Slashdot Mirror
NVidia Releases Linux Drivers Supporting 4K Stacks
Supermathie writes "NVidia has finally released drivers for their chipsets and the 2.6 kernel that support 4K stacks. That means compatability with Fedora Core 2 kernels, people! View the README, visit their driver page, or download the package."
The real story is when they open the source to the drivers.
thisnukes4u.net
Ok... wtf is a 4k stack?
||| I still can't believe Parkay's not butter.
I will miss thee.
Now I can get my ass kicked in Enemy Territory under Fedora Core 2. I was missing that but for some reason, I got so much more work done. :)
sigs are like a box of chocolates, they all suck remove the underscores to email me
For me the best way to test these new drivers is to play Enemy Territory
One of the best online FPS games and it's free-as-in-beer.
Keep up the good work NVIDIA.
http://www.kubuntu.org/
It seems that this driver's OpenGL headers are a little buggy, but the solution was given by NVIDIA employee in this thread at nvnews.net forum.
I've been testing these drivers under Fedora Core 2 for a while, and they appear to work flawlessly.
Thad
I love Mondays. On a Monday, anything is possible.
For people who are building home theater PCs for things like MythTV, this is a major step forward. The last release that supported overscan (so that a TV image doesn't have black stripes on the sides) was many releases back (version 4363). This release not only supports Linux 2.6 with 4K stacks, but has overscan and interlace support, making it ideal for TV and HDTV display.
...with the latest 2.6 kernels, simply turn off 4K stacks. But hey, now it's not necessary. Yay.
4k stacks are a good thing, a first step for Linux to support an insane amount of simultaneous processes on the system.
"Backups are for wimps. Real men upload their data to an FTP site and have everyone else mirror it." -- Linus Torvalds
this is a cut/paste of this article. Unless you actually wrote it, don't copy with no reference.
I submitted this story last night, and it didn't get posted.
Wonder why this story was rejected when I mentioned it 4 days ago and then submitted the story.
>They might have a slightly different perspective.
Yes, the IBM PC XT was a complete POS that couldn't compete with anything else out at the time. Almost nobody used it, apart from a handful of people. That garbage computer didn't even include a decent sound system, for crying out loud!
Then the clones came.
And the XT architecture became popular.
And IBM sold more PCs than they ever thought possible.
If you could be told what you can see or read, then it follows that you could be told what to say or think - BoC
The "zealots" just want to control their own computers. That's what Open Source is about. If you have an nVidia video card in your linux system, and you want full functionality, you have to let nVidia control your computer.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
I'll karma whore, since I read the article linked above.
If you allocate memory in 8k stacks, the kernel's got to find 2 pages of memory together. Which I guess gets to be a pain as uptime increases. Since memory pages on most hardware are 4k, it's easy as pie with 4k stacks. Plus, you separate some of the kernel stuff like software interrupt handlers to their own stack (I think that's what it was), hopefully making the system more stable in the process.
Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
I think I'll wait for this to be tested for more than 24 hours before I try my hand on it.
The greatest experience we can have is the mysterious.
- Albert Einstein
Or go out and revise the nv drivers. Nothing (I would assume) prevents you from doing so. Nothing prevents you from getting a video card from another company. And by the same token, nothing should prevent NVidia from releasing closed-source drivers.
Besides, what would 99.9% of linux people do even if it was open source? Download source, not even look at it, type make install clean, and be done with it. (Or make setup or whatever the build sequence is; point being that most users wouldn't care.) And for the 0.1% of people who do mess with it, unless they discovered some great tweak that would provide a significant feature or speed advantage over the NVidia drivers, I'd just go with them, since I trust them more since the quality of their drivers partially determines their sales, and thus they have a bigger motivation to make them better.
so finally nvidia got its act together. i wonder what took them so long?
They are still ahead of the game with Linux compared to ATI. ATI only just got Linux drivers out a few months ago. NVidia has had Linux drivers for at least around 2-3 years now (I didn't really care about it before then), this is just about them getting the 2.6 kernel drivers (and new chipsets). Also, to my understanding, ATI's Linux drivers arn't all that good, and they have yet to support the 2.6 kernel.
