dispatch_core_chained()

GpuChannelResult MayaFlux::Yantra::GpuDispatchCore::dispatch_core_chained ( const std::vector< std::vector< double > > &  channels, const DataStructureInfo &  structure_info, const ExecutionContext &  ctx  )

protected

Multi-pass (chained) dispatch.

Calls dispatch_batched on GpuResourceManager and reads back once after all passes.

Parameters

channels	Extracted double channels.
structure_info	Dimension/modality metadata.
ctx	ExecutionContext carrying pass_count and pc_updater.

Returns

GpuChannelResult containing primary float readback and aux buffers.

Definition at line 269 of file GpuDispatchCore.cpp.

{
    on_before_gpu_dispatch(channels, structure_info);
    prepare_gpu_inputs(channels, structure_info);
 
    bind_all_descriptors();
 
    const size_t effective = m_staging_floats.empty()
        ? largest_binding_data_element_count()
        : m_staging_floats.size();
    const auto groups = calculate_dispatch_size(effective, structure_info);
 
    m_resources.dispatch_batched(
        dispatch_key(), groups, m_bindings,
        m_gpu_config.push_constant_size, ctx);
 
    GpuChannelResult result;
    result.primary = readback_primary(effective);
    readback_aux(result);
    return result;
}

References bind_all_descriptors(), calculate_dispatch_size(), MayaFlux::Yantra::GpuResourceManager::dispatch_batched(), dispatch_key(), largest_binding_data_element_count(), m_bindings, m_gpu_config, m_resources, m_staging_floats, on_before_gpu_dispatch(), prepare_gpu_inputs(), MayaFlux::Yantra::GpuChannelResult::primary, MayaFlux::Portal::Graphics::GpuComputeConfig::push_constant_size, readback_aux(), and readback_primary().

Referenced by MayaFlux::Yantra::GpuExecutionContext< InputType, OutputType >::execute().

Here is the call graph for this function:

Here is the caller graph for this function: