Technologies

News information

The domestic operating system "surpasses" again, RT-Thread promotes the transformation of the AIoT industry

Release on : Dec 21, 2021

The domestic operating system "surpasses" again, RT-Thread promotes the transformation of the AIoT industry
Domestic operating system RT-Thread promotes AIoT
On December 18th, the RT-Thread Developer Conference was held at the Sheraton Shenzhen Greater China Hotel. The theme of this conference was "Beyond", which means to transcend oneself and break through boundaries. In the morning keynote session, the main person in charge from RT-Thread reviewed the achievements of RT-Thread in 2021, and also introduced the commercialization of RT-Thread in the past few years, as well as the technology trend and future of the next few years. trend.


RT-Thread ecological results summary and future trend
In the opening keynote speech, Mr. Xiong Puxiang, the founder of RT-Thread, summarized the ecological achievements of RT-Thread this year. Up to now, RT-Thread has nearly 400 software packages, and many of them are communities. With the participation of the small partners, the downloads of these software packages are close to 7 million; the contributors of RT-Thread are close to 1,000; at the same time, it has also reached cooperative courses or laboratories with a series of universities, and currently there are more than 150 in the country. There is also an overseas university offering RT-Thread courses; in addition, it has also been recognized by chip manufacturers at home and abroad. Currently, more than 10 chip manufacturers have become paying members of RT-Thread.


Picture: Mr. Xiong Puxiang, founder of RT-Thread

In terms of technological evolution, Mr. Xiong Puxiang hopes to rely on RT-Thread, the underlying core platform technology, to cover general-purpose MCU chips, SoC chips used in IoT, and NPU chips with richer application scenarios through the standard version, nano version, and Smart version. And other chip platforms of different sizes. In other words, they are trying to do a series of integration through the RT-Thread core platform to support multiple platforms, applications and scenarios.


In terms of technical layout, Xiong Puxiang believes that RTOS will move towards componentization and standardization. Specific to the RT-Thread side, first of all, in the future, they will build standardized components around the RT-Thread basic platform, and standardize a series of corresponding components, including the name, author, link, and API interface and API interface. Wait everywhere.

The second is to do programming in code. In the future, RT-Thread Studio will be used to simplify programming and low-code programming to better help developers develop products more efficiently and with low error rates. In other words, RT-Thread Studio will add more efficient and useful functions in the future.

The third is to integrate multiple systems and different systems. At present, the main platform for AI landing is a heterogeneous, multi-core large computing power platform. In the future, RT-Thread will also do more integration for multi-core, large computing power and heterogeneous chips. For example, for armCortexA core multi-core chips, RT-Thread launched VM-Thread, which is a virtualized component, which can be used to virtualize multi-core chips based on the Smart super performance. That is, on a multi-core chip, it can run multiple operating systems, including RT-Thread Smart, RT-Thread RTOS, and Linux or Android. Xiong Puxiang explained that the reason for the development of this component is that in some occasions, multi-computing power and multi-core chips also need to be segmented and cut in different domains, RT-Thread or RT-Thread+vmThread, which can provide better Real-time mechanism, safety mechanism, low power consumption mechanism.

He also revealed that in the Demo display area of ​​the RT-Thread Developer Conference, a Demo board based on Raspberry Pi 4B can demonstrate the effect of the VM-Thread component. Next year Q1 will launch more virtualized products based on multi-core processors.


Regarding the technological trend of RT-Thread, RT-Thread currently has RT-Thread RTOS real-time operating system for real-time operating environment; RT-Thread Smart for complex armCortexA core chips; and VM for multi-core CortexA core chips -Extension of Thread virtualization. VM-Thread is just an extension, it can be well combined with RT-Thread RTOS or RT-Thread Smart to provide a series of virtualization support to the upper layer. In the future, a converged platform will be built through virtualization to meet the needs of more application scenarios.

The commercialization exploration road of RT-Thread
According to Zou Cheng, chief operating officer of RT-Thread, RT-Thread currently has more than 1.4 billion installed capacity and a wide distribution of users, including security, consumer electronics, industrial, medical, automotive, and wireless connections, including cellular connection chips. Manufacturers, etc.; more than tens of thousands of open source enterprise users; more than 100 chip manufacturers support the RT-Thread system, and 30 to 40 chip manufacturers with RT-Thread loaded by default; users of RT-Thread Studio tools More than 150,000.