So really, if you want a brand name video card that supports Linux, NVidia is the way to go (at least for now).
Mods will probably sink this for saying it. But I think they deliberately go against any atempt to predict thier down modding.
Mycroft
https://signup.leagueoflegends.com/?ref=4c3ed6600b6ea
I don't know about you guys, but I think having the source code to recompile it manually would help out immensely.
That's funny, I don't.
First, fixing this stack size problem is not a simple re-compile of the same code. Depending on how the driver is written this is certainly a non-trivial task.
Second, even if you had the source that does not mean that you could distribute a fixed version. Open source != Free Software.
Third, they may be closed source drivers but they are miles ahead of the current FOSS drivers. The Zealots can run their "pure" systems and suffer graphics glitches and poor 3d performance. I'd rather just use something that works. If that meant sticking with by old kernel a bit longer then so be it.
they just don't want to fork it over because somehow you may "magically" make the component up yourself out of basement and not have to buy it.
Not you - their competition. ATI has always been plauged by crap drivers. If ATI had a peak into how NVidia does it you can be sure they'd take something away from it. NVidia would lose a competitive advantage. The GPU war is nasty. The competition is killer - they'll take any advantage they can get.
Okay, I'll take a stab at it. The bedrock of capitalism is simply a market system based on private ownership. Now, most people want to extend that into monopolistic control to maximize their own profit in self-serving interest, but at the same time consumers generally tend to want a lot of competitors that can offer substitutes that give them greater value.
But, the fact is that if IBM hadn't "goofed" and created a mostly open system, it's likely that either another more open system would have succeeded even though it had a lot of obvious fault or no system would have succeeded and the information age wouldn't be near the point it is. Why? Because a more open system allows for programmers, both hobbyist and capitalist, to more easily develop software for the system. This barrier to entry would mean less software overall which would directly decrease the demand for computers. At the same time, monopolistic control would keep prices high, fixing the quality sold at a smaller rate than it is today thanks to the vast number of clones.
So, it's unlikely IBM would have a better market share or sell more products. They might, still, be making more profit due to monopolistic pricing. It does seem unlikely for this to be the case, however, when various other architectures would have likely succeeded in IBM's place and relegated IBM computers into dinosaurs like the Amiga (no offense to the Amiga intended).
As for NVidia, there's at least two principle reasons why they might wish their drivers closed. The first is by closing the drivers they have stronger control over rebranding cards at different price points without modifying hardware which might increase sales without hurting sales on the higher priced cards. The second is NVidia has cross-licensed a variety of patents which probably puts them in the position of not having the authority to license said patentable idea under the GPL.
Without number two, number one could be fixed with creative hardware locking mechanisms. The total cost of such hardware locking would be minimal in comparison to the boosted sales of all the likely free porting and driver work done by volunteers on the NVidia driver. The fact is, NVidia is a hardware company so it is in their best interest to commoditize all software for their hardware to be run on. Open sourcing their driver, if possible, would very likely have this effect (it's hard to argue that it could have the reverse effect, at least).
The claim that trade secrets would somehow be revealed by open sourcing their driver is possible, but I would guess is unlikely as the majority of NVidia's actual trade secrets would be in *hardware*. All a driver is supposed to be is a standard interface for the OS, and if there are tasks beyond this in the driver NVidia would almost certainly advantage by sticking it in hardware as well. It's for this reason I assume NVidia's driver license policy is the main fault for them not open sourcing their driver.
Eurohacker European paranoia, gun rights, and h
It's not like nobody can do it...
Thank you.
This is...
O
U
T
R
A
G
E
O
U
S
!
Dammit, you may be right, I swear it was an accident. I was looking for funny. Crap and I try to avoid most slashdot 'sins'. Well at least I've managed to avoid trollery and Flaimbaitage. I've given up on off-topicness along time ago.
