Android could soon run on RISC-V chips thanks to advances made by Alibaba Cloud in porting the operating system to the emerging instruction-set architecture (ISA).
Last year, the Chinese cloud giant was successfully able to port basic functions including Chrome browsing in Android 10 to a RISC-V chip made by its subsidiary T-Head Semiconductor.
Now though, Alibaba Cloud has taken things a step further by working on porting Android 12 to RISC-V. The company has also figured out to enable third-party vendor modules to facilitate a number of new functions including audio and video playback, Wi-Fi, Bluetooth and camera operation.
- The rise of RISC: 2025 will be the year of the first quasi-mainstream RISC-V laptop as confirmed by the CEO of Framework but I don't think it will be ready for primetime
- British startup launches low-power RISC-V processor design to address the twin problems of memory and energy; promises up to 50% faster calculation speeds
To help facilitate these new functions, Alibaba Cloud has enabled more system enhancement features such as core tool sets, third-party libraries and SoC board support package on RISC-V. Additionally the company has successfully trialed TensorFlow Lite models on RISC-V to support AI functions like image and audio classification as well as Optical Character Recognition (OCR).
Senior director at Alibaba, Jianyi Meng provided further insight on the company’s progress with RISC-V in a press release, saying:
“The support of Android12, vendor modules and the AI framework on RISC-V based devices is another major milestone that we have achieved. We look forward to further contributing to the RISC-V community with our advanced technology and resources, and encouraging more innovation in the community together with global developers.”
MLPerf Tiny benchmark
If porting Android to RISC-V wasn’t enough, Alibaba Cloud’s RISC-V-based Xuantie C906 processor recently attained top marks in the MLPerf Tiny v0.7 AI benchmark that’s focused on IoT devices.
The Xuantie C906 excelled in all four of the benchmark’s core categories: visual wake words, image classifications, keyword spotting and anomaly detection. The performance of Alibaba’s custom-build processor shows the potential of the RISC-V framework in achieving tailored AI functions using extremely low computing power.
At the same time, Alibaba’s RISC-V based processors have already been deployed across a wide range of applications and use cases including smart home appliances, automotive environments and edge computing.
CEO of RISC-V International praised Alibaba’s work in further developing the RISC-V ecosystem especially when it comes to AI for IoT, saying:
“AI for IoT is a highly competitive arena where customisation at every level is critical to achieve new breakthrough results at very low power. Alibaba continues to build RISC-V industry leadership in parallel with their dedication and contribution to the global RISC-V community.”
