OpenBIOS

openbios@openbios.org

4297 discussions

Use of BIOS calls in modern code (Was: Re: [OpenBIOS] Why protected mode?)
by Timothy J. Massey 16 Jul '99

16 Jul '99

On Fri, 16 Jul 1999 14:54:01 +0200 (CEST), Matthias W�chter wrote: >Currently we have the same BIOS interface as the 8086 (except >for the INT16 Bios extensions which are not used by some OSes, sadly, >Linux too in opposition to NT and Win98). Linux can't really take advantage of *any* BIOS routines (OK, outside of the loader and such), even when Win98 can. The reason is simple: Linux is cross-platform. Alphas don't have the same BIOS as PC's, which don't have the same BIOS as Macs which don't have the same BIOS as a Sun. So how could Linux use a BIOS routine that isn't commonly available. Now, there is a way that Linux could make use of BIOS calls (if they really wanted to) without losing cross-platform objectives: let the C compiler libraries use BIOS calls. But there, you run into the same problem. The C compiler used for Linux (GCC) is cross-platform, too. So, they don't want to use the BIOS, either! In any kind of portable OS, any feature that isn't available on all systems is a liability. It's something that will either increase the difficulty of porting the code to a system that doesn't support some feature, or requires additional effort to add that functionality to a particular port. Basically, the BIOS in a modern computer is a hardware configurator and boot loader. Personally, I think that that's fine. I'm not looking for a BIOS to do anything more. Tim Massey - To Unsubscribe: send mail to majordomo(a)freiburg.linux.de with "unsubscribe openbios" in the body of the message

4 3

Re: [OpenBIOS] Why protected mode?
by Matthias Wächter 16 Jul '99

16 Jul '99

On Thu, 15 Jul 1999, Maciej W. Rozycki wrote: > You haven't read Intel processor specifications, have you? No, I have not. But thanks for clarifying! Sehr Wus, - Matthias -- Bunt ist das Dasein und granatenstark. Und: Volle Kanne, Hoschis! aus: "Bill und Teds verrückte Reise durch die Zeit" ----------------------------------------------------------------------------- - To Unsubscribe: send mail to majordomo(a)freiburg.linux.de with "unsubscribe openbios" in the body of the message

1 0

Re: [OpenBIOS] Why protected mode?
by Timothy J. Massey 15 Jul '99

15 Jul '99

On Thu, 15 Jul 1999 17:47:57 -0400 (EDT), Kevin P Lawton wrote: >Well, it would have been a much different situation making >a move between 8086 and 286 and doing it right from the >beginning, than ramming all that IA32 stuff into a >new VLIW architecture. Woof. Two things: 1) I see your point. 2) I don't want to offend anyone else by continuing this increasingly off-topic thread! :) You win... ;) Besides, I never liked the 286 anyway! :) Tim Massey - To Unsubscribe: send mail to majordomo(a)freiburg.linux.de with "unsubscribe openbios" in the body of the message

6 5

Re: [OpenBIOS] Why protected mode?
by Timothy J. Massey 15 Jul '99

15 Jul '99

On Thu, 15 Jul 1999 15:03:45 -0400 (EDT), Kevin P Lawton wrote: >Should have dropped real mode, had the CPU start >out in protected mode, and today we'd all have >protected mode only BIOSs. Oh well, such is life. And how would these computer have run DOS? Or Windows? The only OS available before 1994 or so that didn't use real mode code was OS/2. Did everybody switch over to OS/2? So how could Intel drop real mode? In fact, people's biggest complaint with OS/2 in the late 80's was OS/2 Compatibility Box: the way that it ran real-mode DOS programs. If people weren't willing to use an OS that allowed them to run a decent percentage of their old software, why would they use a computer that allowed them to run *none* of their old software! Think about the difficulty that IBM caused for itself when they dropped ISA in favor of MicroChannel. The decision to go MicroChannel was made for almost identical reasons that you would have wanted Intel to change designs with the 286: a better way of doing things. I think the marketplace voted on which way they wanted to go... One other point to run this into the ground: look at the success that Intel has had so far in changing their architecture with IA-64. We've been waiting for, what? 3 years now? And with what result? In an environment as loose as the PC (as the author of Bochs I'm sure knows), anything that isn't 100% backward compatible is horribly frowned upon. Companies like Apple can change CPU designs with success because they control every detail of the hardware, OS and software. In the PC world, we aren't too open to that level of control. Tim Massey - To Unsubscribe: send mail to majordomo(a)freiburg.linux.de with "unsubscribe openbios" in the body of the message

2 1

RE: [OpenBIOS] Why protected mode?
by Winter Jörg 15 Jul '99

15 Jul '99

> ---------- > From: Timothy J. Massey[SMTP:tmassey@iname.com] > Sent: Donnerstag, 15. Juli 1999 21:08 > To: openbios(a)elvis.informatik.uni-freiburg.de > Subject: Re: [OpenBIOS] Why protected mode? > ...... Sorry, but ... ... hey guys are you shure you´re in the right discussion group? this is bios, not architecture ... (and after all, Int*l keeps doing strange things for years now. Why worry? ;-) ...... greetinx, Jake. - To Unsubscribe: send mail to majordomo(a)freiburg.linux.de with "unsubscribe openbios" in the body of the message