Mycroft
https://signup.leagueoflegends.com/?ref=4c3ed6600b6ea
There is magic in their drivers, and it is explained EVERY SINGLE TIME NVIDIA GETS MENTIONED HERE. It's called a special OpenGL license from SGI and it's also some special in-house code.
Try to remember it this time, it's only the 400 millionth time it's been mentioned.
Closed source software has its places,
But drivers is not one of them. Had they put the closed source code in a user mode library and used just a small open source kernel driver, we wouldn't have all the problems with the driver. It still wouldn't be optimal, but it would be way better than the current situation.
Do you care about the security of your wireless mouse?
Yeah, this is why I use the term zealot. I like Linux (anyone bringing up GNU/Linux can bite me. I know the history and respect GNU. It's unweildy), in fact I prefer it. But I call BS on the altruism of the philosophy behind much of the movement.
If you wanna say "here's our stand, and we stick by it", I respect that. If you say "any stand but ours is unholy and wrong", then you are attempting to control and I have no use fer ya.
I wouldn't violate the GPL, as a programmer I respect other coder's work and time. But I also don't buy into the demand that EVERYTHING be GPL's, or whatever license you prefer.
The world ain't black and white kiddies, time to realize the intelligent people have differing opinions most of the time...
"Faith: Belief without evidence in what is told by one who speaks without knowledge, of things without parallel." - A.B.
Are there any video card manufacturers left who release other than binary only drivers?
Matrox releases open-source drivers for some of their product lines (e.g. the Millenium G series -- G400, G450, G550, etc.). The mga driver that comes along with X is the same as Matrox's, for that reason. And 2D performance under the open-sourced Matrox drivers is actually pretty damned good. This all sounds great, doesn't it? Unfortunately, Matrox's Linux support sucks, and the support for Matrox from the DRI project is fairly nonexistent right now. So if you do have any problems with the driver, or want to get 3D/DRI/hardware acceleration issues solved, you're gonna have to learn to hack the drivers/kernel modules yourself. Good luck.
Man, I just installed these drivers (I was wanting a good excuse to do it, I admit it) on my ancient TNT2 video card, 800mhz Duron, blahdy-blah blah, and now Metisse is running fine. Before, with the nv driver Metisse barely ran. Amazing how much a difference the driver makes. ;)
Like what I said? You might like my music
Except that place is not hardware drivers - all the pain and suffering getting 3D working on Linux when most hardware is a breeze is proof of that.
Methinks the only real reason you'd want to keep your drivers closed off is because you're artificially handicapping your hardware
Um, no.
0) nVidia might not own all the code they compile into their drivers. The license they have the code under might permit binary distribution, but not source.
1) nVidia's drivers contain large amounts of software that is better than any of their competition. They spent money developing this, and they want to milk the competitive edge it gives them. And that is okay.
2) nVidia has more control this way. The Firefox guys are holding control over their cool icons, because they don't want the cool icons slapped onto broken code; only Mozilla-official builds of Firefox get the cool icons. nVidia might want to be sure that no one runs with broken drivers, then thinks nVidia cards are all junk, when in reality some guy made a few "improvements" that broke things, and distributed the changed version anyway.
3) Other reasons are possible. "the only real reason" my left foot.
Personally, I would much much rather have FOSS drivers. But even more than that, I want drivers that work. I switched from a GeForce 4600 to a Radeon 9600 XT, and even though the Radeon is a much better card, it runs slower under Linux than the older GeForce. It's the drivers. ATI's Linux drivers for the 9600 XT are lame. I actually boot into Windows to play Unreal Tournament 2004, because the performance is so much better under Windows. When I had an nVidia card, my Linux 3D gaming performance was just fine.
If nVidia would make a programmable-shaders card that doesn't double as a space heater, I would probably buy it and replace the Radeon. I know that the Unreal Tournament guys check the server stats, and I want to be "voting" for Linux gaming, so I want them to see me running Linux when the check stats on the servers I have been visiting.
steveha
lf(1): it's like ls(1) but sorts filenames by extension, tersely
Actually, this is the idea. The interface with the kernel is open source; the closed source code is a binary object that gets linked into the module. Sorry, user mode doesn't really make much sense here, drivers need full hw access and context switching to a different privilege level would only hurt performance.
