Arm-2D is an open-source project for 2.5D image processing on Cortex-M processors.
Arm-2D consists of:
Position in Ecosystem:
Current Focus and Objectives:
arm_2dp_rgb565_copy_with_src_mask_only() rather than the generic one, i.e. arm_2dp_rgb565_copy_with_src_mask() which takes a mirroring mode as a parameter and keeps all low-level implementations of all possible mirroring modes.Potential users of Arm-2D may have different backgrounds and purposes. Whether you are an application engineer for developing drivers from a silicon vendor, a system engineer of a GUI stack, or an application developer, based on the purpose of using Arm-2D, there are two ways to get started:
Getting Started as an Application Designer
In some resource-constrained systems, the Flash (ROM) and SRAM are limited (e.g. 64K Flash, 12K SRAM), or the memory space left for GUI applications is limited, so such systems cannot afford the cost of a decent GUI stack. Meanwhile, such a low-cost system usually won't have a complex graphical user interface, hence application developers can directly use the arm-2d APIs to draw the panels and scenes.
Getting Started as a Professional User
Application engineers from silicon vendors often need to write drivers for proprietary hardware accelerators. System engineers from GUI software providers may wish to use the APIs provided by Arm-2D to accelerate low-level operations. Engineers focused on embedded video processing may try using the Arm-2D APIs to process the frame buffer just captured from the camera. People from the open-source community may find something interesting to contribute. All developers mentioned above are seen as professional users.
The Arm-2D library provides Low-Level 2D Image Processing Services mainly used in Deep Embedded Display system. The supported features include but not limited to:
Disp0_DrawBitmapNOTE:
As long as the size is smaller than 32767 * 32767.
Transform (i.e. rotation and scaling)
__ARM_2D_HAS_ANTI_ALIAS_TRANSFORM__ to 1 at compile-time.An Unified and User-Friendly Programmers' Model
Helper Services ,Tools and Others
arm_2d_helper_is_time_outarm_2d_helper_time_liner_sliderarm_2d_helper_time_cos_sliderarm_2d_helper_time_half_cos_sliderarm_2d_align_centre, arm_2d_align_bottom_right etc.arm_2d_dock_top,arm_2d_dock_right, arm_2d_dock_vertical, __item_line_dock_horizontal etc.__item_line_horizontal and __item_line_vertical__item_horizontal and __item_verticalimg2c.py for generating arm-2d resources from user specified imagesttf2c.py for generating user customized A1, A2, A4 and A8 fonts from an user specified TrueType Fontjinja2c.py for code generation.Ready and Welcome 3rd party adoption
With more and more smart IoT edge devices introduced to our daily lives, people who are used to the smart-phone-like graphic user interfaces (GUIs) want to have the similar modernized user experience even when using microcontroller-based products. This trend has been long observed and understood by Arm's eco-partners.
As a result, many silicon vendors introduce dedicated hardware accelerators into their microcontroller products to help 2D image processing. Meanwhile, GUI service providers also update their product lines to target microcontroller based GUI applications. Many open-source embedded GUI stacks are hot on GitHub, e.g. LVGL.
In fact, using GUI with microcontrollers isn't new at all. Still, until the recent rise of IoT and AI, people find that the simple and monotonous GUI is really unbearable. The pursuit of user experience similar to that of a smartphone has become a basic requirement for products.
On the other hand, seemingly complicated GUIs are often implemented with simple 2D image-copy and -tiling. Even the translucent effect is not unaffordable for microcontrollers that often run under tens or hundreds of MHz.
Technologies used for 2D graphics have been matured as early as the era of 8-bit Gaming Console. Since 8-bit 6502 can achieve fancy graphics effects, why can't the most advanced Cortex-M processor of the day?
As shown in Figure 1-2, Linux based systems have a complete ecosystem that provides GPU drivers and GUI software services. In contrast, resource-constrained embedded systems do not have such luxury. To enhance the MCU ecosystem, Arm-2D is introduced.
Figure 1-2 Ecosystem Comparison between Rich Embedded and Constraint Embedded System in GUI
When we look at the traditionally embedded GUI architecture(as shown in Figure 1-3 ), there are four layers: 1) the application and designer layer, 2) the GUI software service layer, 3) the rendering layer, and 3) the hardware driver layer.
Arm-2D focuses on accelerating the low-level 2D image processing, and will not touch upper layers of the GUI software stacks, hence avoiding competing with GUI service providers in the same ecosystem. In fact, because Arm has proposed an abstraction layer for the low-level 2D image processing, a whole ecological level of cooperation can be quickly established between chip manufacturers that provide hardware accelerators and software providers that provide GUI services. Everyone can concentrate on enlarging their own values. For example, by adding arm-2d adaptor drivers for their proprietary 2D accelerators silicon vendors can achieve a wide range of support from GUI providers who build/port their GUI stack upon Arm-2D APIs.
Figure 1-3 The Hierarchy of a Typical Embedded GUI System.
The library is targeting ALL Cortex-M processors with/without various hardware 2D image accelerators:
The library is designed with ACI (Arm Custom Instructions) in mind. Accelerations implemented with user-defined instructions can be integrated into the library easily without modifying the existing Arm-2D library or upper-layer software.
The library is designed with 2D image accelerators in mind and following the feature-agnostic principle. The Support for accelerators can be added easily without modifying the existing Arm-2D library or upper-layer software.
The library is designed with resource constraints in mind. For Cortex-M processors with 8K~32K SRAM that cannot afford a complete-framebuffer, Arm-2D introduces a feature called Generic Partial Framebuffer enabling those existing MCUs to run GUI applications in a decent frame rate.
| Projects | Description | Folder | Note |
|---|---|---|---|
| [template][bare-metal] | It is a project template for the bare-metal environment. | examples/[template][bare-metal] | |
| [template][cmsis-rtos2] | It is a project template for the RTOS environment, which use CMSIS-RTOS2 as an example to show how Arm-2D can work with an RTOS. | examples/[template][cmsis-rtos2] | |
| [template][pc][vscode] | It is a project template for PC (i.e. MacOS, Windows and Linux) using VS Code + SDL2 | examples/[template][pc][vscode] | |
| [template][csolution] | It is a csolution project template. | examples/[template][csolution] |
There is no public 2D image processing benchmark available for microcontrollers. To facilitate the library development and optimization, we overcome this problem with following methods:
Figure 1-4 A snapshot of the generic benchmark running on Simulation
Figure 1-5 Performance Comparison among some Cortex-M processors
| Folder and File | Type | Description |
|---|---|---|
| Library | Folder | This folder contains the source files and header files of the library. |
| Helper | Folder | This folder contains the source files and header files of helper functions / services. |
| Acceleration | Folder | This folder contains hardware specific accelerations, e.g. Helium-ACI example in FVP |
| documentation | Folder | This folder contains all the documents. |
| examples | Folder | This folder contains all the examples, controls and templates etc. |
| README | .md | The README.md you are currently reading. |
| LICENSE | License | The Apache 2.0 License |
| tools | Folder | This folder contains useful utilities for using the library. For example, img2c.py is a python script that converts specified pictures into the arm2-d tile data structures. |
| Name | Description | Location |
|---|---|---|
| API Manual | An API manual generated by Doxygen | *  : Documentation for latest official release * Documentation for latest development release |
| README.md | It is the document that you are reading. It provides basic information and guidance for the arm-2d library. | (root) |
| getting_started_as_an_application_designer.md | A guidance for GUI application designers. | documentation |
| getting_started_as_a_professional_user | A guidance for professional users. | documentation |
| how_to_deploy_the_arm_2d_library.md | A step by step guide that helps you to deploy the library to your existing or new projects. | documentation |
| introduction.md | A relatively detailed introduction for the library, including basic concepts, programmers' mode etc. | documentation |
| how_to_use_layout_assistant.md | A detailed document introduce the layout assistant helper service. | documentation |
| how_to_use_tile_operations.md | A detailed document elaborates the APIs dedicated to basic tile operations in the arm-2d library. | documentation |
| how_to_accelerate_arm_2d.md | A detailed document elaborates the methods available for accelerating Arm-2D library. | documentation |
| how_to_use_alpha_blending_operations.md | A detailed document elaborates the APIs dedicated to alpha-blending services provided by the arm-2d library. | documentation |
| how_to_use_conversion_operations.md | A detailed document elaborates the APIs dedicated to colour space conversion services provided by the arm-2d library. | documentation |
| how_to_use_drawing_operations.md | A detailed document elaborates the APIs that provide basic point-drawing and colour-filling services in the arm-2d library. | documentation |
| how_to_contribute | A guidance for contribution. | documentation |
As mentioned before, this project aims to explore and demonstrate the possibilities of delivering modern GUI using low-cost and resource constraint micro-controllers. We expect that arm-2d could inspire more similar initiatives and engineering practices. Hence, your feedback and thoughts are precious to us.
If you can spend some time and try the library, please feel free to leave your thoughts. It would be nice to cover the following perspectives:
Thank you for your time.
Arm-2D Development Team.
23 May 2024