Picture: Zou Cheng, Chief Operating Officer of RT-Thread

These are the results of open source RT-Thread, so how can RT-Thread be commercialized? Zou Cheng said that they have been thinking about this issue and are constantly trying. "Within 3 years, our commercialization model will mainly provide value-added software for vertical markets, including certified OS, commercial components, OS custom development, and chip migration."

"Three years later, we believe that as the RT-Thread ecosystem further expands and the dominant position is further strengthened, there will be more models waiting for us, and we will continue to observe and expand." Zou Cheng further pointed out.

He also emphasized that commercialization will not affect open source. RT-Thread will continue to iterate and evolve in accordance with the original intention of open source, open source model, and open source culture to enhance its competitiveness and contribution to the industry.

RT-Thread's first anniversary, fully embracing developers
Chen Feng, the director of RT-Thread developer community operations, has been working on the RT-Thread ecosystem for many years. In his opinion, RT-Thread will become more robust, more interoperable, and safer in 2021.

One is that RT-Thread basically covers mainstream chips;
Second, the support for POSIX is more complete. At present, 90% of the PSE51 standard has been achieved, and the corresponding functions exceed the support of 200 functions. It is estimated that the support for PSE52 will be achieved next year, even to 53, 54 The support of the function. "In this way, those who like to use Linux in the past and some applications on Linux can be easily transplanted to RT-Thread." Chen Feng said.
Third, Libc has been further increased and improved, and its support for IAR, KEIL, and GCC has also become better.
The fourth is that Kernal code comments are more detailed, and a lot of improvements have been made in terms of grammar and comment style. This means that when developers use RT-Thread, they don't need to find documentation at all, and they can know how the API mentioned in it is used through comments.


Picture: Chen Feng, Director of RT-Thread Developer Community Operations

In terms of partners, RT-Thread has recently cooperated with many chip manufacturers. Many chip manufacturers have become gold or senior members of RT-Thread. To promote cooperation between the two parties. Including Renesas, Jihai Semiconductor, Huada Semiconductor, Xianji, National Technology, Nuvoton, Qinheng Microelectronics." Chen Feng said.

In terms of talent training, RT-Thread conducts many teacher training sessions every year. Chen Feng revealed that by December, the number of teachers participating in RT-Thread teacher training had exceeded 1,000. In addition, RT-Thread is also working on industrial collaboration projects, including engineering practice bases, curriculum reforms, etc. More than 150 colleges and universities are teaching RT-Thread. In terms of activities, RT-Thread sponsored the Smart Car Competition, Nanjing Smart Interconnect Embedded Competition, and the National College Student Operating System Design Competition. In terms of talent qualification certification, RT-Thread cooperated with STMicroelectronics, Fibocom and China Electronics Association to launch the embedded talent qualification certification. This is a brand new upgrade to the talent verification done by RT-Thread in the past. Cover more areas. On the other hand, the operating system used by default for the Loongson 1+X certification is based on RT-Thread.

In terms of overseas promotion, RT-Thread has been in China for 15 years this year. It started overseas promotion last year. After more than a year, there are currently more than 10,000 users overseas. In September of this year, RT-Thread held an online technical conference, two days of speeches, 20 topics, and speeches from English. More than 1,800 overseas fans and engineers registered to participate in RT-Thread activities. Interestingly, I thought that Portuguese lovers also wrote an e-book based on Portuguese RT-Thread.

It is worth mentioning that this year RT-Thread joined the RISC-V Foundation. RT-Thread has made a lot of technical reserves on the operating system level, SMP multi-core chips, and AMP heterogeneous chips. And these codes are all open, developers can find all the codes in the Github repository.

At the same time, RT-Thread also cooperates with other open source projects, such as cooperation with LVGL and so on.

Of course, in addition to these, RT-Thread also cooperates with the media and other chip manufacturers. For example, cooperate with electronic enthusiasts to launch a platform based on RT-ThreadInside, hoping to help partners who use RT-Thread to make products and solutions can find more users; cooperate with Black Sesame, and launch A1000+RT based on Black Sesame Intelligence -Thread Samrt modules and development kits, hoping to meet the industry's demand for large computing power AI.