Microsoft Releases Linux Device Drivers As GPL 362
mjasay writes "Microsoft used to call the GPL 'anti-American.' Now, as Microsoft releases Hyper-V Linux Integration Components (LinuxIC) under the GPL (version 2), apparently Microsoft calls the GPL 'ally.' Of course, there was little chance the device drivers would be accepted into the Linux kernel base unless open source, but the news suggests a shift for Microsoft. It also reflects Microsoft's continued interest in undermining its virtualization competition through low prices, and may suggests concern that it must open up if it wants to fend off insurgent virtualization strategies from Red Hat (KVM), Novell (XEN), and others in the open-source camp. Microsoft said the move demonstrates its interest in using open source in three key areas: 1) Make its software development processes more efficient, 2) product evangelism, and 3) using open source to reduce marketing and sales costs or to try out new features that highlight parts of the platform customers haven't seen before."
Re:Hell called (Score:5, Interesting)
Don't forget Microsoft's strategy: Embrace, Extend, Extinguish
I heard reasonable arguments about that being true for the Ms-PL [opensource.org] but I thought the GPL (v2 and v3) were supposed to be embraceable and extensible but nearly non-extinguishable ... once the code is out there, just fork it. Care to explain to me how this plan can follow after releasing something under the GPL? I'd be shocked that no one's tried it yet if it's possible.
I'm impressed... NOT! (Score:5, Interesting)
Some drivers to make Linux work better inside MS's Windows Server Hyper-V virtualization platform? How altruistic...
I'll be more impressed when MS, for example, helps with the SAMBA project. Or at least, doesn't actively screw up with such interop projects from the FOSS community. No GPL code required, just give people decent, up-to-date, open specs; and no patents bullshit.
Or at very least, when MS stops enforcing such patents (see TomTom / FAT32, or again SMB in MS/Novell "agreement").
Microsoft is going to kill VMWare (Score:5, Interesting)
I think Microsoft gets upset if any other company talks to the BIOS besides them. Here's a page from VMWare [vmware.com] that compares their own product to Microsoft's Hyper-V. Hyper-V only debuted as a beta a year ago and they're already compromising company policy to release Linux kernel level code.
Wikipedia page for Hyper-V [wikipedia.org]
Re:Hell called (Score:3, Interesting)
Given Microsoft's attacks on GPL as a license, I just do not understand why they licensed it GPL instead of LGPL or BSD
Comment removed (Score:2, Interesting)
Re:Microsoft is going to kill VMWare (Score:3, Interesting)
i think most people are missing the point...
the move on microsofts part is to help its partners (citrix and xenserver) product line and provide easier means for transfer of vms between the two. both of which have a goal of crushing vmware.
once vmware is toppled, and microsoft has full interopratability with citrix/xenserver, microsoft can then transition to knock citrix/xenserver out of the market and become the virtual machine leader...
this is just an intermediate step needed to further this goal
Re:The motives are quite simple (Score:3, Interesting)
Yep, I think that's the crux of it. It's a point of control. If they can remain the closest to the hardware, they control the system no matter what's running on it. If they can get Linux to run under Windows and they successfully outpace other VMs, then they can start adding "compatibility" code and making Linux perform worse than Windows Virtual systems.
It's also well documented that Microsoft would rather give away their technology to get people using it than let people use a competitor.
Re:Hell called (Score:2, Interesting)
The Hyper-V patch doesn't affect that; if they could claim patent infringements in other parts of the kernel now, they could do it before too.
Exactly. So there is nothing at all about their contribution that would prevent an embrace-extend-extinguish strategy.
Contributing is the embrace step. More people will use Linux when it's progressively made compatible with Microsoft technologies: extend. Then they can go all SCO and sue the pants off of some strategically selected parties that use it: extinguish. End result: Microsoft owns Linux.
The original question was: how could Microsoft still employ its usual embrace-extend-extinguish strategy while contributing to the Linux kernel? The answer is that it's very simple because there is nothing stopping them. They could do it if they wanted to. And given their track record, it would be silly to trust them not to.
Funny how my first post in this thread is now at -1, Flamebait for accurately answering a question.
Re:Hell called (Score:1, Interesting)
Don't confuse MS with the bank which was on slashdot sueing itself a couple days ago.
They may be evil, but they are not stupid.
Re:Hyper-V? Never heard of it. (Score:3, Interesting)
Hyper-virtualisation. Running OS's under other OS's. In other words, this is a patch for Linux to make it run well on Microsoft systems, so customers will feel less need to actually install Linux on servers. It's not a friendly gesture to make normal Linux systems work better, as the title suggests.
I think I'm way more likely to virtualize *Windows* servers on a Linux host than otherwise. The company I work for doesn't run Windows on bare metal anymore.
A good reason for that is that Windows isn't really administrable via a serial console, so that if networking is blown, you'd require either an iLO/DRAC type hardware solution or would have to go with relatively costly KVM over IP.
Honestly, I don't think this is big news. The host component isn't being opensourced, so you need a Windows-whatever server to run Linux hosts under it. I think I'll stick to Xen for paravirtualization and VMware/Virtualbox for full virtualization, thank you.
Re:Hell called (Score:3, Interesting)
How does this protect against the Hyper-V product being changed to be incompatible with the GPL drivers in the future. What guarentee do we have that compatibility will be maintained or the drivers updated?
GPL drivers don't do much good if they aren't compatible with the hypervisor. Otherwise we'd be using Xen drivers on Hyper-V wouldn't we?
I know I'm jumping to conclusions a bit, but this is a company with a very long history of this kind of behavior. Even after several court cases and many promises.
(Hope I get answers, instead of just getting modded up, then suddenly down into oblivion again)
Re:Hell called (Score:4, Interesting)
Sam, does this mean that this GPLv2 release was actually negotiated and coordinated with Linux kernel developers? If so, it would be interesting to hear more about that side of it. So far I have found this [kroah.com]:
Q: Why release the code?
A: Because we have utilized Linux code, Microsoft has an obligation to open source the device drivers. This is the process outlined by the Linux community.
Q: Why open source the code?
A: Because this is a requirement of the community, and critical in ensuring that as the Linux Kernel evolves, and as Hyper-V evolves, that the Hyper-V Linux Device Drivers evolve as well.
But this is rather vague - it's not clear where the "requirement of the community" comes from; is it implied, or was it specifically talked about?
Re:Hell called (Score:3, Interesting)
Microsoft just wants that people can use virtualized Linux while paying at the same time their windows server licenses. I doubt they will break compatibility...
Re:But there is a real difference. (Score:3, Interesting)
It doesn't have to be about the driver code...
HyperV is distributed for free with every Server Professional (right?) so if they can get the world to run their server and put Linux in VMs under it, they have control of the VM. Now, big corporations see this as a point of sale. Look, we can run our Linux servers on this Windows machine instead of another Linux machine, this VM is pre-installed and free! Microsoft releases or markets HyperV as a faster more reliable solution (even if it isn't) and they convince the corporate types that only trust big M because they use them so widely already. "Let's make our next server a Windows box..." This continues for a few years. MS plays nice and people think all is well. Open projects start to lose interest and support, companies start relying more and more on HyperV. Then MS sneaks in a virtual machine side compatibility patch that makes Linux start to run slower and slower. You won't be able to see this because HyperV isn't open. And all those other VMs that were competing with HyperV before have fallen behind and can't keep up. MS starts sending out comparisons of their OS running in a their modified VM comparing it to Linux running in their VM. Bloggers/Websites eat that up and soon enough, it's on the front page of Slashdot. "Windows runs better in VMs than Linux."
Call it a conspiracy theory if you like, but you couldn't prove one way or another without seeing Microsoft's code... and that's not happening.
Re:Microsoft is going to kill VMWare (Score:4, Interesting)
This is so timely. I spent the last 3 days fighting VMWare. Microsoft don't need to try and kill VMWare. VMware is doing an excellent job of that by itself. We are a small gov branch that spend a measly $40,000 a year with VMWare and then we realised that why shouldn't we get educational discounts from VMWare. We get top level discounts from MS,Novell,Oracle etc.
VMWare, nope sorry. No can do. After 3 days of getting to the right person the answer was no, because your courses you teach aren't long enough. FFS!
So I mentioned that we had a top line educational select agreement with MS, and Hyper V is free! They really didn't give a toss. Give it a year and they will be so hurting. It's a pity because the software is top notch but there is no "getting the business" aspect of it. Once Hyper V is matured, they are done!
Re:Hell called (Score:4, Interesting)
Bingo. As much as I hate MS, I would absolutely make an about face on my opinions if they started joining in on putting things in under GPLv3 (without skirting it's intent). That's all I ask. I don't get why they don't, the whole marketplace would support them if they did so. I guess they don't like expanding marketshare in ways that benefits everyone as opposed to themselves (PS microsoft: this is how you make friends instead of enemies).
Comment removed (Score:3, Interesting)
Re:Hell called (Score:3, Interesting)
Explanation of why people are so untrusting: (Score:5, Interesting)
Microsoft has a long, long history of letting its mid-level managers and employees believe one thing, when the top managers intend something else, something very unfriendly and sneaky.
For example, Microsoft employees believed that they would be allowed to finish their work. But, in spite of strong opposition inside Microsoft, Windows Vista was released.
