Slashdot Mirror
Should Linux Have a Binary Kernel Driver Layer?
zerojoker writes "The discussion is not new but was heated up by a blog entry from Greg Kroah-Hartman: Three OSDL Japan members, namely Fujitsu, NEC and Hitachi are pushing for a stable Kernel driver layer/API, so that driver developers wouldn't need to put their drivers into the main kernel tree. GKH has several points against such an idea." What do you think?
If it means more and stable drivers I'm all for it!
Having a kernel API for drivers allows developers to stay away from the mainstream kernel. This will enhance the stability of the kernel in general and also allow hardware vendors to support Linux with less effort.
To Terminate, or not to Terminate, that's the question - SCSIROB
Microsoft could license code into the drivers or otherwise maneuver the driver makers to license their IP from MS, the drivers could form a layer between Linux and the hardware, Microsoft could then pull the run from under Linux.
Don't go there, it protects Linux from getting tripped up, and devalues any hardware that doesn't support Linux.
Don't underestimate the important of driver support for Linux, you practically can't make any server component without a good solid Linux driver.
I have been waiting for that for a long time. The lack of a stable interface is hampering adoption of Linux. Not all of the manufacturers are willing to open source drivers or for that matter to continuously change them as the APIs change. This is very welcome but unfortunately, I think they'll fail. There is just too much politics surrounding Linux these days.
It's a bad idea because what happens when the driver ABI changes? You have to wait umpteen months for the company to get off thier asses and fix it - like nVidia.
It also precludes anyone else from fixing bugs in the broken, half assed crap most corporates spit out these days.
feh. stuff.
one of the main problems for getting device manufacturers to support linux is the fact that they either have to release a new version of their driver every time the linux kernel changes some esoteric internal API, or be badmouthed for not having good linux support.
would it really hurt so much to guarantee a stable DKI? doesn't have to freeze the whole kernel, just a subset of functions that will be guaranteed to work as they do now in perpetuity.
backwards compatibility is just as important to driver writers as it is to app writers.
doesn't even have to be binary backwards compatible, source level would be sufficient for most.
The difference between Theory and Practice is greater in Practice than in Theory.
One of Linux's great strengths is the flexibility of changing to meet new needs and not being hobbled by rigid backwards compatibility. This can only be done if all source is open and anyone can update drivers to meet new needs. When someone comes up with a patch to streamline a certain minor part of the kernel, it frequently has repercussions elsewhere in kernel land. It is these small changes which have made linux better and better with breathtaking speed. A "stable" binary API removes the possibility of keeping everything up to date and would dramatically show down the adoption of new features and general improvements.
Continual refactoring is worth far more than some supposed binary API which prevents changes. Get rid of binary drivers! If companies are so paranoid that they want binary drivers, then the hell with them. Linux can advance better without that baggage.
Infuriate left and right
I would say yes to a formal driver API, but I also think source for the drivers should still be released. The two concepts (formal API, open source) aren't mutually exclusive, but I guess being able to use closed source drivers is, in reality, why these companies are pushing for a formal API.
These companies want a binary layer so they can build binary drivers.
What people tend to forget about this is that it's a bad idea- from most every perspective.
The Linux kernel was written as a Free Softwate alternative to the existing *nix systems.
We have thousands of drivers in the kernel from a combination of development efforts. Sometimes a driver is written by an independant kernel developer, and sometimes it's written from the company producing the hardware, working alongside the community.
What these companies want is to be able to have thier cake without giving back to the community. This is a very slippery slope at the least, and illegal at best, since these sorts of links to binary kernel drivers have been long known to be illegal to distribute alongside the kernel (unless special previsions are made, such as a userland driver).
Also, binary drivers have been known to be buggy and essentially removie the kernel developers from a position where they have control over the kernel as a whole project. I won't even go into the issues associated with a possible security hole in a binary driver, or a binary driver with, for example, spyware in it.
The arguement for it is, of course, that this might mean more drivers. This is a test of our strength as a community. Doing the right thing is harder. It means we won't have all the hardware at all times, and certainly not the newest thing. But we retain control over our computers.
It's hard to say no, but this looks like a clear case where we have to.
IANAKH, but couldn't more drivers be moved into userspace (or other lower rings) --- especially for things like USB printers and miscellaneous gizmos? I think it would also be nice to not bundle thousands of drivers and support for architectures I don't have with the kernel. The kernel itself could provide a very minimal layer of hardware protection (like an exokernel?) and there'd be libraries exporting generic abstractions for particular classes of hardware. Is the context-switching penalty really so great these days? Educate me!
In my opinion, binary drivers are worse than no drivers at all because they release the pressure on the manufacturer. They can say they support Linux which in case of binary drivers is simply not true.
Save the bandwidth. Don't use sigs!
This is the point though. I never said I couldn't get it to work, I said I was tired of getting punished for it. I don't enjoy wasting a lot of time futzing with kernel patches and hoping for the best, which seemed to happen every time I did a major upgrade. In short, the onus should not be on me to make it work -- the onus should be on the OS to work without my intervention.
Also, for the record, Linux doesn't suck. But I found I could get my command line fix with XP and Cygwin, without the hardware annoyances.
Toronto-area transit rider? Rate your ride.
