Update On Free Linux Driver Development

Remember the offer Greg Kroah-Hartman made earlier this year, to get Linux drivers written for free for any company that wanted them? Now an anonymous reader points us to an article up on linuxworld with an update to this program. Greg K-H, who leads the development of several kernel subsystems including USB and PCI, admits that the January offer was a bit of "marketing hype" — but says it has brought companies and developers together anyway. Twelve companies have said "yes please," one driver is already in the kernel, and five more are in the pipeline.

so did he write them for free or didn't he

[timmarhy]

if he did, good for him, if he didn't he just like every other lieing software house out there.

Re:so did he write them for free or didn't he

[QuantumG]

the volunteers write them without charging the OEMs, yes.

Re:

[bobo mahoney]

Volunteers working for free? Now I'm no genius but isn't that a bit redundant?

Re:

[QuantumG]

You've never heard of a paid volunteer?

Re:

[OldHawk777]

Only in politics, crime, nepotism ... and other stuff.

Sometimes, I am just to redundant with a left and right brain tearing my mind apart.

Re:so did he write them for free or didn't he

[Kelerain]

The marketing hype was the fact that the kernel developers would do this sort of development anyway. This isn't a special program of any kind, it's standard procedure and they were promoting it somewhat like other people promote special "one time offers" and such. So yes marketing hype, and yes they do in fact do that. The helpful part is they have actual hardware samples and/or specs to work with, so it's a real win all around.

Re:

[Atzanteol]

Yes, how dare the bastard exchange currency for goods and services!

List?

[Constantine XVI]

A list of the twelve companies, please?

Re:

[CasperIV]

The list of primary hardware is relatively short when you consider what is making real penetration. If the 12 that asked for drivers are any of the larger companies in the market, it could be a big deal. It would be like if Broadcom were to come to their door and ask for real drivers that work... that right there would cover most newer laptops.

Re:List?

[Zoxed]

> A list of the twelve companies, please?

Maybe this is covered by the NDA mentioned in the article :-)

(Us, no not us: we are a god fearing capitalist company: we would never deal with those commy GPL peeps !)

Re:List?

[rbanffy]

It's easy to imagine a Microsoft exec saying "Nice driver they made for your hardware. But it would be a shame if the device didn't work under Windows anymore, wouldn't it?"

Re:

[ajs318]

I've often wondered about this myself -- could Microsoft actively prevent hardware manufacturers from releasing Linux drivers, just by refusing to certify Windows drivers for any hardware which comes with a Linux driver? And for that matter, have they been doing exactly that?

Now that Windows is moving towards a more locked-down kernel, it's certainly technically feasible. It's probably Anticompetitive Behaviour, but that doesn't seem to be illegal in the USA anymore.

Re:List?

[dpilot]

Let's rephrase this...

There has been enough fuss recently over "Vista-Certified" graphics drivers that didn't really work well with Vista. So all you need is a driver certification plan that pretty much can't possibly be met in every detail. Then go ahead and be relaxed about the certification - most of the time. When you come across a recalcitrant hardware vendor who provides drivers and/or documentation to Linux, it's time to insist on dotting all I's and crossing all T's.

I've suggested a similar possibility in the old 55MpH days on the New York State Thruway. The official speed limit was 55MpH, but the average speed on the road was somewhere above 65MpH. Nearly everyone on the road was a lawbreaker. So at that point, you can use whatever other criteria you wish, and know that whoever you choose to pull over is breaking at least the speed limit. I have no suspicion or evidence that this "selective law enforcement" was actually happening, but never liked the mere possibility.

Re:List?

[ajs318]

..... a problem which only arises because the existing law which already obliges manufacturers to disclose details to hardware owners is not being enforced. Write to your MP and ask for a new law, obliging hardware manufacturers to release, generally and gratis (or at any rate, for no more than the cost of copying and delivery) sufficient documentation so that independent programmers can write drivers for their hardware as a precondition before it is allowed onto the market. Independent expert review of the documentation would be a requirement alongside electrical safety / EMC certification.

The more free drivers, the better

[BluSteel]

Marketing hype or not, I'm always happy to see more hardware supported by free drivers. Thanks, Greg.

Re:

[EsbenMoseHansen]

...free as in beer.

According to the article, free as in GPLv2.

patents, usability

[bcrowell]

To me, the issue isn't so much drivers as patents and usability.

My daughter's mp3 player didn't need any special drivers, because it's simply a standard keychain drive that happens to be able to play mp3's. However, she totally couldn't figure out how to use it on her ubuntu box. There was one problem after another. Ubuntu tried to do the right thing by popping up a gui app when she connected it, but then we couldn't get the gui app to do what we wanted to do. Part of the problem was that getting the mp3 codec to work was a pain, and that springs directly from the fact that mp3 is patented.

My Brother HL-1440 laser printer is 100% supported in Linux. Brother hired the CUPS developers to write GPL-licensed drivers for all their printers. Joy! Unfortunately, I've run into one usability problem after another, all of which are basically problems with the usability of CUPS. I know I'm not the only person in the world who thinks CUPS is a pain, because I've seen other people criticize it for problems that are the same ones I'm experiencing. For instance, CUPS remembers too much of its state, and when it freaks out (e.g., printer spewing page after page of garbage), it's difficult to get CUPS back into a known-good state.

Re:

[QuantumG]

Perhaps you could fund an effort to get the patent invalidated.. or to buy a blanket license for linux.. or to get the patent owner to publically waive their right to sue anyone who distributes the codec or something..

As for CUPS, maybe you could narrow down the exact problem and submit a bug report.. or put together an effort to fund someone to work on it.

It's community software.

Re:

[timmarhy]

who wants to fuck around doing all that just to play some mp3's or print a letter? thats why windows wins.

Re:patents, usability

[cduffy]

who wants to fuck around doing all that just to play some mp3's or print a letter?

Someone does. If I'm bored, I do. And only one person has to fuck around, come up with a fix, submit it upstream and get it merged for everyone else to have their problem solved.

It's very liberating to be able to fix your own problems instead of being at the mercy of a vendor who doesn't care.

Re:patents, usability

[enos]

The thing is, things don't stay fixed. The same old problems constantly get revisited when someone looks at something semi-widely accepted and decides the code is too ugly and makes a rewrite that doesn't add anything from the user's point of view but forces them to relearn another system.

It's one thing to go through several days of Googling and HOWTOs when setting up a new OS for the first time. It's another to do that every single time there's a new version out because they decided to change/rewrite so many things for just some trivial improvements.

Re:

[Hal_Porter]

The thing is, things don't stay fixed. The same old problems constantly get revisited when someone looks at something semi-widely accepted and decides the code is too ugly and makes a rewrite that doesn't add anything from the user's point of view but forces them to relearn another system.

Hmm, that reminds me of a sad story.

http://www.smcc.demon.nl/webcam/ [demon.nl]

Re:

[Hal_Porter]

Did you read the original comment

The same old problems constantly get revisited when someone looks at something semi-widely accepted and decides the code is too ugly and makes a rewrite that doesn't add anything from the user's point of view but forces them to relearn another system.

Nemosoft wrote a GPL driver which called out to a binary decompressor module. All was OK for a couple of years. Then Greg decided to rip out the callback. So suddenly the camera would only work in 160*120. Nemosoft then asked fo

Re:

[DavidNWelton]

"It's one thing to go through several days of Googling and HOWTOs when setting up a new OS for the first time."

That was true a while ago, but Linux has made enormous progress since then. Ubuntu is super easy to set up, as easy as windows is most of the time. My wife wrote her doctoral thesis with virtually no assistance from me on her laptop, which runs Ubuntu.

Re:patents, usability

[grcumb]

it's rediculous to suggest such a practise will be accepted by the masses, thats my point.

Your point draws exactly the opposite conclusion to that of the GP, who says:

And only one person has to fuck around, come up with a fix, submit it upstream and get it merged for everyone else to have their problem solved.

[emphasis mine]

FOSS has worked this way from day 1. And it continues not only to work, but to prove itself superior in many ways to proprietary software approaches. It particularly excels at dealing with software quality. In FOSS, code quality is one of the core metrics[*] of the value of a project, whereas security, debugging and testing are dealt with as externalities (i.e. cost centres to be minimised) by many proprietary software makers. Drivers are a perfect place to make significant investments in FOSS, because then hardware vendors won't be stuck owning the entire problem, and innovative uses of their products will allow them to sell into niches that they never could have afforded before.

----
[*] This is not to say that all FOSS software is quality software. Just like everything else in the world, 95% of it is crap. But the best FOSS software is very high quality indeed in terms of stability, resource usage and suitability to the task.
----

Is Linux ready for the desktop? In managed environments, the answer is an emphatic yes. Ease of administration is many times greater under Linux than under the other offerings, and this means that in-house support and developers can focus on making things better rather than simply fighting fires. A number of organisations have discovered this, and more will do so in the months and years to come. I think time will show that document formats are not nearly the bugbear that people currently think them to be.

