Enabling AI Engine Pipeline - 2023.2 English

AI Engine Tools and Flows User Guide (UG1076)

Document ID
Release Date
2023.2 English

The AI Engine Pipeline view can provide instruction pipeline details. To view AI Engine Pipeline view contents for a core, follow the instructions below.

  1. In the Component Explorer, right click on the AI Engine component to expand the Settings folder.
  2. Select the launch.json to open the Launch Configurations view.
  3. In the Launch Configurations view, open or create an AI Engine AIESIM launch configuration. To create one, click the '+' command in the toolbar menu.

  4. Select the Enable Pipeline View checkbox to enable the feature.

When the AI Engine simulator (AIESIM) is run, the Pipeline view is opened, as shown below.

The Pipeline view displays the program counter vertically and the kernel cycle count horizontally. ID represents instruction decode, and E1-E7 are execution stages. Almost all operations in the scalar unit are scheduled in E1 stage of the pipeline besides non-linear operations. The vector unit scheduling spans from the ID stage to the E6 stage. Address Generation Units (AGUs) span over two pipeline stages. The address is ready in the E2 stage of the pipeline. For load units, the data will be available in the AI Engine from the memory module in the E7 stage. For the store unit, the data will be sent out from the AI Engine to the memory module in the E5 or E6 stage of the pipeline depending on the type of instruction.