The rest is pretty much trolling, at this level. NVidia has been so far quite open source friendly when it comes to producing drivers. But I guess there will always be people to complain. Me, I'm happy NVidia has drivers for platforms where theirs is the only accelerated choice, like amd64. Others would say the same about IA64, or FreeBSD. Windows and Linux on x86 aren't the only games in town, you know.
Finally, how do you know they don't stand to lose something by making the drivers fully open source? look only at the whole 12 pipelines vs. 16 pipelines thing going on between the latest NV and ATI cards, with last minute info prompting new cards on both sides. If NVidia releases drivers for their last generation of cards that take the competition a couple of months to disassemble and analyze, they might keep the edge long enough to move on.
Let's hope ATI follows suit.
It took 2 third party patches and a recompile to get it their driver to install on Fedora Core 2, and it still crashes WineX.
All that we see or seem is but a dream within a dream.
Also, releasing the source would allow the drivers to be compiled on the systems with your gcc optimizations instead of being forced to use binaries, which has nothing to do with whether you're going to modify the source or not. One of the biggest things about my Gentoo box was that you build everything optimized for your hardware, whereas these binaries have to be much more general. Sure, there may not really be a terribly significant difference, but it's just one more reason why it should be open source.
Finally, to get back to your first point (am I going in reverse?) You really can't revise the nv drivers because they're compiled binaries. Nothing is stopping you from modifying them except the little thing that it's not accessible code to modify, since if it was this thread wouldn't have started. As for getting another company's video card, the options are ATi, and Matrox, neither of which are any better in this regard, and in fact ATi is much worse, so while you are right that nothing is preventing me from buying someone else's video card, it's not the point since no one is playing nice with OSS (AFAIK, Matrox might actually be nicer about it all, but they're not really accessible to the public the way the other two companies are,) leaving penguin worshipers with no options. nVidia is the lesser of two evils to be sure, so they get my money (that and the awful ATi driver issues with Windows XP, but that's a different story) for now, but really only because no one is better. Saying nothing is preventing us from going elsewhere really seems to sidestep the actual issue by blaming the users for something we really can't avoid because the best solution is a partial one. Well, anyway, that's just my $0.02
I'm always right and I can prove it, because to the best of my knowledge, I've never been wrong.
You were trying to do very intrinsic things to Linux and then complaint. Such low level changes are hard to do on any system. The good thing - they are unnecessary for novice user. For you, as an inexperienced linux user, I would recommend to use FC2 installer and then just use the system. It has web browser/office suite/etc. Read Fedora support site - some things(NTFS) you were trying to do are not supported by Fedora yet. This would save you some time.
Actually, this is the idea.
Huh? There certainly is a difference between what I described and how the driver is currently working. Maybe my suggestion would require changes of the binary code and/or the X server, but it certainly should be possible.
The interface with the kernel is open source
That statement doesn't make any sense. You can say the interface is open, and you can say the kernel is open source. But an interface is something more abstract than a piece of code. Of course when talking about interfaces to the kernel it is important to keep in mind, that there are two different interfaces. There is the user/kernel mode interface. This API complies (mostly) with various standards: Posix, BSD, Single Unix specification, SysV. But the standards only specify the API, not the ABI which is Linux specific. This ABI is kept as stable as possible even across kernel versions. But this interface is not really important when discussing kernel modules. The functions kernel modules can link against may change, and no attempt is made to keep the ABI stable, only the API is kept stable within each major version as long as the API doesn't turn out to be a major problem. This API is however the same across multiple CPU architectures (unlike the user/kernel ABI discussed before). But this really means that if you want to ship a Linux kernel module, you have to ship it as source. Because it is only at the source level there is a well known interface. A fixed ABI is just not possible, just the differences between CPU architectures is enough to make it impossible, but in addition some data types in the kernel are different depending on the options. And finally there are stuff like the 4K stacks where the current macro had to be changed.
the closed source code is a binary object that gets linked into the module.
