Module drv 

libnrt loader. We need three symbols — nrt_init, nrt_get_total_nc_count, and nrt_get_version — and a best-effort read of the Neuron-specific env vars that AWS containers set.

drop(sym) on a libloading::Symbol ends the borrow on its Library; Symbol doesn’t impl Drop but the lifetime parameter does the work.

Structs

Loaded

Resolved libnrt state: core count, version, and optional entry points.

Enums

NeuronGen

Neuron hardware generation, inferred from instance type or runtime version.

Constants

NRT_SUCCESS

Success status (NRT_SUCCESS).

Functions

detect_generation

Best-effort generation detection. Prefers NEURON_INSTANCE_TYPE env hint (commonly set by AWS Deep Learning AMIs + containers), falls back to the Neuron runtime major version.

is_available

Returns true if libnrt loaded and initialised.

total_cores

Total NeuronCores from nrt_get_total_nc_count, or 0 if libnrt is unavailable.

version

Runtime version string from nrt_get_version, if available.

Type Aliases

NrtAddTensorToSetFn

Signature of nrt_add_tensor_to_tensor_set.

NrtAllocTensorSetFn

Signature of nrt_allocate_tensor_set.

NrtExecuteFn

Signature of nrt_execute.

NrtFreeTensorSetFn

Signature of nrt_free_tensor_set.

NrtLoadFn

Signature of nrt_load.

NrtModelHandle

Opaque nrt_model_t handle.

NrtStatus

NRT_STATUS return code from the Neuron runtime.

NrtTensorHandle

Opaque nrt_tensor_t handle.

NrtTensorSetHandle

Opaque nrt_tensor_set_t handle.

NrtUnloadFn

Signature of nrt_unload.