Support For Old Hardware Is Being Removed From Coreboot

[phoronix]

Phoronix: Support For Old Hardware Is Being Removed From Coreboot

Coreboot developers are taking to their Git tree and dropping support for old motherboards and chipsets...

http://www.phoronix.com/scan.php?page=news_item&px=Coreboot-Dropping-Old-Hardware

[uid313]

A problem is that Coreboot barerly has any support for new hardware.

And for recent hardware I believe the support is often not yet mature and complete.

[gamerk2]

"Because it's not being maintained" is the worst excuse ever. Does it work? Yes. It doesn't need changes.

Because no one ever runs on old hardware; nope, everyone retired their LGA 775 motherboards ages ago.

[WOLF308]

I'll put this out there first. I'm not a coder, have tried to learn many times over the years but it never "sticks" so I have no idea how much work what I'm about to suggest would involve.

Could the PTS be made to run lspci and the superio tool mentioned at http://www.coreboot.org/Support and then compare it to the list at http://www.coreboot.org/Supported_Chipsets_and_Devices to make identifying potentially compatible boards easier or at least giving the coreboot devs more information to work with?

[Tomin]

Because no one ever runs on old hardware; nope, everyone retired their LGA 775 motherboards ages ago.

I think 775 is a bad example. I'm using Core 2 Duo (LGA 775) right now and it works just fine. No reason to retire the motherboard any time soon.

[Michael]

Could the PTS be made to run lspci and the superio tool mentioned at http://www.coreboot.org/Support and then compare it to the list at http://www.coreboot.org/Supported_Chipsets_and_Devices to make identifying potentially compatible boards easier or at least giving the coreboot devs more information to work with?

Easily, PTS already keeps track of PCI information and most of it available on OpenBenchmarking.org where I have detailed information on a combined total of millions of different hardware/software combinations and components, but unfortunately lack the time to do much more with it.

[pgeorgi]

"Because it's not being maintained" is the worst excuse ever. Does it work? Yes. It doesn't need changes.

The main problem is that we have no idea if that stuff still works, and quite some indication that it doesn't. Well maintained old code would be kept (and we're working on criteria to determine "well maintained" better than we do now).

Those boards are just those where there's a relatively high probability that they're broken, no report that they're not and nobody around who volunteers to be a maintainer.

[pal666]

how about adding support for new hardware? like any current gen off-the-shelf amd or intel motherboard?

[SystemCrasher]

So they ditched old MBs support, can't support half of new hardware because of boot guard crap, and ... I wonder what it would support at the end? Just few mere Chromebooks, eventually turning into Google's pet project?

[Luke]

So they ditched old MBs support, can't support half of new hardware because of boot guard crap, and ... I wonder what it would support at the end? Just few mere Chromebooks, eventually turning into Google's pet project?

Even if that happens, it will hopefully guarantee at least one source of boot guard free, locked UEFI-free laptop computers. This goes double so long as Google continues to support a "vanilla" or "developer" version that does not slow down booting things other than ChromeOS. I would not consider buying any of those locked Intel machines unless there was literally no other choice and I had a known working way to hack it far enough to run my OS. That would also require firewalling out any spyware in the UEFI by things like removing the original network device and using a deliberately incompatable USB one instead.

How do we know future versions of Windows won't use UEFI code to reinstall spyware removed by the user? Lenovo has already been caught doing this. The obvious furture use case would be spyware to scan a user's home directory for copyrighted photos, audio and video files etc and report "violations" to some TPP-mandated copyright enforcement agency. It would be like running Picscout on your local machine, with forced reinstallation if you tamper with the binary. This would be done to counter USB filesharing after ISP's block copyrighted content, and to counter files shared using onion routed darknets that ISP's will have great difficulty blocking.