Microsoft's Dirty ACPI Secrets: It's Back! (Updated)

Dr. Roy Schestowitz

2008-07-25 19:12:49 UTC
Modified: 2008-07-26 06:50:44 UTC

Watch this new page, which is already reaching huge levels of circulation.

Complained to the Federal Trade Commission

http://www.ftc.gov

Foxconn 458 E. Lambert Road Fullerton Fullerton, CA 92835

FOXCONN PHONE NUMBER: 714-871-9968

Company sold me a computer motherboard, model G33M-S, claiming that it was compliant with ACPI versions 1.0, 2.0, and 3.0.

Linux and FreeBSD do not work with this motherboard due to it's ACPI configuration, using a disassembler program, I have found that it detects Linux specifically and points it to bad DSDT tables, thereby corrupting it's hardware support, changing this and setting the system to override the BIOS ACPI DSDT tables with a customized version that passes the Windows versions to Linux gives Linux ACPI support stated on the box, I am complaining because I feel this violates an anti-trust provision in the Microsoft settlement, I further believe that Microsoft is giving Foxconn incentives to cripple their motherboards if you try to boot to a non-Windows OS.

We have received your complaint.

Thank you for contacting the FTC. Your complaint has been entered into Consumer Sentinel, a secure online database available to thousands of civil and criminal law enforcement agencies worldwide. Your reference number is: 19642372

[...]

They detect Linux, give it a bad DSDT table, one that looks ok at a glance, but broken in subtle ways so that some of it works, but not correctly.

You call them to ask why their board won't run Linux.

They tell you to buy Vista.

They're basically rubbing Microsoft's back.

Regular readers may have already seen that Microsoft had been exposed for playing dirty tricks with ACPI.

There's this blast from the past [PDF], courtesy of Comes vs. Microsoft:

From: Bill Gates Sent: Sunday, January 24, 1999 8:41 AM To: Jeff Westorinen; Ben Fathi Cc: Carl Stork (Exchange); Nathan Myhrvold; Eric Rudder Subject: ACPI extensions

One thing I find myself wondering about is whether we shouldn't try and make the "ACPI" extensions somehow Windows specific.

It seems unfortunate if we do this work and get our partners to do the work and the results is that Linux works great without having to do the work.

Maybe there is no way to avoid this problem but it does bother me.

Maybe we could define the APIs so that they work well with NT and not the others even if they are open.

Or maybe we could patent something related to this.

The next few days will be interesting because reactions to the Foxconn story will be everywhere. ⬆

Update: There appear to be some skeptics and apologists trying to defend Foxconn. They ought to see this from the high-reliable and always-credible Phoronix.

Foxconn Does Hate Linux Support

[...] The DSDT for Windows is correct, but Foxconn isn't interested in issuing a (simple) update to fix the Linux support. However, this isn't surprising to us. We've known that Foxconn does not wish to support Linux at all. Going back to 2006, Foxconn has told us at Phoronix that they aren't interested in Linux on their motherboards and they have no desire to support it.

This response from Foxconn is ironic too:

You are incorrect in that the motherboard is not ACPI complaint. If it were not, then it would not have received Microsoft Certification for WHQL.

What does the Microsoft Certification say? "The BIOS ain't done until Linux won't run"?

Comments

Yuhong Bao

2008-07-26 06:07:51

The difference is this time they are actually doing it, where last time it was just a suggestion.
David Gerard

2008-07-26 20:57:39

Matthew Garrett attributes it to stupidity rather than malice, and says such stupidity is the norm. (He does a ton of work on ACPI in the Linux kernel.)
Alon David

2008-07-26 21:51:44

Hi,

I have a lenovo 3000 N100 laptop, and after coming back from sleep mode the fans on the motherboard don't work anymore, i think this is an ACPI issue (i've tested this on multiple kernels) but I have no idea on how to investigate it and fix it, can anybody help ?

Thanks.
Roy Bixler

2008-07-27 16:36:29

No doubt that this is really confusing, so it's an issue ripe for different interpretations. The exhibit from the Comes Iowa anti-trust case just adds fuel to the fire. On the other hand, it does sound reasonable that the vendor just never bothered testing their board with Linux because they have no intention of supporting that.

My questions would be the following. Why test for an operating system in the code if there is no intention to support it? Why exactly would a hardware vendor explicitly want to exclude an operating system if supporting it would mean more sales for them?
David Gerard

2008-07-27 16:43:18

Matthew Garrett explains it as: everyone copy'n'pastes their ACPI code, and so the Linux code there is probably leftover fragments of junk code that are never executed anyway; the manufacturer bodged something together, checked it didn't actually catch fire running Vista and shipped. (i.e. such stupidity is the norm.) Here's his full technical writeup as an expert on ACPI code archaeology.
Roy Schestowitz

2008-07-27 16:49:30

What bothers me a lot is that I regularly find people who ditch Linux for something like Vista just because of ACPI. Here is a brand-new example I found just an hour ago:

http://llynix.com/articles/linux/i-bid-adieu-ubuntu/

"Yesterday I threw in the towel. I’ve been having problems with ACPI on my new Ubuntu partition. I’ve got other small problems but the ACPI ones are really the deal breaker. A laptop that can’t hibernate isn’t much of a laptop. [...] Perhaps the LiveCD of the new version will fair better, for now though I’m taking a break and going back to the slower, more sluggish but more compatible Vista system."

Whether there was technical sabotage or not, someone ought to start taking Linux support seriously. That memo from Gates ain't helping. Foxconn's attitude (watch the conversations with them), is not helping, either.
David Gerard

2008-07-27 17:10:46

From the Linux side, Matthew is the most prominent person doing so. From the hardware side, the trick is how to make hardware people want to actively support Linux working (which is why Matthew got so annoyed at the conspiracy theorising in the Foxconn case, when he sees only the usual blithering stupidity and incompetence. Dell pushing its suppliers for bits that'll work well in Ubuntu is basically what we need. The suppliers need to want to support Linux.
Roy Bixler

2008-07-27 17:23:52

@David Gerard:

Thanks for the technical explanation. It seems to confirm what I've already heard about ACPI, which is that it's complex. The answer in a nutshell thusfar seems to be "it's either a Linux bug or a BIOS bug." Either way, I hope it's sorted. Further down in the comments, Ryan (the original person who reported the problem) indicates that Foxconn managers are now willing to work with the Linux ACPI developers. Apparently, they're not oblivious to the argument that they could sell more boards if they achieve better compatibility.
Roy Schestowitz

2008-07-27 20:40:26

Ditto. Thanks, David, for putting a rational and positive spin on it. You should see what Carla at Linux Today wrote about this incident.

Microsoft's Dirty ACPI Secrets: It's Back! (Updated)

Foxconn Does Hate Linux Support

Comments

Recent Techrights' Posts