Announcement

Collapse
No announcement yet.

Microsoft Announces "Project Mu" For Open-Source UEFI Alternative To TianoCore

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Microsoft Announces "Project Mu" For Open-Source UEFI Alternative To TianoCore

    Phoronix: Microsoft Announces "Project Mu" For Open-Source UEFI Alternative To TianoCore

    Microsoft is getting into the open-source UEFI game with today's announcement of Project Mu, which powers their Surface hardware as well as Hyper-V platform...

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    That is really interesting. Up till now, it was always known that Tianocore can only bring up a full system under specific circumstances with Coreboot. That, or as an UEFI implementation for virtual machines.

    The Project Mu fork actually being used to initialize the Surface is serious news.

    Comment


    • #3
      If Microsoft pushes this, could this finally be the end of the bad joke that is vendor UEFI implementations?

      Comment


      • #4
        As long as the UEFI Secureboot mechanism and keys are open, I'm OK with it. No more 'Oops, is that a universal Golden Key?!' bullshit. Fool me once, MFer... and that's all you get.

        Comment


        • #5
          Originally posted by bachchain View Post
          If Microsoft pushes this, could this finally be the end of the bad joke that is vendor UEFI implementations?
          Substitute "Microsoft" and replace it with "Intel" and "UEFI" with "BIOS" and you'd have the same hope some of the more naive held about the switch to the permissively licensed UEFI reference implementation. The reality is far from it, and while I applaud Microsoft opening their code, there's no practical hope that this will solve the problem of broken OEM UEFI implementations in the broader industry.

          It might move more security conscious users towards Surface hardware, so long as they can install something other than Windows on it, since it may become possible to audit parts of the firmware, but it won't have any meaningful effect on the rest of the consumer PC market. OEMs are more likely to go Apple's route and further lock down the firmware with security chips and hash checking instead if it gets to the point where the consumer sector (vendors and consumers) takes more than a fatalistic attitude to security and privacy issues. It's an economic issue rather than a security issue. The only way to change that equation is to make it economically risky to ignore security and privacy. Right now there's no real incentive to do so.

          Also, keep in mind just because it CAN be audited, doesn't mean it ever WILL be audited and those doing the diving may not have the experience to do it right. There's a finite (small) number of people in the world with the experience, time, and inclination to audit board firmware packages.

          Comment


          • #6
            Uefi and all the OEM implementations are a nightmare regarding security, support and standardisation. Everything that will help to improve this is more than welcome and in this case maybe Microsoft is one of the few companies powerful enough to change something.

            Comment


            • #7
              When you see how fast CoreBoot can be why would you possibly want UEFI! Chromebooks use CoreBoot and my ancient Asus C710 boots into Linux Mint very quickly. It's using SeaBIOS at the moment and if it actually used the Linux Kernel payload, the boot would be even quicker still. UEFI is horrible, inefficient, a joke in terms of security and badly designed. I applaud Microsoft in open sourcing Project Mu but Microsoft pushed for UEFI in the first place. u-boot can be even faster. Let's push for a better solution, not prop up a bad one.

              Comment


              • #8
                Originally posted by debianxfce View Post
                There is no clear instructions how to clone an uefi disk in Linux so it is better and easier to use MBR.
                Same way as you would clone MBR disk. The only difference is that you may need to add new boot entry afterwards using efibootmgr.

                Comment


                • #9
                  I fully agree with my friend's statement at this wolf's lair
                  coreboot open source BIOS is already there and supports a lot of real motherboards. https://coreboot.org/status/board-status.html Instead of reinventing the wheel as "mu" it would have been more ef...

                  ( https://web.archive.org/web/20181220...t/mu/issues/52 if they shut it down )

                  coreboot open source BIOS is already there. Why keep reinventing the wheel?

                  Comment


                  • #10
                    Originally posted by Sonadow View Post
                    That is really interesting. Up till now, it was always known that Tianocore can only bring up a full system under specific circumstances with Coreboot. That, or as an UEFI implementation for virtual machines.

                    The Project Mu fork actually being used to initialize the Surface is serious news.
                    This is still technically a respin of Tianocore though so it's still UEFI, but it's a full-blown package with GUI and all, plus there is infrastructure for tracking multiple projects/devices with it.

                    So this is going to compete directly with AMI, Insyde and other UEFI vendor implementations.

                    Comment

                    Working...
                    X