public static int cudaGraphExecHostNodeSetParams(CUgraphExec_st hGraphExec, CUgraphNode_st node, @Const cudaHostNodeParams pNodeParams)
hGraphExec
- - The executable graph in which to set the specified nodenode
- - Host node from the graph which was used to instantiate graphExecpNodeParams
- - Updated Parameters to set::cudaGraphExecNodeSetParams,
::cudaGraphAddHostNode,
::cudaGraphHostNodeSetParams,
::cudaGraphExecKernelNodeSetParams,
::cudaGraphExecMemcpyNodeSetParams,
::cudaGraphExecMemsetNodeSetParams,
::cudaGraphExecChildGraphNodeSetParams,
::cudaGraphExecEventRecordNodeSetEvent,
::cudaGraphExecEventWaitNodeSetEvent,
::cudaGraphExecExternalSemaphoresSignalNodeSetParams,
::cudaGraphExecExternalSemaphoresWaitNodeSetParams,
::cudaGraphExecUpdate,
::cudaGraphInstantiate
@Cast(value="cudaError_t") public static int cudaGraphExecChildGraphNodeSetParams(CUgraphExec_st hGraphExec, CUgraphNode_st node, CUgraph_st childGraph)
hGraphExec
- - The executable graph in which to set the specified nodenode
- - Host node from the graph which was used to instantiate graphExecchildGraph
- - The graph supplying the updated parameters::cudaGraphExecNodeSetParams,
::cudaGraphAddChildGraphNode,
::cudaGraphChildGraphNodeGetGraph,
::cudaGraphExecKernelNodeSetParams,
::cudaGraphExecMemcpyNodeSetParams,
::cudaGraphExecMemsetNodeSetParams,
::cudaGraphExecHostNodeSetParams,
::cudaGraphExecEventRecordNodeSetEvent,
::cudaGraphExecEventWaitNodeSetEvent,
::cudaGraphExecExternalSemaphoresSignalNodeSetParams,
::cudaGraphExecExternalSemaphoresWaitNodeSetParams,
::cudaGraphExecUpdate,
::cudaGraphInstantiate
@Cast(value="cudaError_t") public static int cudaGraphExecEventRecordNodeSetEvent(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, CUevent_st event)
hGraphExec
- - The executable graph in which to set the specified nodehNode
- - Event record node from the graph from which graphExec was instantiatedevent
- - Updated event to use::cudaGraphExecNodeSetParams,
::cudaGraphAddEventRecordNode,
::cudaGraphEventRecordNodeGetEvent,
::cudaGraphEventWaitNodeSetEvent,
::cudaEventRecordWithFlags,
::cudaStreamWaitEvent,
::cudaGraphExecKernelNodeSetParams,
::cudaGraphExecMemcpyNodeSetParams,
::cudaGraphExecMemsetNodeSetParams,
::cudaGraphExecHostNodeSetParams,
::cudaGraphExecChildGraphNodeSetParams,
::cudaGraphExecEventWaitNodeSetEvent,
::cudaGraphExecExternalSemaphoresSignalNodeSetParams,
::cudaGraphExecExternalSemaphoresWaitNodeSetParams,
::cudaGraphExecUpdate,
::cudaGraphInstantiate
@Cast(value="cudaError_t") public static int cudaGraphExecEventWaitNodeSetEvent(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, CUevent_st event)
hGraphExec
- - The executable graph in which to set the specified nodehNode
- - Event wait node from the graph from which graphExec was instantiatedevent
- - Updated event to use::cudaGraphExecNodeSetParams,
::cudaGraphAddEventWaitNode,
::cudaGraphEventWaitNodeGetEvent,
::cudaGraphEventRecordNodeSetEvent,
::cudaEventRecordWithFlags,
::cudaStreamWaitEvent,
::cudaGraphExecKernelNodeSetParams,
::cudaGraphExecMemcpyNodeSetParams,
::cudaGraphExecMemsetNodeSetParams,
::cudaGraphExecHostNodeSetParams,
::cudaGraphExecChildGraphNodeSetParams,
::cudaGraphExecEventRecordNodeSetEvent,
::cudaGraphExecExternalSemaphoresSignalNodeSetParams,
::cudaGraphExecExternalSemaphoresWaitNodeSetParams,
::cudaGraphExecUpdate,
::cudaGraphInstantiate
@Cast(value="cudaError_t") public static int cudaGraphExecExternalSemaphoresSignalNodeSetParams(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, @Const cudaExternalSemaphoreSignalNodeParams nodeParams)
hGraphExec
- - The executable graph in which to set the specified nodehNode
- - semaphore signal node from the graph from which graphExec was instantiatednodeParams
- - Updated Parameters to set::cudaGraphExecNodeSetParams,
::cudaGraphAddExternalSemaphoresSignalNode,
::cudaImportExternalSemaphore,
::cudaSignalExternalSemaphoresAsync,
::cudaWaitExternalSemaphoresAsync,
::cudaGraphExecKernelNodeSetParams,
::cudaGraphExecMemcpyNodeSetParams,
::cudaGraphExecMemsetNodeSetParams,
::cudaGraphExecHostNodeSetParams,
::cudaGraphExecChildGraphNodeSetParams,
::cudaGraphExecEventRecordNodeSetEvent,
::cudaGraphExecEventWaitNodeSetEvent,
::cudaGraphExecExternalSemaphoresWaitNodeSetParams,
::cudaGraphExecUpdate,
::cudaGraphInstantiate
@Cast(value="cudaError_t") public static int cudaGraphExecExternalSemaphoresWaitNodeSetParams(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, @Const cudaExternalSemaphoreWaitNodeParams nodeParams)
hGraphExec
- - The executable graph in which to set the specified nodehNode
- - semaphore wait node from the graph from which graphExec was instantiatednodeParams
- - Updated Parameters to set::cudaGraphExecNodeSetParams,
::cudaGraphAddExternalSemaphoresWaitNode,
::cudaImportExternalSemaphore,
::cudaSignalExternalSemaphoresAsync,
::cudaWaitExternalSemaphoresAsync,
::cudaGraphExecKernelNodeSetParams,
::cudaGraphExecMemcpyNodeSetParams,
::cudaGraphExecMemsetNodeSetParams,
::cudaGraphExecHostNodeSetParams,
::cudaGraphExecChildGraphNodeSetParams,
::cudaGraphExecEventRecordNodeSetEvent,
::cudaGraphExecEventWaitNodeSetEvent,
::cudaGraphExecExternalSemaphoresSignalNodeSetParams,
::cudaGraphExecUpdate,
::cudaGraphInstantiate
@Cast(value="cudaError_t") public static int cudaGraphNodeSetEnabled(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, @Cast(value="unsigned int") int isEnabled)
hGraphExec
- - The executable graph in which to set the specified nodehNode
- - Node from the graph from which graphExec was instantiatedisEnabled
- - Node is enabled if != 0, otherwise the node is disabled::cudaGraphNodeGetEnabled,
::cudaGraphExecUpdate,
::cudaGraphInstantiate
::cudaGraphLaunch
@Cast(value="cudaError_t") public static int cudaGraphNodeGetEnabled(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, @Cast(value="unsigned int*") IntPointer isEnabled)
hGraphExec
- - The executable graph in which to set the specified nodehNode
- - Node from the graph from which graphExec was instantiatedisEnabled
- - Location to return the enabled status of the node::cudaGraphNodeSetEnabled,
::cudaGraphExecUpdate,
::cudaGraphInstantiate
::cudaGraphLaunch
@Cast(value="cudaError_t") public static int cudaGraphNodeGetEnabled(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, @Cast(value="unsigned int*") IntBuffer isEnabled)
@Cast(value="cudaError_t") public static int cudaGraphNodeGetEnabled(CUgraphExec_st hGraphExec, CUgraphNode_st hNode, @Cast(value="unsigned int*") int[] isEnabled)
@Cast(value="cudaError_t") public static int cudaGraphExecUpdate(CUgraphExec_st hGraphExec, CUgraph_st hGraph, cudaGraphExecUpdateResultInfo resultInfo)
hGraphExec
- The instantiated graph to be updatedhGraph
- The graph containing the updated parametersresultInfo
- the error info structure::cudaGraphInstantiate
@Cast(value="cudaError_t") public static int cudaGraphUpload(CUgraphExec_st graphExec, CUstream_st stream)
hGraphExec
- - Executable graph to uploadhStream
- - Stream in which to upload the graph::cudaGraphInstantiate,
::cudaGraphLaunch,
::cudaGraphExecDestroy
@Cast(value="cudaError_t") public static int cudaGraphLaunch(CUgraphExec_st graphExec, CUstream_st stream)
graphExec
- - Executable graph to launchstream
- - Stream in which to launch the graph::cudaGraphInstantiate,
::cudaGraphUpload,
::cudaGraphExecDestroy
@Cast(value="cudaError_t") public static int cudaGraphExecDestroy(CUgraphExec_st graphExec)
graphExec
- - Executable graph to destroy::cudaGraphInstantiate,
::cudaGraphUpload,
::cudaGraphLaunch
@Cast(value="cudaError_t") public static int cudaGraphDestroy(CUgraph_st graph)
graph
- - Graph to destroy::cudaGraphCreate
@Cast(value="cudaError_t") public static int cudaGraphDebugDotPrint(CUgraph_st graph, @Cast(value="const char*") BytePointer path, @Cast(value="unsigned int") int flags)
graph
- - The graph to create a DOT file frompath
- - The path to write the DOT file toflags
