Slashdot Mirror
Greg Kroah-Hartman Gripes About Microsoft's Linux Contribution; MS Renews Effort
dp619 writes "Microsoft's developers were missing in action after the company donated GPL-licensed drivers to the Linux kernel community in July, leaving significant work to the Linux community, according to Linux driver project lead and Novell fellow Greg Kroah-Hartman. The company rekindled its involvement after Kroah-Hartman published a status report this week. Kroah-Hartman said that other companies were also laggards in Linux development, and that Microsoft's lack of involvement was nothing out of the ordinary."
From the blog,
...hmm - hyper-v in Linux?
"hv (Microsoft Hyper-V) drivers. Over 200 patches make up the massive cleanup effort needed to just get this code into a semi-sane kernel coding style (someone owes me a bit bottle of rum for that work!) Unfortunately the Microsoft developers seem to have disappeared, and no one is answering my emails. If they do not show back up to claim this driver soon, it will be removed in the 2.6.33 release. So sad..."
In other words, there is some coding to do. Did the Kernel devs coordinate with the managers at MS to ensure resources would be available to work on these patches? (200 patches is not a lot in my opinion. I have a minor patch coming out on the 21st for my in-house system with 2000+ users and it has over 300 fixes.)
I wonder if there was a minor miscommunication...
Cool!
The Kai's Semi-Updated Website Thingy
I don't get it. Microsoft donates code to the community, licensed under the GPL. Anyone is free to do whatever they like with this code (well, within the constraints of the license), and people are actually bitching about this? I've released quite a few pieces of software under various OSI licenses, and I don't actively maintain the code. If somebody else wants to, fine. Would everyone be happier if Microsoft weren't releasing anything at all?
I use Debian for almost all my server needs, and I'm a big fan of Ubuntu on the desktop. That said, I'm certainly not going to bitch at someone or some company because they aren't slaving away maintaining code I got free of charge.
512 MB RAM, 20 GB disk, 200 GB transfer, five datacenters. $19.95/month.
'scuse me?
They weren't contributing out of the goodness of their hearts?
They were legally obligated to, and since then have avoided doing much of anything further.
It wasn't over zealousness. If they hadn't been reminded of their obligation they wouldn't have contributed at all.
Hmm.. did they forget to thank? It's quite common to cry about this and that... when the only think needed is to shut your mouth and THANK for the little help.
If you're here to build me a bridge, then tell me so and build it. If you're here to bring me a stone, then don't tell me you're here to build me a bridge.
I am the lawn!
Quote from the article:
GKH also said:
If the driver was in such poor shape and didn't follow the kernel coding guidelines, then how did it make it into the kernel in the first place? I was under the impression that the kernel maintainers had pretty strict guidelines about coding style and what is and isn't acceptable.
Cory Doctorow talking about cloud computing makes as much sense as George W Bush talking about electrical engineering.
Two thirds of the summary are lifted directly from the sdtimes link...
Thank them for what? MS's contributed drivers are useless to anyone who isn't running MS's own hypervisor and Linux underneath (i.e., MS's customers). They didn't donate this code out of any altruism, only pure self-interest.
So don't act like they're doing the Linux community any favors, because they're not.
Sounds to me like MS dumped this pile of crap code on the kernel maintainers, and then faded away instead of sticking around to deal with the inevitable problems that result when trying to merge code into the mainline (i.e. fixing all the problems that the maintainers point out).
This is for the drivers/staging tree, which is specifically set aside for drivers that don't meet normal code standards but where the intent is to bring them up to par for merging into the "real" tree.
It only got accepted after the cleanup, done by GKH.
New things are always on the horizon
yup, that is why I would dump this code microsoft donated in to the trash = not even include it in the kernel, and just leave it up to microsoft to to offer it as a third party patch.
Politics is Treachery, Religion is Brainwashing
No, they were violating the GPL.
They had to at least give source to their customers.
Rather than to continue to do that they made this driver the kernel maintainers problem. If they don't want to help maintain it, I say drop it from the kernel.
Greg K-H's public comments about the code being abandoned had the desired effect: people at Microsoft got in touch with him again. The same thing happened with code contributed by several other companies:
This is Microsoft (and other companies) learning how to deal with kernel development. Greg K-H has a good attitude about it, so let's not build a mountain out of this. Perhaps Microsoft will do better next time.
Not everything and everyone associated with Microsoft deserves abuse and scorn. Save it for their next DRM initiative or something.
steveha
lf(1): it's like ls(1) but sorts filenames by extension, tersely
Paying Hyper-V users will probably be running SLES or RHEL, and Microsoft provides support for SLES and RHEL in Hyper-V.
And neither distribution will distributing a kernel new enough to have these drivers in the mainline source for a while.
Yeah, I'd want to do that too. But if the kernel maintainers do that, it would look really bad and arbitrary (no one ever remembers MS's long, long history of unethical behavior), so to be fair, they have to accept the code.
However, that doesn't mean they have to do MS's work for them. I've submitted kernel patches before (pretty minor ones), and part of the process is making your patch meet the standards of the maintainers: the coding style has to match the rest of the kernel (no Hungarian notation crap, which MS is a big fan of), and they usually find all kinds of nitpicky things they want fixed to meet their standards (which is a good thing; this is a place where perfectionism is useful). So it usually takes a few back-and-forth iterations before the patch is accepted and merged.
If I, Joe User, were to submit a patch and then disappear, it would NOT make it into the kernel. They'd write back with their complaints, wait for me to resubmit with the fixes, and then forget about it when I never resubmit.
MS shouldn't be treated any differently. They're not special, and a patchset of this size represents a lot of work to merge into the kernel. If MS wants it merged for the benefit of their customers, it's their responsibility to make the required changes, not expect it to be done for them.
If it has happened, please oh please post the link!
It's NOT me! It's the meds! I'm on 1000mg of Fukitol.
If they don't want to help maintain it, I say drop it from the kernel.
Indeed. Just because they are obligated to release it, that doesn't mean the Linux kernel is obligated to use it.
Well, for one thing, most corporations that contribute code do so so that Linux will work better with their hardware. That's why a lot of the code comes from companies like Intel. As a Linux user with an Intel CPU, that makes me want to buy more Intel CPUs in the future.
MS isn't a hardware company, it's a software company, and it competes directly with Linux.
Anyway, other than this, it really isn't different, but several posters here are acting like it's some kind of useful contribution to Linux. It's not. It's only useful for MS customers running Hyper-V who want to run Linux on top of that.
Just like it's incumbent on other kernel contributors to jump through the hoops necessary to make their code meet the maintainers' standards to be merged into the mainline kernel, MS has to do the same. They can't just dump the code on them and expect them to do all the work. Other companies already do it this way, because they want the code merged into the mainline, for the benefit of their customers. If MS doesn't do this, it's their customers who will suffer, not the Linux community at large.
It's not like they're contributing something that's generally useful to most Linux users, like a codec or a font, or a filesystem. So stop acting like it is.
Ok, honestly, when VMWare released GPLed drivers LOOOOOOOOONG ago, how many blogs did we have talking about how VMWare is contributing code to Linux, and how many people needed to state that those were simply self-serving drivers for VMWare? Nooone... These are self-serving, there is nothing fundamentally wrong with it, they wanted their virtualization stuff to be able to run GNU/Linux, and they released drivers. It is not an 'evil' move but simply a logical one. But for sure it is not a 'contribution' to Linux. It is nothing out of the ordinary and honestly, I am tired of having to read countless of stories about...
Copyright infringement is "piracy" in the same way DRM is "consumer rape"
Don't be stupid.
When Intel contributes a patch, they go through the required process necessary to make the patch meet the maintainers' standards. I actually did this a couple times when I worked at Intel.
If MS isn't going to do the work necessary to make their patches meet the standards, then it shouldn't be merged. I'm actually a little disappointed that they merged it in at all before going through this process fully.
All your posts have been so far have all been complaining about how everyone around here thinks Microsoft is evil, and how the group mindthink is fucking you over. No wonder you got modded down.
Get over yourself... nobody here cares what you think. This isn't the propaganda arm of MS so if people say things that you see as "anti-Microsoft" it is probably because is is true. Certainly a site that is a "Source for technology related news with a heavy slant towards Linux and Open Source issues." clearly identifies itself and needs to be read with a bit of scepticism-even if a significant number of those you see as "anti-Microsoft" are right most of the time. I don't see BestBuy, Staples, PC World, or any other "MS Windows" propaganda arm identifying itself so blatantly as Microsoft for the good of the people.
Thank them for what? MS's contributed drivers are useless to anyone who isn't running MS's own hypervisor and Linux underneath (i.e., MS's customers). They didn't donate this code out of any altruism, only pure self-interest.
Yeah, and they only decide to "donate" this code after it was pointed out to them that keeping the code private was a violation of the GPL, since it's clearly a derivative work of the Linux kernel.
So what do they do? Instead of GPL'ing it and working to maintain and clean up the code themselves, they just dump it on the kernel maintainers. Lame.
In my mind, it shows that Microsoft still doesn't take Linux seriously on some level. They don't bother to build a useful working relationship with the kernel devs because they see this as a one-off interaction just to "get Linux working with Windows."
Contrast this with, say, Intel or AMD or Realtek or IBM or pretty much any hardware company. Of course, contributing to the Linux kernel is a matter of "pure self-interest" for those companies too: they want to make their Linux-using customers to buy and happily use their hardware. But those companies learn to work with the mainstream kernel development process, because they see a long-term interest in a good relationship with the community of Linux developers and users.
My bicyles
Kind of funny how today there was that other announcement here about Microsoft sponsoring CodePlex foundation to help bridge the gap between commercial developers and open source projects, which kind of seems to be what they should have done a bit more here right.
Hmm.. did they forget to thank? It's quite common to cry about this and that... when the only think needed is to shut your mouth and THANK for the little help.
If you're here to build me a bridge, then tell me so and build it. If you're here to bring me a stone, then don't tell me you're here to build me a bridge.
Oh, cry me a river...
coding is life
By default, _really_ heavily downmodded posts are not viewable to users who aren't logged in. If you log in, you can set your view preferences to allow them to be seen.
Somebody call the wambulance please.
That's weird - I have mod points, and your comment is both overrated (5-insightful) and off-topic (as you pointed out).
Damn! I've posted, now I can't mod it. Oh well - I'll go mod some of your posts in another thread...
This one is underrated and hilarious. I'll start there.
You think this constitutes "publicly flame"ing Microsoft? He's just asking them to step it up and contribute. He's much harder on others in that list. It also doesn't seem like he went out of his way to be interviewed. It sounds like he just responded to a few questions that a reporter put to him. "Unfortunately" and "so sad" do not, of themselves, constitute a flame.
Here are a few other choice passages: (these may be interpreted as weak flames)
heci A wonderful example of a company throwing code over the wall, watching it get rejected, and then running away as fast as possible, all the while yelling over their shoulder, "it's required on all new systems, you will love it!" We don't, it sucks, either fix it up, or I am removing it.
me4000 and meilhaus They work on the same hardware, and they duplicate the existing COMEDI drivers. Someone thinks that custom userspace interfaces are fun and required. Turns out that being special and unique is not what to do here, use the COMEDI drivers instead. These will be removed. Heck, I'll go remove them for .32, there is no reason these should still be around, except to watch the RT guys squirm as they try to figure out the byzantine locking and build logic here (which certainly does count for something, cheap entertainment is always good.)
rspiusb A weird, very expensive camera, from a company that does not want to release the specs, and wants custom userspace interfaces. The code hasn't built since the 2.6.20 days. I'll go delete it now from .32, it doesn't deserve to live as no one cares about it, least of all, the original authors of the code :(
In other words: "Though it seems that he has the generosity to not publicly flame them unlike Microsoft." is pure hogwash... on both counts.
