Catch up on stories from the past week (and beyond) at the Slashdot story archive


Forgot your password?

Windows XP and Incompatibilities with Multi-Booting? 201

Morgan asks: " Windows XP (Whistler) won't boot from disks with MBR partition tables, requiring a new GUID Partition Table (GPT). It will still read and write MBR disks. In a cursory search, I find no work to support GPT with LILO or other multi-booting Linux loaders. I'm a 'one OS, one disk' man, especially since disks are so cheap, but what about those who aren't handy with a screwdriver (and an IDE cable)? An easily installable Linux distro that shrinks the Windows partition, but allows multi-booting without requiring a re-install of Windows is a great Trojan horse: 'here, try this real OS, but if you don't like it, or you need a particular app, you can always boot Windows.' Will GPT make this harder?" What reasoning was behind the move to GPT? By making Whistler incompatible with the standard MBR, this could be seen by many as another move to stifle competition in the PC market. How do you feel about it?
This discussion has been archived. No new comments can be posted.

Windows XP and Incompatibilities with Multi-Booting?

Comments Filter:
  • by Anonymous Coward
    According to the FAQ, this is a problem only with the 64 bit versions, not the 32 bit versions (see #'s 13-16 especially 16. So this will only affect "Big Iron" not ordinary dual-boot situations.
  • by Anonymous Coward
    If you read down the page, it sounds like that's exactly what they are doing on x86.

    It would be rather unfortunate to see lots of 1981-style PC crap like the MBR make it onto the IA-64 platform. One obviously bad bit is the whole BIOS/Boot routine which just reeks of the stoneage. Someone make an Open Firmware PC!
  • by Anonymous Coward
    Considering that Intel is funding Linux on IA64 development, presumably the proper folks have been informed. (Since the big boys, Sun DEC etc have dropped UNIX-on-IA64 plans, Intel needs Linux badly to get this stuff into the datacenter.)
  • by Anonymous Coward
    this could be seen by many as another move to stifle competition in the PC market.

    Yeah, and issuing a mandatory federal ID card could be seen by many as government oppression. Which, of course, would not be true.

    Microsoft is not under any obligation to make their operating system to coexist with other, competing operating systems. Insisting that a corporation, which is in business to make money, should respect competing products is just plain lunacy.

  • Yes, it sounds like EFI/GPT is a useful step up from MBR. On the other hand, abrupt transitions like this are generally bad things. If Microsoft cared about the people using their OS (haha) they would have a version or two which supported both, and then dump MBR later.

  • Your CS prof must consider the entire open source movement an exercise in stupdiity, then. For that, his comments warrant no creditiblity whatsoever.

    a) ...not been developed as a result of a requirement analysis.

    Open source software, in general, comes into being as a result of a coder scratching a personal itch. This is about as close to a requirements analysis as you'll get, and it's good enough for me. A coder had the presence of mind to notice a lack, and he set out to meet it.

    b) ...not been designed before coding.

    Well, it's a safe bet a lot of OSS coders actually do have educations, and are at least marginally intelligent. So yes, at least some of the projects are designed.

    c) ...not been audited before coding.

    Extraneous formality that doesn't really fit into the type of development models most OSS projects employ.

    d) ...not been formally block tested.
    e) ...not been formally integration tested.
    f) ...not been formally documented, meaning all the steps mentioned above.

    See above. Do we look like a bunch of coders in suits to you? Reality check time. If the OSS world worried about this sort of formaility, nothing would ever get done. Sheesh, get your head out of the sand.
  • that I wouldn't care in the slightest. Although, as others have pointed out, there is actually no multiboot issue here, it doesn't really matter even if there were. Why? Simple - if you don't like Microsoft's tactics, don't use their products. Nothing forces you to buy or use XP. If it has issues that prevent you from using your system(s) the way you want, just don't use it! All the whining and crying about anticompetitive practices is just so much bull -- you can much more effectively let Microsoft know how you feel by not buying their product.

    Dealing with for-profit corporations is very easy. When you buy their product, you are giving a vote of confidence and approval in them - their products, their people, and their practices. When you don't buy their products, you are doing just the opposite. Since they want to make money, successful for-profit corporations will alter their products, people, and practices in ways that maximize the number of people who buy their products. If you buy a product from a company of which you do not approve, you are voting for Buchanan when you claim you wanted Gore. And, like those who did that, you are being incredibly stupid.

    Proudly 100% Microsoft-free for over 5 years.

  • by defile ( 1059 ) on Sunday May 20, 2001 @08:38AM (#210048) Homepage Journal

    The Extensible Firmware Interface (EFI) is an attempt to bring ancient PC "standards" up to something reasonable. Almost all hardware and OS/driver vendors are _for_ this. This is an attempt to give PC's real firmware, rather than the hack they have now.

    If you've ever used big iron UNIX hardware, you've probably enjoyed the conveniance that the firmware itself provides (or maybe not).

    Replacing the MBR is one aspect of this. This is not something to worry about unless MS pulls the embrace and extend crap, but I doubt they'd stoop to it on such a low-level aspect.

    With some exceptions, all OS, Driver, and Hardware developers/vendors want EFI.

    Possible conspiracy theories:

    • Intel is mostly responsible for EFI. Insert typical big scary corporation fear.
    • EFI lets each hardware vendor provide built-in drivers for all hardware. This could certainly mean that the OS has to use a binary-only driver to transmit something over the network card. Quality manufacturers will probably want an OS specific driver. At least EFI demands that the bundled driver follow a standard interface. The end result is that you may not have the source to some hardware drivers.

    Realistically, the worst that EFI can do is give us something better than what we have now. At best, it could make PC's pleasurable (but don't bet on it).

  • This won't be running on your pc unless you plan on running the 64bit OS on an Itanium on your desktop soon. This only applies to the 64bit version of Whistler. Don't worry about it on your x86 boxen, OK?
    Besides, why would you want that crappy IBM DOS MBR setup on a newly defined hardware platform? This new GUID bs complies with the specs for Intel's Open Firmware (or whatever they call it).
    Everybody take a deep breath... Microsoft isn't out to get you this time. Return to work, nothing to see here.

  • Monopolies are not outlawed in the United States. Where did you here that garbage? In fact, Congress has enshrined one institution in America as safe from any anti-trust litigation:

    Major League Baseball

    Don't believe me? Look it up. I couldn't make anything that stupid up.