- - Flags from cudaGraphDebugDotFlags for specifying which additional node information to write@Cast(value="cudaError_t") public static int cudaGraphDebugDotPrint(CUgraph_st graph, String path, @Cast(value="unsigned int") int flags)
@Cast(value="cudaError_t") public static int cudaUserObjectCreate(@ByPtrPtr CUuserObject_st object_out, Pointer ptr, cudaHostFn_t destroy, @Cast(value="unsigned int") int initialRefcount, @Cast(value="unsigned int") int flags)
object_out
- - Location to return the user object handleptr
- - The pointer to pass to the destroy functiondestroy
- - Callback to free the user object when it is no longer in useinitialRefcount
- - The initial refcount to create the object with, typically 1. The
initial references are owned by the calling thread.flags
- - Currently it is required to pass ::cudaUserObjectNoDestructorSync,
which is the only defined flag. This indicates that the destroy
callback cannot be waited on by any CUDA API. Users requiring
synchronization of the callback should signal its completion
manually.::cudaUserObjectRetain,
::cudaUserObjectRelease,
::cudaGraphRetainUserObject,
::cudaGraphReleaseUserObject,
::cudaGraphCreate
@Cast(value="cudaError_t") public static int cudaUserObjectRetain(CUuserObject_st object, @Cast(value="unsigned int") int count)
object
- - The object to retaincount
- - The number of references to retain, typically 1. Must be nonzero
and not larger than INT_MAX.::cudaUserObjectCreate,
::cudaUserObjectRelease,
::cudaGraphRetainUserObject,
::cudaGraphReleaseUserObject,
::cudaGraphCreate
@Cast(value="cudaError_t") public static int cudaUserObjectRetain(CUuserObject_st object)
@Cast(value="cudaError_t") public static int cudaUserObjectRelease(CUuserObject_st object, @Cast(value="unsigned int") int count)
object
- - The object to releasecount
- - The number of references to release, typically 1. Must be nonzero
and not larger than INT_MAX.::cudaUserObjectCreate,
::cudaUserObjectRetain,
::cudaGraphRetainUserObject,
::cudaGraphReleaseUserObject,
::cudaGraphCreate
@Cast(value="cudaError_t") public static int cudaUserObjectRelease(CUuserObject_st object)
@Cast(value="cudaError_t") public static int cudaGraphRetainUserObject(CUgraph_st graph, CUuserObject_st object, @Cast(value="unsigned int") int count, @Cast(value="unsigned int") int flags)
graph
- - The graph to associate the reference withobject
- - The user object to retain a reference forcount
- - The number of references to add to the graph, typically 1. Must be
nonzero and not larger than INT_MAX.flags
- - The optional flag ::cudaGraphUserObjectMove transfers references
from the calling thread, rather than create new references. Pass 0
to create new references.::cudaUserObjectRetain,
::cudaUserObjectRelease,
::cudaGraphReleaseUserObject,
::cudaGraphCreate
@Cast(value="cudaError_t") public static int cudaGraphRetainUserObject(CUgraph_st graph, CUuserObject_st object)
@Cast(value="cudaError_t") public static int cudaGraphReleaseUserObject(CUgraph_st graph, CUuserObject_st object, @Cast(value="unsigned int") int count)
graph
- - The graph that will release the referenceobject
- - The user object to release a reference forcount
- - The number of references to release, typically 1. Must be nonzero
and not larger than INT_MAX.::cudaUserObjectRetain,
::cudaUserObjectRelease,
::cudaGraphRetainUserObject,
::cudaGraphCreate
@Cast(value="cudaError_t") public static int cudaGraphReleaseUserObject(CUgraph_st graph, CUuserObject_st object)
@Cast(value="cudaError_t") public static int cudaGraphAddNode(@ByPtrPtr CUgraphNode_st pGraphNode, CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st pDependencies, @Cast(value="size_t") long numDependencies, cudaGraphNodeParams nodeParams)
pGraphNode
- - Returns newly created nodegraph
- - Graph to which to add the nodepDependencies
- - Dependencies of the nodenumDependencies
- - Number of dependenciesnodeParams
- - Specification of the node::cudaGraphCreate,
::cudaGraphNodeSetParams,
::cudaGraphExecNodeSetParams
@Cast(value="cudaError_t") public static int cudaGraphAddNode_v2(@ByPtrPtr CUgraphNode_st pGraphNode, CUgraph_st graph, @Cast(value="const cudaGraphNode_t*") @ByPtrPtr CUgraphNode_st pDependencies, @Const cudaGraphEdgeData dependencyData, @Cast(value="size_t") long numDependencies, cudaGraphNodeParams nodeParams)
pGraphNode
- - Returns newly created nodegraph
- - Graph to which to add the nodepDependencies
- - Dependencies of the nodedependencyData
- - Optional edge data for the dependencies. If NULL, the data is
assumed to be default (zeroed) for all dependencies.numDependencies
- - Number of dependenciesnodeParams
- - Specification of the node::cudaGraphCreate,
::cudaGraphNodeSetParams,
::cudaGraphExecNodeSetParams
@Cast(value="cudaError_t") public static int cudaGraphNodeSetParams(CUgraphNode_st node, cudaGraphNodeParams nodeParams)
node
- - Node to set the parameters fornodeParams
- - Parameters to copy::cudaGraphAddNode,
::cudaGraphExecNodeSetParams
@Cast(value="cudaError_t") public static int cudaGraphExecNodeSetParams(CUgraphExec_st graphExec, CUgraphNode_st node, cudaGraphNodeParams nodeParams)
Node type | Allowed changes |
---|---|
kernel | See ::cudaGraphExecKernelNodeSetParams |
memcpy | Addresses for 1-dimensional copies if allocated in same context; see ::cudaGraphExecMemcpyNodeSetParams |
memset | Addresses for 1-dimensional memsets if allocated in same context; see ::cudaGraphExecMemsetNodeSetParams |
host | Unrestricted |
child graph | Topology must match and restrictions apply recursively; see ::cudaGraphExecUpdate |
event wait | Unrestricted |
event record | Unrestricted |
external semaphore signal | Number of semaphore operations cannot change |
external semaphore wait | Number of semaphore operations cannot change |
memory allocation | API unsupported |
memory free | API unsupported |
graphExec
- - The executable graph in which to update the specified nodenode
- - Corresponding node from the graph from which graphExec was instantiatednodeParams
- - Updated Parameters to set::cudaGraphAddNode,
::cudaGraphNodeSetParams
::cudaGraphExecUpdate,
::cudaGraphInstantiate
@Cast(value="cudaError_t") public static int cudaGraphConditionalHandleCreate(@Cast(value="cudaGraphConditionalHandle*") LongPointer pHandle_out, CUgraph_st graph, @Cast(value="unsigned int") int defaultLaunchValue, @Cast(value="unsigned int") int flags)
pHandle_out
- - Pointer used to return the handle to the caller.hGraph
- - Graph which will contain the conditional node using this handle.defaultLaunchValue
- - Optional initial value for the conditional variable.flags
- - Currently must be cudaGraphCondAssignDefault or 0.::cuGraphAddNode,
@Cast(value="cudaError_t") public static int cudaGraphConditionalHandleCreate(@Cast(value="cudaGraphConditionalHandle*") LongPointer pHandle_out, CUgraph_st graph)
@Cast(value="cudaError_t") public static int cudaGraphConditionalHandleCreate(@Cast(value="cudaGraphConditionalHandle*") LongBuffer pHandle_out, CUgraph_st graph, @Cast(value="unsigned int") int defaultLaunchValue, @Cast(value="unsigned int") int flags)
@Cast(value="cudaError_t") public static int cudaGraphConditionalHandleCreate(@Cast(value="cudaGraphConditionalHandle*") LongBuffer pHandle_out, CUgraph_st graph)
@Cast(value="cudaError_t") public static int cudaGraphConditionalHandleCreate(@Cast(value="cudaGraphConditionalHandle*") long[] pHandle_out, CUgraph_st graph, @Cast(value="unsigned int") int defaultLaunchValue, @Cast(value="unsigned int") int flags)
@Cast(value="cudaError_t") public static int cudaGraphConditionalHandleCreate(@Cast(value="cudaGraphConditionalHandle*") long[] pHandle_out, CUgraph_st graph)
@Cast(value="cudaError_t") public static int cudaGetDriverEntryPoint(@Cast(value="const char*") BytePointer symbol, @Cast(value="void**") PointerPointer funcPtr, @Cast(value="unsigned long long") long flags, @Cast(value="cudaDriverEntryPointQueryResult*") IntPointer driverStatus)
symbol
- - The base name of the driver API function to look for. As an example,
for the driver API ::cuMemAlloc_v2, \p symbol would be cuMemAlloc.