I won't join Slashcott. OTOH, If Beta goes live, I just won't be back until it's fixed. Sorry Dice.
"If you're here to build me a bridge, then tell me so and build it. If you're here to bring me a stone, then don't tell me you're here to build me a bridge."
Except they're here to teach you to fish, not to become your unpaid fishermen/code slaves for the rest of eternity.
I actually expect that this was an intern project for one or more interns (intern season just ended), and that there was never any expectation that the people who did the work would be around to maintain it. It was in fact a very large move for Microsoft to release this code at all, and it probably took a lot of begging internally to get management to agree to that much concession. Then they went above and beyond, and released the code patches under the GPL, which meets their requirements under the GPL to donate the code back to the Linux project. You seem to expect them to stick around and maintain that code in perpetuity, which wasn't the contract, and wasn't the license. So you're complaining and threatening to remove the drivers in the next release unless they commit resources in perpetuity to maintaing the drivers vs. *your* code base.
This is really a stupid demand on your part;if the kernel level APIs (what Sun calls their DDI/DKI - Device Driver Interface/Device Kernel Interface) in Linux were stable and not such a moving target, you could just forget the drivers and they'd keep working indefinitely.
As one of the people who struggles internally in a company to keep certain sources out there and available for anyone who wants them, IMO it's really ungrateful of you to look the gift horse in the mouth and demand more just because you can't maintain stable APIs.
-- Terry
People don't have to be under mind control in order to think that Microsoft is evil. They might have enough work experience to know that Microsoft, Adobe, Autodesk and Corel don't care about end customer and care only about own profits and suppressing opposition. Such corporate behavior fits definition of evil in human world.
No offense but you just sound like a rabid anti-Microsoft person who would hate anything they do simply because they are the ones doing it. The reality is that all corporations have self interest at heart when contributing to the Linux community. IBM for one mainly sells software services for Linux/Unix even though they also sell hardware. Oracle sells purely software but have always supported Linux/Unix.
The reality is that I went with XenServer instead of Hyper-V because Linux wasn't a supported guest OS. Since MS has made this change their product is much more competitive with the other players who support multiple operating systems. Microsoft's customers won't suffer because MS doesn't offer drivers for a VM, they will simply migrate to another platform which happens to free.
As far as most Linux users I think you are way the hell off base in that every shop I have ever encountered runs some Windows in their environment and those people will find that virtualizing their infrastructure saves them a ton of money especially since they already bought their Windows license. Get off your anti-MS high horse and come back to reality friend, it's not as bad as you make it out to be. MS is finally starting to embrace the Linux community and they face criticism the whole way, it's a stupid way of doing business and an even less intelligent method of getting what you want which is interoperability between all the major platforms. If only Apple would play so well with others... given the state of networked support in OS X I think your rabid hatred is best reserved for them although I can understand a strong distaste of Windows too.
For those of us out there that run just about everything, we'll continue on and enjoy the small improvements each platform makes and encourage continued improvement from all sides.
So I guess the people who run Linux on MS servers aren't part of the Linux community?
I guess the poor guys that have to do this sort of thing because some manager says so, aren't cool enough to be part of your retarded little club?
Get a grip, god Linux fanboys are a pile of whiney bitches. Add another thing to the list of reasons Linux will always be the third world country of software world, no one likes a bunch of whiney bitches.
Persistent Volume manager for Kubernetes - https://github.com/dwimsey/openshift-pvmanager
Ever heard of back ports? Redhat does this quite a lot. New drivers into their own kernel tree. Redhat's latest 2.6.9 kernel in RHEL4 is way different to the one Linus released all those years ago.