    PS. I am an avid sports fan, but protection for MLB is really stupid. The NBA, NHL and NFL have managed quite well without protection. Heck, the NFL got tagged in the anti-trust suit by the USFL for a whole 3 dollars and they survived!

  • Opera.

    Netscrape and Exploder are 6 of one and half-dozen of the other. Even Lemming shill publications don't paint the picture any differently.

    They're both bloated hellspawn of Mosaic.
  • Just how would any of Microsoft's OS products benefit from that? Microsoft's OSes tend to be limiting enough by themselves that they seldom exceed the limitations of the current BIOS scheme.

    The only real immediate or forseeable impact will be to make AltOS vendors refit their products to catch up with Microsoft's standard du jour.
  • 16. Can the 32-bit version of Windows Whistler read, write, and boot from MBR disks?

    So yes, you can happily boot your 32 bit version of whistler with Linux.

    As to why the 64 bit version is different, I have no idea. However, I am thinking 64 bit whistler is going to flop anyway, due to lack of decent hardware (for at least 12 months in any case)....

  • Why are there so many stupid people... read the article before posting bullshit.

    I know someone who is dual booting Linux and Windows XP right now using grub with no problems on ia32, which as the article states it has no problems with...

    On the topic of ia64 not being able to dual boot, does any person who would want to dual boot even own one yet? IA64 will mainly be used for servers at least for next few years and so in most (all?) instances won't even be in a dual boot scenario.
  • Any time Microsoft does something which is an improvement it is an attempt to stifle competition.

    Look at the recent whining regarding .Net, etc. :(

    It's just a fact of life, we move on and ignore these people.
  • by booch ( 4157 )
    Actually, VAX is a hardware architecture, not an OS. Normally, you'd run VMS on it, which I agree was a pretty good OS. But you could also run Ultrix (a UNIX variant) on it. There is a project porting Linux to it as well. But I doubt that you could set up any multi-boot on it.
  • Yes, this sucks. I highly doubt there's any good technical reason to go to this other boot record system. Honestly it just seems mean.

    But it is competitive, not "stifling to competition". A better choice of words would be "stifling to THE competition," similar words, but diametric meanings.

    Despite what economics textbooks say, competition isn't a market state, it's a process. When one firm drives another out of business with lower prices (ala Wal Mart), that's competition. When one firm makes its product more attractive than the other guys' to buyers, that's competition.
    What MS is doing here is basically an odd case of that second point. By making it more difficult to use their product alongside a competing product, they are (to some degree) making their product relatively more attractive, essentially by making the /other/ product /less/ attractive.
    For another example, consider if a car maker had a policy that if you used aftermarket parts in significant repairs (eg: new brakes, cylinder heads, etc), it voided part of the warranty. This causes the buyer to lean toward using the dealer's repair services or at least official factory parts rather than (possibly cheaper and/or "better") aftermarket parts. It's not nice, it may actually be detrimental to consumer utility, but it's COMPETITIVE.

  • What are you? EMACS?

  • There is no technical defense for what they've done

    Well, none unless one wants to remove many of the more arcane constraints of MBR. Frankly everyone who knows anything about drives and layouts agrees MBR was obsolete over a decade ago and should been retired back then.

    Furthermore for most users this won't be a big deal, certianly not for Jane-hobbyist. If you actually bothered to do a bit of research you'll discover that this implementation is fairly backwards-amendable except for some high-end configurations where it's unlikely to be an issue anyhow.

    Finally, hard drives retail for ~US$100/20GB, anyone installing XP is likely to be able to pony up the cash. This change won't affect ye olde 486s.

    This would be something for the DOJ to investigate...
    By your logic is any hardware advancement possible or shall the industry remain forever stuck in 1980's technology? At least Apple moved to OpenFirmware along with the rest of the workstation market or is that somehow part of an evil plot too?
    I can tell you, other than the Kindergarden level GUI (that looks like it was designed by Miss Shirley for Romper Room) there is NO advantage over 2000 at all. The new GUI has been found to be VERY annoying to most of the experienced network admins/engineers who have been testing it in my lab.
    Ahh, so it's not StUdLy enough for yeah, huh? Luckly you'll be able to change the chrome to a camoflauge background and Matrix-themes to assauge your offended aesthetics.
    This move to "break" the MBR is nothing but the usual Microsoft anti-competitive malice at it's worst.
    Just as the new `Doze 2000 SP2 broke most popular software firewalls on the market
    Yeah, patching things Baaaaadddd... right.
    Simply put, Microsoft is more interested in protecting their marketshare...
    ...than in giving the market what it wants:

    1. An OS that is stable.
    2. An OS that is secure.

    The market gets what it buys. IS Depts & consumers continue to buy on sizzle & psuedo-feature sets then MS will continue selling it to them. It built a megacompany so they're doing something right.

  • "arbitrary limitations ... early 80s."

    Check out hardware prices, and what users were willing or able to pay, and the state of the art at the time, and you'll see that those limitations weren't so arbitrary.

    Of course that doesn't make them any less infuriating, but if Henry Ford had tried to design the Model T so that you could easily retrofit disc brakes, air bags, electronically controlled fuel injection, and emission control plumbing, he'd have burned through all his venture capital before he ever got any product out the door.

  • How many primary partitions can you have on a SCSI hard drive? How many extended? How many logical in an extended?
  • WD drives (but not Maxtor, annoyingly) will take a 10 pin socket (like on serial port header connectors) which will make it easier to get the wire length that you need, but why screw around with printed circuit boards, headers, etc., when you can just solder the other ends to a double-pole, double throw switch?
  • Actually "...cause more problems than Win2K." makes just as much sense in that context, maybe more.
  • Ah but there not insisting that a corporation respect competing products. They're asking a MONOPOLY to back down. The anti-trust laws can and will enforce actions such as making MS's next big OS play nice with others. (See the DOJ injunction against IBM for an example)

  • By the time this becomes an issue, LILO and/or GRUB (which is a better bootloader anyway, if you haven't tried it yet), will support MBT. Try not to forget one of the few propaganda items about open source that's actually unambiguously true: we can and do respond quickly to changes.

    Is it anti-competitive? Probably. But it's likely to be as successful as MS' attempt to crush us with the DAV protocol. Remember that one? A Linux beta came out the same week MS mentioned it. So screw 'em. What's holding us back on the desktop isn't Microsoft's standards twiddling, it's the fact we can't convince the major vendors to port their apps to X. (Which is a subset of the problem that, for several classes of desktop apps, MS is the only vendor.)

