Stable Arrays - 2022.2 English

The stable pragma can be used to mark input or output variables of a dataflow region. Its effect is to remove their corresponding task-level synchronizations, assuming that the user guarantees this removal is indeed correct.

void dataflow_region(int A[...], ...
#pragma HLS stable variable=A
#pragma HLS dataflow
    proc1(...);
    proc2(A, ...);

Without the stable pragma, and assuming that A is read by proc2, then proc2 would be part of the initial synchronization for the dataflow region where it is located. This means that proc1 would not restart until proc2 is also ready to start again, which would prevent dataflow iterations to be overlapped and induce a possible loss of performance. The stable pragma indicates that this synchronization is not necessary to preserve correctness.

With the stable pragma, the compiler assumes that:

• If A is read by proc2, then the memory locations that are read are still accessible and will not be overwritten by any other process or calling context, while the dataflow_region is being executed.
• If A is written by proc2, then the memory locations written will not be read, before their definition, by any other process or calling context, while dataflow_region is being executed.

A typical scenario is when the caller updates or reads these variables only when the dataflow region has not started yet or has completed execution.

In summary, the Dataflow optimization is a powerful optimization that can significantly improve the throughput of your design. As there is reliance on the HLS tool to do the inference of the available parallelism in your design, it requires the designer's help to ensure that the code is written in such a way that the inference is straightforward for the HLS tool. Finally, there will be situations where the designer might see the need to deploy both the Dataflow model and the Task-Channel model in the same design. The next section describes this hybrid combination model that can lead to some interesting designs.