Beggars cannot be choosers and like it or not, Linux community = beggars.
They didn't donate this code out of any altruism, only pure self-interest.
Everyone using the GPL is doing so out of self-interest. If it was altruism, they'd be released their code into the public domain.
But aren't we just seeing this process dramatized before us.
Microsoft: "Here you go. Hyper-V Code."
Kernel Maintainer: "Wait. There are some problems with it that you need to deal with or we aren't going to include it in the kernel."
[Slashdot story]
Microsoft: "Oh we'll look into that."
If Microsoft cares enough about getting it into the Kernel they'll fix it. Otherwise they wont... problem solved.
No, they were violating the GPL.
They had to at least give source to their customers.
Rather than to continue to do that they made this driver the kernel maintainers problem. If they don't want to help maintain it, I say drop it from the kernel.
It was a twofer. MS weaseled out of punishment for license violation ( GPL ) and at the same time just shat in the kernel maintainers' collective pocket.
Denial of Service attacks work in meatspace, too. The maintainers have no obligation to burn up hours coding and supporting someone else's abandonware.
For that matter, so do injection attacks. For example, find out who gave the order to install any given Windows server, assuming you can still find one these days. No one will 'fess up.
Beta is broken and the link to classic doesn't work. Stop wasting our time or there won't be anybody left here.
From the blog,
"hv (Microsoft Hyper-V) drivers. Over 200 patches make up the massive cleanup effort needed to just get this code into a semi-sane kernel coding style (someone owes me a bit bottle of rum for that work!) Unfortunately the Microsoft developers seem to have disappeared, and no one is answering my emails. If they do not show back up to claim this driver soon, it will be removed in the 2.6.33 release. So sad..."
I'm not shocked. M$ reputation is so bad that this kind of behavior is not a surprise.
Was is a bit of a shock is that GKH actually wasted one minute longer than rejecting the fake drivers with a missive to "hold off on re-submission of the code until it actually works" What happened to the idea of showing something that works? There used to be pretty strict guidelines about coding style and what is and isn't acceptable, including having some working code first.
Anyway, what are these so-called developers from M$ off doing while the Linux team is cleaning up from this hit and run?
Beta is broken and the link to classic doesn't work. Stop wasting our time or there won't be anybody left here.
Don't you mean;
Microsoft: Here is a binary Linux kernel module for Hyper-V. .32 drivers.]
Someone: But wait... it is using GPL-only kernel interfaces.
Microsoft: Oh, er...
Grek KH: You need to release the code to be compliant.
Microsoft: Here you go. Here is the code for the driver.
Microsoft PR: Microsoft has generously donated a lot of code to the Linux kernel under the GPL license. We did this because it is the right thing to do.
[Slashdot story #1]
[Greg KH spends a lot of time thanklessly getting the code to the point where it is ready for inclusion with the kernel.]
[Greg KH sends e-mail to Microsoft asking for help.]
[Greg KH posts a summary for the upcoming
[Slashdot story #2]
Microsoft: Ok, we'll help maintain it.
Microsoft PR: Microsoft is actively involved with the Linux community.
You seem like one paranoid individual.
I think the mods are on crack again. This is a plausible scenario -- not thru malicious action on MS part but just the inevitable turpitude of a large uncaring monopoly.
Every mans' island needs an ocean; choose your ocean carefully.
Don't be stupid.
When Intel contributes a patch, they go through the required process necessary to make the patch meet the maintainers' standards. I actually did this a couple times when I worked at Intel.
If MS isn't going to do the work necessary to make their patches meet the standards, then it shouldn't be merged. I'm actually a little disappointed that they merged it in at all before going through this process fully.
It hasn't been merged to drivers/ proper, only drivers/staging/. This is the normal procedure these days for subpar driver code: it gets merged to staging/ in the hopes it will be cleaned up and can be merged to mainline proper.
MediaWiki developer, Total War Center sysadmin