Is Apple using BIOS?

[shamino]

Unfortunately, "Macintosh computers using Intel microprocessors do not use Open Firmware."

from: http://developer.apple.com/documentation/MacOSX/Conceptual/universal_binary/universal_binary.pdf page 47

Re-read that. The developer docs say that the current transition systems don't use it. They tell developers to not assume OF's presence.

It says nothing about what will ship in production x86 boxes. Apple has already said that they will not say what those systems use.

 

[shamino]

(Big, big hint: A standard PC BIOS can't boot an HFS+ volume.)

Whatever gives you that idea?

The boot sequence of a PC doesn't depend in any way on how the disk partitions are formatted, or even on how the drive is partitioned.

The BIOS loads absolute-disk-block zero into memory (the master boot record, or MBR) and executes whatever it finds there. That code is typically a stub-loader that reads the partition table and identifies the bootable partition.

The MBR code then loads block-zero of the bootable partition (the so-called "boot sector") and executes that. This code knows just enough about the OS's file system to find and load the OS's first-stage boot loader. (This was IBMBIO.COM or IO.SYS back in the MS-DOS days. Other files on other OS's).

You can use any kind of partition table you want, as long as the MBR code understands it. You can use any kind of file system as long as the boot sector understands it.

This is why a PC can boot from FAT, vFAT, FAT32, NTFS, HPFS (OS/2), ext2 (Linux) or ext3 (Linux). There is no reason why HFS/HFS+ couldn't also be booted - all it takes is a compatible boot-sector in the partition - something that should be easy for anybody familiar with HFS to be able to write.

 

[y0zza]

Did any one actually read the linked dev-list message?

Something interesting in addition to what was quoted by this thread:

> I'm cautiously hoping for EFI as the firmware.

The general consensus I've heard from other developers is:
1) They don't want us to use BIOS
2) If they haven't heard of EFI, they want us to use OF
3) If they have heard of EFI, they want us to use EFI

This is not a statement about what Apple will use, just what I've heard from developers that have an opinion on the subject.

I would certainly expect EFI from Apple if they abandon Open Firmware. They are in the position to build a fully legacy-free system, and it would be surprising if they didn't take advantage of that fact.

 

[shamino]

Another bit on EFI: "It allows vendors to create drivers which cannot be reverse engineered."

And I'll believe that insane clame when I see it. There is nothing in the world that can't be reverse engineered.

If you can observe the code's behavior, you can reverse engineer it. The only code that can't be reverse engineered is code that appears to serve no purpose.

 

[shamino]

What does a BIOS apple mean for end users? Probably the most significant would be really excessive times for sleep/unsleep functions. We're talking 10-15 seconds here, more more of that sub-1 second unsleep time that you get currently when you open and close your iBook or Powerbook.

[i[That[/i] is why you wouldn't want a BIOS apple as an end user. Oh, and you'd likely lose features such as option key at boot to select the boot volume, firewire disk mode, and several others.

What makes you think this is due to BIOS and not Windows?

 

[shamino]

What is BIOS?

Basic Input/Output System.

It's a ROM in every PC. It contains the code necessary for booting the OS. It also provides simple I/O services, which were used extensively back in the days of MS-DOS (but are not used very much today, since modern operating systems provide much better equivalent services.)

While some parts of BIOS are very modern (like power management code and chipset-setup facilities), some parts are downright ancient (like the code for reading/writing strings to a text-mode console), dating back to the very first IBM PC's. (Although even this code has been rewritten/optimized over the years.)

While some claim that the BIOS is a horrible thing that cripples computer systems, that's not really true. A modern OS almost never uses any BIOS code, after the boot sequence completes. The only real exception to this that I can think of is power management (including sleep mode). And this is only because it's considered easier to use the BIOS code than to write an OS-level device driver to replace that code.

 

[iMeowbot]

Whatever gives you that idea?

It wasn't possible until Darwin 8.0.1. I see that they've now mangled up a workaround.

 

[IJ Reilly]

What makes you think this is due to BIOS and not Windows?

Haven't you said as much yourself, more or less -- that energy management is handled by the BIOS? This is presumably why my XP box freezes more often than not when I try to wake it from standby mode and my Mac never does.

Anyway, I don't think we need to fear PC-BIOS. Apple would never take such a huge step backwards when they've already got Open Firmware and could use EFI.

 

[jhu]

Re-read that. The developer docs say that the current transition systems don't use it. They tell developers to not assume OF's presence.