1 0

Re: [OpenBIOS] Why protected mode?
by Timothy J. Massey 15 Jul '99

15 Jul '99

On Thu, 15 Jul 1999 12:31:48 -0400 (EDT), Kevin P Lawton wrote: >Also IMHO, by far their greatest mistake in history >was when they went to a new protected mode environment >in the 286, they carried forward the segmentation >architecture. But they had to. The 286 is a 16-bit processor. Without segmented memory, you would be limited to 64k of memory... Leaving protected mode out of it, the rest of the 286 was merely an evolution of the 80186: a change from a 20-bit memory bus to a 24-bit. That's the only real change. Of course, there *is* protected mode... Only with a 32-bit processor can you realistically eliminate segmented memory. Tim Massey - To Unsubscribe: send mail to majordomo(a)freiburg.linux.de with "unsubscribe openbios" in the body of the message

4 3

Re: [OpenBIOS] Why protected mode?
by Timothy J. Massey 15 Jul '99

15 Jul '99

On Thu, 15 Jul 1999 20:42:27 +0200 (CEST), Matthias W�chter wrote: >On Thu, 15 Jul 1999, Timothy J. Massey wrote: > >> Only with a 32-bit processor can you realistically eliminate segmented >> memory. > >Don't say that: There are some Xeon motherboards around (well, most of >them are sold in complete supported systems) supporting 16GB of memory and >more. Since you need more than 32 bits to address this space, they had to >design work-arounds like segmentation (I don't know anything more about >it). Even worse: The switching is done externally, by the chipset... That only makes my point even more. In that case, even 32-bit memory paths are not enough. If a 32-bit bus won't do it, why would you even attempt it with a 16-bit CPU? I'm not saying that 32 bits is all you need. What I'm saying is that you need at *least* 32 bits! Tim Massey - To Unsubscribe: send mail to majordomo(a)freiburg.linux.de with "unsubscribe openbios" in the body of the message

1 0

Re: [OpenBIOS] Why protected mode?
by Phil King 15 Jul '99

15 Jul '99

At 05:28 PM 7/7/99 -0700, you wrote: >>> 2) Go into protected mode for initial configuration, then exit it to load >>> the boot code (note that the BIOS service routines would have to run in >>> real mode in this case). >> >>This is how other BIOSs today do it. That is also how you have to do if >>you strive for IBM-PC BIOS compability. > >And this continues the necessity for either a keyboard or a keyboard >simulator, does it not? My recollection is that the only way out of >protected mode (back in 286 days) was through a keyboard reset sequence. That was only on the '286, which required a processor reset to get back into real mode and was assisted by the keyboard controller, one of the Great PC Hacks. It is not as traumatic with '386 and beyond. Cheers! Phil King pking(a)netschools.net - To Unsubscribe: send mail to majordomo(a)freiburg.linux.de with "unsubscribe openbios" in the body of the message

3 3

Re: [OpenBIOS] Why protected mode?
by William Meyer 15 Jul '99

15 Jul '99

>I also wanted to add that all this nonsense was necessary because int he >design of the 80286 there was a method of ENTERing protected mode, but no >clean way to get back. I'm curious to know whether the double/triple fault >was used as a way back because that's how Intel had *designed* it, or if the >lack of method to come back to Real mode was an oversight. Thinking back, I recall an article which discussed the one way process, and offered no alternative to the keyboard kludge. As the author seemed well versed in the inner essence of the 286, I tend to thing that your multiple fault method was discovered later, as a head-smacking sort of experience. William Meyer MATCO www.matco-video.com - To Unsubscribe: send mail to majordomo(a)freiburg.linux.de with "unsubscribe openbios" in the body of the message

1 0

Re: [OpenBIOS] Why protected mode?
by Timothy J. Massey 15 Jul '99

15 Jul '99

On Thu, 15 Jul 1999 08:39:45 -0400, Andrew Kohlsmith wrote: >> > That was only on the '286, which required a processor reset to get back >> > into real mode and was assisted by the keyboard controller, one of the >> > Great PC Hacks. It is not as traumatic with '386 and beyond. > >I also wanted to add that all this nonsense was necessary because int he >design of the 80286 there was a method of ENTERing protected mode, but no >clean way to get back. I'm curious to know whether the double/triple fault >was used as a way back because that's how Intel had *designed* it, or if the >lack of method to come back to Real mode was an oversight. > >Looking at pages like x86.org, I tend to think the latter. Definitely a mistake. Well, let's say shortsighted thinking from Intel. They thought, "Once in protected mode, who would want to go back to real mode?" In fact, that was one big reason why OS/2 version 1.x was slow and unreliable when it came to DOS programs. Resetting the processor to real mode on the 286 had been described by Intel as "Putting a car's transmission into reverse at 50 miles per hour". (That and the lack of hardware virtualization via the IO map...) Of course, unless we're writing a bios for 286 computers, that's all a moot point! :) But an interesting piece of trivia nonetheless. Tim Massey - To Unsubscribe: send mail to majordomo(a)freiburg.linux.de with "unsubscribe openbios" in the body of the message

1 0

Jump to page: