Slashdot Mirror

← Back to Stories (view on slashdot.org)

Intro To Intel's Next-Gen BIOS Architecture

Posted by timothy on from the redefining-basic dept.

An anonymous reader writes "This article introduces the Intel Platform Innovation Framework for the Extensible Firmware Interface (EFI), which is intended to provide an alternative to BIOS that will allow for faster booting, manageability, and additional features. According to the author, there has been rapid evolution of the personal computer platform since the 1980s. These advances have included order-of-magnitude increases in performance, ease-of-use, storage capacity, and connectivity. But there is one element of the PC that has not changed for the past 23 years -- namely, the BIOS (basic input/output system)."

36 comments

  1. Why is that bad? by AKnightCowboy · · Score: 4, Insightful

    The awesome backward compatibility of the PC architecture is one of the main reasons it's the most popular personal computer platform. Besides, there have been tons of advances in BIOS technology over the past 20 years. Sure, the basics are still in there, but even 5 years ago you couldn't boot off a USB key fob or even over the network (PXE) on most motherboards. 10 years ago power management was basically non-existent and things like ACPI and APIC were years away. Extend, don't just throw it away. If you're going to get rid of it then at least adopt Openboot firmware like Macs and Suns have.

    1. Re:Why is that bad? by zulux · · Score: 4, Interesting

      If you're going to get rid of it then at least adopt Openboot firmware like Macs and Suns have.

      The fun thing about Openboot is that it's in Forth:

      On the Openboot prompt you can do this:

      $ 42 69 + .

      and it will spit out

      $ 111

      --

      Moneyed corporations, non-working 'poor' and criminal prisoners are turning productive citizens into tax-slaves.

    2. Re:Why is that bad? by aster_ken · · Score: 4, Informative

      OpenFirmware (defined in IEEE 1275 as updated in 1994) is no longer being maintained. It was withdrawn by the IEEE Standards Organization in, I belive, 1999. The rights to the document were sold to Global Engineering.

      This does not mean that OpenFirmware is dead. On the contrary, there are several commercial implementations available. It does mean that it is no longer considered a "standard", though.

      If you'd like some more information on OpenFirmware, you can visit the OpenBIOS web site.

  2. progress? by prof187 · · Score: 5, Funny

    According to the author, there has been rapid evolution of the personal computer platform since the 1980s

    What?! When did this happen and why wasn't I told? Time to trade in the vacuum tubes

    --

    My other sig is an import.
    1. Re:progress? by djdanlib · · Score: 4, Funny

      You had vacuum tubes? Back in my day, we didn't have electricity. We had to use colored rocks, and those were hard to find because the world was in black and white, and sometimes one of us would get eaten by a mountain lion! And we were thankful!

  3. why? by pilot1 · · Score: 0, Flamebait

    What is there to gain from destroying backward compatibility by replacing the BIOS?
    I wouldn't be suprised if Intel tried to 'lock out' Linux, too.

    1. Re:why? by obeythefist · · Score: 4, Informative

      EFI allows really big enhancements to the bootup procedures. Everything that happens from POST onwards in BIOS's today is a huge waste of time. (Especially on windows PCs, haha). All that time spend ennumarating ISA slots (what ISA slots?), floppy drives, hard drives, etc etc is a big time blowout.

      One of the biggest steps in moving from that 1-2 minute bootup time on a PC is getting rid of the 10 or so seconds we spend letting the BIOS do its thing.

      Why shouldn't you hit the "on" button and see the logon screen instantly? With more and more advances like EFI and solid state storage this might be closer than you think.

      --
      I am government man, come from the government. The government has sent me. -- G.I.R.
    2. Re:why? by Anonymous Coward · · Score: 0

      I wouldn't be suprised if Intel tried to 'lock out' Linux, too.

      That would be a strange thing to do, considering Intel pays part of Linus' salary.

    3. Re:why? by Anonymous Coward · · Score: 1, Informative

      Huh? Try reading the article. Backwards compatability is explicitly addressed; modules provide a legacy interface for OSes. Additionally, Intel is not going to lock out Linux (which they support) or any UNIX-like OS through EFI. Just read the article...

    4. Re:why? by Guy+Harris · · Score: 3, Informative
      I wouldn't be suprised if Intel tried to 'lock out' Linux, too.

      Presumably you mean locking out with something other than EFI, given that Linux supports EFI, as this ArsTechnical article notes ("Nevertheless, one thing is certain: Linux already runs on EFI boxen, so this isn't some evil ploy to kick Linux off of the PC.")

    5. Re:why? by Anonymous Coward · · Score: 0

      Linux runs on _current_ EFI boxen. The article mentions that EFI will be integrating "Trustworthy computing". So future versions of EFI will be capable of locking out anything except unmodified versions of signed operating systems.

  4. Faster Booting? by Captain+Splendid · · Score: 1

    Who the hell cares? Even with Windows, I'm up and running in under a minute, and besides, who turns off their computer anymore?

    --
    Linux, you magnificent bastard, I read the fucking manual!
    1. Re:Faster Booting? by Skyfire · · Score: 2, Insightful

      If you are running a server, and have a restart (due to failure, new kernel, or whatever) you want to be able restart as fast possible to minimize downtime.

      --
      Do not go gentle into that good night. Rage, rage against the dying of the light.
    2. Re:Faster Booting? by Anonymous Coward · · Score: 0

      Add a couple SCSI controllers and WoL NIC cards and your boot time goes through the roof.

      Plus there's the stupidity about pressing Ctrl+A a the correct time to get into the "SCSI BIOS" and stuff like that.

    3. Re:Faster Booting? by kinnell · · Score: 1
      Even with Windows, I'm up and running in under a minute

      When I was a lad, we had computers which would boot instantaneously.

      --
      If I seem short sighted, it is because I stand on the shoulders of midgets
    4. Re:Faster Booting? by droid_rage · · Score: 1

      You know what I've noticed on that? It's not the PC bios but the damn SCSI bios that takes the longest time during startup on our servers. I'm not really a hardware guy, and I don't know if that's normal or not, but these are mostly fairly new Dells.

  5. Boo... Hiss... by nathanh · · Score: 1, Flamebait

    Just do Open Firmware you stupid NIH Intelidiots.

    1. Re:Boo... Hiss... by Thing+1 · · Score: 3, Insightful
      Just do Open Firmware you stupid NIH Intelidiots.

      Well, I think the answer's obvious. The same reason that drug companies scorn rain forest treatments and marijuana: because it can't be patented.

      --
      I feel fantastic, and I'm still alive.
    2. Re:Boo... Hiss... by Anonymous Coward · · Score: 0

      The same reason that drug companies scorn rain forest treatments and marijuana: because it can't be patented.

      Um, if by "rain forest treatments" you mean new drugs based on previously unused plants and the like, then they *can* be patented. Heck, someone's even patented that antifreeze compound found in antarctic fish.

      You can patent anything these days. In fact, I just patented YOU, so you owe me royalties for existing. (And we aren't even in Soviet Russia!)

  6. Mama told me only buy 100% IBM compatible by Radical+Rad · · Score: 3, Insightful

    Otherwise you have to patch Lotus 1-2-3 to make it work.

    Seriously though, it seemed to me that at first this should break backward compatability but why couldn't BIOS emulation be plugged in as an EFI driver? All it has to do is provide the same software interrupts and it wouldn't even have to be loaded unless your OS needs BIOS to boot. Just because this "framework" isn't natively backwards compatible doesn't mean it couldn't be made backwards compatible.

    1. Re:Mama told me only buy 100% IBM compatible by Anonymous Coward · · Score: 5, Informative

      RTFA -- it will emulate the PC BIOS (see figure 3).

    2. Re:Mama told me only buy 100% IBM compatible by Radical+Rad · · Score: 1

      Ok. I went back and carefully re-read above figure 3 and re-read below figure 3. I searched the article for "emulate" and for "compat". I studied figure 3 in great detail. I even unfocused my eyes and tried to look *into* figure 3. ...No joy.

      Apparently all an anonymous troll must do to whore karma is to tell someone else to to RTFA and pretend to point them to the right spot. And since the moderators don't bother to look... TADA! +5 informative. Sheesh.

    3. Re:Mama told me only buy 100% IBM compatible by aanantha · · Score: 1
      I studied figure 3 in great detail. I even unfocused my eyes and tried to look *into* figure 3. ...No joy.

      Really? Figure 3 has a black box that says "Compatibility Support Module". And then a blue box on top of that saying "Legacy OS Loader". And then the text above the figure says "In addition, the Framework provides support for legacy OS interfaces via a set of drivers."