The typical comment of a Slashdot user. We should all buy new hardware and spend hours on top of hours getting things to work, rather than just "having" things work. Call this a troll if you want, but this is the exact reason that the masses AREN'T running Linux.
You people need to figure out exactly what you want, Linux for the masses (read: grandma, mom, etc) or an O/S where you have to spend valuable time just getting it to work with regular hardware. You bashed the poster for buying "random hardware" and expecting it to work even though you don't know what the hardware in question was, yet in your own message you bought 3 "bleeding edge" Gateway laptops (a fairly well known manufacturer) and you had to (in your words) "_try_ to make it work."
I was going to post this anonymously, but it would of course be modded a troll at that point. Let my karma burn for all I care.
One of Linux's biggest problems is the lack of device drivers for common devices, especially newer video cards. Let's face it, companies like ATI and NVIDIA aren't going to release fully open-source drivers. It would be wonderful if they would, but it would also be wonderful if we had flying cars.
Having a stable binary driver interface would make it easier for hardware manufacturers to embrace Linux, give things like wireless chipsets more usability on Linux and drive further adoption of Linux as a viable competitor to more proprietary solutions
The perfect is the enemy of the good, and the more Linux gains a foothold the better it is for open source. Insisting that device manufacturers need to have on-staff kernel hackers in order to keep ahead of a frequently-changing kernel makes it that much harder for manufacturers to support Linux as a viable alternative.
Provided Linux can have a stable binary driver infrastructure that doesn't harm stability, it would greatly help in the adoption of Linux worldwide.
But we retain control over our computers.
Tell me, what BIOS do you run ? Do you have the source to the firmware in your IDE disk drive ? In your CD-ROM/DVD-ROM drive ? Do you have the source to your SCSI controller's firmware ?
If you think you have control over your computer you are suffering under a delusion.
If a company is developing an embedded Linux ap for their own hardware. All of a sudden, all of the communications with the board-specific hardware is being done through binary drivers, resulting in an effectively closed system.
No more hacking WRT54G's for you, chump.
"Eve of Destruction", it's not just for old hippies anymore...
The idea is not just a formal API, but a formal API that doesn't change for a long (VERY VERY long if it is to be usefull at all) time.
/. is not the place to do it) or by throwing money at them.
The problem is that you will be stuck with the mistakes for the same time.
You can't improve the kernel anymore because you have to keep supporting those mistakes.
You would have to add huge glue layers to your kernels to keep emulating thos mistakes.
It is a bad idea. It has only advantages to those driver writers who want to stay far away from the kernel and want to keep using their crap for a long time. The kernel and the kernel writers are not going to get better from it (they only get extra work supporting lazy driver writers) why should they support it?
Linux development works very easy: do it yourself or convince someone else to do it.
You can convince other people either by motivating them with proper reasoning (and
Nobody has been throwing enough money yet to get the kernel developpers to do it. (And very likely nobody ever will throw enough)
Jeroen
Secure messaging: http://quickmsg.vreeken.net/
Driver certification means just that Microsoft received the driver, and agreed to confirm that the driver comes from one of its business partners, and not from a suspicious open-source hacker. You don't even have a guarantee that the driver is free from rootkits.
The company isn't really interested in fixing any issues with the drivers -- if you have problems with a bug, you have already paid them all the money they are going to get for that particular piece of hardware.
An example: the !@#$%^&* nVidia proprietary X driver. On some older cards, it will cause kernel oopses, within my usage patterns around once per several hours. Is there anything we can do to fix the problem? I'm not a master kernel hacker, but I do have some rudimentary skills there -- I would have at least some chances to make a fix myself; if I wouldn't succeed, reporting a data-loss bug would make us have it fixed by someone with more knowledge in no time.
On my current desktop, switching to text mode the first time X is run after a boot puts the console into 80x25 mode without even doing a TIOCSWINSZ. Somehow, if I kill the X server, reset the video mode then start X again, everything is fine, until the next boot/resume. What can I do to fix this annoyance? Begging to nVidia does nyet work.
The creatures outside looked from Alt-Right to Antifa; but already it was impossible to say which was which.
Linux was, is and hopefully will always be "open". I don't want closed drivers in the kernel (even via an API layer) any more than I want a Sony rootkit masquerading as DRM.
It isn't about "politics". It's about policy and philosophy.
If the hardware doesn't work with Linux, don't buy the hardware/pester the vendor for an open driver, or don't run Linux.
This is the problem with the open source movement. Putting the code before the user.
And this is why you fail.
-everphilski-
It is not welcome. Linux is about Open Source, and allowing people to link-in binary closed drivers goes against this.
Bypassing the dogma of the above, there are numerous pragmatic reasons why this would be better for linux, even if you don't include support for binary third-party drivers.
Sure, some of these are extreme cases. You can usually get away with just re-compiling the driver, and occasionally, you can even use the binary from the existing version.
The point is you should *always* be able to do this wihtin the same major kernel version. There is no technical reason, aside form the politicis of not wanting to ever allow binary drivers, to not have a stable driver API.
Imagine if the Mozilla plugin API changed with every new version of Firefox. And look at all the complaints when a new Firefox version doesn't work with all the old extentions. It is the exact same.
If you are going to take the strategy of "We will do things to attempt to force you to do thigns our way," don't be supprised if the response of companies is "Fine, then we will ignore you."
The simple fact of the matter is that most companies are not willing to go open source, for software or drivers. You can argue that's a bad thing, but it is the reality of the situation. So, if open source is out in their book, either because of contractual obligations or mentality or whatever, they are left with two choices:
1) Do Linux drivers, and update them every time the interface changes, which can be as often as every minor kernel revision.
2) Ignore Linux, and let the community write the drivers if they want.
The problem is that Linux is a bit player. They are larger than the other bit players, but they are still tiny, less than 10%. Given that the continous rewrites can get expensive, the choice for many will simply be not to write the driver.
So if you are ok with that, then great, but don't get mad at companies when they won't play by your rules. Are they being unaccomidating? Sure, but so are you.
In the end, it comes down to needing to make a decision of what you want Linux to be. If you want Linux to try and become the next big thing in OSes and start to really make an entrance in the home market, standardisation is needed. Standard APIs, standard UIs, inter-version consistencies, etc. In essence, it needs to become more like OS-X. Now if you are ok with Linux being more of a geek/server OS then that's not necessary, but you can't demand the world change around you.
Assuming that we had a stable kernel source interface for the kernel, a binary interface would naturally happen too, right? Wrong. Please consider the following facts about the Linux kernel:
But see, the thing is... a "stable" binary interface requires that structures used specify padding, alignment, and fields to be fixed! If these can vary, then by definition , it's not stable. Ditto the variations that depend on kernel build options.
Now, if you want to make the case that it's not possible / practical to make an interface that can cover all of these conditions adequately, well, by all means, do so (though I'd say that the hundreds of existing operating systems with binary interfaces show that this isn't the case in the general sense).
But what I see here is a relatively weak technical argument that is being used to justify an ideological decision.
I am not a Kernel Developer, but I know some. ;) I guess that my open question is how this would benefit kernel users? Yes, I see it would reduce the workload of kernel devs. Yes, I see it would allow driver developers to not have to go through the kernel code vetting process. But, the kernel code vetting process is what is a strong benefit of using Linux, from a user perspective, as I know that the code is well tested by an army of users and developers.
Once you push driver development out of the kernel, yet give access to kernel internals in this way, you introduce a level of uncertainty in so far as stability and robustness is concerned. One must question why these big comapnies are pushing for this, but most human kernel devs are not.
On the driver certification part: it does only certify that the driver came from a Microsoft business partner... but the characterization of why this is a bit off. It is to certify that this is not from an unknown third party. It is done for the same reasons that most linux iso's are distributed with a MD5 hash next to them (which most people ignore anyways): it allows you to be sure of where the code is coming from, and that it is not coming from a nefarious third party.
I do wish that linux and MacOS X has similar setups, but it is difficult to setup such a system without having a central authority with a great deal of power (to abuse).
GKH raises good points about how a stable binary driver interface will open the floodgates to both security problems and to update/maintenance problems. As it stands right now, Linux kernel developers can quickly respond to threats because they are able to fix all instances of a given problem, in all drivers, at the same time. If they do not maintain this flexibility, either some drivers stop working unexpectedly when security fixes are made and the interfaces are forced to change (making Linux appear "unstable") or backwards compatibility must be maintained making the Linux kernel grow over time (whenever a new interface has to be written to address flaws in the old interface).
Yes, abstraction is good...but, in this case, stability, the perception of the user and maintainability (where the *real* costs lie) must win over abstraction. Most of the kernel developers are not being compensated; how often do you think that backwards compatibility is going to be maintained? Its not. Right now, fixes are accomplished because it is easy to accomplish - global search and replace, etc. Make it difficult and it just won't be done.
Manufacturers want binary drivers because they want to play for free - they want all of the benefits of open source without any of the costs. Not cool.
Because of this, I'm 100% not convinced making binary driver developers lives harder changes anything. Are large businesses (the type who make hardware that's difficult to reverse engineer) likely to say, hey, gosh, you know this Greg KH guy kind of doesn't like closed drivers, maybe we should open them up to please him? Nope. They'll just work around the difficulties or not provide drivers at all.
I've been a Wine developer for years and have spent many hours doing this impossible thing of which you speak, and your average copy of MS Word or Steam is a LOT larger than your average driver. Yes, it's hard. No, it's not impossible. I've heard various excuses as to why kernel development is just different!! to userland software development, and don't buy it. Yes, having to reboot when a crash occurs is a royal pain in the ass, but so is not being able to get a backtrace because the game you're investigating treats any attempt at attaching a debugger as an attempt to hax0r its copy protection. Different space, different challenges. It's still possible.
Ignoring for a moment, the ideological points, lets consider a frequently raised practical point: the idea that an API would either get out of sync with the kernel, or be a drag on innovation.
/lib/modules/binary/" than "you need to recompile and reinstall the whole kernel with this patch"?. For obscure hardware which will never be in the official kernel, it would be nice to have the ability to easily use it with any linux distro. No need for a slew of precompiled RPMs, DEBs, and a user-unfriendly source tarball, just one driver binary and you're ready to roll.
I agree that when the Linux kernel was young and untried, standardizing a binary API was bound to become a millstone in a short period of time, as the kernel internals were in a constant state of churn and iterative improvement. Nowadays though, surely, the kernel has been "shaken down" enough that it could afford to commit to binary APIs that are stable at least throughout each minor version number?
Returning to ideology, I can see how a stable binary API would be useful even to open source hardware. How much easier is it, to say "drop this file under
In itself, that says nothing, either pro or anti, about the availability of driver source.
It doesn't mean much formally, but if a vendor doesn't bother to certify the driver (it's not that expensive after all) it's a good indication that they might not care about driver improvement as well. Or to put it in different words: Would you believe that a company which can't afford certification would invest more than necessary into driver development? I know that this doesn't apply in many cases, but it's a good rule of thumb.
I never had significant problems with certified drivers, but I guess that some of you know good examples.
I don't read replies by ACs.
They won't release now, but eventually they will. You don't cave in to "terrorist threats", ever. And that's what this is. If those vendors want in on the open source market that revolves around the linux kernel, then they can play ball, too, end of story.
Linux is at a really important part in its evolution. Caving in to closed source interests would be counter productive in the long term. It is better to force/cajole the vendors to finally "see the light" with open source and the GPL. Maybe eventually another graphics card vendor will appear and become the champion, precisely from having open source drivers. The manufacture of electronics in Asia in particular is exploding, where there is a market, a vendor will appear. Patience.
the question I have for this is "why?" wouldn't a stable binary API likely result in far more third party hardware support for linux? possibly more laptops that are actually compatible with linux?
This seems like a case of open source programmers shooting themselves in the foot because they want everything to be open source... not every application and driver is going to be made open source just to suit the desires of the linux development community. It seems to me that sticking to a hard party line against closed source software instead of trying to co-exist with said software is bound to keep linux in relative obscurity and pretty much ensure that it never becomes a viable competitor in the desktop market.
"In America, first you get the sugar, then you get the power, then you get the women..." -H. Simpson
Some manufacturers don't want to give away their specs or source code. They will always give binary drivers, like nVidia or softmodems. As a user, I want to be sure, 100%, that these drivers will work out of the box, even if, let's say, I run kernel 2.6.24 while the driver was written for 2.6.15. Right now, I don't have this guarantee. And as a user, I need it.
Firstly, I get paid to do driver development. Secondly, I have never worked for a company that willingly said, ok lets open source the drivers we spend thousands of dollars writing and optimizing.
That said, this whole discussion is as silly as the one about not putting a kernel debugger in the main kernel source code. Frankly, linux desperatly needs both a kernel debugger, and an ABI to be a REAL alternative for many customers. It also needs the ABI for driver developers so that we can write a single driver and expect it to work on the dozens of flavors of linux we are expected to support. Saying that everyone should opensource their drivers is like saying food should be free. It isn't going to to happen and wishing for it, won't make it happen any sooner. In the interm, almost all the hardware out there has better support on windows (Our sysadmin can't even get support for major linux distributions, from major hardware vendors, even when they have little linux logos on their hardware and websites) the windows drivers tend, to accually work, and they almost always have better features sets. This isn't going to change as long as the "opensource" community treats hardware vendors that think they have IP in the driver as second class citizens.
Oh, and for people who don't accually work for hardware companies that ship drivers, driver development is often times an expensive process, not because the software engineers are expensive, but because the hardware and software needs to be tested and certified in particular enviroments. There are orders of magnitude more linux distributions this makes it cost orders of magnitude more to test and support than a half dozen windows enviroments, most of which can be tested at microsoft, or one of the major OEM's if the hardware isn't avialable onsite. Putting 10x the money into a market that may be less than 1/10 of sales is not always a good idea, especially when resources are limited. Creating an proper ABI helps to solve this problem.
That said, if the damn linux kernel accually had a real architecture, it could support an ABI, and even isolate itself from rogue drivers. As it is, the kernel arch is pretty much non existant and just a pile of code that tends to behave like a real kernel, except when you try to do something a little outside of the mainstream desktop or small web server enviroment. This was fine when the whole kernel was just a few hundred thousand lines, but given its current size its getting massivly unmaintainable. This is proven by the fact that linux system stability seems to have gotten really bad over the last few years. Getting to a stable system, takes a lot of vendor testing by the likes of Suse, Redhat, etc.
Lastly, the tainted concept works fine for the kernel developers, why not carry it forwared so that any binary driver simply marks the kernel as having a binary module loaded, and uses the standard abstract interfaces instead of linking against all kinds of unneeded kernel crap that just provides the posibility to screw something up.
This is necessary if we ever want Linux to be ready for the desktop. The ability to have driver modules is certainly more advanced than having everything compiled into the kernel, but it's severely lacking in many regards. The module has to be custom made for each kernel, making binary distribution useless because there are 2^100 kernels out there. So unless the manufacturer open sources the driver, they can't make a driver for Linux. Or you could go with an open source interface to a closed source driver.
Here's a sample of what I put up with. I downloaded Agnula Demudi 1.2.1 and installed it with the 2.6 kernel. I was ready to install some Nvidia drivers. But after some searching, I couldn't find any binary driver interfaces compatible with my kernel. Fine, I can compile my own. So I download the interface sources and launch module-assistant. It complains that riva driver support in my kernel conflicts with the nvidia driver, and I need to recompile the kernel. (I then went through the joy of trying to find the hidden demudi sources and figuring out how to patche them and configure them, ultimately failing to compile it, but this is getting away from the topic.) Finally, I said screw it.
You might blame the distro, but it's really the kernel at fault here. Recompiling the kernel to support a driver is NOT something that a user should have to do. Windows does not require you to recompile your kernel to install drivers.
First of all, I completely understand that drivers with source are better than binary-only kernel drivers. However, the fact of the matter is that companies do not feel comfortable freely releasing their intellectual property. Given the choice between a piece of hardware in a laptop which doesn't work with Linux at all and one that works with a binary-only driver, I would rather have the binary-only driver.
As long as Linux kernel developers complain that binary-only drivers are "illegal", Linux will have less hardware support. One of the major complaints people have against Linux is that a lot of devices that one can attach to a Windows machine plain simply do not work in Linux (I still think Linux is far behind Windows when it comes to wireless drivers, for example). I want to see a true alternative to Windows on the desktop; GPL fanaticism and an inability to understnad how big corporations work harms this.
What you seem to forget, is that for the majority of users, Linux is a 'free' unix like system, and they could give a rats ass about "Free". If "Free" starts making things difficult to use, or preventing companies from supplying drivers for hardware users buy, then not only do users not care about "Free", but they will actively dislike "Free".
The bottom line is that having code be open is only important to a fraction of *developers*, and an extremely small small fraction of the general populance. Ultimately, "Free" software people want to push their ideoligy on others, they don't care about makeing functional easy to use systems.
So preach all you want. Very few people care.
So in other words, Linus has decided to deliberately damage the kernel (by introducing unnecessary instability) in order to make it harder for people to write software that uses it.
I thought that Linus usually put technical correctness above political correctness, but it looks like I was wrong.
The most rabid believers in American Exceptionalism are the exact same people whose policies are destroying it.
Oh, wait, it's been done. He didn't listen.
I am TheRaven on Soylent News
The BIOS is indeed an issue, and there are efforts underway to make a Free BIOS.
But why not try our best to have as much control as we can?
How about you try installing a copy of WindowsXP from the store on your bleeding-edge laptop. I'll bet you dollars to donuts that you still have to go hunt for drivers and _try_ to make it work. I'll bet you even more that it'd take less time to get Linux to work on it than it would a version of Windows.
Wait, we're lambasting slashdot "groupthink" here, though. My bad. I should say "You only like Linux because you hate Microsoft. You hatey hater man. Stop hating because you're jealous"
My blog. Good stuff (when I remember to update it). Read it.
They don't release the source. There is no source. That's the whole point. They just release the programming docs so anyone can write a driver. There is no IP for them to protect.
If you give companies a way to use closed source drivers, that is all they will ever use, because that is the way they've always done it. Linux presents a serious change to the way hardware vendors do business, and in the long run I think it would be better for everyone if they would embrace it.
It is just that once you open the gates, they won't ever be closed again.
There's a joke- programming is a lot like sex. One mistake, and you're supporting it for the rest of your life. By this measure, writting a device driver for Linux is a mistake. Because even after the feature set and the hardware itself is stable, you are letting yourself into a never-ending task of constantly changing the driver to keep track of the current API changes. At a previous job we spec'd it out at about 1/2 of a full time position to keep up with all the kernel changes. This is your job, from now until eternity, or at least until you're willing to have the driver declared obsolete and abandoned (after all, we all know that there is just three types of software- vapor, beta, and obsolete).
Well, congratulations. You have your religous purity. And guess what- it's comming at a cost. You wonder why Linux isn't more popular on the desktop? Well, here's part of the reason. It's hardware support will always- ALWAYS- be behind that of Windows. Why? Because when the hardware ships, it ships with Windows drivers that the hardware vendor wrote with it. Note that Windows pulls the same sort of API changing crap that Linux does. The difference is that the hardware vendors look at Windows, and the half man-year per year cost of supporting Windows costs, and go "but we have to support Windows if we want to sell more than 3 units." They then look at Linux, and the half man-year per year cost of support Linux, and go "Supporting Linux is not cost-effective at this time." I know, because I've seen this happen. So now the hardware is out there. And now we wait, for someone willing to step up and volunteer the time to write, and maintain indefinately, the driver. Someone less capable of doing it than the hardware manufacturer (this isn't to question the capabilities of the current kernel developers, but the fact of the matter is that there is a huge advantage to being three cubes down from the hardware developers, and capable of wandering over and asking direct questions, instead of having to reverse engineer what is really going on, having worked both ways).
So this is the fundamental question: which is worse. Having binary-only proprietary drivers, or being forever behind in hardware support and not having people contribute simply because they don't feel like having to constantly update the driver once they finish it, they'd like to be able to move on. I come down on one side, Linus and the kernel developers down on the other.
Fine. Their kernel. Their problem.
"Basically, I want people to know that when they use binary-only modules, it's THEIR problem. I want people to know that in their bones, and I want it shouted out from the rooftops. I want people to wake up in a cold sweat every once in a while if they use binary-only modules."
- Linus Torvalds on linux-kernel
And many people forgets that non-gpl drivers may be very well impossible to write at all (at least some lawyers think this), drivers are not at all like an app is WRT to gtk, drivers are more like "plugins". Plus, a closed driver module makes MUCH HARDER to debug bugs if the driver is doing bad things, and you can't know that (which makes harder to stabilize and/or develop the kernel. Several closed drivers can make it a hell or impossible at all.
So I wrote a nice driver for it, with help from the excellent Rubini book since there's no frickin' driver documentation and the Linux sources are uncommented (putting your name at the top doesn't count as commenting on any team I've been on). It worked for a few months, until the next incompatible change in the kernel. Then I poked around in the kernel until I thought I'd figured out what had changed. Got it kind-of working (DMA is unreliable and I don't know why). Then the 2nd edition of the Rubini book came out, so I rewrote the driver and got it semi-OK again. For a few months.
The kernel has had about a zillion incompatible changes in the driver interface since then, ALL UNDOCUMENTED, there's no 3rd edition of Rubini's book to sort it out for me, and I'm just lost digging through the kernel sources. It's always the same pattern: customer complains, I spend a few days fixing it, and it sort-of works for a few months until the next gratuitous interface change.
OK so I'm a for-profit vendor and therefore deserve to die (how dare I want to be paid for being a programmer, it's only people who don't create new IP who should be allowed to choose what they do for a living). But I'm not the vendor. I wrote the driver for their device (and plenty of other open source stuff) because it needed writing and no one else was doing it. The driver is distributed as fully commented source (obviously), and I've given it out to anyone who asked for it (uhh ... both of them -- this hasn't made me a nickel in extra sales for the package itself, I tell users who want reliable Q-bus interfacing to use my DOS version because there's no OS to break my driver there, it's worked unchanged for YEARS). The users haven't touched a single line of code on their own that I know of, all they do is complain to me. So what's wrong with this picture? Why isn't Open Source magically making everything work, at no cost to anyone?
It's just plain arrogant to assume that Groovy Open Source will always fix itself, and therefore it's OK to make the kernel interface be a rapidly moving target with no real manual. I'm sure that when we're talking about some extremely commonplace device that hundreds of thousands of people have, it's not hard at all to raise a posse and go after new bugs when the kernel changes. But this device I'm talking about costs $2575 a pop and is of no interest to anyone who isn't keeping a 30-year-old minicomputer system alive with a brain transplant. Absolutely no one is helping me maintain the driver even though they have everything they need to do it. I'm doing a lousy job of maintaining it because it's 0.0001% of my job description, and it's just not my strength (I know how to program bare metal, keeping up with Linux internals would have to be its own full-time job).
I couldn't care less whether drivers are required by law to be open source (although I really can't believe that's enforcible on a loadable driver, licenses exist to make exceptions to the default copyright restrictions, and interoperation is not copying). If people want to see my I/O code they're welcome to it. Just please stop making it break!!! Especially if you're not going to bother helping me maintain it. I've got my hands full just keeping the user-mode code working (WTF happened to /dev/vcsa7 and up?! and /proc/meminfo?!).
Anyway my point is, if Linux would pick one driver interface, and commit to supporting it (OK, possibly among other
By the way- another advantage to having a stable kernel API layer is the ability to write a dead-tree book about how to write Linux device drivers and not have it be obsolete by the time it gets published. I have half a dozen books on this subject- half of them were obsolete by the time they were published, they're all obsolete now.
A few year ago I wrote and maintained the Aironet driver. The years of maintaining the driver and supporting users (I still support them) has illustrated to me the importance of an ABI. I didn't really notice it at first, but it hit me in stages.
.config. It was hell. It was something that neither I nor they wanted to go through. I figured the problem would go away once I was integrated with kernel source.
1) Experimental driver stage. I distributed the source and had online instructions, but you would be surprised at the large number of people that I had to walk through compiling their kernel with the correct
2) Late experimental stage. I had all the problems of the previous stage, but now the kernel API was changing under me so I had to put #ifdefs in to deal with it. I figured the problem would go away once I was integrated with kernel source.
3) In the kernel. Yeah I made it into the kernel. Okay first it was the pcmcia package and then the kernel. But now I had to strip out all the #ifdefs because we don't want that cruft in the kernel, but I still had to maintain the #ifdefs for other kernels. So now I had all the previous problems, but now I had to make patches with and without the #ifdefs. I figured the problem would go aways once everyone moved to the new kernel.
4) Firmware changes. Oh no! Cisco changed the firmware which changed a bit the I/O interface. Oh and look they are still changing the API in the kernel. So I can patch the new kernel code to support the new firmware, but I can't expect everyone to upgrade kernel just for my driver. (I wouldn't even do that because the XXXX driver doesn't work so well in the latest kernel.) Now I have even more problems to deal with including everything from before.
5) Throwing in the towel. It became just too much of a time sink. Both sides of my driver was changing like mad (the hardware and the kernel API) and the poor users that were trying to make it all work with kernels that they wanted to use. All my time was being sucked up in maintaining the status quo and I couldn't work on anything new, so I turn the driver over to good hands and moved on.