    Anyway, it will take a lot more than a partition table format to sink us.


  • Nice joke, but in reality I'm getting a little frightened of the momentum Linux is gathering. I mean it's great as a server OS and getting to be a nice desktop OS as well, but I certainly don't want all that amateur code in embedded, possibly life-critical systems.

    I hope you're not thinking that Windoz would be better... At least Linux can be trimmed down to a very small amount of code to fit the purpose. If someone is dumb enough to blindly take Linux code and run a Nuke sub or powerplant with it then we are in trouble. I don't think engineers working on this type of equipment are that dumb. Though I have seen the BSOD on way too many non critical devices ( scoreboards, billboards, airline screens, etc ). If engineers are dumb enough to use Windoz on mission critical devices.....

    NOW THAT SCARES ME! Oh wait, isn't Bill Gates paying the US DOD to use Windoz on Aegis class ships and a new aircraft carrier?
    Time to head for the hills. ;)


  • by K. ( 10774 ) on Sunday May 20, 2001 @01:03PM (#210067) Homepage Journal
    Or, indeed, problems such as making sure your cruise missiles don't hit non-combatant's embassies...
  • Prepare for flamage from the UNIX on VAX faction...

    Let me help it along:

    Unix was written on our machines and for our machines many years ago. Today, much of UNIX being done is done on our machines. Ten percent of our VAXs are going for UNIX use. UNIX is a simple language, easy to understand, easy to get started with. It's great for students, great for somewhat casual users, and it's great for interchanging programs between different machines. And so, because of its popularity in these markets, we support it. We have good UNIX on VAX and good UNIX on PDP-11s.

    It is our belief, however, that serious professional users will run out of things they can do with UNIX. They'll want a real system and will end up doing VMS when they get to be serious about programming.

    With UNIX, if you're looking for something, you can easily and quickly check that small manual and find out that it's not there. With VMS, no matter what you look for -- it's literally a five-foot shelf of documentation -- if you look long enough it's there. That's the difference -- the beauty of UNIX is it's simple; and the beauty of VMS is that it's all there.

    - Ken Olsen 1984

  • Windows NT's bootloader can pick up a SCSI or other disk driver (.SYS file) from a BIOS-addressable FAT partition and use it to find the OS without using the BIOS. I have a feeling this what the MSR is for, but who knows, especially if there's no DOS-style drive detection.

    I think GRUB works in a similar manner to the NT loader. LI(jump to the partition and see what happens)LO is cruder but in someways easier.
  • If someone ever fixed PCs so that they had real firmware and you didn't have to play BIOS mindreader to multi-boot, you'd probably feel that swapping disks was kinda a kludgy solution. It's an electronic computer, goddammit! Why should you have to touch it once it's set up?

    Anyway, I have a box with 3 SCSI controllers and a bunch of disks on various internal and external chains. I've felt the pain, but it can be made to work (good understanding of various OS boot processes and SystemCommander helps.)
  • From your first link, it looks like the new boot process requires some small partitions (ESP Extensible Firmware Interface System Partition and MSR Microsoft Reserved Partition.)

    Shades of the old OS/2 boot manager partition or the EISA Config partition that seemed to confound certain users with itchy FDISK fingers.

    Anyway, it's nice to see something that looks like real boot firmware and not a CP/M-compatible kludge. Any hope of support for this on standard IA32 boxes?
  • After the marketing implosion of the PIII CPUID, the BIOS spec was altered so that it provides a unique ID number for each motherboard (which is a better solution than a CPU ID even for the assent management departments that wanted this sort of thing).

    This is really nothing new either, custom BIOSes from IBM and Compaq have provided serial numbers for years and years, it's just now a standard call so that Windows can get this info and use it for it's registration voodoo.

    Anyway, it's only in the last couple years that PCs have had a GUID number other than the NIC.
  • Very nice. Got any more from the fireplace series? :) <hopeful look>


  • by Zico ( 14255 ) on Sunday May 20, 2001 @10:35AM (#210074)
    Dear Slashdot editors,

    Is it your intention to make the Linux community look like idiots? Seriously. You post these half-baked articles so frequently, often incorrect and always misleading, and they always get the community fired up to the point of seeing who can make the dumbest post about something that's not even true.

    Not sure how many people here are familiar with it, but it's almost like a digital equivalent of the Phil Hendrie Radio Show. It's a hilarious show, btw, if anyone has a chance to listen to it. He'll have guests on (actually voiced by himself) who make dumb and subtly inflammatory comments, but keeping it halfway believable. So every show, you end up with all these people calling up the show and getting irate arguing with some fictional character. (Hendrie = ./ editors, Guest = story submission, Callers = Linux advocate posters). In both places the comedy comes in large part from listening to or reading the opinions of the idiots who believe what they just heard and subsequently make incredible asses of themselves.


  • and people didnt see this comeing standard practice make it incompatitable if you dont like your comptition like the word formats keeps linux distros playing catchup and not IMPROVEING but one thing that linux has got is the ability to run on many differant CPU Arch but they are lagging behind come on linus sort out the patchs I only see mr cox doing this (ARM /S390 merges) please this has got to get better regards john jones
  • by Sloppy ( 14984 ) on Sunday May 20, 2001 @09:00AM (#210076) Homepage Journal

    When you look at other partitioning schemes (such as the RDB system that the Amiga has used since the 80s) the whole MBR/fdisk thing is a complete joke. The anachronistic partitioning system on x86 PeeCees has caused me to waste significant time on something that should be trivial, and there's a whole sub-industry of bootloaders [] and utilities just to get around the arbitrary limitations that were built in by some visionless idiot in the early 80s.

    Dumping it is a good move. And I note that this time (for a change) MS isn't replacing it with a proprietary decommoditized MS-only defacto standard de jour. This one is open and anyone can be compatable with it. What's not to like?

  • by Midnight Thunder ( 17205 ) on Sunday May 20, 2001 @08:31AM (#210077) Homepage Journal
    It looks like this is all part of the modernization of the intel platform. The GPT specification is actually documented as part of the Extensible Firmware Interface (EFI), which appears to be a replacement to the BIOS. There is a boot manager as part of the EFI, so in conjunction with a GPT based drive, there is no need for LILO or its equivalent, as this feature is provided for you.