Is Linux ready for the desktop at home? It's ready in potentio, but it will take time for vendors to work out how to package it on new machines. This will be a tough slog, not for technical reasons, but because Microsoft will do its very best to ensure that they have every incentive not to move from a Windows-only sales model. Having open source drivers provides one more bit of leverage against this inertia.

Re:patents, usability

[the_womble]

You could say the same for Windows:

Windows is fine in an office environment, with IT there to fix things.

However:

1) Making Windows secure requires work and knowledge
2) When things go wrong they can be hard to fix, and even when fixed they have a tendency to mysteriously go wrong again.
3) Software is hard to find, install and update. There are no repositories of software that is safe (not malware), will install with a click, and will all be auto updated.

In short: there is no OS that is really suitable for the home, and there are at least some ways in which Linux is better than the competition.

Re:patents, usability

[AJWM]

until basic basic shit like this works without a problem, pushing the linux desktop is a wasted effort.

This stuff nearly all works without a problem on my Linux desktop -- Suse, which is quite a nice distro regardless of what you think of their parent Novell. I've certainly had a hell of a lot fewer problems with plugging and playing stuff into a Linux box than with Windows (got driver disks for that? that support your particular version? and don't require you to download something from Microsoft's web site that ends up requiring you to register for Windows Genuine Advantage?). By that measure, Windows isn't ready for the desktop.

Re:

[timmarhy]

"stuff nearly all works"

and what happens with the stuff that doesn't? can you request your money back or ask for technical support because your trying to use linux? checkmate, i win. the distro that can focus 100% on getting everyday useage right will win IMHO

Re:patents, usability

[Russellkhan]

Yeah right. You can request your money back and ask for technical support all you want, but when exactly was the last time you actually received your money back or useful technical support from Microsoft?
 
...And do you have some delusion that Microsoft is focused 100% on getting everyday usage right? Then why is it that nearly every home user's Windows box that I see is constantly alerting about the security updates they haven't downloaded? And why do these users complain about the problems their computers give them instead of rejoicing with the ease and perfection of having a 100% everyday usage oriented OS?

As a simple example, a couple weeks ago, when I was visiting a realtor, she tried for 5 or 10 minutes to get her computer to bring up some MLS site, and was about to give up before I had to step in and get her connected to the office's wireless network. If Windows is such an ideal, usable operating system, why would the user have such a problem? Should she have called Microsoft, would they have walked her through getting the system on the network?

Re:

[freeweed]

checkmate, i win

It's a good thing this isn't a popularity contest, or a football game, or a presidential election.

If you have problems with Linux, so be it. Those of that don't, and/or are willing to work around otherwise minor issues, are all happily running it. And usually experiencing fewer problems than we were before we switched.

Why you seem hell-bent on insisting that we're all "losing" is beyond me. Like I said, this ins't a popularity contest. Use what works for YOU. The rest of us are quite alright

Re:

[mhall119]

Name me one OS where everything works, then you can say you win.

Re:

[Bent Mind]

who wants to fuck around doing all that just to play some mp3's or print a letter? thats why windows wins.

Someone does. If I'm bored, I do. And only one person has to fuck around, come up with a fix, submit it upstream and get it merged for everyone else to have their problem solved.

it's rediculous to suggest such a practise will be accepted by the masses, thats my point. until basic basic shit like this works without a problem, pushing the linux desktop is a wasted effort.

Such a practice is already accepted by the masses. I've often come across codecs that don't ship with Windows. Several of them use patented technologies. With Windows you can download codecs from sites that sell them, offer them free with advertising, or are hosted in countries that do not recognize software patents. With Linux, it's exactly the same. To purchase codecs for Linux, look at Fluendo's site [fluendo.com]. Several Linux distributions are based in countries that do not recognize software patents. These di

Re:patents, usability

[QuantumG]

I wasn't able to get a printer working under Windows XP 64 bit Edition for about 9 months. Only last week did I figure out how to force windows not to try to use the drivers from the print server (which is 32 bit XP) and to stop it from overwriting the 64 bit drivers with the 32 bit ones. I guarentee it is because the date-time stamps on the drivers were out of wack.. in 3 weeks time they'll probably magically stop working again.

Re:

[ClosedSource]

There's a difference between having a discussion and having a consensus conclusion. If you were really so sure that FOSS was destiny, you wouldn't have bothered to post at all. The fact is that you're still trying to convince people in 2007, you're just using a very poor argument to do it.

Re:

[QuantumG]

Oh, we're still trying to convince people.. unfortunately we have explain why "anyone" would be interested in working on Open Source to these people. They make the same arguments from 10+ years ago and when we say "go read X" they refuse. Or, alternatively, we say "well obviously they are, otherwise we wouldn't have all the Open Source software we have now" and they start denying that we have any software.. or just ignore us.

Re:

[ClosedSource]

Thanks for the straw arguments.

Re:patents, usability

[arodland]

Maybe I'm a bit simple but I don't see where CUPS even has "usability" to complain about. You install it (if, oddly, it isn't already), you tell it what and where your printer are (preferably using the KDE print config thing because it's amazingly simple, but the CUPS web jigger isn't bad either), and then from then on you print, and you forget that CUPS exists. Where's the hangup?

Re:

[QuantumG]

I think he's refering to gnome-cups-manager.. which is pretty weak in some parts at the moment. I believe there are some improvements in the GNOME CVS tree that have yet to filter down.

Re:

[Anonymous Coward]

CUPS was written long before USB was a twinkle in Intel's eye and behaves like it. I had a Linux box that I used as a print server via CUPS. One printer was LPT, one was USB. The LPT, man, I could turn it off, turn it on, unplug it, send it into the next timezone and bring it back, no problems. If my server went to sleep, still no problems. The USB printer? I had to write a fucking hotplug script for it that removed it from CUPS every time it was disconnected (either unplugged or turned off) and then

Re:

[jrumney]

Why did you write a script to remove it? The print jobs will be queued until it is next available.

Re:

[bcrowell]

Maybe I'm a bit simple but I don't see where CUPS even has "usability" to complain about.[...] Where's the hangup?
Problems I've had:

1. the problem I described in my original post
2. Upgrading ubuntu to a new version made cups stop working.
3. The web interface says administrative functions are disabled. Because of that, I tried editing the config file to accomplish what I wanted. I also downloaded drivers directly from Brother, because that was what people on the ubuntu wiki suggested, but that actually didn'

Re:

[arodland]

Was your USB printer plugged-in and powered-up when you started your system? No? Oh well. No printing for you, then. Reboot to print, or work out your own hack for CUPS, or manually get in there fix it every time you run into this stupid problem.

No such problem here, ever.

Re:

[jrumney]

I only ever turn my printer on when I want to print. Never had a problem. Try manually loading the usb printer module (modprobe printer). If it works after that, investigate to see why it isn't being loaded automatically when your printer is plugged in or switched on. It certainly isn't CUPS that has a problem with printers coming and going.

Re:

[ajs318]

You shouldn't need to reboot just to make it detect the printer -- that's Windows thinking! Linux allows you to stop and restart misbehaving subsystems on an individual basis. Just

# /etc/init.d/cups restart

which will stop and restart CUPS, thereby forcing it to reread its configuration files and check for connected devices. (On Debian, Ubuntu and derivatives, it's cupsys not cups). If that doesn't work, try

# modprobe printer

and restart CUPS again. If that makes it work, then

# [ -z `egrep ^[[

Time for an insightful car analogy

[Dan Ost]

Car analogy:

The last time I used a car, I had to hand crank it myself to start it. It was completely unacceptable!

a few replies later...

From the replies, I'm glad to hear this problem fixed, but approx. 70 years ago, it was a well-known limitation of cars.

Seriously, you make broad criticisms and then admit you really don't know the current state of things? How fair is that?

Next time, be honest about the last time you used the system, state your concerns about how the system behaved then, and then ask if you

Re:

[the_womble]

Usability: I used to use Mandrake at home and Windows at work. Every time I had a problem with Mandrake and though "maybe Windows would be better" I would soon have a worse problem with Windows.

Codecs: There are instructions on how to install the extra codecs. Its the first thing most people do after installing Ubuntu: its very much a one off.

CUPS: I agree, but GUI tools for managing it are getting better.

Re:

[repvik]

Codecs: There are instructions on how to install the extra codecs. Its the first thing most people do after installing Ubuntu: its very much a one off.

Really? How do you know that "most people" do this? And which instructions are you talking about?

Mad Propz to van Doorn, Wu and Realtek

[smittyoneeach]

This post brought to you by these two patches, against 2.6.22-rc2:
http://thread.gmane.org/gmane.linux.kernel.wireles s.general/2368 [gmane.org]
http://thread.gmane.org/gmane.linux.kernel.wireles s.general/2369 [gmane.org]
The little WG11v2 [netgear.com] is a happy interface. Figure I'll need to stockpile a couple them critters.
Now, how is it that I'm off the hook for managing any of that bad, bad firmware with this wee beastie?
Ivo or Michael, though I'm nowhere near as cool as you dudes, I'll buy you a beverage if I see you in Ottawa next month.
Dunno if GKH's driver program actually helped in this matter, but the general trend in hardware is positive, and I feel Realtek and Netgear deserve a free shill.
Best,
Chris

Fishing for Drivers

[Doc Ruby]

How about a driver for this ATI All-In-Wonder 3D Rage II +DVD PCI card I can't find drivers for?

