Stories
Slash Boxes
Comments

SoylentNews is people

posted by janrinok on Friday February 28 2020, @02:04PM   Printer-friendly

UEFI Boot Support Published For RISC-V On Linux

Western Digital's Atish Patra sent out the patch series on Tuesday for adding UEFI support for the RISC-V architecture. This initial UEFI Linux bring-up is for supporting boot time services while the UEFI runtime service support is still being worked on. This RISC-V UEFI support can work in conjunction with the U-Boot bootloader and depends upon other recent Linux kernel work around RISC-V's Supervisor Binary Interface (SBI).

Building off the common (U)EFI code within the Linux kernel, the RISC-V bring-up so far is just over four hundred lines of code. Depending upon how quickly this code is reviewed, the initial UEFI RISC-V support could land for the Linux 5.7 cycle. So far this RISC-V UEFI boot support has been tested under QEMU.

Unified Extensible Firmware Interface (UEFI).

See also: Linux EFI Going Through Spring Cleaning Before RISC-V Support Lands

Related:
Western Digital to Transition Consumption of Over One Billion Cores Per Year to RISC-V
Western Digital Publishes RISC-V "SweRV" Core Design Under Apache 2.0 License


Original Submission

Related Stories

Western Digital to Transition Consumption of Over One Billion Cores Per Year to RISC-V 17 comments

From a Western Digital press release:

Western Digital Corp. (NASDAQ: WDC) announced today at the 7th RISC-V Workshop that the company intends to lead the industry transition toward open, purpose-built compute architectures. In his keynote address, Western Digital's Chief Technology Officer Martin Fink expressed the company's commitment to [...] transitioning its own consumption of processors – over one billion cores per year – to RISC-V.


Original Submission

Western Digital Publishes RISC-V "SweRV" Core Design Under Apache 2.0 License 9 comments

Western Digital's RISC-V "SweRV" Core Design Released For Free

Western Digital has published a register-transfer level (RTL) design abstraction of its in-house designed SweRV RISC-V core. The SweRV core is one of several RISC-V projects the company as undertaken as part of their effort to spearhead the ISA, its ecosystem, and foster their own transition away from licensed, royalty-charging CPU cores. In accordance with the more open design goals of RISC-V, the publication of the high-level representation of SweTV means that third parties can use it in their own chip designs, which will popularize not only the particular core design, but also the RISC-V architecture in general.

The RTL design abstraction of Western Digital's RISC-V SweRV core is now available at GitHub. The design is licensed under the Apache 2.0 license, which is a very permissive (and non-copyleft) license that allows the core to be used free of charge, with or without modifications, and without requiring any modifications to be released in-kind. In fact the requirements of the license are quite slim; besides requiring appropriate attribution, the only other notable restriction is that third party developers cannot use Western Digital's brands to mark their work.

Previously: Western Digital to Transition Consumption of Over One Billion Cores Per Year to RISC-V
Western Digital Unveils RISC-V Controller Design


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.
(1)
  • (Score: 2) by Rich on Friday February 28 2020, @04:03PM (4 children)

    by Rich (945) on Friday February 28 2020, @04:03PM (#964146) Journal

    One could very well live with a plain device tree and chainloading kernels if any weird shit has to go on before the main kernel is booted. For new hardware, power management drivers might as well be done as kernel module (or just SBI ("BIOS") code), rather than being put into the separate UEFI environment - and eventually being reimplemented in the kernel, because the lower-level stuff is unmaintained and broken for any given piece of hardware older than a year.

    I wonder why WD is doing this and who foots the bill.

    • (Score: 1, Insightful) by Anonymous Coward on Friday February 28 2020, @06:29PM (1 child)

      by Anonymous Coward on Friday February 28 2020, @06:29PM (#964249)

      That is the sole purpose for UEFI: providing enhanced backdoor opportunities and less user control over the firmware stack for their hardware, as well as mandating (in)secure signed platform boot. Of course for now SOME hardware can disable it, but you will note almost no hardware actually supports user signed or unsigned trusted boot bringup to allow implementing the signing key in flash while bringing one's system up to trusted status with a user compiled open source or audited trusted execution environment.

      RISC-V has turned from a platform claimed to liberate us all to another part of the problem. In order for us to regain control of the hardware and software ecosystem it is time hardware and software engineers who still believe get together and design a successor to the AT/ATX PC clone architecture with trusted platform security intended to protect the end user, not the governments and cabals of media and technology companies which have trojaned the security of the entire globe.

      • (Score: 2) by Rich on Saturday February 29 2020, @03:10PM

        by Rich (945) on Saturday February 29 2020, @03:10PM (#964600) Journal

        If WD would see gains to be made from locking users out, why would they give away the keys? Making something like BIOS/OpenFirmware/UEFI isn't black magic and rather than using FORTH oder some bytecode, they could just stick to R-V binary, after all it's what they want to push through. So they'd cook their own soup and force a standard that the container crowd needs to use. When the world has adopted it, they'd do a minor revision and announce to the world that they are now the single signing authority for R-V. Of course, in.. er... order to guarantee coherent platform management to data center users, you'd have to bundle their platform snoop... supporting blob to get signed. Or something the like.

        For their NAS stuff, it has no benefits. They control the whole stack and can tivoize it (german saying:) until the doctor comes. Only effect from UEFI is longer boot times.

        The only one I see benefitting is Microsoft, because they couldn't manage drivers/images for the emerging zoo of R-V hardware in-house. With UEFI they can ship a single image. So I could very well imagine that MS contracted that job with WD for, er... Azure cloud server requirements.

    • (Score: 0) by Anonymous Coward on Friday February 28 2020, @08:29PM (1 child)

      by Anonymous Coward on Friday February 28 2020, @08:29PM (#964314)

      I'm thinking it's to do with the RISC-V chips they're developing for their HDD controllers.

      • (Score: 1) by petecox on Friday February 28 2020, @11:53PM

        by petecox (3228) on Friday February 28 2020, @11:53PM (#964409)

        400 lines of code to support what was already there for ARM, whose SBBR specification mandates UEFI.

        If WD want to release a riscv-powered NAS, compliance with whatever's expected on ARM64...

  • (Score: 0) by Anonymous Coward on Friday February 28 2020, @09:19PM (1 child)

    by Anonymous Coward on Friday February 28 2020, @09:19PM (#964338)

    World domination plan:

    Step 0: Create an interesting new ISA that is in some ways quite a classical RISC platform, but learns from the past in all sorts of important implementation details.
    Step 1: Port a part of the boot process from the arcane history of x86 PC's.
    Step 2: ???
    Step 3: Profit!

    Seriously, EFI is a weird hack of some (but not all) of the good ideas of OpenFirmware, awkwardly grafted onto PC's, with a ton of complexity and implementation requirements for things like "portable drivers" that will never actually be used by the OS for anything. I get that there's utility in "principle of least surprise" by having Linux boot in similar ways across platforms, but EFI is probably the wrong sort of designed-by-committee monstrosity to try and make universal on a clean-sheet rethink of a platform.

    • (Score: 2) by Debvgger on Friday February 28 2020, @09:54PM

      by Debvgger (545) on Friday February 28 2020, @09:54PM (#964364)

      Step 2 is building an Amiga accelerator with a RISC-V CPU.

(1)