It says nothing about what will ship in production x86 boxes. Apple has already said that they will not say what those systems use.

which makes sense. besides, what kind of programs, other than some specialized tools, really need to know what type of firmware is present on the machine?

 

[iMeowbot]

which makes sense. besides, what kind of programs, other than some specialized tools, really need to know what type of firmware is present on the machine?

It's not so much the kind of firmware but that IODeviceTree is reportedly b0rked. That's going to make asset manglement fun, and bug feedback programs a pain.

 

[mischief]

And I'll believe that insane clame when I see it. There is nothing in the world that can't be reverse engineered.

If you can observe the code's behavior, you can reverse engineer it. The only code that can't be reverse engineered is code that appears to serve no purpose.

Okay. Shall we say then that, in this case:

Since the hardware environment the EFI code will be written to will likely encorporate enough unique parts and drivers that reproducing that environment would require either custom fabrication or very cunning emulation the act of reverse engineering it could prove a very amusing (and time consuming) "Rubik's Cube" to the fifteen most talented Hackers in the world while proving daunting enough to everyone else that the act of reverse engineering approaches diminished returns with about the same impact noise as a pigeon hitting a window?

See also: How long until we have completely reverse engineered the Human Genome? Ten years? Does that fit on a standard Hacker's timetable of "to do"'s?

 

[mischief]

Did any one actually read the linked dev-list message?

Something interesting in addition to what was quoted by this thread:


I would certainly expect EFI from Apple if they abandon Open Firmware. They are in the position to build a fully legacy-free system, and it would be surprising if they didn't take advantage of that fact.

Bingo. This is the IDEAL opportunity for Apple to build an uncontestably superior product using Technology Intel can't get the rest of the industry to adopt because it would make their machines cost more. EFI is just one of easily fifteen extremely cool Intel technologies that are currently under represented in the PC market due to the industry emphasis on "Cheaper, Faster, Legacy Friendly".

Apple will go from being an overpriced four-star sushi house in a Macdonalds world to being a competitively priced four-star steakhouse in a Macdonalds world.

 

[SiliconAddict]

BIOS can burn in hell. Open Firmware is a million times better...

Option on Startup -graphical view of all possible ways to start up. (may take a bit). Can use mouse.
F3 on Startup -no startup sound
C on Startup -force startup from CD
X on Startup -Force startup into OS X (on dualboot system, is not permanent)
T on Startup -FireWire Target Disk Mode
Command+Option+P+R on Startup -Zap PRAM
Command+S on Startup -Single User Mode
Command+V on Startup -Verbose Startup
Command+Option+N+V on Startup -Open Firmware prompt

Eat that BIOS!

Ummm I can do 90% of that on my Thinkpad.


graphical view of all possible ways to start up- Thinkpad has a sep partition that is hidden from the OS at a hardware level. This partition has a small footprint and can be accessed by hitting the ACCESS button on the keyboard when booting up.
Startup sound - Its called mute.
Boot from CD - F12 gives you a boot list. from CD - Network
FireWire Target Disk Mode - I would LOVE to have this feature. I know Apple could do this with BIOS.
Verbose Startup - F8
Zap PRAM - Not needed to begin with since this is an Open Firmware thing.
Single User Mode - What is single user mode? Is that like safe mode on Windows? If so...F8 again.
Open Firmware prompt - Is this the command line? Apple's is prob more robust but you have the command console for Windows but that (Much to MS's stupidity) isn't installed by defauly. winnt32.exe /comcons

Bet you $20 that Apple could do a good enough job with BIOS that they completely fool everyone into thinking that its OF.

 

[Lepton]

EFI + LeGrande = Secure Mac

Read up on Intel's LeGrande technology. It is hardware based and among other things lets a process run in a secure environment. Combine it with EFI and this is very likely the mechanism Apple will use to insure 1) A great startup experience, and 2) High security and an inability to run Mac OS on boxes other than Apple's.

This stuff is going to be very hard to hack, being hardware based. Which is good, because I believe that having some way to run Mac OS on any old PC box would be a disaster for Apple's business.

Of course, hackers are really, really smart and getting Mac OS to run on regular PCs might be the new holy grail soon...

Read about LeGrande here

 

[SiliconAddict]

Haven't you said as much yourself, more or less -- that energy management is handled by the BIOS? This is presumably why my XP box freezes more often than not when I try to wake it from standby mode and my Mac never does.

Even if they use PC BIOS don't you think Apple would make sure whatever they release would have instant resume? Apple isn't going to release ANYTHING that is halfbaked and I can attest to coming out of standby being fast on my Thinkpad. What hoses it is 3rd party drivers. Example. The thinkpad has been using an orinoco WIFI card and it causes a 30-60 second delay in coming out of a sleep mode. I just upgraded to an internal centrino WIFI card and I've had zero problems since. Takes about 2 seconds to come out of standby. Finally not to start a flame war but I've read my fare share of posts on macrumors that read "HELP! Standby not working on my PowerBook!) So OF isn't the end all be all solution.

 

[anonymous161]

(Big, big hint: A standard PC BIOS can't boot an HFS+ volume.)

[SKYFALLING] Well that's it! Apple is switching to NTFS!!![/SKYFALLING]

 

[Sun Baked]

(Big, big hint: A standard PC BIOS can't boot an HFS+ volume.)

[SKYFALLING] Well that's it! Apple is switching to NTFS!!![/SKYFALLING]

C'mon, didn't you guys click on the link... the quoted arn left is from the developers basically asking what partition format Apple plans on using.

Mark Eaton writes:

says that the partition format is different for Macs running on Intel
processors.

Is there any more detailed info/documentation available on what
specifically is different?

 

[IJ Reilly]

Even if they use PC BIOS don't you think Apple would make sure whatever they release would have instant resume? Apple isn't going to release ANYTHING that is halfbaked and I can attest to coming out of standby being fast on my Thinkpad. What hoses it is 3rd party drivers. Example. The thinkpad has been using an orinoco WIFI card and it causes a 30-60 second delay in coming out of a sleep mode. I just upgraded to an internal centrino WIFI card and I've had zero problems since. Takes about 2 seconds to come out of standby. Finally not to start a flame war but I've read my fare share of posts on macrumors that read "HELP! Standby not working on my PowerBook!) So OF isn't the end all be all solution.

I don't do flamewars, so not to worry.

My technical knowledge here is limited, so I can't say with confidence either way, but from what I understand, PC-BIOS is PC-BIOS. It's primitive, and nearly 25 years old, and can't be tricked around to be much more than what it is. Open Firmware is a step or two up in technical sophistication and from what I've read so far, ESA is another step or two up. It's an Intel technology besides, so that's what I'd expect Apple to use.

 

[MaCaDDiCT21]

I would love my mac to have a BIOS. If you know how to use it correctly, it can be extremely powerful.

I would love it if I could OC my mac...

 

[WM.]

I would love my mac to have a BIOS. If you know how to use it correctly, it can be extremely powerful.

I would love it if I could OC my mac...

Software overclockability doesn't have anything to do with BIOS vs. OF vs. EFI either. The reason you can't overclock your Mac is that Apple has designed the hardware so that you can't (without at least unsoldering and resoldering zero-ohm resistors). There were some iBooks a few years ago that you could overclock in software (well, Open Firmware), but you had to type commands into the OF command prompt. This again is not a limitation of OF itself, since it would be easily possible to create an interface for OF that would be much easier to use than the semi-graphical one on most BIOSes I've seen.

See here for some very cool things you can do in Open Firmware (scroll down near the bottom). I got the Tower of Hanoi one to work, but the GUI one mangles the window and doesn't show a mouse pointer at all (on my 1.5 GHz PB12--perhaps my version of OF is too new...).

To reiterate, the only reason you can mess with CPU clocks and RAM timings and such things in many PC BIOSes is that both the hardware and the BIOS have been specifically designed to support it. It's not that having a BIOS magically enables that functionality, and it would be easy to provide in Open Firmware (and with a better interface to boot). The issue is that Apple doesn't want and probably will never allow people to mess with the hardware in that fashion.

 

[WM.]

Ummm I can do 90% of that on my Thinkpad.

graphical view of all possible ways to start up- Thinkpad has a sep partition that is hidden from the OS at a hardware level. This partition has a small footprint and can be accessed by hitting the ACCESS button on the keyboard when booting up.

The fact that that requires a separate partition (!!!) should tell you something about the limits of the classic PC BIOS...

The graphical bootloader in OF doesn't reside on any bootable devices; it's written into the firmware itself.

Open Firmware prompt - Is this the command line? Apple's is prob more robust but you have the command console for Windows but that (Much to MS's stupidity) isn't installed by defauly. winnt32.exe /comcons

No. The Open Firmware prompt is something rather different. As the name implies, it doesn't have anything to do with the OS. You can access it by holding down cmd-opt-O-F on boot on any NewWorld (i.e. has built-in USB) Mac.

As with the graphical boot-device selection thingy, this is not resident on any attached devices; it's part of the firmware itself. You can disconnect all hard drives and other storage devices from a Mac and still have access to Open Firmware (including the graphical selection thing).

Bet you $20 that Apple could do a good enough job with BIOS that they completely fool everyone into thinking that its OF.

I doubt it--not if we're talking about the current, standard PC BIOS...

 

[Mechcozmo]

Doesn't Open Firmware fit this definition???
Isn't Open Firmware then another BIOS?

Open Firmware is a BIOS, but not the BIOS.

 

[jauh]

The fact that that requires a separate partition (!!!) should tell you something about the limits of the classic PC BIOS...

The graphical bootloader in OF doesn't reside on any bootable devices; it's written into the firmware itself.


No. The Open Firmware prompt is something rather different. As the name implies, it doesn't have anything to do with the OS. You can access it by holding down cmd-opt-O-F on boot on any NewWorld (i.e. has built-in USB) Mac.

As with the graphical boot-device selection thingy, this is not resident on any attached devices; it's part of the firmware itself. You can disconnect all hard drives and other storage devices from a Mac and still have access to Open Firmware (including the graphical selection thing).


I doubt it--not if we're talking about the current, standard PC BIOS...

Not quite sure where the thing with ThinkPad comes in, but on my T23, there's no separate partition that's used for Graphic Screen (that's all it really is) at startup, it's a part of T23's BIOS. AFAIK, lots of companies, MB manufacturers are puthing this graphical screen into their BIOSes recently... It's a royal PITA, as I personally prefer to watch the numbers roll at startup, so I have to go round switching this (nothing more than political propaganda on manufacturers' behalf) off...

In fact, why on earth do we care OF vs. any other flavour of BIOS??? I need a system that is upgradable (something that current PowerMacs currently aren't really) and stable! If Apple can provide that on an Intel platform I'd buy one tomorrow! Of course, I get to check out their transition kit and see how that plays out...

 

[GregA]

Don't you love the vagueness of a statement like the Quote in there? "We can't tell you yet ... but we will as soon as we reasonably can." It does appear to indicate (at least leaning towards) that the new systems will not be configured anywhere near the transition boxes. JMO <though I guess it also leaves open the possibility that the present BIOS in the transition boxes MAY actually be in the shipping computers ... LOL>

Nah, it's a good statement. It leaves them open to several things

1. They can use Intel's open standard "EFI", with an Apple feel.
OR
2. They can remain open to using BIOS and release a machine more like the current development system (if their EFI developments or Intel's Yonah chip take longer than expected)
AND
3. If they do go straight to EFI with more efficient motherboard, they can make some good publicity about Apple quality, since their motherboard won't need all the legacy chips that old BIOSes expect.

 

[Abdesai]

"Encapsulated" Drivers in the Firmware

A great feature of the current Macintosh ROM (firmware) which is based on Open Firmware (OF, http://playground.sun.com/1275/) is the ability for PCI adapter cards to include both the Open Firmware driver, if required for booting, and also the Mac OS driver "encapsulated" for the PCI adapter card. This allows the Macintosh platform to have the true "plug and play" elegance of a PCI adapter card appearing to be "driverless" because the user simply puts the card in the Macintosh and it "just works". The current Windows environment is vastly interior by requiring some type of BIOS driver to reside in the PCI adapter card ROM to boot, but then during boot, the Windows environment will "detect new hardware" and require a disk to be available to install the Windows OS driver.

If Intel's Extensible Firmware Interface (EFI, http://www.intel.com/ technology/efi/index.htm) used as the ROM code for a Macintosh based on an Intel processor would still allow PCI adapter cards to include not only the pre-OS boot drivers, but also the actual Mac OS driver "encapsulated", then EFI would seem to provide as good as solution as the current OF. Otherwise it would seem that OF has very distinct "end user experience" advantage over EFI. Can anyone clarify if EFI would support PCI adapter cards with their own ROM's having both the pre-OS boot drivers and also the Mac OS driver "encapsulated" that would give the user experience of "driverless" PCI adapter cards?


Joel

http://lists.apple.com/archives/Darwin-drivers/2005/Jun/msg00024.html

What do you people think about this?