Re:

[RobertLTux]

doc you may want pipe the output of lspci to a file and see what chips it has (your driver may be included in X.org 6.9/7.0) for example
01:00.0 VGA compatible controller: ATI Technologies Inc Radeon RV200 QW [Radeon 7500]
is my card and i think that there is a couple rage drivers in the gatos project

Linux hardware selection rules, from digg

[Werrismys]

Apparently people don't read the rules of Linux hardware:
1. No ATI
2. Check the hardware compatibility list
3. "Partially supported" means "barely functional if you can get it to install."
4. No ATI
5. No, really, don't use ATI.
6. Really, I don't care if you're a fanboy, don't use ATI!!!

Discontinued projects

[phorm]

Yeah. The GATOS [sourceforge.net] project would have worked with this, however it seems to be discontinued for modern kernels. Even if it weren't, the documentation seems pretty horrid, so I couldn't tell either way.

vaporware is hype

[phrostie]

this is productive

Criticism from the peanut gallery

[tyler_larson]

FTA:

While one developer of a competing open source operating system has criticized the NDA approach...

While no mention was made of the identity of the criticizing developer, 10:1 says that the "competing" operating system has the letters *BS* right there in the name.

Re:

[robbak]

It is only linux that considers NDAs acceptable for open source. They say they espouse Openess of information, but actively seek NDAs that prevent it. That makes no sense.
A NDA would be acceptable if it allows you to release fully commented code, i.e. sufficient for anyone to rewrite the driver for any OS or in any language.
All products should have their specs as html on public servers.

BURN all NDAs. Now.

(I wonder who has burnallndas.com? No-one? Wow. Makes me want to take a trip to my nearest registrar!)

Re:Criticism from the peanut gallery

[QuantumG]

The purpose of the OEM getting the driver developer to sign an NDA is so they can just give the developer all their documentation, including stuff they would prefer their competitors never see. If they don't do this, they have to get someone to sit there and go through everything that is being released and censor it. You can't just expect a company to hand over their product secrets so you can write a driver without any assurance that you're not going to immediately sell their documentation to their competitors.

A[n] NDA would be acceptable if it allows you to release fully commented code, i.e. sufficient for anyone to rewrite the driver for any OS or in any language.

And that's exactly what the Software Freedom Law Centre will be requiring from OEMs.. poorly commented code doesn't live long in the kernel tree.

Re:

[QuantumG]

If only there was a casino that would cover this kind of action.. you could make a lot of money betting on things that happened 5 months ago.

Can't copy GPL code?

[Skapare]

From TFA:

While one developer of a competing open source operating system has criticized the NDA approach, developers are free under the GPL to use the Linux driver as the documentation for a new one as long as they don't copy the actual code. "The drivers are generally better written than the specs," Kroah-Hartman says.

What? If the driver code is GPL, why can't I copy it?

I suspect he means "copy" as in "make a derived work" that would have chunks of code taken from the original. But still, this is what GPL is about ... being able to take an existing source and make a derived work from it (that presumably would be better), and redistribute that derived work also under GPL (so someone else can derive from that later on ... and on ... and on).

Re:

[Wonko the Sane]

What? If the driver code is GPL, why can't I copy it?

Presumably, the people he is talking about want to release code in under a GPL-incompatible license. So they can't just copy the code.

Re:Can't copy GPL code?

[QuantumG]

Yes, you are absolutely right. This statement:

While one developer of a competing open source operating system has criticized the NDA approach, developers are free under the GPL to use the Linux driver as the documentation for a new one as long as they don't copy the actual code.

in attempt to be diplomatic has just added confusion. Allow me to clarify:

      one developer = Theo de Raadt.
      competing open source operating system = OpenBSD
      criticized = profanity

So to rewrite the sentence so it actually make sense:

While Theo de Raadt, has slung profanities at the NDA approach, he is free to write a driver for OpenBSD if he wants by using the Linux driver as documentation, but he best not copy any of the code from the Linux driver if he wants to avoid having to GPL it (which he almost certainly does).

