process()

void MayaFlux::Buffers::BufferProcessingChain::process

(

const std::shared_ptr< Buffer > &

buffer

)

Applies the transformation pipeline to a buffer with intelligent execution.

Parameters

buffer

Buffer to transform

Applies each processor in the buffer's sequence using an optimized execution strategy. The chain leverages processor capabilities for maximum performance:

• Backend Optimization: Uses processor-recommended backends when beneficial
• Parallel Execution: Executes compatible processors in parallel when possible
• Resource Management: Optimally allocates CPU, GPU, and memory resources
• Error Handling: Provides robust error recovery and fallback mechanisms

This does not include the final processor, which must be applied separately with process_final() to ensure proper pipeline completion.

Definition at line 175 of file BufferProcessingChain.cpp.

{
    bool expected = false;
    if (!m_is_processing.compare_exchange_strong(expected, true,
            std::memory_order_acquire, std::memory_order_relaxed)) {
        return;
    }
 
    if (m_pending_count.load(std::memory_order_relaxed) > 0) {
        process_pending_processor_operations();
    }
 
    auto it = m_buffer_processors.find(buffer);
    if (it == m_buffer_processors.end() || it->second.empty()) {
        m_is_processing.store(false, std::memory_order_release);
        return;
    }
 
    for (auto& processor : it->second) {
        bool should_process = true;
 
        if (m_enforcement_strategy == TokenEnforcementStrategy::OVERRIDE_SKIP) {
            auto processor_token = processor->get_processing_token();
            should_process = are_tokens_compatible(m_token_filter_mask, processor_token);
        }
 
        if (should_process) {
            processor->process(buffer);
        }
    }
 
    if (m_enforcement_strategy == TokenEnforcementStrategy::OVERRIDE_REJECT) {
        cleanup_rejected_processors(buffer);
    }
 
    m_is_processing.store(false, std::memory_order_release);
}

References MayaFlux::Buffers::are_tokens_compatible(), buffer, cleanup_rejected_processors(), m_buffer_processors, m_enforcement_strategy, m_is_processing, m_pending_count, m_token_filter_mask, MayaFlux::Buffers::OVERRIDE_REJECT, MayaFlux::Buffers::OVERRIDE_SKIP, and process_pending_processor_operations().

Referenced by process_complete().

Here is the call graph for this function:

Here is the caller graph for this function: