Russia pivots to Chinese CPUs that aren't subject to US sanctions — Russia's homegrown Linux-based Alt OS now supports Chinese LoongArch chips

Loongson
(Image credit: Loongson)

The Alt operating system developed by Moscow, Russia's Basalt SPO has been recompiled to support Chinese Loongson processors based on the LoongArch architecture. Blacklisted Loongson has actively supported the porting process. The OS is available as a distro with a basic set of programs (Firefox, GIMP, LibreOffice) that can be installed on desktops, workstations, and servers. The company says the product is offered 'as is,' and its commercial distributives will be available later.

Alt is now the first Russian operating system capable of running on Loongson's processors based on the 64-bit LoongArch architecture, such as LS5000 and LS6000 series, which some in Russia consider alternatives to x86 CPUs from AMD and Intel. Recently, China lifted the export ban on Loongson's latest CPUs to Russia, which opened doors to using these processors in desktops, laptops, servers, storage equipment, and other applications.  

The distro is based on the unstable, experimental branch of the Sisyphus project repository, which developers use to test new ideas and developments, including support for various processor architectures, reports CNews. The distro is available for download, but a stable release with LoongArch support is set for Q1 2024. 

"In early 2024, Basalt SPO will release a new stable branch of the Sisyphus project repository — the Eleventh platform," said Alexei Novodvorsky, an advisor to the CEO of Basalt SPO. "We plan to release Alt OS p11 distributions, including distributions for the LoongArch platform in Q1 2024."

The Alt OS is only the first step towards broader support of LoongArch-based CPUs in Russia, but it is undoubtedly essential.

"Basalt SPO is presenting a unique technology — the first Russian operating system for the Loongarch64 architecture," said Sergey Trandin, CEO of Basalt SPO. "Loongarch64-based processors have recently entered the Russian market, which means organizations and enterprises face a large-scale task of adapting software to work with the Chinese architecture. The Alt operating systems can become a platform for testing such products."

Basalt SPO developers adapted the Alt OS for full-fledged work with Loongarch64 in just nine months. This quick result was possible thanks to the automation of the assembly process using its unique process flow called catch-up package assembly. 

"The catch-up assembly technology significantly accelerates the process of adapting the OS for new architectures," said Trandin. "Originally, packages for Loongarch64 were assembled 'manually' — the assembly of the first thousand took about half a year. In July, an assembly line was set up for this platform, and the process proceeded much faster. Today, more than 16,000 software packages for the Chinese architecture have been compiled in the Sisyphus project repository, and their base is constantly replenished. The source codes of the Loongarch64 storage packages are also published and are developed as part of the Sisyphus project."

Software specialists from Loongson actively supported Basalt SPO by providing detailed documentation and modern Loongson-based servers for testing the OS.

Russia's adaptation of the Linux-based Alt to support the LoongArch architecture from China's Loongson company is a crucial development. On the one hand, by mating the Alt OS with Loongarch64 processors, Russia is moving towards greater technological independence from Western technologies and diversifying its use of processor technology. On the other hand, it introduces a dependency on Chinese technology.

Anton Shilov
Freelance News Writer

Anton Shilov is a Freelance News Writer at Tom’s Hardware US. Over the past couple of decades, he has covered everything from CPUs and GPUs to supercomputers and from modern process technologies and latest fab tools to high-tech industry trends.

  • abufrejoval
    I guess we'll see more of these non-US controlled architectures and Linux variants along the road, because US governments and agents (just like everbody else, I guess) can't resist weaponising technology and IT.

    Where it becomes really interesting is when these forks are upstreamed and vulnerability-management is being done on both sides of the political trenches.

    There is a good chance, collaboration will just continue in the spirit of open source, but if course this could get weaponized and core developers may be coerced into doing so even against their will.

    Ever since I heard how Ross-Perot-Systems, which used to run Iran's health care system during the times of the Shah, used software that would basically send critical health data to the US embassy across the street via electronic emissions much like a telegraph, modulating CPU produced noise using modified idle loops, I've become rather paranoid.

    Yet within open source also lies some of the biggest potential to regain bilateral trust and peace.
    Reply
  • Jerry_W14
    abufrejoval said:
    I guess we'll see more of these non-US controlled architectures and Linux variants along the road, because US governments and agents (just like everbody else, I guess) can't resist weaponising technology and IT.

    Where it becomes really interesting is when these forks are upstreamed and vulnerability-management is being done on both sides of the political trenches.

    There is a good chance, collaboration will just continue in the spirit of open source, but if course this could get weaponized and core developers may be coerced into doing so even against their will.

    Ever since I heard how Ross-Perot-Systems, which used to run Iran's health care system during the times of the Shah, used software that would basically send critical health data to the US embassy across the street via electronic emissions much like a telegraph, modulating CPU produced noise using modified idle loops, I've become rather paranoid.

    Yet within open source also lies some of the biggest potential to regain bilateral trust and peace.
    I see ample opportunities to inject malicious malware in this process. Oh man a hackers dream, who will be the 1st to inject ransomware into all those russian made Chinese computers. LOL. I can see them laughing all the way to the water cooler at the NSA. My bet is North Korea though, Kim would love this opportunity to double dip into Putin's pockets! It's a cruel, cruel world in the axis of evil.
    Reply
  • RedBear87
    Jerry_W14 said:
    I see ample opportunities to inject malicious malware in this process. Oh man a hackers dream, who will be the 1st to inject ransomware into all those russian made Chinese computers. LOL. I can see them laughing all the way to the water cooler at the NSA. My bet is North Korea though, Kim would love this opportunity to double dip into Putin's pockets! It's a cruel, cruel world in the axis of evil.
    We live in the world where the German chancellor was being spied by the NSA, so it's not exactly limited to the "axis of evil"...
    Reply
  • Jerry_W14
    Yeah but Germany did start WWII after losing WWI. So history says you have to keep your eyes upon them. Now it's Russia and China we must keep eyes upon.
    Reply
  • bit_user
    Jerry_W14 said:
    I see ample opportunities to inject malicious malware in this process.
    If there's a vulnerability in just the Loongarch path, then you pretty much know there's little chance for it to impact "friendly" users or their networks. That's the main downside I see to having your own architecture, assuming they don't just permanently fork the Linux kernel and tightly control all contributions.
    Reply
  • buzeze
    abufrejoval said:
    Where it becomes really interesting is when these forks are upstreamed and vulnerability-management is being done on both sides of the political trenches.
    I don't expect it'll be upstreamed in near future.
    There are precedent on rejecting kernel patch by political reasons: https://lore.kernel.org/all/20230314103316.313e5f61@kernel.org/
    Reply
  • bit_user
    buzeze said:
    I don't expect it'll be upstreamed in near future.
    There are precedent on rejecting kernel patch by political reasons: https://lore.kernel.org/all/20230314103316.313e5f61@kernel.org/
    The longer they maintain out-of-tree patches, the more effort it's going to be. If they're blocked from upstreaming for long enough, it'll be creating a defacto schism.

    That's a somewhat extreme take, however. There are patchsets which have been maintained out-of-tree for many years, so it's not impossible they could take this route.
    Reply