Enum TrtMsg 

pub enum TrtMsg {
    Build {
        source: NetworkSource,
        config: Box<IBuilderConfig>,
        reply: BuildReply,
    },
    Deserialize {
        plan: EnginePlan,
        reply: DeserializeReply,
    },
    CreateContext {
        engine: Arc<TrtEngine>,
        reply: CreateContextReply,
    },
    EnqueueOnStream {
        stream: Arc<CudaStream>,
        context: ExecutionContext,
        bindings: ExecutionBindings,
        reply: EnqueueReply,
    },
    Refit {
        engine: Arc<TrtEngine>,
        weights: Vec<RefitWeights>,
        reply: RefitReply,
    },
}

Public message surface for TrtActor.

The variant EnqueueOnStream accepts the Arc<CudaStream> from atomr-accel-cuda::DeviceActor so the TensorRT runtime shares the device’s stream timeline (no cross-stream synchronisation, no extra event hops).

Variants

Build

Build a TensorRT engine from a network source + config. Returns the serialised plan on success.

Fields

source: NetworkSource

config: Box<IBuilderConfig>

reply: BuildReply

Deserialize

Deserialise a plan blob into a shared engine handle.

Fields

plan: EnginePlan

reply: DeserializeReply

CreateContext

Create a fresh IExecutionContext from an existing engine. Returns the new context (caller owns it).

Fields

engine: Arc<TrtEngine>

reply: CreateContextReply

EnqueueOnStream

Submit enqueueV3 on the supplied CUDA stream. The actor returns immediately after submission; real GPU completion is observed by atomr-accel-cuda’s completion strategy on the shared stream.

Fields

stream: Arc<CudaStream>

context: ExecutionContext

bindings: ExecutionBindings

reply: EnqueueReply

Refit

Refit a built engine in-place with new weights. Requires the engine to have been built with RefitPolicy::OnDemand or WeightsStreaming.

Fields

engine: Arc<TrtEngine>

weights: Vec<RefitWeights>

reply: RefitReply