Note that the API will use the CUDA runtime version to return the
address to the most recent ABI compatible driver symbol, ::cuMemAlloc
or ::cuMemAlloc_v2.funcPtr
- - Location to return the function pointer to the requested driver functionflags
- - Flags to specify search options.driverStatus
- - Optional location to store the status of finding the symbol from
the driver. See ::cudaDriverEntryPointQueryResult for
possible values.::cuGetProcAddress
@Cast(value="cudaError_t") public static int cudaGetDriverEntryPoint(@Cast(value="const char*") BytePointer symbol, @Cast(value="void**") @ByPtrPtr Pointer funcPtr, @Cast(value="unsigned long long") long flags)
@Cast(value="cudaError_t") public static int cudaGetDriverEntryPoint(@Cast(value="const char*") BytePointer symbol, @Cast(value="void**") @ByPtrPtr Pointer funcPtr, @Cast(value="unsigned long long") long flags, @Cast(value="cudaDriverEntryPointQueryResult*") IntPointer driverStatus)
@Cast(value="cudaError_t") public static int cudaGetDriverEntryPoint(String symbol, @Cast(value="void**") @ByPtrPtr Pointer funcPtr, @Cast(value="unsigned long long") long flags, @Cast(value="cudaDriverEntryPointQueryResult*") IntBuffer driverStatus)
@Cast(value="cudaError_t") public static int cudaGetDriverEntryPoint(String symbol, @Cast(value="void**") @ByPtrPtr Pointer funcPtr, @Cast(value="unsigned long long") long flags)
@Cast(value="cudaError_t") public static int cudaGetDriverEntryPoint(@Cast(value="const char*") BytePointer symbol, @Cast(value="void**") @ByPtrPtr Pointer funcPtr, @Cast(value="unsigned long long") long flags, @Cast(value="cudaDriverEntryPointQueryResult*") int[] driverStatus)
@Cast(value="cudaError_t") public static int cudaGetDriverEntryPoint(String symbol, @Cast(value="void**") @ByPtrPtr Pointer funcPtr, @Cast(value="unsigned long long") long flags, @Cast(value="cudaDriverEntryPointQueryResult*") IntPointer driverStatus)
@Cast(value="cudaError_t") public static int cudaGetDriverEntryPoint(@Cast(value="const char*") BytePointer symbol, @Cast(value="void**") @ByPtrPtr Pointer funcPtr, @Cast(value="unsigned long long") long flags, @Cast(value="cudaDriverEntryPointQueryResult*") IntBuffer driverStatus)
@Cast(value="cudaError_t") public static int cudaGetDriverEntryPoint(String symbol, @Cast(value="void**") @ByPtrPtr Pointer funcPtr, @Cast(value="unsigned long long") long flags, @Cast(value="cudaDriverEntryPointQueryResult*") int[] driverStatus)
@Cast(value="cudaError_t") public static int cudaGetExportTable(@Cast(value="const void**") PointerPointer ppExportTable, @Const cudaUUID_t pExportTableId)
@Cast(value="cudaError_t") public static int cudaGetExportTable(@Cast(value="const void**") @ByPtrPtr Pointer ppExportTable, @Const cudaUUID_t pExportTableId)
@Cast(value="cudaError_t") public static int cudaGetFuncBySymbol(@ByPtrPtr CUfunc_st functionPtr, @Const Pointer symbolPtr)
functionPtr
- - Returns the device entry functionsymbolPtr
- - Pointer to device entry function to search for@Cast(value="cudaError_t") public static int cudaGetKernel(@ByPtrPtr CUkern_st kernelPtr, @Const Pointer entryFuncAddr)
kernelPtr
- - Returns the device kernelentryFuncAddr
- - Address of device entry function to search kernel for::cudaGetKernel(cudaKernel_t *kernelPtr, const T *entryFuncAddr) "cudaGetKernel (C++ API)"
@ByVal public static cudaPitchedPtr make_cudaPitchedPtr(Pointer d, @Cast(value="size_t") long p, @Cast(value="size_t") long xsz, @Cast(value="size_t") long ysz)
d
- - Pointer to allocated memoryp
- - Pitch of allocated memory in bytesxsz
- - Logical width of allocation in elementsysz
- - Logical height of allocation in elementsmake_cudaExtent, make_cudaPos
@ByVal public static cudaPos make_cudaPos(@Cast(value="size_t") long x, @Cast(value="size_t") long y, @Cast(value="size_t") long z)
x
- - X positiony
- - Y positionz
- - Z positionmake_cudaExtent, make_cudaPitchedPtr
@ByVal public static cudaExtent make_cudaExtent(@Cast(value="size_t") long w, @Cast(value="size_t") long h, @Cast(value="size_t") long d)
w
- - Width in elements when referring to array memory, in bytes when referring to linear memoryh
- - Height in elementsd
- - Depth in elementsmake_cudaPitchedPtr, make_cudaPos
@ByVal public static uchar2 make_uchar2(@Cast(value="unsigned char") byte x, @Cast(value="unsigned char") byte y)
@ByVal public static uchar3 make_uchar3(@Cast(value="unsigned char") byte x, @Cast(value="unsigned char") byte y, @Cast(value="unsigned char") byte z)
@ByVal public static uchar4 make_uchar4(@Cast(value="unsigned char") byte x, @Cast(value="unsigned char") byte y, @Cast(value="unsigned char") byte z, @Cast(value="unsigned char") byte w)
@ByVal public static ushort2 make_ushort2(@Cast(value="unsigned short") short x, @Cast(value="unsigned short") short y)
@ByVal public static ushort3 make_ushort3(@Cast(value="unsigned short") short x, @Cast(value="unsigned short") short y, @Cast(value="unsigned short") short z)
@ByVal public static ushort4 make_ushort4(@Cast(value="unsigned short") short x, @Cast(value="unsigned short") short y, @Cast(value="unsigned short") short z, @Cast(value="unsigned short") short w)
@ByVal public static uint2 make_uint2(@Cast(value="unsigned int") int x, @Cast(value="unsigned int") int y)
@ByVal public static uint3 make_uint3(@Cast(value="unsigned int") int x, @Cast(value="unsigned int") int y, @Cast(value="unsigned int") int z)
@ByVal public static uint4 make_uint4(@Cast(value="unsigned int") int x, @Cast(value="unsigned int") int y, @Cast(value="unsigned int") int z, @Cast(value="unsigned int") int w)
@ByVal public static ulong2 make_ulong2(@Cast(value="unsigned long int") long x, @Cast(value="unsigned long int") long y)
@ByVal public static ulong3 make_ulong3(@Cast(value="unsigned long int") long x, @Cast(value="unsigned long int") long y, @Cast(value="unsigned long int") long z)
@ByVal public static ulong4 make_ulong4(@Cast(value="unsigned long int") long x, @Cast(value="unsigned long int") long y, @Cast(value="unsigned long int") long z, @Cast(value="unsigned long int") long w)
@ByVal public static ulonglong1 make_ulonglong1(@Cast(value="unsigned long long int") long x)
@ByVal public static ulonglong2 make_ulonglong2(@Cast(value="unsigned long long int") long x, @Cast(value="unsigned long long int") long y)
@ByVal public static ulonglong3 make_ulonglong3(@Cast(value="unsigned long long int") long x, @Cast(value="unsigned long long int") long y, @Cast(value="unsigned long long int") long z)
@ByVal public static ulonglong4 make_ulonglong4(@Cast(value="unsigned long long int") long x, @Cast(value="unsigned long long int") long y, @Cast(value="unsigned long long int") long z, @Cast(value="unsigned long long int") long w)
@ByVal @Cast(value="cuFloatComplex*") public static float2 make_cuFloatComplex(float r, float i)
@ByVal @Cast(value="cuFloatComplex*") public static float2 cuConjf(@ByVal @Cast(value="cuFloatComplex*") float2 x)
@ByVal @Cast(value="cuFloatComplex*") public static float2 cuCaddf(@ByVal @Cast(value="cuFloatComplex*") float2 x, @ByVal @Cast(value="cuFloatComplex*") float2 y)
@ByVal @Cast(value="cuFloatComplex*") public static float2 cuCsubf(@ByVal @Cast(value="cuFloatComplex*") float2 x, @ByVal @Cast(value="cuFloatComplex*") float2 y)
@ByVal @Cast(value="cuFloatComplex*") public static float2 cuCmulf(@ByVal @Cast(value="cuFloatComplex*") float2 x, @ByVal @Cast(value="cuFloatComplex*") float2 y)
@ByVal @Cast(value="cuFloatComplex*") public static float2 cuCdivf(@ByVal @Cast(value="cuFloatComplex*") float2 x, @ByVal @Cast(value="cuFloatComplex*") float2 y)
@ByVal @Cast(value="cuDoubleComplex*") public static double2 make_cuDoubleComplex(double r, double i)
@ByVal @Cast(value="cuDoubleComplex*") public static double2 cuConj(@ByVal @Cast(value="cuDoubleComplex*") double2 x)
@ByVal @Cast(value="cuDoubleComplex*") public static double2 cuCadd(@ByVal @Cast(value="cuDoubleComplex*") double2 x, @ByVal @Cast(value="cuDoubleComplex*") double2 y)
@ByVal @Cast(value="cuDoubleComplex*") public static double2 cuCsub(@ByVal @Cast(value="cuDoubleComplex*") double2 x, @ByVal @Cast(value="cuDoubleComplex*") double2 y)
@ByVal @Cast(value="cuDoubleComplex*") public static double2 cuCmul(@ByVal @Cast(value="cuDoubleComplex*") double2 x, @ByVal @Cast(value="cuDoubleComplex*") double2 y)
@ByVal @Cast(value="cuDoubleComplex*") public static double2 cuCdiv(@ByVal @Cast(value="cuDoubleComplex*") double2 x, @ByVal @Cast(value="cuDoubleComplex*") double2 y)
@ByVal @Cast(value="cuComplex*") public static float2 make_cuComplex(float x, float y)
@ByVal @Cast(value="cuDoubleComplex*") public static double2 cuComplexFloatToDouble(@ByVal @Cast(value="cuFloatComplex*") float2 c)
@ByVal @Cast(value="cuFloatComplex*") public static float2 cuComplexDoubleToFloat(@ByVal @Cast(value="cuDoubleComplex*") double2 c)
@ByVal @Cast(value="cuComplex*") public static float2 cuCfmaf(@ByVal @Cast(value="cuComplex*") float2 x, @ByVal @Cast(value="cuComplex*") float2 y, @ByVal @Cast(value="cuComplex*") float2 d)
@ByVal @Cast(value="cuDoubleComplex*") public static double2 cuCfma(@ByVal @Cast(value="cuDoubleComplex*") double2 x, @ByVal @Cast(value="cuDoubleComplex*") double2 y, @ByVal @Cast(value="cuDoubleComplex*") double2 d)
@ByVal public static __half __double2half(double a)
a
- [in] - double. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __float2half(float a)
a
- [in] - float. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __float2half_rn(float a)
a
- [in] - float. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __float2half_rz(float a)
a
- [in] - float. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __float2half_rd(float a)
a
- [in] - float. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __float2half_ru(float a)
a
- [in] - float. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternalpublic static float __half2float(@Const @ByVal __half a)
a
- [in] - float. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __float2half2_rn(float a)
a
- [in] - float. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __floats2half2_rn(float a, float b)
a
- [in] - float. Is only being read.b
- [in] - float. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternalpublic static float __low2float(@Const @ByVal __half2 a)
a
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternalpublic static float __high2float(@Const @ByVal __half2 a)
a
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternalpublic static byte __half2char_rz(@Const @ByVal __half h)
h
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned char") public static byte __half2uchar_rz(@Const @ByVal __half h)
h
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternalpublic static short __half2short_rz(@Const @ByVal __half h)
h
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned short int") public static short __half2ushort_rz(@Const @ByVal __half h)
h
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternalpublic static int __half2int_rz(@Const @ByVal __half h)
h
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned int") public static int __half2uint_rz(@Const @ByVal __half h)
h
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternalpublic static long __half2ll_rz(@Const @ByVal __half h)
h
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned long long int") public static long __half2ull_rz(@Const @ByVal __half h)
h
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 make_half2(@Const @ByVal __half x, @Const @ByVal __half y)
x
- [in] - half. Is only being read.y
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __float22half2_rn(@Const @ByVal float2 a)
a
- [in] - float2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static float2 __half22float2(@Const @ByVal __half2 a)
a
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __int2half_rn(int i)
i
- [in] - int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __int2half_rz(int i)
i
- [in] - int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __int2half_rd(int i)
i
- [in] - int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __int2half_ru(int i)
i
- [in] - int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __short2half_rn(short i)
i
- [in] - short int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __short2half_rz(short i)
i
- [in] - short int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __short2half_rd(short i)
i
- [in] - short int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __short2half_ru(short i)
i
- [in] - short int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __uint2half_rn(@Cast(value="const unsigned int") int i)
i
- [in] - unsigned int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __uint2half_rz(@Cast(value="const unsigned int") int i)
i
- [in] - unsigned int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __uint2half_rd(@Cast(value="const unsigned int") int i)
i
- [in] - unsigned int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __uint2half_ru(@Cast(value="const unsigned int") int i)
i
- [in] - unsigned int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __ushort2half_rn(@Cast(value="const unsigned short int") short i)
i
- [in] - unsigned short int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __ushort2half_rz(@Cast(value="const unsigned short int") short i)
i
- [in] - unsigned short int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __ushort2half_rd(@Cast(value="const unsigned short int") short i)
i
- [in] - unsigned short int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __ushort2half_ru(@Cast(value="const unsigned short int") short i)
i
- [in] - unsigned short int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __ull2half_rn(@Cast(value="const unsigned long long int") long i)
i
- [in] - unsigned long long int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __ull2half_rz(@Cast(value="const unsigned long long int") long i)
i
- [in] - unsigned long long int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __ull2half_rd(@Cast(value="const unsigned long long int") long i)
i
- [in] - unsigned long long int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __ull2half_ru(@Cast(value="const unsigned long long int") long i)
i
- [in] - unsigned long long int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __ll2half_rn(long i)
i
- [in] - long long int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __ll2half_rz(long i)
i
- [in] - long long int. Is only being read.@ByVal public static __half __ll2half_rd(long i)
i
- [in] - long long int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __ll2half_ru(long i)
i
- [in] - long long int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __half2half2(@Const @ByVal __half a)
a
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __lowhigh2highlow(@Const @ByVal __half2 a)
a
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __lows2half2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __highs2half2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __high2half(@Const @ByVal __half2 a)
a
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __low2half(@Const @ByVal __half2 a)
a
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternalpublic static int __hisinf(@Const @ByVal __half a)
a
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __halves2half2(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __low2half2(@Const @ByVal __half2 a)
a
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __high2half2(@Const @ByVal __half2 a)
a
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternalpublic static short __half_as_short(@Const @ByVal __half h)
h
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned short int") public static short __half_as_ushort(@Const @ByVal __half h)
h
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __short_as_half(short i)
i
- [in] - short int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __ushort_as_half(@Cast(value="const unsigned short int") short i)
i
- [in] - unsigned short int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __hmax(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __hmin(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hmax2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hmin2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __heq2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hne2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hle2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hge2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hlt2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hgt2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hequ2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hneu2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hleu2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hgeu2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hltu2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hgtu2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __heq2_mask(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hne2_mask(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hle2_mask(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hge2_mask(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hlt2_mask(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hgt2_mask(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hequ2_mask(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hneu2_mask(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hleu2_mask(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hgeu2_mask(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hltu2_mask(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hgtu2_mask(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hisnan2(@Const @ByVal __half2 a)
a
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hadd2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hsub2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hmul2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hadd2_rn(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hsub2_rn(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hmul2_rn(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __h2div(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __habs2(@Const @ByVal __half2 a)
a
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hadd2_sat(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hsub2_sat(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hmul2_sat(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hneg2(@Const @ByVal __half2 a)
a
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __habs(@Const @ByVal __half a)
a
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __hadd(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __hsub(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __hmul(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.@ByVal public static __half __hadd_rn(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __hsub_rn(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __hmul_rn(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.@ByVal public static __half __hdiv(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __hadd_sat(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __hsub_sat(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __hmul_sat(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __hneg(@Const @ByVal __half a)
a
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hbeq2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hbne2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hble2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hbge2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hblt2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hbgt2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hbequ2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hbneu2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hbleu2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hbgeu2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hbltu2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hbgtu2(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __heq(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hne(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hle(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hge(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hlt(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hgt(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hequ(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hneu(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hleu(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hgeu(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hltu(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hgtu(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hisnan(@Const @ByVal __half a)
a
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __hmax_nan(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half __hmin_nan(@Const @ByVal __half a, @Const @ByVal __half b)
a
- [in] - half. Is only being read.b
- [in] - half. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hmax2_nan(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __half2 __hmin2_nan(@Const @ByVal __half2 a, @Const @ByVal __half2 b)
a
- [in] - half2. Is only being read.b
- [in] - half2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal @Name(value="operator +") public static __half add(@Const @ByRef __half lh, @Const @ByRef __half rh)
@ByVal @Name(value="operator -") public static __half subtract(@Const @ByRef __half lh, @Const @ByRef __half rh)
@ByVal @Name(value="operator *") public static __half multiply(@Const @ByRef __half lh, @Const @ByRef __half rh)
@ByVal @Name(value="operator /") public static __half divide(@Const @ByRef __half lh, @Const @ByRef __half rh)
@ByRef @Name(value="operator +=") public static __half addPut(@ByRef __half lh, @Const @ByRef __half rh)
@ByRef @Name(value="operator -=") public static __half subtractPut(@ByRef __half lh, @Const @ByRef __half rh)
@ByRef @Name(value="operator *=") public static __half multiplyPut(@ByRef __half lh, @Const @ByRef __half rh)
@ByRef @Name(value="operator /=") public static __half dividePut(@ByRef __half lh, @Const @ByRef __half rh)
@ByRef @Name(value="operator ++") public static __half increment(@ByRef __half h)
@ByRef @Name(value="operator --") public static __half decrement(@ByRef __half h)
@ByVal @Name(value="operator ++") public static __half increment(@ByRef __half h, int ignored)
@ByVal @Name(value="operator --") public static __half decrement(@ByRef __half h, int ignored)
@ByVal @Name(value="operator +") public static __half add(@Const @ByRef __half h)
@ByVal @Name(value="operator -") public static __half subtract(@Const @ByRef __half h)
@Cast(value="bool") @Name(value="operator ==") public static boolean equals(@Const @ByRef __half lh, @Const @ByRef __half rh)
@Cast(value="bool") @Name(value="operator !=") public static boolean notEquals(@Const @ByRef __half lh, @Const @ByRef __half rh)
@Cast(value="bool") @Name(value="operator >") public static boolean greaterThan(@Const @ByRef __half lh, @Const @ByRef __half rh)
@Cast(value="bool") @Name(value="operator <") public static boolean lessThan(@Const @ByRef __half lh, @Const @ByRef __half rh)
@Cast(value="bool") @Name(value="operator >=") public static boolean greaterThanEquals(@Const @ByRef __half lh, @Const @ByRef __half rh)
@Cast(value="bool") @Name(value="operator <=") public static boolean lessThanEquals(@Const @ByRef __half lh, @Const @ByRef __half rh)
@ByVal @Name(value="operator +") public static __half2 add(@Const @ByRef __half2 lh, @Const @ByRef __half2 rh)
@ByVal @Name(value="operator -") public static __half2 subtract(@Const @ByRef __half2 lh, @Const @ByRef __half2 rh)
@ByVal @Name(value="operator *") public static __half2 multiply(@Const @ByRef __half2 lh, @Const @ByRef __half2 rh)
@ByVal @Name(value="operator /") public static __half2 divide(@Const @ByRef __half2 lh, @Const @ByRef __half2 rh)
@ByRef @Name(value="operator +=") public static __half2 addPut(@ByRef __half2 lh, @Const @ByRef __half2 rh)
@ByRef @Name(value="operator -=") public static __half2 subtractPut(@ByRef __half2 lh, @Const @ByRef __half2 rh)
@ByRef @Name(value="operator *=") public static __half2 multiplyPut(@ByRef __half2 lh, @Const @ByRef __half2 rh)
@ByRef @Name(value="operator /=") public static __half2 dividePut(@ByRef __half2 lh, @Const @ByRef __half2 rh)
@ByRef @Name(value="operator ++") public static __half2 increment(@ByRef __half2 h)
@ByRef @Name(value="operator --") public static __half2 decrement(@ByRef __half2 h)
@ByVal @Name(value="operator ++") public static __half2 increment(@ByRef __half2 h, int ignored)
@ByVal @Name(value="operator --") public static __half2 decrement(@ByRef __half2 h, int ignored)
@ByVal @Name(value="operator +") public static __half2 add(@Const @ByRef __half2 h)
@ByVal @Name(value="operator -") public static __half2 subtract(@Const @ByRef __half2 h)
@Cast(value="bool") @Name(value="operator ==") public static boolean equals(@Const @ByRef __half2 lh, @Const @ByRef __half2 rh)
@Cast(value="bool") @Name(value="operator !=") public static boolean notEquals(@Const @ByRef __half2 lh, @Const @ByRef __half2 rh)
@Cast(value="bool") @Name(value="operator >") public static boolean greaterThan(@Const @ByRef __half2 lh, @Const @ByRef __half2 rh)
@Cast(value="bool") @Name(value="operator <") public static boolean lessThan(@Const @ByRef __half2 lh, @Const @ByRef __half2 rh)
@Cast(value="bool") @Name(value="operator >=") public static boolean greaterThanEquals(@Const @ByRef __half2 lh, @Const @ByRef __half2 rh)
@Cast(value="bool") @Name(value="operator <=") public static boolean lessThanEquals(@Const @ByRef __half2 lh, @Const @ByRef __half2 rh)
@MemberGetter public static int CUDART_INF_FP16()
@MemberGetter public static int CUDART_NAN_FP16()
@MemberGetter public static int CUDART_MIN_DENORM_FP16()
@MemberGetter public static int CUDART_MAX_NORMAL_FP16()
@MemberGetter public static int CUDART_NEG_ZERO_FP16()
@MemberGetter public static int CUDART_ZERO_FP16()
@MemberGetter public static int CUDART_ONE_FP16()
@Cast(value="unsigned short") public static short __internal_float2half(float f, @Cast(value="unsigned int*") @ByRef IntPointer sign, @Cast(value="unsigned int*") @ByRef IntPointer remainder)
@Cast(value="unsigned short") public static short __internal_float2half(float f, @Cast(value="unsigned int*") @ByRef IntBuffer sign, @Cast(value="unsigned int*") @ByRef IntBuffer remainder)
@Cast(value="unsigned short") public static short __internal_float2half(float f, @Cast(value="unsigned int*") @ByRef int[] sign, @Cast(value="unsigned int*") @ByRef int[] remainder)
public static float __internal_half2float(@Cast(value="const unsigned short") short h)
@ByVal public static __nv_bfloat16 __double2bfloat16(double a)
a
- [in] - double. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __float2bfloat16(float a)
a
- [in] - float. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __float2bfloat16_rn(float a)
a
- [in] - float. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __float2bfloat16_rz(float a)
a
- [in] - float. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __float2bfloat16_rd(float a)