    Until we get mother boards that includes the new EFI approach, we will definetly see transition solutions to support GPT drives.

    It will be great to have this technology - sure it will break a lot of the old systems, but then again sometimes when you go forward there are sacrifices to be made.

    If MS has decided to limit the addition of GPT support for 64-bit XP, then this is probably because 64-bit motherboards would break most 32-bit OS anyhow, and thus they are unlikely to have people complaing that Windows 98 doesn't work on those machine.

    For more info on EFi and GPT, check ou the following link: .htm []

  • I think we need to suspend paranoia mode for a second here.

    At the risk of sounding like "user friendly" software, Are you sure? :)

    From the GPT/EFI FAQ:

    Each GPT partition has a unique identification GUID and a partition content type, so no coordination is necessary to prevent partition identifier collision.

    With a unique identification, is it possible that this could be used for some basic, broken form of DMCA-style "content protection"?
  • I'd just like them to stop posting ask /.'s that are obviously flamebait. I mean, even ignoring the fact that this submission wasn't news worthy, (who's going to be running a dual boot on a IA64? And where are they so I can punch them in the face?) It's worded to get the hot-heads going.

    /. could go a long way by not baiting the morons out there, But hey, maybe that's what they're after anymore. Maybe prepping for a new fox special they're trying to push, "When Angry, Misinformed, Geeks Attack!!"?
  • Sorry, your BeOS theory doesn't fly -- the same bug (have to login twice before it sticks) exists under Win2k w/ IE5x as well. Drives me crazy. astro
  • As stated in the Win XP FAQ, GUID partitioning is a replacement for MBR partitioning that allows more partitions on the disk, of any size, without the silly BIOS limitations. It's a part if EFI, the Extensible Firmware Interface that is replacing the old hairball that is the BIOS. The IA-64 Linux distributions already have a version of LILO that works with EFI.

    In short, don't worry, Linux adapts, as it always does.

  • I have 2 dual boot boxes running XP Beta 2 and Mandrake 8.0. It's using Mandrake's graphical LILO as a bootloader, and it's working just fine. Is this something that will be "implemented" for the final release of XP?

  • They're generated using the MAC address of your NIC (if one is present); so reading the GUID table would be a way to track computers - an ID like Intel's much disliked CPU ID was.
  • I for one am very happy about this development, and I hope the MBR system dies a very quick death, no matter how painful. Anything would be better, even if we had to reverse engineer it, which we don't in this case.

    I know much more than I ever wanted to know about extended partitions and how to fix them when the chain gets corrupted.

  • by kubrick ( 27291 ) on Sunday May 20, 2001 @08:23AM (#210085)
    Note that this only applies to the 64-bit (Itanium) version of XP, not the 32-bit.

    Just goes to show how many slashbots are ready to flame without even reading the material linked to (about 30% at the time I was reading, comments at +1 and above). That's a scary amount of political power if CmdrTaco is ever looking to mobilize the ignorant forces...


  • by Why2K ( 29813 ) on Sunday May 20, 2001 @07:59AM (#210086)
    13. Can the 64-bit version of Windows Whistler read, write, and boot from GPT disks?
    14. Can the 64-bit version of Windows Whistler read, write, and boot from MBR disks?
    The 64-bit version of Windows Whistler can read and write MBR disks, but cannot boot from MBR disks.
    15. Can the 32-bit version of Windows Whistler read, write, and boot from GPT disks?
    No. The 32-bit version will see only the Protective MBR. The EE partition will not be mounted or otherwise exposed to application software.
    16. Can the 32-bit version of Windows Whistler read, write, and boot from MBR disks?

    Note that this only applies to the 64-bit (Itanium) version of XP, not the 32-bit. Frankly, I wished it worked with the 32 bit version too, since the current MBR-based scheme sucks. Try having FAT + NTFS + Linux + FreeBSD + Plan 9, etc on one disk and you'll wish for something better too.

  • Thats probably the worst troll ever to befoul mine eyes.

    'amateur code' indeed.
  • AC wrote:
    Microsoft is not under any obligation to make their operating system to coexist with other, competing operating systems. Insisting that a corporation, which is in business to make money, should respect competing products is just plain lunacy.

    False. As a monopoly, Microsoft has to be careful that its actions do not act exclude competitors from the marketplace. In this case they deviated from a long-time practise for no other reason than to make things more difficult for its competitors. It took EFFORT for Microsoft to make this change, a change that hardly affects anyone outside of Powerquest, the Linux distributions and the *BSD flavours. It is a clear-cut abuse of their market power, and yet another annoyance we can chalk up to them. But its not so big a deal, because it's a temporary annoyance. Red Hat, Debian et. al. will find some way to deal with it, and things will be back to normal.
  • i agree that it is the responsibilty of the readers to read the stories, but the editors should bare some of this responsibility also. when you include the following in a story:

    Windows XP (Whistler) won't boot from disks with MBR partition tables, requiring a new GUID Partition Table (GPT).

    and add no clarification, what do you expect? my inital impression was not good, but i read further befor i started typing. hell, i'm not sure cliff read the entire story either.

    use LaTeX? want an online reference manager that
  • Normally, you'd run VMS on it

    Except for those who ran Unix on it. In fact, for a long time, the Vax was the dominant platform for running Unix, so much so that it even got made into one of the C commandmants [] that not all the world's a Vax.

  • wheres the time to market ?

    it doesnt matter how you design stuff...if you dont have time to implement it properly your commercial software will suck. it doesnt matter if you have roadmaps documentation, bug lists or any other crap. your time to implementation will govern how your final output works.
    unfortunately very few commercial vendors have either the time or inclination to let their software teams work at the optimum pace -- 7 lines of code or less per day. thats the speed which mission critical systems (think space shuttle control code or aircraft control code) are written at. and dont tell me that youre willing to let your programmers code only 7 lines of code per day.
  • *cough*

    It's an Intel proposal for their ia64 architecture (Itanium) for which the Linux port already has patches to handle.
  • *cough*

    It's an Intel proposal for their ia64 architecture (Itanium) for which the Linux port already has patches to handle.

    This does not affect the 32-bit version of XP (which still uses MBR).
  • The MICROS~1 problem is a filesystem of FAT and FAT32. They already "solved" the problem with NTFS. Note that ext2 doesn't have the MICROS~1 problem, so how can it be a partition table limitation?

