List of all items

Structs

• fact::DeviceFact
• memory::DeviceMemSchema
• memory::DeviceMemoryPool
• memory::DeviceResolvedMemSchema
• ops::apply_rope::GpuApplyRope
• ops::binary::GpuBinOp
• ops::broadcast::GpuMultiBroadcastTo
• ops::cast::GpuCast
• ops::change_axes::GpuAxisOp
• ops::concat::GpuConcat
• ops::dyn_kv_cache::FrozenGpuDynKVCacheState
• ops::dyn_kv_cache::GpuDynKVCache
• ops::dyn_kv_cache::GpuDynKVCacheState
• ops::element_wise::GpuElementWise
• ops::gelu_approximate::GpuGeluApproximate
• ops::iff::GpuIff
• ops::leaky_relu::GpuLeakyRelu
• ops::pulse::GpuDelay
• ops::pulse::GpuDelayState
• ops::pulse::GpuPulsePad
• ops::reduce::GpuReduce
• ops::rms_norm::GpuRmsNorm
• ops::rotate_half::GpuRotateHalf
• ops::scaled_masked_softmax::GpuScaledMaskedSoftmax
• ops::slice::GpuSlice
• ops::softmax::GpuSoftmax
• session_handler::DeviceSessionHandler
• sync::DeviceSync
• tensor::DeviceArenaView

Enums

• fact::DeviceTensorOrigin
• ops::reduce::Reducer
• sync::DeviceSyncKind
• tensor::DeviceTensor
• utils::BroadcastKind

Traits

• device::DeviceBuffer
• device::DeviceContext
• fact::DeviceTypedFactExt
• tensor::DeviceTensorExt
• tensor::IntoDevice
• tensor::OwnedDeviceTensor

Macros

• rule_ensure

Functions

• device::get_context
• device::set_context
• ops::change_axes::permute_output_shape
• ops::copy_based::try_make_copy_based_op
• rewrite_rules::rewire_sdpa::causal_mask_as_extern
• rewrite_rules::rewire_sdpa::create_sdpa_mask_graph
• rewrite_rules::rewire_sdpa::neutral_mask_for_full_attn
• rewrite_rules::rewire_sdpa::rewire_sdpa
• rewrite_rules::rewire_sdpa::rewire_sdpa_op
• rewrite_rules::rewire_syncs::rewire_back_and_forth_sync
• rewrite_rules::rewire_syncs::rewire_sync_after_const
• rewrite_rules::rewire_syncs::rewire_syncs
• rewrite_rules::rms_norm::remove_rms_norm_cast
• session_handler::make_scalar_exotic_tensor_for_node
• session_handler::make_tensor_for_node
• sync::sync_inputs_if_required
• sync::sync_model_outputs_if_required
• utils::as_q40_tensor
• utils::as_quant_fact
• utils::check_strides_validity
• utils::compute_broadcast_strides
• utils::facts_to_device_facts
• utils::get_device_fact
• utils::get_device_facts
• utils::get_quant_fact
• utils::reshape_to_rank_2
• utils::reshape_to_rank_3

Type Aliases

• ops::apply_rope::DispatchApplyRopeFn
• ops::binary::DispatchBinOpFn
• ops::cast::DispatchCastFn
• ops::element_wise::DispatchElementWiseFn
• ops::gelu_approximate::DispatchGeluApproximateFn
• ops::iff::DispatchIffFn
• ops::leaky_relu::DispatchLeakyReluFn
• ops::reduce::DispatchReduceFn
• ops::rms_norm::DispatchRmsNormFn
• ops::rotate_half::DispatchRotateHalfFn
• ops::scaled_masked_softmax::DispatchScaledMaskedSoftmaxFn
• ops::softmax::DispatchSoftmaxFn

Statics

• device::DEVICE_CONTEXT