Now imagine how nice would be if in the experimental phase I could release the source and a binary for everyone to use. I wouldn't have to tweak and recompile for every new kernel. Anyone would be able to just grab the binary and use it if they wanted to. (Kinda like Windows... Ironically I use ndiswrapper for my new laptop with a broadcom driver and it rocks! I've used the same windows driver in linux for the past year across many versions of the kernel. It sucks that the windows network driver ABI is the only driver ABI that linux has.) If the firmware changed, I or anyone else could fix it and everyone could use it.
Whether or not we Linux allows closed source drivers is orthogonal to an ABI. Technically you can write closed source drivers now and if you want to, you can prohibit closed source drivers with your new ABI.
It's funny how the masses always complain about windows being instable and crashing. I have news for you, windows does not (not XP, anyway). All the crashes I've seen in the past years were cause of crappy drivers (mainly nVidia and Ati). That's because Microsoft developers are stupid and can't develop a decent driver API? (because we all know that the average IQ in microsoft is something around 15, and that the legend of the 1000 typing monkeys... ok, but I ain't bein funny)
I'm perfectly aware of the reasons mentioned in the previous posts. But the decision isn't trivial. There is a compromise/tradeoff to make, and it ain't so simple. Is it worth to attract the masses? I doubt. the real deal for the masses is Apple... masses, in the IT world, don't need to be "free to do things". Because that translates in "free to damage things". So the real deal is a vendor that offers you everything from its own factory, all shipped up in a cool package, with just 4 models to choose from (ok, maybe even 20, doesn't make much difference). You run your thing, you don't need to know nothing, and IT'S FUCKING BETTER YOU DON'T. Everything works, as long as your needs conforms to the standard needs. I'm sure that 99% of the people would be damned happy with this. And to them I say, go Apple.
For me, this is just a cool version of the Soviet Republic. But I'm a mad anarchist, so just leave me alone with my crappy Linux, that needs days to be configured with mainstream hardware, but that does not "limit" me in any way
(Many would say mine is just a bunch of politic arg... idiocies. True. Because Linux and Open Source is a political matter. Being opensource does not give a project any mere technical advantage over a closed source one. It has its upsides in how the software interacts with the community of users/developers, and hence is a social/politic matter)
nbody2002:If you can read this you may be addicted to the internet
Some rebuttals:
1) A lot of the work involved in maintaining a driver is keeping up with in-kernel API changes. If the driver is in the tree, the kernel developers will handle these sorts of sweeping changes, and the hardware vendor doesn't need to worry about them. Also, the kernel developers are pretty easy-going about driver coding style, as it tends to be isolated.
2) If you can't see the code for the module, you don't know what the module did. If you load the module, and it tramples some memory, and you unload the module, that memory is still corrupt. This is why loading a module taints the kernel, even after the module is unloaded. It's impossible for the kernel community at large to debug kernels which have had closed-source modules loaded.
3) Getting the end-user to build and install a kernel module is not hard. Vmware manages to make it an automated part of the installation script. The advantage of shipping source to end-users is that those who are knowledgable will be able to make your module work on other kernel versions, while a binary version is by definition limited in terms of compatibility.
In other words, it would turn Linux into the same kind of piece of shit that Windows is, and defeat the entire purpose of using it!
"[Regarding the 'cloud,'] ownership was what made America different than Russia." -- Woz
Initial costs associated with a manufacturer-supported driver:
Ongoing costs associated with a manufacturer-supported driver:
These costs exist even if a version of the driver is merged into the mainline kernel. The only problem solved by such source-level merging is compatibility with the latest kernel version. It is not acceptable to the manufacturers' customers to be required to update to the latest kernel/distribution to be able to use the device.
Here's the key point: If there is no binary interface between the driver and the kernel, all of the above costs skyrocket. You have M kernel versions against N distributions, with the total increasing over the life of the product. If there is a binary interface guarantee from the kernel development team to change only very slowly and only extremely rarely breaking compatibility -- like the guarantee Windows provides -- then the incremental costs are containable. It is reasonable to expect that 95% of their testing on 2.6.5 is valid on 2.6.14.
The perfectly reasonable response from kernel developers is that with closed-source drivers they get stuck debugging problems that are't kernel-related (I don't hold ideology to be economically significant so I'll ignore it here, without insult to people's strong opinions on the subject). Their proposed solution is to require the driver's source before they'll help with the debugging.
From the manufacturers' point of view that's a very draconian requirement. They are justifiably concerned about intellectual property (availability of the source makes it much easier for competitors to reverse-engineer the hardware/firmware). Surely there must be a middle ground. Is there some way to have a relationship between the device manufacturers and the kernel developers that minimizes everyone's costs?
I think there is. Note that all of the above costs and issues are just as valid in the Windows world as in the Linux world. Microsoft doesn't want to deal with bad drivers crashing their systems, costing them both development/debugging time and reduced perceived stability (--> lower sales). Their solution is the Windows Hardware Quality Lab (WHQL).
