List of all items

Structs

• completion::HostFnCompletion
• completion::PolledCompletion
• completion::SyncCompletion
• device::AllocReq
• device::ContextActor
• device::CopyFromHostReq
• device::CopyToHostReq
• device::DeviceLoad
• device::DeviceState
• device::device_actor::DeviceActor
• device::device_actor::DeviceConfig
• device::device_actor::EnabledLibraries
• device::device_actor::KernelChildren
• device::device_actor::SgemmRequest
• dispatcher::GpuDispatcher
• dtype::C32
• dtype::C64
• error::DeviceSupervisor
• event::Event
• event::EventActor
• event::IpcEventHandle
• gpu_ref::GpuRef
• graph::GraphActor
• graph::GraphHandle
• graph::GraphRecordCtx
• graph::SendGraph
• graph::child::ChildGraphInsertion
• graph::child::ChildGraphOp
• graph::conditional::IfNodeDescriptor
• graph::conditional::WhileNodeDescriptor
• graph::dot::DotFlags
• graph::record::cudnn::ActivationOp
• graph::record::cudnn::ConvForwardOp
• graph::record::cudnn::SoftmaxOp
• graph::record::cusparse::SpMmOp
• graph::record::cusparse::SpMvOp
• graph::record::fft_r2c::FftR2COp
• graph::record::memcpy::MemcpyOp
• graph::record::rng_fill_uniform::RngFillUniformOp
• graph::record::sgemm::SgemmOp
• hopper::cluster::ClusterDim
• hopper::cluster::LaunchSpec
• hopper::tma::TensorMapDescriptor
• host::PinnedBuf
• host::PinnedBufferPool
• host::PinnedBufferPoolConfig
• host::PinnedPoolStats
• kernel::CsrMatrix
• kernel::SparseActor
• kernel::blas::BlasActor
• kernel::blas::gemm::GemmRequest
• kernel::blas::gemm_strided_batched::GemmStridedBatchedRequest
• kernel::blas::l1::AsumRequest
• kernel::blas::l1::AxpyRequest
• kernel::blas::l1::CopyRequest
• kernel::blas::l1::DotRequest
• kernel::blas::l1::IamaxRequest
• kernel::blas::l1::IaminRequest
• kernel::blas::l1::Nrm2Request
• kernel::blas::l1::RotRequest
• kernel::blas::l1::ScalRequest
• kernel::blas::l1::SwapRequest
• kernel::blas::l2::GemvRequest
• kernel::blas::l2::GerRequest
• kernel::blas::l3::GeamRequest
• kernel::blas::l3::SyrkRequest
• kernel::blas::l3::TrsmRequest
• kernel::blas::scaling::PerRowScale
• kernel::blas::scaling::PerTensorScale
• kernel::blas_lt::BlasLtActor
• kernel::blas_lt::heuristic::HeuristicCacheRef
• kernel::blas_lt::heuristic::HeuristicEntry
• kernel::blas_lt::heuristic::HeuristicKey
• kernel::blas_lt::matmul::MatmulRequest
• kernel::blas_lt::scaling::ScaleSet
• kernel::blas_lt::workspace::WorkspaceLease
• kernel::blas_lt::workspace::WorkspacePool
• kernel::collective::CollectiveActor
• kernel::collective::all_to_all::AllToAllRequest
• kernel::collective::all_to_all::AllToAllvRequest
• kernel::collective::allgather::AllGatherRequest
• kernel::collective::allreduce::AllReduceRequest
• kernel::collective::broadcast::BroadcastRequest
• kernel::collective::capabilities::NcclCapabilities
• kernel::collective::custom_op::PreMulSumOp
• kernel::collective::group::GroupGuard
• kernel::collective::group::GroupTracker
• kernel::collective::p2p::RecvRequest
• kernel::collective::p2p::SendRequest
• kernel::collective::reduce::ReduceRequest
• kernel::collective::reduce_scatter::ReduceScatterRequest
• kernel::cudnn::ActivationRequest
• kernel::cudnn::ConvForwardRequest
• kernel::cudnn::ConvParams
• kernel::cudnn::CudnnActor
• kernel::cudnn::SoftmaxRequest
• kernel::cudnn::activation::ActivationFwdRequest
• kernel::cudnn::activation::DropoutFwdRequest
• kernel::cudnn::activation::LrnFwdRequest
• kernel::cudnn::activation::LrnParams
• kernel::cudnn::activation::SoftmaxFwdRequest
• kernel::cudnn::attention::AttentionParams
• kernel::cudnn::attention::MultiHeadAttnBwdRequest
• kernel::cudnn::attention::MultiHeadAttnFwdRequest
• kernel::cudnn::conv::ConvBwdDataRequest
• kernel::cudnn::conv::ConvBwdFilterRequest
• kernel::cudnn::conv::ConvDescParams
• kernel::cudnn::conv::ConvFwdRequest