And that is a problem. Not only does it only work on one architecture, but it makes assumptions about the kernel, which may not be satisfied. The amount of stack space is not the only problem here. The code can break the kernel in various ways, which means you can no longer trust your kernel.
context switching to a different privilege level would only hurt performance.
Some years ago I did some meassurements of this on a computer, that is now five years old. It could do one million switches from user mode to kernel mode and back again per second. I believe newer machines can do a bit more than that. I guess very few people have a monitor refresh rate of more than 100Hz. That means you will have time for about 10000 switches. Of course you can't use all your CPU time just to be switching, but let's say you can do a single frame with less than 1000 switches, then you certainly wouldn't have a performance problem. And if more than 1000 switches are required to do a single frame, then you have a broken design that needs to be fixed. It is not the amount of data needing to be transfered that is a problem, because you could either map board memory directly into the user mode process, or (a litle more complicated) do DMA directly to user space. So I won't believe your talking about performance problems, until I see a proof that it can't be avoided.
NVidia has been so far quite open source friendly when it comes to producing drivers.
NVidia have not really been that friendly. They may seem friendly when compared to other 3Dgfx manufacturers. This really just means there is a market, where no vendor give a damn about their customers. I hope some vendor will realize this, because if they do, and make the product the customers want, then I believe they can make some money.
Do you care about the security of your wireless mouse?
I must admit-I am a bit suprised that SLASHDOT didn't pick up on it. It might just be a little insignificant thing which doesn't warrant much attention anyway-who knows. Of course everyone is mentioning the support for 4k stacks. And of course this is important. Anyone who has used Andrew Morton's patch set knows what a PITA this issue was. But nvidia even did more than fix the single most blocking issue regarding their drivers and the 2.6.x kernels.
They also:
Added support for ACPI
Fixed problem that prevented 32-bit kernel driver from running on certain AMD64 CPUs.
Added support for GLSL (OpenGL Shading Language).
along with the new nvidia-settings utility-GPL'ed and written in GTK2....
and finally they added:
Added a new Xv adaptor on GeForce4 and GeForce FX which uses the 3D engine to do Xv PutImage requests.
Now I am not an expert on such things-25 years of experience and I am still left asking more questions than my ability to answer. _But I noticed this little innocuous "xv" thing and was like WOW-cool. I leave it up to those who know more to shoot me down-but doesn't this little "xv" thing mean that all those Linux users who use nvidia GeForce4 and FX cards suddenly got a a tremendous boost when doing much of anything with video ? After all XV is what all of the video players under Linux use for good quality full-screen video(mplayer, xine, totem, gxine, helixplayer etc.)
Now if I understand this correctly everytime a PutImage() request comes along under XV this is handed over to the 3D engine-automatically. It seems as if this would be a very, very significant reduction in CPU usage-particularly for older generation(PII/PIII) machines which happen to have fairly modern graphic cards. Full-screen divx under mplayer with the new drivers uses 12% CPU on average on my machine-I unfortunately did not do a benchmark to test this-but if my memory serves me correctly this is significantly less than what is was with the older drivers.
Now the downside to this-at least for the time being- is that some apps don't quite work with these new changes-Xine-and it's siblings(totem,gxine, kxine etc.)
But I assume these will be fixed pronto.
Well where am I going qoing with this train of thought:
Putting this kind of support for XV in the NVIDIA drivers -is really simple for the NVIDIA guys-perhaps even trivial-but it can mean a tremendous improvement for the users of these cards. NVIDIA has always treated Linux like a second class citizen-but hey who can complain-at least they acknowledge that Linux exists-compared to the BSD's Linux support is great-of course only if you are using x86 CPU's. Now everyone knows that the graphic workstation market has all but disappeared. But what if NVIDIA was to decide to simply really take advantage of the X11 windowing system and it's features.