    According to the article, some of the things that GPT will allow are better partition names and Unicode support. Now why we actually need Unicode for partition names is anyone's guess, but if you're gonna do it, do it all the way.
  • Not an issue. The 32bit version of XP uses MBR. The Linux-Itanium port supports GPT. Dual booting is still possible for those who are inclined (and have an Itanium).

    I wouldn't be too surprised to see a Linux patch (updated utilities) for GPT on ia32 in the future. As long as it's a Linux-only box, you could get rid of the limitations to MBR without having to get an Itanium processor.
  • The real question is if Whistler will support more than 26 partitions on a single disk.
  • If Microsoft cared about the people using their OS (haha) they would have a version or two which supported both, and then dump MBR later

    Isn't this how we got stuck w/ MS-DOS for so long? :-)

    I like nonsense, it wakes up the brain cells.
  • by seanw ( 45548 ) on Sunday May 20, 2001 @09:13AM (#210100)

    yes, it is possible. and, when you pay for phone service, they also give you a "unique identification" number. this could no doubt be used for some form of DMCA enforcement, somehow. or maybe even personal tracking! quick, run! hide from the government and corporations alike! you have been tagged, numbered, and identified--you have...a PHONE NUMBER!

    those cheeky bastards.


  • by seanw ( 45548 ) on Sunday May 20, 2001 @08:02AM (#210101)
    I think we need to suspend paranoia mode for a second here. according to the FAQ, the GPT disk partitioning specification is a documented Intel standard. and, really, isn't it about time for the old MBR scheme to die? the architectural limit to 4 primary partitions is getting old. GPT can scale far better and has a less arcane internal structure (cylinder and head arithmetic, anyone?)

    and if you read more carefully you will see that it is only the 64-bit version of Whistler that won't boot on MBRs. the 32-bit version should boot just fine.

    I think MS is just scaling with their new OS architecture. anything can be viewed through an anti-competitive lense. the important question is: will we be getting a better OS? I think so.

  • I am not expecting /. to send investigative reporters to far off techie conventions and the like. I just want them to read through an article before they post it. I don't think I am asking too much here. This doesn't really require too much extra money. While I would love real technical editors, I understand that /. may not be able to afford it. I find it hard to believe they can't afford to read an article before they post it though.
    What really bothers me is that often it seems that /. is trying to manipulate readers with outrageous headlines just to get people to read on. I can't stand when traditional news sources do this and /. is no different.
    Perhaps I am just too pessimistic and /. doesn't do it on purpose. Perhaps it is just that their editors don't have the time or money to check stories before rushing them out the door. I just don't understand how someone can read a post enough to believe it is worth an AskSlashdot but not bother to read the article sourced in the post.
    Don't get me wrong, I really like /. on the whole. I just think that it could be even better. The only way the powers that be will know that people want change is if they say it. So I am saying it.
  • by soldack ( 48581 ) <soldacker AT yahoo DOT com> on Sunday May 20, 2001 @09:52AM (#210103) Homepage
    First about this post. As others have commented, this only applies to Windows XP/2002 running on IA-64 processors. All operating systems wishing to boot on these systems will need changes to work with Intel's firmware for IA64 based systems.
    This brings a fact about /. that keeps raising its ugly head. All too often stories are posted that are not checked up on. I understand that those who run /. can not be expected to be experts in everything but stories are often posted that only required a simple reading of a FAQ or two to see that the post is just plain wrong. I had high hopes of /. getting much more professional when it was purchased by Andover. But nothing has changed. The staff seems about the same size and quality that it was before the corporate move. /. really has such great potential. If it could start using expert editors who only post stories that have been properly checked. One reason I love magazines like Discover and Popular Science because the stories are usually checked out pretty well. I understand the need to post things that generate interest but the site also claims to be a news site. No self respecting news site would allow such a low level of checking into the background of the stories they present. Is /. a news site at its heart or just entertainment? It has the potential to be both a high quality and entertaining news site and discussion forum. I understand that /. is a business and the object of a business is to make profits for their shareholders but I don't understand why this has to be in direct opposition to high quality.

    from a /.er who hopes things get better...
  • Nobody was suggesting that Windows is better. Just because Windows is a POS doesn't mean that Linux is the best OS on the planet.

    Linux has its (many) uses, but it's still not the be-all-and-end-all of operating systems.

  • You don't need to throughly research the topic. Fifteen minutes of reading the site and doing some quick searches on Google would turn up enough information. Its not like these are minor errors here. These are full blown the article is attributed the wrong meaning errors. These don't take extensive research to catch.
  • by be-fan ( 61476 ) on Sunday May 20, 2001 @12:40PM (#210114)
    Is it just me, or does anti-MS fud have a fast path to the main page somewhere in the Slash code? I'm still trying to prove that Slash has an anti-BeOS hack that requires Net+ users to type their password twice before the system accepts the login, but my new project is finding the filtering software that gives priority to MS anticompetitive stories. Seriously, though, the volume of traffic on the /. main page isn't that big. Maybe ten stories per day max. You'd think they'd actually check out the stories that make it far along in the submission processes for them to consider posting it.
  • Of course we dual booted our vaxen - VMS and Unix .... plus we ran Unix and RSX virtually under VMS (sort of a per-process dual booting :-)
  • Of course, the problem with that is that people will use the old way until its no longer an option.
  • GTP is a good thing and frankly its not Microsoft's responsibility to worry about linux dual booters. In the long run this will be good for linux, because GTP is far better than MBR which is older than the pyramids. However, in the near term this will be a huge hassel for those who know and love the current configuration of LILO and want to dual boot XP. This will really hit the small developer shop that relies on dual boots for office compatibility.
  • For details on the 128 Gig limit, the 1024 cylinder limit and other horrors from our DOS legacy, see the Large-Disk-HOWTO [], particularly History of BIOS and IDE Limits [].

