Crate edgefirst_tensor 

EdgeFirst HAL - Tensor Module

The edgefirst_tensor crate provides a unified interface for managing multi-dimensional arrays (tensors) with support for different memory types, including Direct Memory Access (DMA), POSIX Shared Memory (Shm), and system memory. The crate defines traits and structures for creating, reshaping, and mapping tensors into memory.

Examples

use edgefirst_tensor::{Error, Tensor, TensorMemory, TensorTrait};
let tensor = Tensor::<f32>::new(&[2, 3, 4], Some(TensorMemory::Mem), Some("test_tensor"))?;
assert_eq!(tensor.memory(), TensorMemory::Mem);
assert_eq!(tensor.name(), "test_tensor");

Overview

The main structures and traits provided by the edgefirst_tensor crate is the TensorTrait and TensorMapTrait traits, which define the behavior of Tensors and their memory mappings, respectively. The Tensor enum encapsulates different tensor implementations based on the memory type, while the TensorMap enum provides access to the underlying data.

Structs

BufferIdentity

Unique identity for a tensor’s underlying buffer.

DmaMap

DmaTensor

A tensor backed by DMA (Direct Memory Access) memory.

MemMap

MemTensor

PboMap

PboMapping

Raw mapped pointer from a PBO. CPU-accessible while the buffer is mapped. The pointer is only valid between map and unmap calls.

PboTensor

A tensor backed by an OpenGL Pixel Buffer Object.

ShmMap

ShmTensor

Enums

Error

Tensor

TensorMap

TensorMemory

Traits

PboOps

Trait for PBO GL operations, implemented by the image crate.

TensorMapTrait

TensorTrait

Functions

is_dma_available

Check if DMA memory allocation is available on this system.

is_shm_available

Check if POSIX shared memory allocation is available on this system.

Type Aliases

Result