Slashdot Mirror
Torvalds Slams NVIDIA's Linux Support
New submitter jppiiroinen writes "Linus Torvalds received the Millennium prize last week for his work on Linux operating system. He was already in Finland, so Aalto University arranged a talk session with him (video). During the Q&A, a person asks why NVIDIA does not play well with Linux. Torvalds explained shortly that NVIDIA has been one of the worst companies to work with Linux project — which makes it even worse that NVIDIA ships a high number of chips for Android devices (which use Linux inside). Torvalds even summarized that ('Nvidia, f*** you!') in a playful manner. What has been your experience on NVIDIA drivers with Linux?"
AMD provides specifications and a small developer team that actually works on open source drivers.
Intel provides open source drivers.
NVIDIA makes good binary drivers, but those have problems when a new kernel version comes out with changed interfaces:
Only NVIDIA can adapt them, and until they get around to it, NVIDIA may not work with the latest kernel version.
C - the footgun of programming languages
Linus did give Nvidia the bird and a "fuck you" but he never slammed the quality of the Nvidia's hardware. His gripe with Nvidia is how hard it is for Linux to work with the company since they only provide a binary blob driver which makes bug fixing for it dependent on Nvidia's whims. Plus they refuse to even provide specs and API's for their hardware which make writing open drivers much more difficult and time-consuming because of having to reverse-engineer everthing to get a workable driver. In this case, Linus is absolutely correct.
I think you and many here misunderstood what Linus was talking about. Linux specifically said NVIDIA is one of the worst companies with whom they worked. Period. There is no need to bring up ATI into the table. There is no need to bring up that your card has always worked beautifully. This is talking about his and the kernel maintainer's experience in dealing with hardware vendors, something that we ourselves never have to deal with. Their proprietary drivers may be the best ever but that has nothing to do with!
If the kernel maintainers have a question about the hardware, they can't ask NVIDIA they have to test and reverse engineer to find the answer whereas with other companies, they may get an answer directly from the manufacturer. Get it? "...NVIDIA just made the damn drivers. Now that is not good enough." Not from a kernel maintainer's or Stallman's point of view, I'm pretty sure.
"Top notch" might be overstating it a bit, there are still a few instances where they seem to be lacking. For instance, Windows users who have SLI and multiple monitors have been able to enable SLI and use both of their monitors at the same time since about 2008. But under Linux, no dice.
So if I had two monitors (which I do), and two Nvidia GPUs in SLI mode (which I do), and I wanted to run some 3D app that took advantage of SLI, I would have to: reconfigure X to disable my second monitor and enable SLI, restart X, play the game/use the app I wanted, when I was done I would have to reconfigure X again to enable my second monitor and disable SLI, restart X again, and reopen all my apps. Hardly ideal.
http://crummysocks.com
NVidia isn't required to, as you put it, "subscribe to the projects goals and visions".
But Linus Torvalds is also not required to enjoy or approve of NVidia's policies, particularly when they generate bad publicity for Linus Torvalds' project and also cause a number of people to complain to Linus Torvalds about a problem which he didn't caused nor can he do anything at all about it. Hence, a very appropriate and sorely required "Fuck you, NVidia".
Slashdot, fix your code or at least hire someone who is competent at it to do it for you.
I wasn't aware that NVidia had their own OS. So NVidia makes no Windows or OS X drivers then?
Well, there's spam egg sausage and spam, that's not got much spam in it.
> If it costs them money to support Linux and they're damned if they do, damned if they don't, then what incentive is there to continue even supporting them at all?
It doesn't.
I think people nowadays just don't understand what Linux is anymore.
This discussion has nothing to do with desktop, performance or even servers.
Nowadays, Linux is shipped on 900000 phones every days and that's a very lucrative business.
Nvidia has a share of this with its Tegra chips but their way of supporting it makes it a pain in the ass for others inside the kernel development community.
From that point, Linux complaining is totally fair. PR is basically his only weapon.
It isn't NVIDIA's job to make drivers for someone else's OS.
If nVidia does not make drivers for someone else's OS who, how do they make it for? They don't have their own OS. Their entire business relies on someone else's OS. So does nVidia make drivers for Windows and OS X or not?
Well, there's spam egg sausage and spam, that's not got much spam in it.
Nvidia drivers were actually the driving force behind the concept of kernel tainting where an oops dump reports that non-free modules were in use. Nvidia drivers were creating a huge number of oops reports that couldn't be debugged due to being closed source.
This story starts more than a decade ago. There was a hugely popular software vendor concerned that maybe one day people might choose to not use their software. They had vast sums of money and controlled access to the immediate future for software and hardware vendors alike.
Foreseeing a potential difficult future they chose to defend themselves in a particular way. They formed subsidiaries they controlled and gave them patents, and filled them with developers skilled in the finer (and secret) nuances of how to interact with their software, and they kept them informed with advance knowledge of how it would work in the future.
These subsidiaries approached hardware designers with a simple message: they would accept the patented technologies and use them; they would let the subsidiaries write the drivers that had special hooks into the software; they would do this under non-disclosure and never tell - or they wouldn't. If they accepted they would not be able to publish open specifications about how their own hardware worked because that would be exclusively cross-licensed with the subsidiaries in exchange for access to the patents. The hardware makers who wouldn't play along wouldn't get as good compatibility with the big company's software, nor inclusion in their distribution CD and OEM images. The refusers would be plagued with difficult installation, buggy drivers and unhappy customers and fail in the market. The software would change in ways the refusers could not predict, but the accepters could. Some accepted, and some refused. Those who accepted survived, those who refused mostly died.
This has continued to the present day and as the hardware has evolved the agreements persist in ways that are now not removable.
Nobody involved in Linux wants hardware manufacturers to write the device drivers for them. They only want open and clear specifications for how the hardware works so they can make their own drivers. They aren't going to get that from NVidia, nor ATI, nor any others whose technology is intertwined with this compromise from yesteryear. This boon is now beyond their ability to grant without starting again from the beginning.
Help stamp out iliturcy.