      I think that's you're looking for? That Compatibility Support Module must be a BIOS emulation layer.

    4. Re:Mama told me only buy 100% IBM compatible by Radical+Rad · · Score: 1

      Yes really. You have made assumptions that may not be true. Nowhere does the article state that the PC BIOS is emulated.

  7. Surely this brings DRM to life by rueger · · Score: 4, Insightful

    It strikes me that losing the hardware BIOS will almost certainly lead to some kind of DRM scheme - probably hardware related - that can't be bypassed by something like a boot floppy.

    In fact, I really can't believe that DRM won't be built into whatever replaces the BIOS chip.

    Course maybe I'm i'm just paranoid...

    --
    Three Squirrels
    1. Re:Surely this brings DRM to life by ctr2sprt · · Score: 3, Informative

      Look at the diagram. The first stage is "Security." That's where DRM would go. Its purpose according to the diagram is to verify the next stages, which would be to eliminate boot viruses. Personally, I think you're paranoid. But hey, even paranoid people are right sometimes!

    2. Re:Surely this brings DRM to life by IntlHarvester · · Score: 4, Informative

      You can do DRM in a conventional BIOS (see IBM systems now shipping)

      You can have a DRM-free EFI implementation (see Itanium systems now shipping)

      DRM and EFI aren't directly related at all -- although they will both become mainstream at about the same point in time.

      --
      Business. Numbers. Money. People. Computer World.
    3. Re:Surely this brings DRM to life by zamboni1138 · · Score: 1

      On the same figure check out the last stage of the "flow". It goes straight from "Run Time" to "After Life" with a question mark. The entire diagram is spent on the boot phase of this new system.. Who cares/knows how it all turns off. :-)

    4. Re:Surely this brings DRM to life by MrLint · · Score: 1

      i havent seen a boot block virus in nearly 10 years. In fact if you were to run down the list of recent viruses id have you say you'd be hard pressed to find one. All of the recent ones rely on windows. So 'security' is all about DRM

    5. Re:Surely this brings DRM to life by Anonymous Coward · · Score: 0

      Rather than viruses, I think their fear is user-developed firmware modules.

  8. Some thoughts... by notsoclever · · Score: 3, Informative
    Okay, so EFI can emulate standard BIOS through a module. Which means that it can also obviously emulate OpenFirmware.

    Since OpenFirmware is also totally programmable, does that mean that OF could emulate EFI?

    It'd be interesting to see if future video cards decided to support EFI instead of x86-specific BIOS code. One of the big problems with getting video cards on the Mac is that they use x86 BIOS code which means that vendors need to make two versions, one PC and one OpenFirmware, and since the PC market is so much bigger the OpenFirmware version is usually 2-3x as expensive (and because of the development effort in making the OpenFirmware version, ATI et al do whatever they can to prevent people from just flashing the x86 version of the card with the OpenFirmware firmware).

    It sure would be nice to not have to pay $200 for the Mac version of a $75 Radeon...

    --
    There are 10 kinds of people: ones who understand ternary, ones who don't, and ones who think this joke is about binary
  9. Oh, and better yet: by notsoclever · · Score: 1

    What would be even better is if the video card makers would just do it all in OpenFirmware to begin with, and get Intel to put an OF module into EFI by default. :)

    --
    There are 10 kinds of people: ones who understand ternary, ones who don't, and ones who think this joke is about binary
  10. Do we need faster booting? by mixmaster · · Score: 2, Funny

    We don't need faster booting, everyone turns on their computer, and then goes for coffee like a zombie.

  11. Servers by tepples · · Score: 1

    But while your server is booting, it's not making money for your company.

  12. So, the first phase is security... by drakaan · · Score: 1
    and "It should be noted that each phase builds upon the other.", but the first phase they mention is PEI...

    Intel mentions (about the "Security" phase) here that:

    The objective is to ensure that the first code executed by the processor is trustworthy and that this code has sufficient resources in and of itself to determine the trustworthiness of any subsequent code. What "authenticate" and "trustworthy" mean can evolve over time and across platforms

    Trustworthy computing anyone?

    --
    "Murphy was an optimist" - O'Toole's commentary on Murphy's Law
  13. BIOS? by dominick · · Score: 0

    BIOS, what's it about, is it good or is it whack?