• kernel::cudnn::graph::CachedPlan
• kernel::cudnn::graph::OperationGraphSpec
• kernel::cudnn::graph::PlanCache
• kernel::cudnn::graph::PlanCacheKey
• kernel::cudnn::graph::TensorSpec
• kernel::cudnn::norm::BatchNormRequest
• kernel::cudnn::norm::GroupNormRequest
• kernel::cudnn::norm::InstanceNormRequest
• kernel::cudnn::norm::LayerNormRequest
• kernel::cudnn::norm::NormBwdRequest
• kernel::cudnn::pool::PoolBwdRequest
• kernel::cudnn::pool::PoolFwdRequest
• kernel::cudnn::pool::PoolParams
• kernel::cudnn::rnn::RnnBwdRequest
• kernel::cudnn::rnn::RnnFwdRequest
• kernel::cudnn::rnn::RnnParams
• kernel::dispatch::BlasDispatchCtx
• kernel::dispatch::BlasLtDispatchCtx
• kernel::dispatch::CollectiveDispatchCtx
• kernel::dispatch::CudnnDispatchCtx
• kernel::dispatch::FftDispatchCtx
• kernel::dispatch::NvrtcDispatchCtx
• kernel::dispatch::SendSparseHandle
• kernel::dispatch::SparseDispatchCtx
• kernel::dispatch::TensorDispatchCtx
• kernel::dispatch::WorkspacePool
• kernel::envelope::KernelEnvelope
• kernel::envelope::KernelInfo
• kernel::fft::FftActor
• kernel::fft::FftPlan
• kernel::fft::FftPlanMany
• kernel::fft::FftRequest
• kernel::fft::PlanKey
• kernel::nvrtc::KernelHandle
• kernel::nvrtc::NvrtcActor
• kernel::nvrtc::NvrtcOpts
• kernel::record::BlasRecorder
• kernel::record::BlasSgemmOp
• kernel::record::FftR2COp
• kernel::record::FftRecorder
• kernel::record::MemcpyOp
• kernel::record::MemcpyRecorder
• kernel::record::RngFillUniformOp
• kernel::record::RngRecorder
• kernel::rng::RngActor
• kernel::rng::dist::FillRequest
• kernel::solver::SolverActor
• kernel::solver::SolverCells
• kernel::solver::batched::GesvdjBatchedRequest
• kernel::solver::batched::GetrfBatchedRequest
• kernel::solver::batched::PotrfBatchedRequest
• kernel::solver::dense::CholeskyRequest
• kernel::solver::dense::LuRequest
• kernel::solver::dense::LuSolveRequest
• kernel::solver::dense::QrRequest
• kernel::solver::dense::SvdRequest
• kernel::solver::dense::SyevdRequest
• kernel::solver::generalized::HegvdRequest
• kernel::solver::generalized::SygvdRequest
• kernel::tensor::SendHandle
• kernel::tensor::TensorActor
• kernel::tensor::contract::ContractRequest
• kernel::tensor::contract::OperandSpec
• kernel::tensor::elementwise::ElementwiseBinaryRequest
• kernel::tensor::elementwise::ElementwiseTrinaryRequest
• kernel::tensor::permute::PermutationRequest
• kernel::tensor::plan_cache::CachedPlan
• kernel::tensor::plan_cache::PlanCache
• kernel::tensor::plan_cache::PlanKey
• kernel::tensor::reduce::ReductionRequest
• memory::ipc::IpcMemHandle
• memory::ipc::OpenedMem
• memory::managed::ManagedAllocatorActor
• memory::managed::ManagedRef
• memory::managed::ManagedStats
• module::FunctionHandle
• module::ModuleActor
• module::ModuleHandle
• multi_device::NcclWorldActor
• multi_device::NcclWorldConfig
• nvrtc_cache::CachedKernel
• nvrtc_cache::NvrtcCache
• nvrtc_cache::NvrtcCacheKey
• observability::GpuProbeState
• observability::GpuProbes
• p2p::P2pGraph
• p2p::P2pTopology
• pipeline::PipelineExecutor
• pipeline::PipelineExecutorN
• pipeline::PipelineSink
• pipeline::PipelineSource
• pipeline::StageBox
• placement::DeviceChoice
• placement::LeastLoadedPolicy
• placement::PlacementActor
• placement::PlacementHints
• placement::RoundRobinPolicy
• placement::sharded::DeviceExtractor
• placement::sharded::PlacementShardingAdapter
• placement::sharded::RoutedDeviceMsg
• replay::ReplayHarness
• stream::ActorHints
• stream::PerActorAllocator
• stream::PooledAllocator
• stream::SingleStreamAllocator
• sys::cublaslt::Preference
• sys::cudnn::BackendDescriptor
• sys::cusparse::DnMatGuard
• sys::cusparse::DnVecGuard
• sys::cusparse::SpGemmDescGuard
• sys::cusparse::SpMatGuard
• sys::cusparse::SpSvDescGuard
• sys::cutensor::CutensorError

Enums

• device::ContextMsg
• device::HostBuf
• device::device_actor::DeviceMsg
• device::device_actor::WorkRequest
• dtype::DType
• dtype::DTypeKind
• error::GpuError
• event::EventMsg
• graph::GraphMsg
• graph::GraphOpLegacy
• graph::conditional::ConditionalKind
• graph::exec_update::GraphExecUpdateOutcome
• hopper::cluster::ClusterError
• hopper::cp_async::PipelineStages
• hopper::tma::TensorMapDataType
• hopper::tma::TensorMapInterleave
• hopper::tma::TensorMapL2Promotion
• hopper::tma::TensorMapOobFill
• hopper::tma::TensorMapSwizzle
• hopper::tma::TmaEncodeError
• hopper::wgmma::WgmmaShape
• host::PinnedPoolMsg
• kernel::Activation
• kernel::ReduceOp
• kernel::SparseMsg
• kernel::blas::BlasMsg
• kernel::blas_lt::Activation
• kernel::blas_lt::BlasLtMsg
• kernel::blas_lt::epilogue::Epilogue
• kernel::collective::CollectiveMsg
• kernel::collective::ReduceOp
• kernel::cudnn::CudnnMsg
• kernel::cudnn::activation::ActivationKind
• kernel::cudnn::activation::SoftmaxMode
• kernel::cudnn::attention::AttentionMask
• kernel::cudnn::conv::EpilogueKind
• kernel::cudnn::graph::DtypeTag
• kernel::cudnn::graph::NormMode
• kernel::cudnn::graph::NormPhase
• kernel::cudnn::graph::OpSpec
• kernel::cudnn::graph::PointwiseMode
• kernel::cudnn::graph::PoolKind
• kernel::cudnn::graph::ReduceOp
• kernel::cudnn::graph::TensorLayout
• kernel::cudnn::pool::PoolMode
• kernel::cudnn::rnn::RnnDirection
• kernel::cudnn::rnn::RnnMode
• kernel::dispatch::DispatchDType
• kernel::dispatch::SparseOp
• kernel::fft::FftCallbackKind
• kernel::fft::FftDirection
• kernel::fft::FftKind
• kernel::fft::FftMsg
• kernel::nvrtc::CppStd
• kernel::nvrtc::KernelArg
• kernel::nvrtc::NvrtcMsg
• kernel::nvrtc::SmArch
• kernel::rng::RngMsg
• kernel::rng::dist::Distribution
• kernel::solver::SolverMsg
• kernel::solver::Uplo
• kernel::solver::generalized::EigType
• kernel::tensor::TensorMsg
• kernel::tensor::compute_desc::ComputeDesc
• kernel::tensor::plan_cache::OpKind
• memory::advise::MemAdvice
• memory::managed::ManagedFlags
• memory::managed::ManagedMsg
• memory::managed::PrefetchTarget
• module::ModuleMsg
• multi_device::NcclWorldMsg
• p2p::P2pMsg
• placement::PlacementMsg
• prelude::Activation
• prelude::ReduceOp
• replay::JournalEntry
• replay::ReplayMode
• replay::ReplayMsg
• stream::Priority
• stream::WorkloadKind
• sys::cufft::CufftXtCallbackType
• sys::curand::RngGeneratorKind
• sys::cusparse::SddmmAlg
• sys::cusparse::SpGemmAlg
• sys::cusparse::SpMmAlg
• sys::cusparse::SpMvAlg
• sys::cusparse::SpSvAlg

Traits

• completion::CompletionStrategy
• device::AllocDispatch
• device::CopyFromHostDispatch
• device::CopyToHostDispatch
• dtype::AccelDtype
• dtype::AxpyDotNrm2Supported
• dtype::CudaDtype
• dtype::CudnnSupported
• dtype::FftSupported
• dtype::GeamSupported
• dtype::GemmSupported
• dtype::GemvSupported
• dtype::GerSupported
• dtype::NcclReduceSupported
• dtype::RngFloatSupported
• dtype::RngIntSupported
• dtype::SolverSupported
• dtype::SparseIndex
• dtype::SparseSupported
• dtype::SyrkSupported
• dtype::TensorSupported
• dtype::TrsmSupported
• graph::GraphOp
• graph::GraphOpRecord
• kernel::collective::NcclReduceSupported
• kernel::dispatch::BlasL1Dispatch
• kernel::dispatch::BlasL2Dispatch
• kernel::dispatch::BlasL3Dispatch
• kernel::dispatch::BlasLtDispatch
• kernel::dispatch::CollectiveDispatch
• kernel::dispatch::CudnnDispatch
• kernel::dispatch::DevSliceArg
• kernel::dispatch::FftDispatch
• kernel::dispatch::GemmDispatch
• kernel::dispatch::GemmStridedBatchedDispatch
• kernel::dispatch::NvrtcLaunchDispatch
• kernel::dispatch::RngDispatch
• kernel::dispatch::ScalarArg
• kernel::dispatch::SparseDispatch
• kernel::dispatch::TensorDispatch
• kernel::envelope::KernelTrace
• kernel::record::RecordMode
• kernel::rng::dist::NormalParam
• kernel::solver::SolverDispatch
• pipeline::BoxedStage
• pipeline::PipelineStage
• placement::PlacementPolicy
• replay::ReplaySink
• stream::StreamAllocator
• sys::cusolver::SolverScalar