a
- [in] - float. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __float2bfloat16_ru(float a)
a
- [in] - float. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternalpublic static float __bfloat162float(@Const @ByVal __nv_bfloat16 a)
a
- [in] - float. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __float2bfloat162_rn(float a)
a
- [in] - float. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __floats2bfloat162_rn(float a, float b)
a
- [in] - float. Is only being read.b
- [in] - float. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternalpublic static float __low2float(@Const @ByVal __nv_bfloat162 a)
a
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternalpublic static float __high2float(@Const @ByVal __nv_bfloat162 a)
a
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __float22bfloat162_rn(@Const @ByVal float2 a)
a
- [in] - float2. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static float2 __bfloat1622float2(@Const @ByVal __nv_bfloat162 a)
a
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternalpublic static byte __bfloat162char_rz(@Const @ByVal __nv_bfloat16 h)
h
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned char") public static byte __bfloat162uchar_rz(@Const @ByVal __nv_bfloat16 h)
h
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternalpublic static int __bfloat162int_rz(@Const @ByVal __nv_bfloat16 h)
h
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __int2bfloat16_rn(int i)
i
- [in] - int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternalpublic static short __bfloat162short_rz(@Const @ByVal __nv_bfloat16 h)
h
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __short2bfloat16_rn(short i)
i
- [in] - short int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned int") public static int __bfloat162uint_rz(@Const @ByVal __nv_bfloat16 h)
h
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __uint2bfloat16_rn(@Cast(value="const unsigned int") int i)
i
- [in] - unsigned int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned short int") public static short __bfloat162ushort_rz(@Const @ByVal __nv_bfloat16 h)
h
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __ushort2bfloat16_rn(@Cast(value="const unsigned short int") short i)
i
- [in] - unsigned short int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned long long int") public static long __bfloat162ull_rz(@Const @ByVal __nv_bfloat16 h)
h
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 make_bfloat162(@Const @ByVal __nv_bfloat16 x, @Const @ByVal __nv_bfloat16 y)
x
- [in] - nv_bfloat16. Is only being read.y
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __ull2bfloat16_rn(@Cast(value="const unsigned long long int") long i)
i
- [in] - unsigned long long int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternalpublic static long __bfloat162ll_rz(@Const @ByVal __nv_bfloat16 h)
h
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __ll2bfloat16_rn(long i)
i
- [in] - long long int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __bfloat162bfloat162(@Const @ByVal __nv_bfloat16 a)
a
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __lowhigh2highlow(@Const @ByVal __nv_bfloat162 a)
a
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __lows2bfloat162(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __highs2bfloat162(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __high2bfloat16(@Const @ByVal __nv_bfloat162 a)
a
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __low2bfloat16(@Const @ByVal __nv_bfloat162 a)
a
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternalpublic static int __hisinf(@Const @ByVal __nv_bfloat16 a)
a
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __halves2bfloat162(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __low2bfloat162(@Const @ByVal __nv_bfloat162 a)
a
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __high2bfloat162(@Const @ByVal __nv_bfloat162 a)
a
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternalpublic static short __bfloat16_as_short(@Const @ByVal __nv_bfloat16 h)
h
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned short int") public static short __bfloat16_as_ushort(@Const @ByVal __nv_bfloat16 h)
h
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __short_as_bfloat16(short i)
i
- [in] - short int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __ushort_as_bfloat16(@Cast(value="const unsigned short int") short i)
i
- [in] - unsigned short int. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __heq2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hne2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hle2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hge2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hlt2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hgt2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hequ2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hneu2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hleu2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hgeu2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hltu2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hgtu2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __heq2_mask(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hne2_mask(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hle2_mask(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hge2_mask(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hlt2_mask(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hgt2_mask(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hequ2_mask(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hneu2_mask(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hleu2_mask(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hgeu2_mask(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hltu2_mask(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="unsigned") public static int __hgtu2_mask(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hisnan2(@Const @ByVal __nv_bfloat162 a)
a
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hadd2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hsub2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hmul2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hadd2_rn(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hsub2_rn(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hmul2_rn(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __h2div(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __habs2(@Const @ByVal __nv_bfloat162 a)
a
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hadd2_sat(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hsub2_sat(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hmul2_sat(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hneg2(@Const @ByVal __nv_bfloat162 a)
a
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __habs(@Const @ByVal __nv_bfloat16 a)
a
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __hadd(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __hsub(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __hmul(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.@ByVal public static __nv_bfloat16 __hadd_rn(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __hsub_rn(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __hmul_rn(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.@ByVal public static __nv_bfloat16 __hdiv(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __hadd_sat(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __hsub_sat(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __hmul_sat(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __hneg(@Const @ByVal __nv_bfloat16 a)
a
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hbeq2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hbne2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hble2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hbge2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hblt2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hbgt2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hbequ2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hbneu2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hbleu2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hbgeu2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hbltu2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hbgtu2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __heq(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hne(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hle(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hge(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hlt(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hgt(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hequ(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hneu(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hleu(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hgeu(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hltu(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hgtu(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@Cast(value="bool") public static boolean __hisnan(@Const @ByVal __nv_bfloat16 a)
