Stories
Slash Boxes
Comments

SoylentNews is people

posted by chromas on Tuesday September 15 2020, @10:23PM   Printer-friendly
from the send-me-a-sample dept.

SiFive to Debut RISC-V PC for Developers based on Freedom U740 next-gen SoC

In recent years, people have discussed the need to have Arm-based PCs or workstations for developers to work directly on the target hardware, and there are now several options including SynQuacer E-Series 24-Core Arm PC, Ampere eMAG 64bit Arm Workstation, and HoneyComb LX2K 16-core Arm Workstation.

Now it appears we'll soon get something similar for RISC-V architecture with SiFive to debut the first RISC-V PC for developers at the Linley Fall Processor Conference 2020 taking place on October 20-22 and October 27-29. The PC will be powered by Freedom U740 next-generation RISC-V processor that will also be introduced at the event.

We have very few details about this point in time, but the company points the SiFive Freedom U740 (FU740) SoC will enable professional developers to create RISC-V applications from bare-metal to Linux-based. The processor is said to combines[sic] a heterogeneous mix+match core complex with modern PC expansion capabilities, which probably means PCIe, SATA etc.., and the company will provide tools to ease professional software development.

Freedom U740 details are unknown, but Freedom U540 is a quad-core CPU that was used in the HiFive Unleashed single-board computer.

Related: SiFive Introduces RISC-V Linux-Capable Multicore Processor
SiFive HiFive Unleashed Not as Open as Previously Thought
SiFive Announces a RISC-V Core With an Out-of-Order Microarchitecture
GlobalFoundries and SiFive Partner on High Bandwidth Memory (HBM2E)


Original Submission

 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 2, Interesting) by Anonymous Coward on Tuesday September 15 2020, @10:44PM (3 children)

    by Anonymous Coward on Tuesday September 15 2020, @10:44PM (#1051467)

    Computer architecture has basically been boring for the last decade or two. X86 is fast, and ARM is cheap and low power. It has pretty much just been a story of consolidation.

    But if you can remember the 80's and 90's, there was a much wider diversity, and constant excitement with new architectures that were supposedly going to solve all of the world's problems and make you breakfast. MIPS! Intel trying to replace 8086 with iAPX 432! SPARC! Intel trying to replace x86 with i860! Alpha! PowerPC! Intel trying to replace x86 with IA64! Etc.

    Obviously, FU740 isn't going to kill x86. SiFive doesn't have as large a budget as Intel, and they don't have the economies of scale of x86 and ARM to drive down prices. But RISC-V is a really interesting architecture in some ways. I think the design of the vector system is a lot more interesting than Intel's AVX512 -- Look at Linus Torvalds' rant about it. The special purpose fixed width vector units in Intel chips burn a ton of power, but software needs to be explicitly written to take advantage of the exact vector register width. The RISC-V design is a lot more flexible, so software made today should also run great in a few years when new RISC-V HPC chips come out with super wide vector registers analogous to AVX-512 or 1024. Meanwhile, when Intel moves on past AVX-512, all of the current code that has been lovingly optimized for it won't work as well on the new chips.

    Plus if you've ever tried writing an OS from scratch as a learning exercise, modern x86 is a ton of complicated legacy gotchas. I am curious to learn more about the non-processor parts of the RISC-V ecosystem, because it could be a much more convenient target for "modern" hobby OS development projects.

    Starting Score:    0  points
    Moderation   +2  
       Interesting=2, Total=2
    Extra 'Interesting' Modifier   0  

    Total Score:   2  
  • (Score: 4, Informative) by bzipitidoo on Wednesday September 16 2020, @04:57AM (1 child)

    by bzipitidoo (4388) on Wednesday September 16 2020, @04:57AM (#1051575) Journal

    Commercially successful though it is, x86 has been a terrible architecture from the beginning. It sucks less now, but it's still lacking. The reason Linux does not support anything older than the 486 is the lack of support for semaphores. (Yes, Linux used to work on the 386, but they dropped that because of that lack.) Support for virtualization is still poor, though they're improving. If the architecture had good support for virtualization, there would be no need for special software to run a virtual machine. Wouldn't need VMware or VirtualBox. A problem that just gets worse and worse over time is legacy support. x86 has a whole bunch of crud that isn't needed, but they have to keep it all to maintain backward compatibility. Really can't reassign opcodes, can only add new ones.

    Also, things have changed an awful lot since the 1980s. For instance, ASCII is giving way to UTF-8. Although UTF-8 is still byte based, it's a variable number of bytes. It may be worthwhile to have some hardware handling for it. Computer architecture is due for a major reboot.

    So, yes, I'm keenly interested in RISC-V.

    • (Score: 1, Informative) by Anonymous Coward on Wednesday September 16 2020, @08:37AM

      by Anonymous Coward on Wednesday September 16 2020, @08:37AM (#1051626)

      The reason Linux does not support anything older than the 486 is the lack of support for semaphores

      Not semaphores directly, but atomic instructions (CMPXCHG), which are a requisite for implementing them efficiently. It is possible to implement semaphores on 386 using critical sections, but that would require a lot of hairy custom code that nobody would use, and could easily bitrot.

      If the architecture had good support for virtualization, there would be no need for special software to run a virtual machine

      You mean process isolation, not virtualization, I guess. Running virtual machines would always require custom software to manage the virtual hardware and the VM consoles.

  • (Score: 2) by Wootery on Wednesday September 16 2020, @10:44PM

    by Wootery (2341) on Wednesday September 16 2020, @10:44PM (#1052000)

    Computer architecture has basically been boring for the last decade or two.

    Including GPUs and APUs?