Other products released before they were finished:
Windows XP (Okay after SP2, a lot of grief before)
Windows ME
DOS 3.0
Since Microsoft has acted against the best interests of its customers in many ways in the past, people think that will happen this time.
I listened to this interview of you: Sam Ramji of Microsoft Tells all [sun.com]. It's obvious that you are intelligent and well-meaning. I would tend to trust anything you say if you have control over it. However, I think it is likely that you have no control. I'm guessing that it is likely that some vicious Microsoft top manager has some plan to cause trouble.
Why do I think that? Because sneaky behavior by Microsoft has cost me tens of thousands of dollars over the years.
show us the patents (Score:2, Interesting)
I think before there can be any dialog with you, you should show the community the 200+ patents you claim Linux infringes, or else retract that claim.
What about it?
Re:Hell called (Score:3, Interesting)
coLinux [colinux.org] seems to work well for when I do. My wife insists on booting her PC into Windows due to some stupid bug with Flash not showing some stupid menu in some stupid Facebook game. I've used coLinux on there so that computer isn't a completely useless box while being a Facebook terminal.
Not that I'm bitter or anything. (I'm looking at you Adobe)
Re:Hell called (Score:3, Interesting)
Microsoft did indeed purposefully sabotage Lotus (Score:2, Interesting)
It is too bad that magazines like Infoworld cannot (or do not) bring their old material into the World Wide Web. As the debacle unfurled, Infoworld sent an editor to interview the highest guy at Microsoft in charge of DOS. The interview went like this:
IW: "Lotus 1-2-3 is the most popular application running on PCs today. Your new version of DOS does not work with it. Didn't you do any testing?"
MS: "Yes of course we did testing."
IW: "What were the results of your tests?"
MS: "We knew there would be problems."
It was only afterward that the phrase "DOS isn't done until Lotus won't run" was exposed.
In case you didn't live through this history (I did):
Microsoft was a partner in the L.I.M. specification that allowed programs to access extended memory. L.I.M. = Lotus, Intel, and Microsoft.
Microsoft changed Windows and (and through it their new spreadsheet product, Excel) to do LIM access on word boundaries instead of byte boundaries. Then they changed the LIM driver to only work on word boundaries, and to cause a fault in the programs that attempted access on byte boundaries. Super conveniently, they didn't bother to notify Lotus (or Intel) that they implemented a we-are-going-to-break-all-your-programs change to the L.I.M. spec.
They shipped DOS first, and apologized later. Except they didn't apologize. They ran advertisements picturing a jet fighter pilot crash helmet. "Crash proof. Doesn't it make sense to get your applications from the people who make your OS?"
More history: the first time you launched Lotus 1-2-3 in Windows with the new DOS, the dialog box said "This program has violated system integrity. You should reboot to ensure proper operation of the system. If it happens again, consult with your application vendor."
Clearly the blame was pointed at Lotus 1-2-3 by Microsoft in Windows. But what changed?
DOS was finally done when Lotus wouldn't run.
Infoworld also interviewed people at Lotus. Infoworld asked if Lotus was going to sue, and the Lotus person said no, for two reasons. One, that Lotus was still dependent on Microsoft and DOS (reading between the lines, it looked like they were saying they've sabotaged us once already, and could do it again). Two, the lawyers at Lotus asked the engineers about the change, and came to the conclusion that Microsoft would claim they made the change because "it is better". Word boundaries for memory access are easier than byte boundaries.
The evil here is that the change was made with malice aforethought toward Lotus, AND, the notification of the change was withheld from Lotus.
Lotus would probably have agreed that word boundaries were better. The crime was they were denied an opportunity to prepare for the change.
But proving to a judge (and this was before judges were at all tech-savvy) that Microsoft didn't innocently bungle a line of communication or two was not a case the Lotus lawyers thought they could win. The technical argument "it is better" would have to be offset by "no it's not. it is memory wasteful" which in the age of 2MB RAM machines meant something.
As for your claim that people wouldn't buy the new DOS - they didn't. Microsoft slip-streamed the new version of DOS to Hewlett-Packard, Compaq, et. al. They told HP (and everybody) "Here is a new version of DOS. Include it with your new machines instead of the old version of DOS." As the debacle unfurled, HP had to quick issue old versions of DOS to everyone that that got screwed. (I was one of those HP customers at the time).
I realize that you are probably a Microsoft shill that will always attempt to discredit the deliberate damage Microsoft inflicted on it's biggest competitor. Which is why I am going to call you out on it. If you don't want to be called a shill, then you need to not be a shill.
The truth is Microsoft changed DOS and knew it would completely screw over Lotus. They had tested it. They wrote a Windows error message to shi