a
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __hmax(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __hmin(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __hmax_nan(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat16 __hmin_nan(@Const @ByVal __nv_bfloat16 a, @Const @ByVal __nv_bfloat16 b)
a
- [in] - nv_bfloat16. Is only being read.b
- [in] - nv_bfloat16. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hmax2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hmin2(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hmax2_nan(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal public static __nv_bfloat162 __hmin2_nan(@Const @ByVal __nv_bfloat162 a, @Const @ByVal __nv_bfloat162 b)
a
- [in] - nv_bfloat162. Is only being read.b
- [in] - nv_bfloat162. Is only being read.-guarantee
- no-throw guarantee
\behavior reentrant, thread safe
\endinternal@ByVal @Name(value="operator +") public static __nv_bfloat16 add(@Const @ByRef __nv_bfloat16 lh, @Const @ByRef __nv_bfloat16 rh)
@ByVal @Name(value="operator -") public static __nv_bfloat16 subtract(@Const @ByRef __nv_bfloat16 lh, @Const @ByRef __nv_bfloat16 rh)
@ByVal @Name(value="operator *") public static __nv_bfloat16 multiply(@Const @ByRef __nv_bfloat16 lh, @Const @ByRef __nv_bfloat16 rh)
@ByVal @Name(value="operator /") public static __nv_bfloat16 divide(@Const @ByRef __nv_bfloat16 lh, @Const @ByRef __nv_bfloat16 rh)
@ByRef @Name(value="operator +=") public static __nv_bfloat16 addPut(@ByRef __nv_bfloat16 lh, @Const @ByRef __nv_bfloat16 rh)
@ByRef @Name(value="operator -=") public static __nv_bfloat16 subtractPut(@ByRef __nv_bfloat16 lh, @Const @ByRef __nv_bfloat16 rh)
@ByRef @Name(value="operator *=") public static __nv_bfloat16 multiplyPut(@ByRef __nv_bfloat16 lh, @Const @ByRef __nv_bfloat16 rh)
@ByRef @Name(value="operator /=") public static __nv_bfloat16 dividePut(@ByRef __nv_bfloat16 lh, @Const @ByRef __nv_bfloat16 rh)
@ByRef @Name(value="operator ++") public static __nv_bfloat16 increment(@ByRef __nv_bfloat16 h)
@ByRef @Name(value="operator --") public static __nv_bfloat16 decrement(@ByRef __nv_bfloat16 h)
@ByVal @Name(value="operator ++") public static __nv_bfloat16 increment(@ByRef __nv_bfloat16 h, int ignored)
@ByVal @Name(value="operator --") public static __nv_bfloat16 decrement(@ByRef __nv_bfloat16 h, int ignored)
@ByVal @Name(value="operator +") public static __nv_bfloat16 add(@Const @ByRef __nv_bfloat16 h)
@ByVal @Name(value="operator -") public static __nv_bfloat16 subtract(@Const @ByRef __nv_bfloat16 h)
@Cast(value="bool") @Name(value="operator ==") public static boolean equals(@Const @ByRef __nv_bfloat16 lh, @Const @ByRef __nv_bfloat16 rh)
@Cast(value="bool") @Name(value="operator !=") public static boolean notEquals(@Const @ByRef __nv_bfloat16 lh, @Const @ByRef __nv_bfloat16 rh)
@Cast(value="bool") @Name(value="operator >") public static boolean greaterThan(@Const @ByRef __nv_bfloat16 lh, @Const @ByRef __nv_bfloat16 rh)
@Cast(value="bool") @Name(value="operator <") public static boolean lessThan(@Const @ByRef __nv_bfloat16 lh, @Const @ByRef __nv_bfloat16 rh)
@Cast(value="bool") @Name(value="operator >=") public static boolean greaterThanEquals(@Const @ByRef __nv_bfloat16 lh, @Const @ByRef __nv_bfloat16 rh)
@Cast(value="bool") @Name(value="operator <=") public static boolean lessThanEquals(@Const @ByRef __nv_bfloat16 lh, @Const @ByRef __nv_bfloat16 rh)
@ByVal @Name(value="operator +") public static __nv_bfloat162 add(@Const @ByRef __nv_bfloat162 lh, @Const @ByRef __nv_bfloat162 rh)
@ByVal @Name(value="operator -") public static __nv_bfloat162 subtract(@Const @ByRef __nv_bfloat162 lh, @Const @ByRef __nv_bfloat162 rh)
@ByVal @Name(value="operator *") public static __nv_bfloat162 multiply(@Const @ByRef __nv_bfloat162 lh, @Const @ByRef __nv_bfloat162 rh)
@ByVal @Name(value="operator /") public static __nv_bfloat162 divide(@Const @ByRef __nv_bfloat162 lh, @Const @ByRef __nv_bfloat162 rh)
@ByRef @Name(value="operator +=") public static __nv_bfloat162 addPut(@ByRef __nv_bfloat162 lh, @Const @ByRef __nv_bfloat162 rh)
@ByRef @Name(value="operator -=") public static __nv_bfloat162 subtractPut(@ByRef __nv_bfloat162 lh, @Const @ByRef __nv_bfloat162 rh)
@ByRef @Name(value="operator *=") public static __nv_bfloat162 multiplyPut(@ByRef __nv_bfloat162 lh, @Const @ByRef __nv_bfloat162 rh)
@ByRef @Name(value="operator /=") public static __nv_bfloat162 dividePut(@ByRef __nv_bfloat162 lh, @Const @ByRef __nv_bfloat162 rh)
@ByRef @Name(value="operator ++") public static __nv_bfloat162 increment(@ByRef __nv_bfloat162 h)
@ByRef @Name(value="operator --") public static __nv_bfloat162 decrement(@ByRef __nv_bfloat162 h)
@ByVal @Name(value="operator ++") public static __nv_bfloat162 increment(@ByRef __nv_bfloat162 h, int ignored)
@ByVal @Name(value="operator --") public static __nv_bfloat162 decrement(@ByRef __nv_bfloat162 h, int ignored)
@ByVal @Name(value="operator +") public static __nv_bfloat162 add(@Const @ByRef __nv_bfloat162 h)
@ByVal @Name(value="operator -") public static __nv_bfloat162 subtract(@Const @ByRef __nv_bfloat162 h)
@Cast(value="bool") @Name(value="operator ==") public static boolean equals(@Const @ByRef __nv_bfloat162 lh, @Const @ByRef __nv_bfloat162 rh)
@Cast(value="bool") @Name(value="operator !=") public static boolean notEquals(@Const @ByRef __nv_bfloat162 lh, @Const @ByRef __nv_bfloat162 rh)
@Cast(value="bool") @Name(value="operator >") public static boolean greaterThan(@Const @ByRef __nv_bfloat162 lh, @Const @ByRef __nv_bfloat162 rh)
@Cast(value="bool") @Name(value="operator <") public static boolean lessThan(@Const @ByRef __nv_bfloat162 lh, @Const @ByRef __nv_bfloat162 rh)
@Cast(value="bool") @Name(value="operator >=") public static boolean greaterThanEquals(@Const @ByRef __nv_bfloat162 lh, @Const @ByRef __nv_bfloat162 rh)
@Cast(value="bool") @Name(value="operator <=") public static boolean lessThanEquals(@Const @ByRef __nv_bfloat162 lh, @Const @ByRef __nv_bfloat162 rh)
@MemberGetter public static int CUDART_INF_BF16()
@MemberGetter public static int CUDART_NAN_BF16()
@MemberGetter public static int CUDART_MIN_DENORM_BF16()
@MemberGetter public static int CUDART_MAX_NORMAL_BF16()
@MemberGetter public static int CUDART_NEG_ZERO_BF16()
@MemberGetter public static int CUDART_ZERO_BF16()
@MemberGetter public static int CUDART_ONE_BF16()
@Cast(value="unsigned int") public static int __internal_float_as_uint(float f)
public static float __internal_uint_as_float(@Cast(value="const unsigned int") int u)
@Cast(value="unsigned short") public static short __internal_float2bfloat16(float f, @Cast(value="unsigned int*") @ByRef IntPointer sign, @Cast(value="unsigned int*") @ByRef IntPointer remainder)
@Cast(value="unsigned short") public static short __internal_float2bfloat16(float f, @Cast(value="unsigned int*") @ByRef IntBuffer sign, @Cast(value="unsigned int*") @ByRef IntBuffer remainder)
@Cast(value="unsigned short") public static short __internal_float2bfloat16(float f, @Cast(value="unsigned int*") @ByRef int[] sign, @Cast(value="unsigned int*") @ByRef int[] remainder)
public static float __internal_double2float_rn(double x)
public static double __internal_float2double(float x)
public static float __internal_bfloat162float(@Cast(value="const unsigned short") short h)
public static int __internal_bfloat162int_rz(@Const @ByVal __nv_bfloat16 h)
@Cast(value="unsigned int") public static int __internal_bfloat162uint_rz(@Const @ByVal __nv_bfloat16 h)
@Cast(value="CUresult") public static int cuGraphicsGLRegisterBuffer(@ByPtrPtr CUgraphicsResource_st pCudaResource, @Cast(value="GLuint") int buffer, @Cast(value="unsigned int") int Flags)
pCudaResource
- - Pointer to the returned object handlebuffer
- - name of buffer object to be registeredFlags
- - Register flags::cuGraphicsUnregisterResource,
::cuGraphicsMapResources,
::cuGraphicsResourceGetMappedPointer,
::cudaGraphicsGLRegisterBuffer
@Cast(value="CUresult") public static int cuGraphicsGLRegisterImage(@ByPtrPtr CUgraphicsResource_st pCudaResource, @Cast(value="GLuint") int image, @Cast(value="GLenum") int target, @Cast(value="unsigned int") int Flags)
pCudaResource
- - Pointer to the returned object handleimage
- - name of texture or renderbuffer object to be registeredtarget
- - Identifies the type of object specified by \p imageFlags
- - Register flags::cuGraphicsUnregisterResource,
::cuGraphicsMapResources,
::cuGraphicsSubResourceGetMappedArray,
::cudaGraphicsGLRegisterImage
@Cast(value="CUresult") public static int cuGLGetDevices(@Cast(value="unsigned int*") IntPointer pCudaDeviceCount, @Cast(value="CUdevice*") IntPointer pCudaDevices, @Cast(value="unsigned int") int cudaDeviceCount, @Cast(value="CUGLDeviceList") int deviceList)
pCudaDeviceCount
- - Returned number of CUDA devices.pCudaDevices
- - Returned CUDA devices.cudaDeviceCount
- - The size of the output device array pCudaDevices.deviceList