The WHQL is a separate entity from Microsoft. Device manufacturers are required to submit their driver source (effectively under NDA) along with their device. The WHQL staff runs the driver through a battery of tests, probably mostly automated. If the device and driver meet stability standards set by Microsoft, the driver is signed by WHQL. Windows checks for this signature at installation time and warns the administrator if it is not present. Microsoft can reasonably refuse to support non-WHQL-signed drivers when crashes occur, for exactly the same reasons that Linux kernel developers refuse to support drivers without the source. This system has been the single most important factor in Windows' significan
Bandannarama
Have you actually read the HCT test spec?
s .mspx
http://www.microsoft.com/whdc/whql/resources/spec
E.g for one of a real time virus checkers that layer over filesystem, they test things obscure filesystem features like defragmentation support, and oplocks and reparse points. Plus there is a 14 day stress test. And they update the test to catch common bugs. Of course a decent company will do this sort of thing, but in practice most companies will release undertested code if they are under pressure.
So the HQL certification means that you run these tests. Then you fix the test failures and submit the log and binary to Microsoft and they sign the binary.
It's not perfect - as the Raymond Chen post says, but it's a lot better than nothing. But the cool thing is it doesn't really need to be perfect, if the drivers are visibly unstable, people will stop buying the hardware and the company will go bust. Both NVidia and ATI hardware is rock solid with the latest drivers, so who cares if they cheat a bit with 3dmark and HQL.
echo -e 'global _start\n _start:\n mov eax, 2\n int 80h\n jmp _start' > a.asm; nasm a.asm -f elf; ld a.o -o a;
Then tell us how they should solve IP issues connected with the knowledge resting in the driver. Tell us how ATI and nVidia should handle this. More and more functionality is in the driver nowadays, and the IP in there is littered with patents, royalties, licenses.... and there is also the fact that the drivers often contain company secrets, so better forget about the idea of companies putting their stuff as open source.
This sig does not contain any SCO code.
When the source code is available for modification, *everyone* benefits, whether they're reading it, hacking it, or just using precompiled binaries. When I need something new in software, I may not be able to code it myself, but I can put it in bugzilla, where others in the community can discuss it, assign it, and eventually implement it.
Of course the process is not perfect. It is a human process. That is part of its appeal -- the community supports its own software, rather than relying on a company to pump out binaries... who may not be around next week.
There is nothing elitist about saying that I am perfectly satisfied with what Linux provides, and there is nothing wrong with valuing the freedom that its open nature gives to all users -- remember, those freedoms benefit them whether they read the code or not.
From this user's point of view, the extra freedom, stability, and community that Linux provides are well worth the wide selection of ordinary hardware -- at ordinary prices -- that I choose to run it on. For me, the value of an operating system is that it provides a platform which offers freedom to me and to my community while we do our work. Putting off-brand generic wireless cards with quick-hacked binary-only drivers into a system does not help me do my work -- even on "supported" systems, that has usually just given me a headache and required a trip back to the shop -- and it would probably force some compromises in the freedom we value.
For you, it seems that an operating system is nothing more than a collection of software to be used unmodified and unexamined, while you do other tasks, without the need for a community to support you if you need to change it. Those needs are perfecly valid, and perfectly incompatible with the Free Software movement. I would be happy to help you select and purchase such a system from one of the many vendors who will sell it to you. You'll be able to plug cheap hardware with crashy drivers into it 'til your heart's content. For support, you'll be able to call the company, whom you'll soon find have invested even less in support than they did in the cheap hardware.
And if your needs should change, our community will still be here, and we'll be glad to help you find a system that meets them.
In fact, since linux already implements 3d support (dri/drm) inside the kernel, you might explain us what the hell you're talking about?
/proc/kcore!
Unless the DRI people are lying to me, no, that's not what DRI is. DRI is "framework for allowing direct access to graphics hardware under the X Window System". (And DRM is the "kernel module that gives direct hardware access".) Hmm, nothing about 3d support there -- just opening it up so user apps can play with the hardware directly.
Yes, it "integrates with Mesa, an open source implementation of the OpenGL API". That does not make it "3d support inside the kernel".
If you had a kernel module that allowed user programs direct access to your sound card, and could write a userspace program that implemented 3d audio for your particular sound card, would you claim that the kernel had "3d audio support inside the kernel"? This seems a truly bizarre definition. By this logic, the kernel has "support" for *everything* "inside the kernel": just su root and twiddle with
No, they don't, unless you are putting Windows XP and Windows 95 in the same category. Just about any Windows 2000 driver will work on any version through service pack 4, but with Linux I need to recompile my drivers whenever a minor version number changes. Joy.
Now before I get modded down, I be to remind whoever might read this that what I am saying is FACT. - bogaboga
See the long version of this post for expanded discussion of these points...
[By the way -- AKAImBatman, please send me some email, I'd like to chat with you further about Project UDI...]
Deven
"Simple things should be simple, and complex things should be possible." - Alan Kay
You do realize that MacOS X does not have userspace drivers, and is not a microkernel in the usual sense of the word?
Just because they used the VM layer from mach doesn't make it a microkernel. Though at least the graphics driver runs mostly in userspace (much just like X11 on UNIX, in fact).
Oddly enough, NT4 is still considered a microkernel by some, despite having GDI, all drivers and all filesystems in kernel. Linux+X11 is more of a microkernel than NT4!