    Mike []

  • The article is a bit on the flamy side to bad we can't mark them as such. Reading the faq supprisingly left me with a much different attitude than I went in with. It seems if you are using XP now you can use the XP bootloader in the same way people use the NT boot loader now. This setup should work for 64bit XP and 32Bit XP. If your using 32bit xp it looks like you have a few more options however as it plainly says it can read,write and boot from MBR disks. It doesn't look all that complicated after reading the faq which is pretty straight forward. My only concern is the GUID. While I can't quesion mob mentality I can question why someone didn't read this prior to posting the story as all questions seem to be answered there in black white and blue.
  • This GPT thing sounds good, a way to get rid of the legacy MBR. However, one thing that scared me a bit while reading through their FAQ was:

    36. What is a Microsoft Reserved Partition (MSR)?
    The Microsoft Reserved Partition (MSR) reserves space on each disk drive for subsequent use by operating system software. GPT disks do not allow hidden sectors. Software components that formerly used hidden sectors now allocate portions of the MSR for component-specific partitions. For example, converting a basic disk to a dynamic disk causes the MSR on that disk to be reduced in size and a newly created partition holds the dynamic disk database. The MSR has the Partition GUID:
    DEFINE_GUID (PARTITION_MSFT_RESERVED_GUID, 0xE3C9E316L, 0x0B5C, 0x4DB8, 0x81, 0x7D, 0xF9, 0x2D, 0xF0, 0x02, 0x15, 0xAE)

    *Shiver* <paranoia> Mental note: Do not install XP, for fear of Microsoft gathering information and storing it in a hidden partition.</paranoia>


  • Replacing the MBR is one aspect of this. This is not something to worry about unless MS pulls the embrace and extend crap, but I doubt they'd stoop to it on such a low-level aspect.

    How low can you go? The very name of the standard, "Extensible...", makes the hair on my neck stand up. You can put all your fears into this one.

    BIOS extentions break stuff. I've got an Athalon board with virus protection built right in. The stupid thing recognizes LILO as a boot sector virus. I can turn it off, and I'd return it if I could not. The motherboard drivers that came with a couple of computers I've built give me the shudders.

    There are two reasons I could care less about this one, however. The first is that there will be work arounds for any software I care about far sooner than for MS BS. The second reason is that I will never ever install MS BS on another computer I own.

  • Don't use Microsoft products.

    it's been working out great for me.
    Let's see....(hacking in to /.'s webserver...finding access_log...ah, good, combined format...extract user-agent):
    "Mozilla/4.0 (compatible; MSIE 5.5; Windows 98; Win 9x 4.90)"

    It's Funny...laugh

  • The articles are starting points for discussion. Even though this article is off base, a quick read through the posts will point readers to good links for more information.

    Slashdot isn't the NYTimes, which "pronounces" truth to the masses, nor does it need to be. Nor would most of us want it to be. Nor could it be. Get over it.

  • Indeed, a simple Google search [] reveals that the Linux community is on this.
  • By making Whistler incompatible with the standard MBR, this could be seen by many as another move to stifle competition in the PC market.

    I am so anti-M$ that it oozes from my pores on a daily basis. However, this is the sort of shit that the Free Software® movement needs least! We keep hearkening for Microshaft to drop their FUD campaigns, yet here we have Linux zealots blatently trumpeting the same thing. (Apropos, Cliff, I hear isopropanol does wonders for cleaning FUD out of your horn.)

    The MBR-style partition table has been in use for, what, fifteen years? At least? It is a roadblock in the exploding mass-storage market. The entire concept of a maximum of four partitions (there are, after all, only four slots in an MBR partition table) was smoothed out by the molestation^Wchanging of the standard to permit "extended partitions", and within those, "logical drives". However, with hard drive capacities soon to break 100 gigabytes (!), it's high time we, as members of the computing indutry, came up with something better.

    I've never heard of GUID partition tables until now, and therefore know absolutely nothing about them; therefore I can't comment on whether or not GPT is truly better. But can anything be much worse than what we're currently stuck with?

  • Looking at the FAQ page linked to, I think this is a false alarm. Microsoft has several legitimate reasons for introducing this new partitioning scheme. It is an open standard; they link to the definition here [].

    It looks like a lot of the reason they are doing this is support for larger systems and disks. See these Q/As:

    Q: If the disk is larger than the maximum size an MBR can report, will the entire disk contents be protected? A:The EE partition in the Protective MBR is specified to be the maximum size allowable in an MBR.

    So the maximum size supported by the new format is greater than the old.

    Q: Can the 32-bit version of Windows Whistler read, write, and boot from GPT disks?
    A: No. The 32-bit version will see only the Protective MBR. The EE partition will not be mounted or otherwise exposed to application software.

    This doesn't affect your current system at all. If you install Windows Whistler on your ia32 machine, it will use the old format.

    Linux already has the support for many different partitioning schemes. I don't see this as different. There is no support for it now, but there will be when someone decides to develop it.

  • GPT isn't proprietary, is it? If not, Linux should support it. MBR was a pain, anyway.
  • I've XP beta 2 and Mandrake 8.0 on one harddrive, and both boot fine, using the nice mandrake bootmenu. So where's the beef?
  • If you're the type that says "if it ain't broke, don't fix it", consider that good-ole MBR partition tables have 16 bytes per partition, and all 16 of those bytes have a defined purpose already... there's no room to add anything. There's a byte for the partition type, one for the "active flag". Six bytes define the partition start and end in terms of CHS, and eight bytes define the start (32 bits) and length (32 bits) of the partition in terms of LBA. There's still plenty of CHS usage today (but only with drives under 8.4 gigs of course), so you can't reclaim those bytes.

    Now admittedly, that's 32 bits of 512 byte sectors, which works out to be two terrabytes... the 128 gig limit is due to ATA's definition of LBA being 28 bits, which is due to byte-wide registers for cylinder and sector but only a single nibble for the head, all of which got recycled for LBA, with a then unthinkable 128 gig limit!

    Rumor has it (from Hale Landis, a pretty good source) that ATA/ATAPI-6 will allocate 48 bits (somehow?) for LBA numbers. It's gonna be pretty damn hard to fit those 48 bit LBA start and length numbers into the 32 bit wide slots of legacy MBR partition table.

    Thus continues the saga of hard drive size barriers and the long PC legacy of kludge upon kludge, 512 megs (CHS bios call), 2 gig (fat16), 8 gigs (kludged CHS bios call), numerous bios bugs along the way... just when it seems like all that's behind us, >128 gig IDE drives will be upon us soon since you can already get SCSI larger than that, and someday two terrabytes hard limit within the partition table will also become an issue. Hopefully by then we'll all have fiber-to-the-home, so it'll be easier to fill up a couple terrabytes with mp3z, p0rn, DivXs, etc.

    In an ideal world, one might expect that the PC industry could collectively plan ahead and make a smooth migration to new standardized formats long before the legacy ones how signs of obsolescence... Oh well.

  • by locutus074 ( 137331 ) on Sunday May 20, 2001 @08:41AM (#210155)
    It looks like Microsoft is using a publicly available standard to do this ( d.htm []) and there are very good reasons to use this scheme.

    Maybe I'm just a little dense, but I have no idea why this is "from the another-attempt-to-stifle-competition dept." If you can get the specification, how are they attempting to be incompatible?

    What reasoning was behind the move to GPT?

    Look here []. More than 4 partitions without hacks like extended partitions..... Personally, I'm looking forward to this becoming mainstream.


  • While I agree with your points on editing somewhat, I'd also like to point out that you can always cancel your subscription. Once those dollars for your subscription stop rolling in, and other do the same, I'm sure Taco will shell out a few buckeriniees for a technical editor. 'Course, once that happens, the speed of reporting will go way down, now that background checks on the submitter have to be checked for bias.

    OK, you really do have a valid point, but the fact the /. doesn't have a great deal of money to do this sort of stuff. I guess it would be possible to break up the submission process. Say, one person filters out the goat sex crap, passes up what's left to a team of others, they filter out the crap and pass it to someone to do source checking.

    Taco could ask the community to pitch in, were he so inclined. Personally, I take every thing I read with a grain of salt. If a story impacts my work directly, I will source check as much as I'm able, be it the New York Times, The Register, SlashDot, Live 5 News, or NEWS 4 U. (gag).

  • by binford2k ( 142561 ) on Sunday May 20, 2001 @12:27PM (#210157) Homepage Journal
    13. Can the 64-bit version of Windows Whistler read, write, and boot from GPT disks?
    14. Can the 64-bit version of Windows Whistler read, write, and boot from MBR disks?
    The 64-bit version of Windows Whistler can read and write MBR disks, but cannot boot from MBR disks.
    15. Can the 32-bit version of Windows Whistler read, write, and boot from GPT disks?
    No. The 32-bit version will see only the Protective MBR. The EE partition will not be mounted or otherwise exposed to application software.
    16. Can the 32-bit version of Windows Whistler read, write, and boot from MBR disks?
    17. Can Windows 2000, Windows NT 4, or Windows 95/98 read, write, and boot from GPT?
    No. Again, legacy software will see only the Protective MBR.
    18. What about mixing and matching GPT and MBR disks on the same system?
    GPT and MBR disks can be mixed only on 64-bit systems, and the following restrictions apply:

    * The Windows Whistler loader and the boot partition must reside on a GPT disk. Other hard disks can be either MBR or GPT.
    * Both MBR and GPT disks can be present in a single dynamic disk group. Volume sets can span both MBR and GPT disks, However, the MBR cylinder alignment restriction might cause some difficulties mirroring or striping MBR and GPT disks.

  • I'm not punishing myself by giving MS more of my money in exchange for the headache of reinstalling another Windows, reinstalling apps, etc. It's crap. I'll do whatever I can to prevent new versions of Windows from entering my workplace (approx 1000 pcs, approx 12 knowledgeable users). There really is NO need for it. I say we all band together and say, Win9x is fine for Gomer. Office97 is fine for Gomer. Only order pcs that work with win9x, order them without the latest stinky version of Windows, ghost them all identically, and leave Gomer alone. Maybe Gomer will learn how to use what he/she has if we leave it alone long enough?
  • by 3247 ( 161794 ) on Sunday May 20, 2001 @08:46AM (#210165) Homepage
    "The MBR is considered a legacy agent, mainly by the fact to get to it, you have to use software interrupts."
    Huh? The MBR is just another sector on the harddisk. There's nothing special about it. As soon as you have your disk driver loaded, you do not need the BIOS (and before that, you will have to use BIOS INTs for all disk accesses).
  • by TeknoHog ( 164938 ) on Sunday May 20, 2001 @07:57AM (#210167) Homepage Journal
    I'm a 'one OS, one disk' man, especially since disks are so cheap, but what about those who aren't handy with a screwdriver (and an IDE cable)?

    ..and those with a laptop as their main computer? I've just switched to a bigger HD, and left a 2GB partition for testing alternative OSes. This is bad news, although I don't have plans to run any version of Windoze.

    Even so, I know I have to be ready for the other OS screwing up MBR, so I've got a boot disk ready. This is a perfectly working option for trying out Linux, albeit slower to boot. But it may take time to convince a typical Windows user that floppies still exist and are actively used by many of us... which reminds me, there's always Loadlin.


  • The MS FAQ makes it very clear...the GPT only applies to the upcoming 64-bit version of Whistler, not to the imminent 32-bit version.

    I'll bash Microsoft as much as the next luser, but once again this is FUD spread by Slashdot's inability to read the stories it's linking from.

  • by dj_whitebread ( 171775 ) on Sunday May 20, 2001 @07:55AM (#210171) Homepage
    with the move towards having EFI as the intermediate between the BIOS and the OS, the GPT is a step to get rid of the one more legacy element in the PC. The MBR is considered a legacy agent, mainly by the fact to get to it, you have to use software interrupts. These are trying to be phased out. Now, with the GPT, you can install as many OSs as you would like, they each just need an entry in the table (which is very easy to do) and a file on the harddrive that actually tells it where to go.
  • The real question is if Whistler will support more than 26 partitions on a single disk.

    While apparently yes, why would you need more than 1 partition per MS OS? And why wouldn't you need more than the latest and greatest MS OS on your system. (recalling a certain comment about 604K)

    Why I can see the benefit and logic to expanding and changing out the partition system. Anything with MS finger prints makes me glad I am getting better with *nix boxes. not prejudiced, of course, just my opinion based on the past track record. Of course you could put any partition system you want on a box, or make one up. The question is what would run on it.

    It would be cool if the *nixen could read all partition systems, instead of locking one out of one system vs another. Of course it is just easier if you have a 64 bit OS to deal with a 64bit file system, instead of having to be completely backwards compatible. Less work.

    Check out the Vinny the Vampire [] comic strip

  • Unfortuatly, that's not exactly true. When I was consulting to the IT department of a steelworking company, we would order Win98 machines (as 98 was the only OS our supplier would ship) and then whipe and install Win95. The reason behind it was the version of CA-OpenIngres we were using, along with a custom written app, wouldn't run correctly on Win98. Everything was going fine until some third party bitch was asked to audit our office site license. Apparently, according to the auditors, we were violating license agreement by transfering Win98 licenses to Win95 installations.