- - The set of devices to return.::cuWGLGetDevice,
::cudaGLGetDevices
@Cast(value="CUresult") public static int cuGLGetDevices(@Cast(value="unsigned int*") IntBuffer pCudaDeviceCount, @Cast(value="CUdevice*") IntBuffer pCudaDevices, @Cast(value="unsigned int") int cudaDeviceCount, @Cast(value="CUGLDeviceList") int deviceList)
@Cast(value="CUresult") public static int cuGLGetDevices(@Cast(value="unsigned int*") int[] pCudaDeviceCount, @Cast(value="CUdevice*") int[] pCudaDevices, @Cast(value="unsigned int") int cudaDeviceCount, @Cast(value="CUGLDeviceList") int deviceList)
@Cast(value="CUresult") @Deprecated public static int cuGLCtxCreate(@ByPtrPtr CUctx_st pCtx, @Cast(value="unsigned int") int Flags, @Cast(value="CUdevice") int device)
pCtx
- - Returned CUDA contextFlags
- - Options for CUDA context creationdevice
- - Device on which to create the context::cuCtxCreate, ::cuGLInit, ::cuGLMapBufferObject,
::cuGLRegisterBufferObject, ::cuGLUnmapBufferObject,
::cuGLUnregisterBufferObject, ::cuGLMapBufferObjectAsync,
::cuGLUnmapBufferObjectAsync, ::cuGLSetBufferObjectMapFlags,
::cuWGLGetDevice
@Cast(value="CUresult") @Deprecated public static int cuGLInit()
::cuGLMapBufferObject,
::cuGLRegisterBufferObject, ::cuGLUnmapBufferObject,
::cuGLUnregisterBufferObject, ::cuGLMapBufferObjectAsync,
::cuGLUnmapBufferObjectAsync, ::cuGLSetBufferObjectMapFlags,
::cuWGLGetDevice
@Cast(value="CUresult") @Deprecated public static int cuGLRegisterBufferObject(@Cast(value="GLuint") int buffer)
buffer
- - The name of the buffer object to register.::cuGraphicsGLRegisterBuffer
@Cast(value="CUresult") @Deprecated public static int cuGLMapBufferObject(@Cast(value="CUdeviceptr*") LongPointer dptr, @Cast(value="size_t*") SizeTPointer size, @Cast(value="GLuint") int buffer)
dptr
- - Returned mapped base pointersize
- - Returned size of mappingbuffer
- - The name of the buffer object to map::cuGraphicsMapResources
@Cast(value="CUresult") @Deprecated public static int cuGLMapBufferObject(@Cast(value="CUdeviceptr*") LongBuffer dptr, @Cast(value="size_t*") SizeTPointer size, @Cast(value="GLuint") int buffer)
@Cast(value="CUresult") @Deprecated public static int cuGLMapBufferObject(@Cast(value="CUdeviceptr*") long[] dptr, @Cast(value="size_t*") SizeTPointer size, @Cast(value="GLuint") int buffer)
@Cast(value="CUresult") @Deprecated public static int cuGLUnmapBufferObject(@Cast(value="GLuint") int buffer)
buffer
- - Buffer object to unmap::cuGraphicsUnmapResources
@Cast(value="CUresult") @Deprecated public static int cuGLUnregisterBufferObject(@Cast(value="GLuint") int buffer)
buffer
- - Name of the buffer object to unregister::cuGraphicsUnregisterResource
@Cast(value="CUresult") @Deprecated public static int cuGLSetBufferObjectMapFlags(@Cast(value="GLuint") int buffer, @Cast(value="unsigned int") int Flags)
buffer
- - Buffer object to unmapFlags
- - Map flags::cuGraphicsResourceSetMapFlags
@Cast(value="CUresult") @Deprecated public static int cuGLMapBufferObjectAsync(@Cast(value="CUdeviceptr*") LongPointer dptr, @Cast(value="size_t*") SizeTPointer size, @Cast(value="GLuint") int buffer, CUstream_st hStream)
dptr
- - Returned mapped base pointersize
- - Returned size of mappingbuffer
- - The name of the buffer object to maphStream
- - Stream to synchronize::cuGraphicsMapResources
@Cast(value="CUresult") @Deprecated public static int cuGLMapBufferObjectAsync(@Cast(value="CUdeviceptr*") LongBuffer dptr, @Cast(value="size_t*") SizeTPointer size, @Cast(value="GLuint") int buffer, CUstream_st hStream)
@Cast(value="CUresult") @Deprecated public static int cuGLMapBufferObjectAsync(@Cast(value="CUdeviceptr*") long[] dptr, @Cast(value="size_t*") SizeTPointer size, @Cast(value="GLuint") int buffer, CUstream_st hStream)
@Cast(value="CUresult") @Deprecated public static int cuGLUnmapBufferObjectAsync(@Cast(value="GLuint") int buffer, CUstream_st hStream)
buffer
- - Name of the buffer object to unmaphStream
- - Stream to synchronize::cuGraphicsUnmapResources
@Cast(value="cudaError_t") public static int cudaGLGetDevices(@Cast(value="unsigned int*") IntPointer pCudaDeviceCount, IntPointer pCudaDevices, @Cast(value="unsigned int") int cudaDeviceCount, @Cast(value="cudaGLDeviceList") int deviceList)
pCudaDeviceCount
- - Returned number of CUDA devices corresponding to the
current OpenGL contextpCudaDevices
- - Returned CUDA devices corresponding to the current
OpenGL contextcudaDeviceCount
- - The size of the output device array \p pCudaDevicesdeviceList
- - The set of devices to return. This set may be
::cudaGLDeviceListAll for all devices,
::cudaGLDeviceListCurrentFrame for the devices used to
render the current frame (in SLI), or
::cudaGLDeviceListNextFrame for the devices used to
render the next frame (in SLI).::cudaGraphicsUnregisterResource,
::cudaGraphicsMapResources,
::cudaGraphicsSubResourceGetMappedArray,
::cudaGraphicsResourceGetMappedPointer,
::cuGLGetDevices
@Cast(value="cudaError_t") public static int cudaGLGetDevices(@Cast(value="unsigned int*") IntBuffer pCudaDeviceCount, IntBuffer pCudaDevices, @Cast(value="unsigned int") int cudaDeviceCount, @Cast(value="cudaGLDeviceList") int deviceList)
@Cast(value="cudaError_t") public static int cudaGLGetDevices(@Cast(value="unsigned int*") int[] pCudaDeviceCount, int[] pCudaDevices, @Cast(value="unsigned int") int cudaDeviceCount, @Cast(value="cudaGLDeviceList") int deviceList)
@Cast(value="cudaError_t") public static int cudaGraphicsGLRegisterImage(@Cast(value="cudaGraphicsResource**") PointerPointer resource, @Cast(value="GLuint") int image, @Cast(value="GLenum") int target, @Cast(value="unsigned int") int flags)
resource
- - Pointer to the returned object handleimage
- - name of texture or renderbuffer object to be registeredtarget
- - Identifies the type of object specified by \p imageflags
- - Register flags::cudaGraphicsUnregisterResource,
::cudaGraphicsMapResources,
::cudaGraphicsSubResourceGetMappedArray,
::cuGraphicsGLRegisterImage
@Cast(value="cudaError_t") public static int cudaGraphicsGLRegisterImage(@ByPtrPtr cudaGraphicsResource resource, @Cast(value="GLuint") int image, @Cast(value="GLenum") int target, @Cast(value="unsigned int") int flags)
@Cast(value="cudaError_t") public static int cudaGraphicsGLRegisterBuffer(@Cast(value="cudaGraphicsResource**") PointerPointer resource, @Cast(value="GLuint") int buffer, @Cast(value="unsigned int") int flags)
resource
- - Pointer to the returned object handlebuffer
- - name of buffer object to be registeredflags
- - Register flags::cudaGraphicsUnregisterResource,
::cudaGraphicsMapResources,
::cudaGraphicsResourceGetMappedPointer,
::cuGraphicsGLRegisterBuffer
@Cast(value="cudaError_t") public static int cudaGraphicsGLRegisterBuffer(@ByPtrPtr cudaGraphicsResource resource, @Cast(value="GLuint") int buffer, @Cast(value="unsigned int") int flags)
@Cast(value="cudaError_t") @Deprecated public static int cudaGLSetGLDevice(int device)
device
- - Device to use for OpenGL interoperability::cudaGraphicsGLRegisterBuffer, ::cudaGraphicsGLRegisterImage
@Cast(value="cudaError_t") @Deprecated public static int cudaGLRegisterBufferObject(@Cast(value="GLuint") int bufObj)
bufObj
- - Buffer object ID to register::cudaGraphicsGLRegisterBuffer
@Cast(value="cudaError_t") @Deprecated public static int cudaGLMapBufferObject(@Cast(value="void**") PointerPointer devPtr, @Cast(value="GLuint") int bufObj)
devPtr
- - Returned device pointer to CUDA objectbufObj
- - Buffer object ID to map::cudaGraphicsMapResources
@Cast(value="cudaError_t") @Deprecated public static int cudaGLMapBufferObject(@Cast(value="void**") @ByPtrPtr Pointer devPtr, @Cast(value="GLuint") int bufObj)
@Cast(value="cudaError_t") @Deprecated public static int cudaGLUnmapBufferObject(@Cast(value="GLuint") int bufObj)
bufObj
- - Buffer object to unmap::cudaGraphicsUnmapResources
@Cast(value="cudaError_t") @Deprecated public static int cudaGLUnregisterBufferObject(@Cast(value="GLuint") int bufObj)
bufObj
- - Buffer object to unregister::cudaGraphicsUnregisterResource
@Cast(value="cudaError_t") @Deprecated public static int cudaGLSetBufferObjectMapFlags(@Cast(value="GLuint") int bufObj, @Cast(value="unsigned int") int flags)
bufObj
- - Registered buffer object to set flags forflags
- - Parameters for buffer mapping::cudaGraphicsResourceSetMapFlags
@Cast(value="cudaError_t") @Deprecated public static int cudaGLMapBufferObjectAsync(@Cast(value="void**") PointerPointer devPtr, @Cast(value="GLuint") int bufObj, CUstream_st stream)
devPtr
- - Returned device pointer to CUDA objectbufObj
- - Buffer object ID to mapstream
- - Stream to synchronize::cudaGraphicsMapResources
@Cast(value="cudaError_t") @Deprecated public static int cudaGLMapBufferObjectAsync(@Cast(value="void**") @ByPtrPtr Pointer devPtr, @Cast(value="GLuint") int bufObj, CUstream_st stream)
@Cast(value="cudaError_t") @Deprecated public static int cudaGLUnmapBufferObjectAsync(@Cast(value="GLuint") int bufObj, CUstream_st stream)
bufObj
- - Buffer object to unmapstream
- - Stream to synchronize::cudaGraphicsUnmapResources
Copyright © 2024. All rights reserved.