OSNews

Now this sounds like a perfect fit for open source if there ever was one. Something everyone regardless of the OS can relate to and reap the benefits from.

I like the old fashion BIOS..

"I like the old fashion BIOS.. "

its pretty much outdated. a replacement is sorely needed.ask anyone doing system applications and os development or the hardware manufacturers

Now this sounds like a perfect fit for open source if there ever was one. Something everyone regardless of the OS can relate to and reap the benefits from.

Just curious, how is this terribly different than say, Open Firmware? Really, I'm clueless here, and would like to know.

how much of an advance will this be?
will it do away with
IRQ's?
DMA's?
VGA palatte snoop?
serial, parallel, ps/2 ports?
640k "reserve"
etc

The CPL has been getting pretty popular lately

is this the tech that linus claimed he didnt want to see ever as it was another metalanguage similar to acpi and that was allready takeing up way to mutch time to get supported properly in the kernel?

Any thougths on what this means for DRM on the BIOS level? This is definitely a step in the right direction, open source BIOS means less chance of DRM.

It's good to see that Intel's going to be competing on quality with LinuxBIOS - even better to see that they'll be competing full stop.

Maybe the two projects'll take idea and even code form each other, EPI for standard machines, LinuxBIOS for high-performance ones like the Beowulfs.

Things have just got interesting! Well done, Intel - I'll be interested in seeing the code.

Maybe the two projects'll take idea and even code form each other

Won't happen the CPL is incompatible with the GPL and visa versa, that is atleast for the coding part.

is this the tech that linus claimed he didnt want to see ever as it was another metalanguage similar to acpi and that was allready takeing up way to mutch time to get supported properly in the kernel?

Do you actually realise that support is already in the Linux kernel via the Itanium port?

As for comparison with OpenBoot, there was an article a while back, best bet is to google for it. In a nutshell, you could call it OpenBoot v2.

"is this the tech that linus claimed he didnt want to see ever "

will we ever see a thread that dosent have something related to linux posted :-(

I wish they would have picked up OpenFirmware rather than make their own.

Oh Well, nothing from keeping Apple and SUN from running over to it for their RISC Machines, it would make the hardware Manufacturers lives a heck of a lot easier since the Drivers can just be written to EFI and ignore the OS all together so any OS that runs an EFI system will have the support of the EFI hardware.

Come on IBM it is your platform? Why not set the standard for your own machine? I wish IBM and others used OpenFirmware, just to be contrarian.

Yes, why didn't they use OpenBoot or OpenFirmware? Those aren't hacks like the PC BIOS was, and I guess quite some people are already familiar with it. And the Apple implementation shows that you can also add a module for legacy support, so I guess it would be no problem to create a module to boot real-mode bootloaders like those of Windows 95-XP.

I took a look at the EFI website at intel, and it seems that one can add additional applications to EFI, like a partitioner and a formatting utility. That might sound great, but do those belong in the firmware? Shouldn't the OS handle such things? This because I suspect EFI won't support every existing filesystem, including Sun UFS, FFS, XFS, JFS, ReiserFS, NTFS, HPFS, Fat12 and so on, so that for some systems you still need a bootdisk to format the partition.

Now I have a little question: would a PCI card made for the Apple, thus supporting OpenFirmware, also support Sun's OpenBoot without changing the firmware of the card? If so, then it would be even more a reason to use Open* on the PC, as that would do away with the need to make separate Mac versions of cards.

This is basically the run up to LaGrande. Intel is open sourcing the 'Foundation Code' and appease people, but of course once it is used this won't matter. It's getting it out there that matters.

how much of an advance will this be?
will it do away with
IRQ's?
DMA's?
VGA palatte snoop?
serial, parallel, ps/2 ports?
640k "reserve"
etc

Configuring IRQs and such hasn't been a problem since DOS, because all modern OSs on the PC override the BIOS configuration. As for 640K... what are you talking about? Ever heard of protected mode??? And the ports: that is not a matter of firmware, but they will slowly slip away to USB and Firewire, but not for a long time yet.

If this is what will be used in the Trusted Computing initiative then at least open source OS'es will be able to compete and properly implement TCPA without reverse engineering or breaking patents. DRM per se isn't bad by definition, it's the lockout tactics that big companies like Microsoft and media publishers could (and more than likely would) employ on a closed system that are.

I don't know how to connect the DOTS.
Microsoft bought a motherboard BIOS manufacturer.
Intel embeds a serial number in their chips.
Maybe AMD, too????????
Sometimes you have a hard time installing device drivers in MS because the driver isn't " Digitally Signed ". Eventually, I can force it to use the driver I selected, and it works, so that must mean that the manufacturer refused to pay an exorbatant amount of money for MS to test/certify it???????
If you change too many devices, you have to get another authorization number/code.
Will new devices be required to be certified in the BIOS/by the BIOS? So they can be trusted.........
How many new phone home and trackers will be in Longhorn???
This might make a good read for any concerned.....
http://www.cl.cam.ac.uk/~rja14/tcpa-faq.html

extensible linkable C-interface BIOS code... sounds to me like another way to get even deeper viruses