Announcement

Collapse
No announcement yet.

Nouveau To Go Into Linux 2.6.33 Kernel!

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • #11
    "Up to this point unless building the Nouveau components from source, the only place to find this free software NVIDIA driver in action has been with Fedora."

    Not actually true; Mandriva has had nouveau available as an optional package for a while, now. You could always install it with a single command, on Mandriva.

    NVIDIA should be able to write their kernel module to work correctly even if the nouveau module is present - by forcibly unloading it if nothing else. So far, in Fedora, testing is somewhat unclear: it seems the two modules can co-exist in certain cases but not in others.

    Comment


    • #12
      Originally posted by HokTar View Post
      Now I think there's no point in backporting Nouveau to the 2.6.32 kernel for Lucid. Plus it carries a lot more new features for ATi/Intel users too, so I hope that the guys at Canonical will change their minds and ship .33 instead.
      I agree. Reading this news post, my main thought was that it would be great if Canonical shipped 2.6.33 in 10.04 LTS. Nouveau has been the default Nv driver in two Fedora releases now, so it has received a decent amount of testing while under development. The number of DRM changes pulled in for ATI/Intel graphics chips is also very enticing (at least to me).

      It's still early enough in the release cycle that this decision might be made (I hope so). They were planning on back-porting Nouveau anyway, and by switching to 2.6.33, they'd get a much more functional ATI OSS stack as well (interrupt support makes me very happy). And with 2.6.33 being released in January (right?), they still have 3-4 months worth of time to test it and work out any kinks.

      Comment


      • #13
        Originally posted by timofonic View Post
        Despite Nvidia denying the reality about FOSS, their "binary blob only" strategy will make them look at enemies in the FOSS world once AMD plan gets mature and feature complete. They are not collaboring to a project that did born as a phillantrophic project and now evolved to be partially financed by one of the most active companies and even capitalistic pioneers behind the Free Software movement, instead the makers of the hardware financing it as AMD is doing it smartly as a mid plan term goal.

        I think binary blobs are coming to an end, it's just a bunch of annoying drivers that are just making development of critical systems in Linux and other FOSS kernels/operating systems a TOTAL NIGHTMARE.

        There are projects out there just as a way of solving problems with binary blobs, they are seen as enemies because being totally contrary to the Free/Open Source philosophy and becoming too annoying in the development way too.

        I hope companies learn that outsourcing can be a way of self damaging themselves if not owning the knowledge received, this is pure strategy because those chains make them attached to "the evil" instead evolving in a free environment where their hardware can be used. But maybe they prefer their evil lords because they can make old hardware less usable by artificially removing support of them instead giving the tools to the community.

        This world of "intellectual property" and excessive consumism must come to an end, Free Software shows it's possible to do more with less and even able to get benefits in both a social and capitalist manner. Collaboring more and competing less aggresively is the way to go, companies just need to make the best products for consumers and facilitate their products being usable on any operating system available.

        Nvidia, please evolve and company. Learn from AMD and Intel, your competitors that are also microprocessor makers (and you don't, haha).
        Nvidia should learn from ATI? Last I checked ATI didn't have a real solution. Nvidia needs to learn to not provide a working solution?

        If ATI starts providing a viable open source solution that is feature competitive (Hint: the features have to actually WORK!) I'll start listening to you freetards. Until then all your arguments about "artificially removing support" are null and void. First ATI needs to artificially ADD support for that argument to even come close to working. You can't proclaim victory if you aren't even in the race.

        ATI get up off your ass and actually give NVidia some incentive to open up. Show them that an open strategy can actually pull it off. ATI, put your money where your freetard's mouth is. Stop developing your closed blob entirely and start actually sending patches to the open driver.

        Show my freedom hating ass how it's done... please for the love of god!

        Comment


        • #14
          Originally posted by Jimmy View Post
          Nvidia should learn from ATI? Last I checked ATI didn't have a real solution. Nvidia needs to learn to not provide a working solution?

          If ATI starts providing a viable open source solution that is feature competitive (Hint: the features have to actually WORK!) I'll start listening to you freetards. Until then all your arguments about "artificially removing support" are null and void. First ATI needs to artificially ADD support for that argument to even come close to working. You can't proclaim victory if you aren't even in the race.

          ATI get up off your ass and actually give NVidia some incentive to open up. Show them that an open strategy can actually pull it off. ATI, put your money where your freetard's mouth is. Stop developing your closed blob entirely and start actually sending patches to the open driver.

          Show my freedom hating ass how it's done... please for the love of god!
          AMD is actually sending patches. Where do you think work Richard Li and Alex Deucher?

          Comment


          • #15
            Right. AMD/ATI are heavily involved in the development of the free stack. Which already works pretty solidly for r500 and lower and surprisingly well for r600. r700+ should be working very soon, as well.

            Comment


            • #16
              Originally posted by Jimmy View Post
              Nvidia should learn from ATI? Last I checked ATI didn't have a real solution. Nvidia needs to learn to not provide a working solution?

              If ATI starts providing a viable open source solution that is feature competitive (Hint: the features have to actually WORK!) I'll start listening to you freetards. Until then all your arguments about "artificially removing support" are null and void. First ATI needs to artificially ADD support for that argument to even come close to working. You can't proclaim victory if you aren't even in the race.

              ATI get up off your ass and actually give NVidia some incentive to open up. Show them that an open strategy can actually pull it off. ATI, put your money where your freetard's mouth is. Stop developing your closed blob entirely and start actually sending patches to the open driver.

              Show my freedom hating ass how it's done... please for the love of god!
              If you are talking about the ATI fglrx binary driver then that makes sense. Fglrx never really has worked that well at all, while nVidia's has worked better.

              Nvidia should support the nouveau project in my opinion as the code's come a long way and very promising and much of that code can be reworked if documentation is ever supplied from nVida

              Comment


              • #17
                Originally posted by stikonas View Post
                AMD is actually sending patches. Where do you think work Richard Li and Alex Deucher?
                And Cooper Yuan...

                Comment


                • #18
                  Originally posted by DeepDayze View Post
                  If you are talking about the ATI fglrx binary driver then that makes sense. Fglrx never really has worked that well at all, while nVidia's has worked better.

                  Nvidia should support the nouveau project in my opinion as the code's come a long way and very promising and much of that code can be reworked if documentation is ever supplied from nVida
                  It's ok. It'll just make nvidia look like jokes because their stuff will run slow just because they didn't provide the documentation. I don't think it will be a 4550 and S3's 540GT embarrasing 8800GT but it's going to be noticable.

                  Comment


                  • #19
                    Originally posted by Jimmy View Post
                    ATI get up off your ass and actually give NVidia some incentive to open up. Show them that an open strategy can actually pull it off. ATI, put your money where your freetard's mouth is. Stop developing your closed blob entirely and start actually sending patches to the open driver.

                    Show my freedom hating ass how it's done... please for the love of god!
                    You don't understand. The kernel is going to suck up every GPU on the planet and put it inside. Because the kernel is going to be not only a CPU kernel but a GPU kernel as well and it's going to call either or both or everything anytime it wants to. Linux is not evolving to a mature OS with good graphics support. It's evolving to a new processor usage system. And unfortunately AMD NVIDIA and Intel are too stupid to do this on their own. Intel starts drooling out it's mouth everytime it meets complicated floating point calculations, nvidia wants to make everything just the way IT wants it to be and be a total monopoly and AMD just wants make it through another year without buying a vulcan cannon and opening fire on it's sick twisted competitors.

                    Comment


                    • #20


                      I like the last one

                      Comment

                      Working...
                      X