Which makes this comment:

"The drivers are generally better written than the specs," Kroah-Hartman says.

make a lot more sense. But what the hell, I'll translate that too:

Theo, stop moaning about specs.. these companies are not going to give us blueprints to the damn hardware.. and whatever they do give us is going to be confidential. That's the reality. Deal with it. If you refuse to enter into a non-disclosure agreement with these companies then don't complain when the only documentation you have is a Linux kernel driver. The specs aint that great anyway.

Or, at least, that's what I read.

Re:

[Skapare]

Thanks for updating the facts on this. I didn't know Theo was the party involved. Now that makes sense. And of course he doesn't want OpenBSD to end up being GPL. Microsoft might have the same issue if they didn't already have their own means to get the technical details :-)

Re:Can't copy GPL code?

[Darby]

didn't know Theo was the party involved.

Heck, the OP might not "know" that (maybe he does) but I barely even noticed that I read the summary more or less how he translated it ;-)

That's very relevant to why, while I use Linux for my web, file and database servers, when it comes to my firewalls it's OpenBSD. *Every* *fucking* *time*.

Yes, Theo can be abrasive. Yes, he's an absolutist on a lot of topics. Absolutely yes, that's the type of person I'll trust for the security of my network and my business. I'll install a binary blob driver for my desktop so I can run games, but security is not a game.

Given that the primary focus of his distribution is security, he's 100% absolutely *right* to refuse to allow binaries which he and his team can not audit to the extent that they do every other part of their releases.

So, they might be behind on support for some hardware, but when it's done, your confidence in its security is rightfully higher.

So, it's not just that he doesn't want it GPL. He doesn't trust people whose goal isn't security to write his code for him. He sure as shit isn't going to put his reputation and the security of the people who trust his OS in large part because of that reputation in the hands of some third party. So, maybe a lot of people think he has a reputation as a dick, but let's see them go up against him purely in a security context. They've had issues, but vulnerability for vulnerability he wins against damn near anything else. In the context of anything a normal person could get ahold of, I don't think anything else is even in the same league as OpenBSD. Not Linux, sure as hell not Windows.

Security and useability are in an inverse relationship. Some people are willing to adjust their balance on that scale and that's fine. A lot of good things can come out of that. It is absolutely a great thing that there is somebody out there who refuses to shift it away from security too.

Re:

[huiac]

So, from what you say:

You're fine with firmware that's soldered onto the board as a non-FRU mask-programmed ROM, but if it's loaded as a vendor-supplied blob that can (at least in principle) be updated as issues are identified, that's bad?

Interesting tradeoff...

John.

Re:

[Darby]

You're fine with firmware that's soldered onto the board as a non-FRU mask-programmed ROM, but if it's loaded as a vendor-supplied blob that can (at least in principle) be updated as issues are identified, that's bad?

Actually, I never said I was "fine" with anything.

Given the same piece of hardware, it's better to have drivers written by people who care about security than by those who don't. If issues are found, I want the people fixing it to be those whose incentive is securing my infrastructure as oppose

Re:

[jZnat]

Only the American companies who confuse being hardware manufacturers with being software developers force these NDAs which result in drivers which are basically decompiled blobs. Projects like OpenBSD (and Linux even) have had better experience with Asian companies for example on getting documentation regarding hardware. Blueprints to the hardware are not needed to write a driver; just the op codes and messages you can send to the hardware to control it. For example, CPU architectures are quite documente

Re:

[dbIII]

Now if GNU had actually written a kernel themselves rather than adopted Linux back in the early 90's

Hang on. Gnu has the hurd - and until the late 1990's RMS was joyfully saying "linux, never heard of it. ha ha" in repeated interviews. I suppose it was funny to some at the time and nobody really thought he had never heard of it paticularly since it kept coming up in interviews. The LiGnuX renaming suggestion afterwards that later turned into the gnu/linux renaming suggestion could never be taken as a jo

Re:

[QuantumG]

Shame about the HURD.. that's what happens when you try to do something experimental when you're playing catch up. They should have just focused on a monolithic kernel from day one.

Re:

[dbIII]

I think it came down to not many people working on it. It still exists but is not the popular choice for development. Linux always seemed to be a more inclusive project but I can't really quantify why I got that impression. Would Donald Becker have been welcomed with open arms by hurd or would they have been upset that he was writing ethernet drivers for commercial reasons? It was drivers like that that ultimately made linux very useful.

Re:

[QuantumG]

Linux always seemed to be a more inclusive project but I can't really quantify why I got that impression.

Linus, that's why.

He has great humility.

Re:

[pyrrhonist]

They should have just focused on a monolithic kernel from day one.

But then they couldn't have named it Hurd!