    The resulting payout the company was forced to do for licenses cost me and a few other contractors our jobs. The budget could no longer afford us. Several other projects got scrapped. Big pain in the arse.

    So, according to licenses, you are forced to upgrade. Thought I'd give you a heads up of potential problems you could run into if this is a buisness you are installing the old software on.

    Kant speel, don't kare.
  • by angry old man ( 211217 ) on Sunday May 20, 2001 @07:56AM (#210187)
    Back in my day, we didn't need all these fancy schmancy boot loaders and tables. There was only one real operating system to boot and that was VAX. Nowadays all you kids think that you need these fancy schmancy OSes because they support 3D games or because they support free speech!

    If you had real computing problems to worry about such as soviet missle trajectory prediction, then you wouldn't even think about booting into multiple operating systems.

  • by AFCArchvile ( 221494 ) on Sunday May 20, 2001 @12:49PM (#210192)
    Just install XP on one drive, install Linux on another (with both hard drives on the primary IDE channel), and then connect two old HDD LED wires onto the Master jumper pins (I know it'll work for WD and Maxtor hard drives). Now, cut the LED end of those wires off and solder them to a PCB with either two 2x1 .100" male headers and a jumper, or a rotary switch that connects one of the pairs but not the other, et cetera. And bingo, dual booting with absolutely NO software intervention; instead, you'd make the boot choice before power-up.

    I could see a company selling a premade version of this.

  • This is a halfway decent strategy, but it of course required Linux boot loader developers to accomodate it.

    I havn't been following this issue closely enough so I can't day with authority that Microsoft made an affort to keep this change in their OS quiet, so as to make it more difficult for other OS providers to accomodate it; although that wouldn't suprise me at all...


  • boot floppy.
  • When I was younger, I thought putting a few OS's on the same hard drive should/would be a snap. But little did I know the enemy MBR was silently laughing in my face! My eyes were opened when I tried to get LILO to boot NT, win9x, OS/2 and Linux all on the same HD. I learned about MBR alright! I then added more HD's and learned about Master, secondary Master, slave and more MBR issues! I learned about hardrive jumpers and why I would need tweasers to install a hard drive! Ever since, I have been trying to figure out who invented the whole MBR thing and SHOOT THEM. The next thing that has to go is master/slave nonsense on HD's.
  • Well, first off, I don't think anyone capable of running today's Linux distros will have much problem jumpering a drive to slave, dropping it in and plugging in two cables. I can't see it taking very long for someone to get a multiboot manager working, even if it requires a patch to XP's boot sequence.

    MBR has been a thorn in the side of OS development ever since people wanted to create more than 4 partitions, and has also chronically stood in the way of hard drives' exponential capacity growth. It has been begging to be replaced for years, and I'm glad someone is finally doing it.
  • The FAQ says that only the 64-bits versions of Whistler will require GPT to boot from, 32-bits versions of Whistler will use the normal MBR.
    15. Can the 32-bit version of Windows Whistler read, write, and boot from GPT disks?
    No. The 32-bit version will see only the Protective MBR. The EE partition will not be mounted or otherwise exposed to application software.
  • I certainly don't want all that amateur code in embedded, possibly life-critical systems.

    On the one hand, we have graduate students from various universities, industrial and academic researchers, and industrial software developers. On the other hand, we have some nebulous people deep inside the bowels of some defense, medical, and aerospace contractors that have been writing "life-critical systems" for decades. The kind of people who managed to blow up the Osprey and the Ariane in software and punch holes into patients with radiation machines. They use and create low-volume, proprietary software with little or no peer review. A review of the kinds of blunders these professionals make shows that there are quite a few bad apples in the bunch and just leaves people scratching their heads. So, no, it isn't obvious to me that the "amateur" open source code is any worse in principle than the professional code.

    Off-the-shelf Linux itself is probably not a good choice for "life-critical systems" because it wasn't developed with that goal in mind. I'd still trust it more than some of the supposedly "reliable" proprietary systems being sold. If you wanted to make Linux truly reliable, you would probably start by throwing out 90% of the drivers and other extensions and carry out a lot more additional code review and testing. In userland, you'd also limit yourself to a small set of very strictly tested tools.

    More generally, however, given the same goals and the same resources for testing and development, I'd trust the open source community considerably more to develop reliable and safety critical systems than a bunch of proprietary professionals: I think the open source community has better people in it, and the open source process brings more resources to bear on code review and bug fixing.

  • As author of both the Linux IA-64 kernel patch to add GPT support, and of the GNU Parted GPT patch, it's appropriate for me to comment.

    Intel's Extensible Firmware Interface specification [] contains the spec for GPT. All systems incorporating EFI (currently, this is only IA-64 systems) need use GPT, for others it is optional.

    I wrote the Linux IA-64 kernel GPT support, and it has been included in the IA-64 kernel port since September 2000. It allows use of either MBR-style or GPT-style partition records per disk. With minor tweaking, it could be used on Linux IA-32, but it is not currently.

    Boot loaders such as LILO remain unaffected. LILO takes a file system location (i.e. /boot/vmlinuz), from which it asks the kernel for the physical location of that file on the disk (i.e. where in /dev/sda is the file?). The file location is entirely independent of the partitioning strategy, thus LILO works with both MBR-style and GPT-style partitioning. Linux IA-64 uses its own boot loader (elilo.efi) which again does not need to be aware of the partitioning strategy of a given disk.

    Partitioning tools such as fdisk and GNU Parted need to become GPT-aware. I have created a patch for GNU Parted and have made it available to the IA-64 distributions. It could use a bit of cleanup, but is quite functional. I don't believe that anyone has written GPT support into fdisk, and would encourage anyone who wishes to please do so.

    I've also written a new Linux IA-64 application "efibootmgr" which manages the EFI Boot Manager options (as defined in the EFI spec above). This tool is both MBR and GPT-aware, and allows for booting your OS off of any disk/partition in your system provided the partition contains a FAT-12,-16, or -32 file system. Additional work will go into EFI in the future to add ext2 file system awareness to avoid even this restriction.

    All of these topics have been discussed on the Linux IA-64 developer mailing list. If you care to join the Linux IA-64 developer community, please see [].

Only God can make random selections.