Imagine if NVIDIA would actually provide good RENDER support-wow what a difference that would make for 2D desktop support-particularly under GNOME which uses RENDER extensively in VTE/PANGO-ie. why text scrolling in gnome-terminal is so abysmal. I am still stumped by the fact that the open-source X11 nvidia drivers support RENDEr far, far better than NVIDIA's own in-house drivers.....
Imagine if NVIDIA would really support the libfixes, libdamages and libcomposite extensions which are currently being developed at Xorg-X11. Sun's Looking Glass is already using libdamages and libfixes-I got it up and running on my machine yesterday-and yes it is still pre-alpha-but I have never, ever seen such a fluid desktop environment. This tech is almost *evil*- the promise which it presents is simply baffling-rendering all previous X11 windowing experiences to the days of the stone age. I don't really care that much about Looking Glass-if NVIDIA properly supports the X11 extensions we will have cairo-enabled desktops inside of the next year which will fundamentally alter the X11 experience for X users.
Ok. So here is the point of this little essay: If NVIDIA would simpl
Actually, this is the idea. The interface with the kernel is open source; the closed source code is a binary object that gets linked into the module.
That works great if you can guarantee separation. Otherwise debugging is a nightmare, knowing that there are some black boxes in your system which can manipulate the whole system.
Sorry, user mode doesn't really make much sense here, drivers need full hw access and context switching to a different privilege level would only hurt performance.
Right, that wouldn't work too good - but if everything runs in kernel mode then there is no border control between the driver and the rest of the kernel. The driver has to be trusted to play nice and not to fuck up the kernel data structures, because there's nothing that can stop him doing that. It would be different if the driver ran in user mode, because then the driver would throw segmentation faults and the like if it does something illegal.
The conclusion is that source code should be available for everything that runs in kernel mode.
"Unfettered, unrestricted capitalism is a fantasy" - only because any time it shows up, regulation follows along behind. (Similar dilution happens with other "pure" implementations of economic theories.)
"only in the last couple of decades"? what about the great big monopolistic empires of the late 19th Century? Standard Oil? United Steel? J P Morgan and Carnegie? The railroads? These are the reasons the original anti-trust laws were passed. (And before that, go back to the East India Company and the other government-licensed charter companies.)
Corporatism is a political system which is not at all at odds with "pure" or even regulated capitalism (an economic system).
People are always going to label people or arguments they dislike with names for other things they dislike, whether it's "you poopoo-head!" on the playground, "communist" or "Nazi" (or "capitalist") for adults. Are you hoping that's a "sign of the end days of this world view"? Keep hoping, 'cause people have been dismissing (or attacking) other people as "socialist" or "communist" pretty much since those terms were coined.
I agree with you! Closed source software has its places, just as open source software does.
I have been arguing this for years. Part of "Freedom" is choice, and having the choice to release your source code or not, just as I have the choice to use open or closed source applications. Abuse of a monopoly is not the same thing as closed source.
It is ironic that some (but not most) of the advocates of Open Source rail against anything that is not Free. This intolorance is why they get compared to "commies" and socialists, taking a position that "either software is Free or it should not exist". Fortunately, most of us who are Free software fans don't share their intolorant views.
If a company wants to keep their source closed and try to actually make money SELLING it, fine. If someone wants to make a Free version that does basically the same thing, even better, because then we have a choice, and the MARKETPLACE decides.
Tequila: It's not just for breakfast anymore!
There are still serious bugs left from the previous revision, which was six months ago. That's a bit long to wait. While GLSL and 6800 support is nice, an interm bugfix wouldn't be unwelcome...
A problem that leaves the console framebuffer blank after X is started remains. You need to work around it by adding
Option "IgnoreDisplayDevices" "TV"
to your xorg.conf. If you are actually using TV out, this could be a bit annoying.
Even worse, it hasn't been more than 24 hours since I've installed them, and these drivers have already hung X twice. When an OpenGL process segfaulted, that process assumes state D (uninteruptable sleep), and becomes completely unkillable, along with X itself. I haven't figured out how to reboot cleanly once this happens. All I can do is ssh in, sync the disks, and hit the power button.
Fred
"A fool and his freedom are soon parted"
-RMS