Re:

[Kjella]

I guess it boils down to "How complete can a driver be documented and still comply with an NDA?". I mean, if the code is full of setting magic memory addresses to magic values then it's not really open source because you can't make a clue of it without NDA'd docs. On the other hand, if the bit registers and opcodes are all clearly laid out and documented in the source, isn't that exactly what's covered by the NDA? Software developers don't need any circuit layouts in the first place, they just need the inte

Re:Can't copy GPL code?

[QuantumG]

The point of the NDA is so that the OEM can just hand over all their documentation without having to sit down and censor it all so they're not handing over anything that might give their competitors an advantage. They clear the developer to disclose as much information as needed to make a device driver that is well documented and works. They don't clear the developer to turn over the secret algorithms that might be used in the hardware or the production methods that may be of interest to the OEM's competitors.. but they might give that stuff to the developer because it was in the same filing cabinet.

If the driver isn't well written, commented and documented, it will not be accepted into the tree. The NDAs are being drafted by the Software Freedom Law Centre.. you don't think they're going to get the best possible deal?

Otherwise, what would you prefer? Would you prefer the OEM hired a developer to make a binary-only driver? Which they'll stop supporting as soon as it is economically justifiable? Would you prefer they just don't release any drivers for Linux? Don't say you would prefer if they just sat down and wrote perfect developer documentation cause there's no such thing.

Software developers don't need any circuit layouts in the first place, they just need the interface.

Sometimes you do.. sometimes "the interface" just isn't defined and you need to sit there with an osciloscope to figure out what the hell this piece of hardware is doing.. and knowing what line is what kinda helps.

Re:

[evilviper]

While Theo de Raadt, has slung profanities at the NDA approach, he is free to write a driver for OpenBSD if he wants by using the Linux driver as documentation, but he best not copy any of the code from the Linux driver if he wants to avoid having to GPL it (which he almost certainly does).

That's really a rather baseless attempt to confuse the issue. Theo deRaadt has attempted to use GPL'd code as hardware documentation many times, and he knows full well the limitations of it.

There's ample, in-depth explan

Re:

[QuantumG]

There's three ways to get an open source driver written:

1. Get the company to release docs so that independant programmers can write drivers. This is the approach Theo advocates.
2. Reverse engineer the hardware and write the driver independantly. This is the approach taken by a number of people, including the OpenBSD developers at times.
3. Get a company employee or contractor, under NDA, to write the driver. Often the result of this is not an open source driver.. it's a binary driver, and that's what Gre

Driver Groups

[NaNO2x]

I wonder if support could get out to little groups who are trying hard. I personally have a webcam with no driver and the group trying to develop one just doesn't seem to be there enough. If someone is offering this support then it would be nice if he found a group like this and helped out. It would be nice to have a website that brings together all drivers that are being worked on and make them easy to find for someone who really wants to help. Here is the driver I was talking about by the way: http:// [actiongames.co.uk]

Re:

[QuantumG]

Is your webcam in this list [mxhaard.free.fr].. if not, maybe you could contact that guy.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Not NVIDIA

[elronxenu]

Unfortunately the openness doesn't extend to NVIDIA, who refuse to release specifications or other assistance to developers working on the ULi M920x chipset, which is used for receiving Digital Television.

NVIDIA bought ULi and then cancelled development of the M920x, but you can (still) buy DVB receivers which use this chipset.

Requests for assistance or interface specifications have been refused by NVIDIA.

Re:Why...

[QuantumG]

1. They're loadable modules.
2. You should maybe leave the kernel development to the kernel developers.

Re:

[QuietLagoon]

Maybe you should try to understand that some people ask questions in an attempt to learn, and not to troll.

Re:

[QuantumG]

Maybe you should try some humility instead of phrasing your question like you're a know-it-all.

Re:

[QuietLagoon]

The question mark at the end of the sentence indicates humility. I admit that I do not know, therefore I ask. Thus is the reason for the question mark.

Re:

[pembo13]

Then you need to learn how to ask questions better.

Re:

[jZnat]

I believe it's because you mentioned that they are "Two things that a reliable kernel should avoid?" That makes you sound like you know better, so that's the troll-like bit. Also, everyone on /. is assumed to know everything about everything when posting unless said otherwise...

Re:

[QuietLagoon]

I believe it's because you mentioned that they are "Two things that a reliable kernel should avoid?"

I think that ever increasing size and complexity are things that a reliable kernel should avoid. What is wrong with that?

Re:

[jZnat]

If you had said "I think" or "I thought", you wouldn't have sounded like a "knows-better-than-you" kind of person. Linguistics and all that.

Re:

[dbIII]

Nothing really - you've just missed that the drivers are the kernel - that is where they belong. An operating system is there to let the applications talk to the hardware after all. With linux it has been possible for a long time to have them in seperate files (a modular kernel) as distinct from one big file (a monolithic kernel), which is how a linux distribution can install on a wide range of hardware and how you can often upgrade stuff just by moving the hard drive into a newer box. Compiling in sup

Re:

[Knuckles]

What is wrong with that?

The believe that more driver modules make the kernel more complicated.

Re:Why...

[Darby]

Seriously, I don't know crap about kernel development, but:
1) I knew the answer to your question since the first time I even tried to compile a kernel. By "compile a kernel", I mean run make menuconfig, flip through idiot proof menus and say yes when it tells me to.
2) You proposed a bunch of dumb ideas implying that the people who actually do know how to develop one are idiots.
3) asking questions in a dick way and then appending a question mark in no way indicates humility, or even politeness.

Seriously, asking dumb questions is fine, but *you* need to actively treat them as dumb questions if you want them to be treated as legitimate questions in a problem space in which you're ignorant. Don't treat the people you want answers from as dumb preemptively.

Your sig

[WindBourne]

BTW, you may wish to go easy on john. On the Saturn, they lost a few of the engines on the way up. Fortunately, they did not blow, but plain and simple, they did fail. As you pointed, Spacex and Armadillo is looking to place a large number of identical engines along the line of parallel server. But they are not the first. In addition to USA, the Russian have been and still do. [wikipedia.org] As long as the engine is well developed, this makes sense. In fact, I think that both Amadillo And spacex are doing it right.

Re:

[QuantumG]

I think I did go easy on him :)

It's not really Carmack who is the one on the pipe, it's the people who are suggesting that you could make 100x100 module first stages to support 25x25 module second stages. But still, the reason why that is insanity (unobtainium would be needed to hold the modules together) is also a good reason why 5x5 or 8x8 configurations are a bit loopy.

But hey, I don't mind.. Armadillo is a fun effort to watch.

Re:

[QuantumG]

Why do people continually make the same argument for a binary interface?

It's been addressed again and again and again.

Here's the definitive response [kroah.com].

Give it up.

Modules

[Tharkban]

Modules. Pretty much all drivers are modules and not compiled directly into the kernel. They don't increase the kernel size unless you load them. Although they do increase the kernel source size (in their own files generally) so it is taking a little longer to compile all kernel modules, but that's a price I'm willing to pay for things just working.

Re:

[QuietLagoon]

OK, that makes sense. Thanks for the rational answer.

Outlandish statements

[The Monster]

Try researching before making outlandish statements

Welcome to Slashdot. If there weren't any outlandish statements, there'd only be a couple of stories a day on the front page.

Re:

[Breakfast Pants]

Yes and no. For one, they can be dynamically loaded as kernel modules. Or you can custom compile your kernel without any dynamic loading, to have just the drivers you need. As long as driver foo isn't touching many systems, it doesn't drive up complexity to have driver bar that touches the same number of systems.

Re:

[Chirs]

If you compile your own kernel you can choose to either leave out the functionality entirely, build it as a runtime-loadable module, or build it into the kernel.

So the only permanent size increase is in the kernel source code. Assuming that the driver is part of a class of similar devices, there is basically no complexity increase as the driver will bind into the standard API for that class of devices.

So generally there is very little downside to adding new drivers to the tree.

Re:

[Wonko the Sane]

Translation from marketing-speak to English: "Greg K-H was lying, knew that he was lying, doesn't feel bad about it, and will do it again".

wrong, wrong, right, right
You're only half wrong [slashdot.org].

Re:Translation

[QuantumG]

Uhhh.. no he wasn't. He wasn't lying at all.

Why would you feel the need to post a "translation" when you have no idea what you are talking about?

The fact that people are willing to write Open Source software without charging a fee for their services is hardly a new concept, but Greg did the smart thing of treating it like it is and, in doing so, attracted the attention of people who thought that it wasn't the case.

This was one of the biggest problems with the Free Software movement before Open Source came along.. no-one talked about the benefits that businesses could get from the community. For a while, no-one talked about anything else, and then it went quiet again. RMS will tell you that we need to talk about freedom. I happen to agree, but we also need to talk about the practical advantages of open software development too.

Re:

[Rycross]

How exactly was he lying? They said that they would code drivers for free for companies that released their specs, and they did. The "marketing hype" was that they were making a big deal and advertising for something that they already did, and would have done anyway