Functions

• error::decider
• error::device_supervisor_strategy
• graph::child::child_graph_op
• graph::conditional::build_params
• graph::conditional::driver_supports_conditional
• graph::dot::export_dot
• graph::exec_update::exec_update
• hopper::atomr_hopper_header_path
• hopper::cluster::dsm_total_bytes
• hopper::cp_async::mbarrier_arrival_count
• kernel::blas_lt::scaling::null_scale_ptr
• kernel::blas_lt::workspace::size_class
• kernel::collective::capabilities::probe_capabilities
• kernel::cudnn::activation::build_activation_fwd_graph
• kernel::cudnn::attention::build_mha_bwd_graph
• kernel::cudnn::attention::build_mha_fwd_graph
• kernel::cudnn::conv::build_conv_bwd_data_graph
• kernel::cudnn::conv::build_conv_bwd_filter_graph
• kernel::cudnn::conv::build_conv_fwd_graph
• kernel::cudnn::graph::cache_key
• kernel::cudnn::norm::build_norm_fwd_graph
• kernel::cudnn::pool::build_pool_bwd_graph
• kernel::cudnn::pool::build_pool_fwd_graph
• kernel::cudnn::rnn::build_rnn_fwd_spec
• kernel::dispatch::reply_unsupported
• kernel::envelope::access_all_2
• kernel::envelope::access_all_3
• kernel::envelope::access_all_4
• kernel::envelope::run_kernel
• kernel::nvrtc::compile_to_ptx
• kernel::nvrtc::default_disk_cache_path
• kernel::rng::_capability_marker_compile_fail_doc
• kernel::tensor::compute_desc::compute_desc_tag
• kernel::tensor::compute_desc::resolve_compute_desc
• kernel::tensor::contract::build_contract_key
• kernel::tensor::contract::default_compute_for
• kernel::tensor::elementwise::build_binary_key_raw
• kernel::tensor::elementwise::build_trinary_key_raw
• kernel::tensor::permute::build_permutation_key_raw
• kernel::tensor::plan_cache::hash_i32s
• kernel::tensor::plan_cache::hash_i64s
• kernel::tensor::reduce::build_reduction_key_raw
• memory::advise::advise
• memory::ipc::get_mem_handle
• memory::ipc::open_mem_handle
• memory::prefetch::prefetch_async
• nvrtc_cache::hash_options
• nvrtc_cache::hash_source
• observability::install
• pipeline::run_pipeline
• pipeline::spawn_pipeline
• replay::replay_via_sink
• streams_pipeline::gpu_stage
• streams_pipeline::run_collect
• streams_pipeline::source_from_unbounded
• sys::cublas::asum_ex
• sys::cublas::axpy_ex
• sys::cublas::copy_ex
• sys::cublas::dgeam
• sys::cublas::dgemv
• sys::cublas::dger
• sys::cublas::dot_ex
• sys::cublas::dsyrk
• sys::cublas::dtrsm
• sys::cublas::gemm_ex
• sys::cublas::gemm_strided_batched_ex
• sys::cublas::iamax_ex
• sys::cublas::iamin_ex
• sys::cublas::nrm2_ex
• sys::cublas::rot_ex
• sys::cublas::scal_ex
• sys::cublas::sgeam
• sys::cublas::sgemv
• sys::cublas::sger
• sys::cublas::ssyrk
• sys::cublas::strsm
• sys::cublas::swap_ex
• sys::cublaslt::check
• sys::cublaslt::set_desc_i32_attr
• sys::cublaslt::set_desc_pointer_attr
• sys::cuda_driver::ipc_close_mem_handle
• sys::cuda_driver::ipc_get_event_handle
• sys::cuda_driver::ipc_get_mem_handle
• sys::cuda_driver::ipc_open_event_handle
• sys::cuda_driver::ipc_open_mem_handle_v2
• sys::cuda_driver::launch_cooperative_kernel
• sys::cuda_driver::launch_kernel
• sys::cuda_driver::mem_advise_v2
• sys::cuda_driver::mem_prefetch_async_v2
• sys::cuda_driver::module_get_function
• sys::cuda_driver::module_load_data
• sys::cuda_driver::module_unload
• sys::cuda_driver::runtime_ipc_get_mem_handle
• sys::cudnn::backend_execute
• sys::cufft::xt_clear_callback
• sys::cufft::xt_set_callback
• sys::curand::create_generator
• sys::curand::destroy_generator
• sys::curand::generate_poisson_u32
• sys::curand::generate_u32
• sys::curand::generate_u64
• sys::curand::set_offset
• sys::curand::set_seed
• sys::curand::set_stream
• sys::cusolver::status_to_result
• sys::cusparse::ok
• sys::cutensor::c_32f
• sys::cutensor::create_elementwise_binary
• sys::cutensor::create_elementwise_trinary
• sys::cutensor::create_permutation
• sys::cutensor::elementwise_binary_execute
• sys::cutensor::elementwise_trinary_execute
• sys::cutensor::permute
• sys::cutensor::plan_preference_set_algo
• sys::cutensor::r_32f
• sys::cutensor::r_64f
• sys::cutensor::r_min_16bf
• sys::cutensor::r_min_16f
• sys::cutensor::r_min_32f
• sys::cutensor::r_min_64f
• sys::cutensor::r_min_tf32
• sys::cutensor::reduce

Type Aliases

• kernel::dispatch::GemmDispatchCtx
• kernel::tensor::TensorSpec

Constants

• error::CONTEXT_POISONED_TAG
• error::OUT_OF_MEMORY_TAG
• error::UNRECOVERABLE_TAG
• hopper::ATOMR_HOPPER_HEADER_REL_PATH
• hopper::cp_async::macro_names::CP_ASYNC_BULK
• hopper::cp_async::macro_names::CP_ASYNC_CA_4
• hopper::cp_async::macro_names::CP_ASYNC_CG_16
• hopper::cp_async::macro_names::CP_ASYNC_COMMIT_GROUP
• hopper::cp_async::macro_names::CP_ASYNC_WAIT_GROUP
• kernel::blas_lt::heuristic::DEFAULT_HEURISTIC_CAPACITY
• kernel::blas_lt::heuristic::DEFAULT_TOP_K
• kernel::blas_lt::workspace::DEFAULT_POOL_CAPACITY_PER_CLASS
• kernel::cudnn::graph::DEFAULT_PLAN_CACHE_SIZE
• kernel::tensor::plan_cache::DEFAULT_PLAN_CACHE_SIZE
• sys::cusolver::LIB