Announcement

Collapse
No announcement yet.

VIA's DRM Patches Finally Get Some Feedback

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

  • VIA's DRM Patches Finally Get Some Feedback

    Phoronix: VIA's DRM Patches Finally Get Some Feedback

    A month ago we shared that VIA tried again to push some new DRM code into the mainline Linux kernel. This was months after VIA Technologies had already tried multiple times pushing new Direct Rendering Manager code for its hardware into the kernel, but failed for various reasons...

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    It's funny when you say:

    "VIA tried again to push some new DRM code into the mainline Linux kernel."

    It's as if no one wants their code

    Comment


    • #3
      Originally posted by bulletxt View Post
      It's funny when you say:

      "VIA tried again to push some new DRM code into the mainline Linux kernel."

      It's as if no one wants their code
      Last i knew this DRM code was only going to work with the closed source Xorg driver, and VIA had said there were no intentions to make it open source. So i can see why the Kernel devs would care more about other things than this code. I may be wring though it's been a while since i followed this via hardware.

      Comment


      • #4
        Originally posted by bulletxt View Post
        It's funny when you say:

        "VIA tried again to push some new DRM code into the mainline Linux kernel."

        It's as if no one wants their code
        Well...

        (in no particular order)

        ~) The kernel has very strict style guidelines. There's a helper script, checkpatch.pl, that all incoming patches should pass, and additionally, there's other rules concerning versioning, ioctls, etc. VIA hasn't been the worst offender, by far, but they're not perfect on this front yet.

        ~) Code in the kernel should expose at least one of two kinds of interfaces: Higher-level kernel interface, or userspace interface; and either way, that interface should have some real-world consumers, for testing purposes if nothing else. A fair amount of this code only interfaces with VIA's closed-source userspace driver; that's pretty lame, since it means maintenance can't really happen.

        ~) Frankly, VIA's not really a member of the community these days. Even nVidia has been more forthcoming towards Xorg and Linux; Aaron Plattner @ nVidia is an Xorg member and shows up to conferences. Off the top of my head, I don't think there are any VIA employees connected to Xorg; Harald Welte has advised them on code, but he's not an Xorg guy; Thomas and Luc have written scads of KLOCs of VIA-related code, but they're not really connected to VIA AFAIK.

        ~) Slightly related to the previous note: I have no idea what VIA's roadmap is WRT Xorg. I know everybody else's: Intel is squeezing yet more chars/s out of their little IGP chips, AMD's trying to finish off OpenGL 2.0 caps on most of their chipsets, nVidia's getting libvdpau integrated with DRI2 and making it more robust, VMWare is continually improving Gallium. No idea what VIA's doing, and Google is coming up empty-handed, suggesting I'm not the only one in the dark here.

        If Luc, Thomas, or Harald want to correct me, they're more than welcome, but as it stands, I honestly find myself kind of at a loss as to why I should care about this stuff.

        ~ C.

        Comment


        • #5
          Originally posted by MostAwesomeDude View Post
          and Google is coming up empty-handed, suggesting I'm not the only one in the dark here.
          It is probably a very safe bet, that Google's interest is the GPU's that the netbook uses, as their Google Chrome OS is based on Linux and GNU.

          Comment


          • #6
            And i guess Google wants to use ARM powered devices later, no idea what gfx chip they would use then - psb maybe? Would at least be with h264 accelleration, so they only need to convince Adobe to use libva for flash player Tegra would be possible too of course.

            Comment


            • #7
              Originally posted by bulletxt View Post
              It's funny when you say:

              "VIA tried again to push some new DRM code into the mainline Linux kernel."

              It's as if no one wants their code
              This is sadly kind of true, especially since their drm code broke everything. But there are always some people who take in such crap and then refuse to deal with it correctly: cfr. https://bugzilla.novell.com/show_bug.cgi?id=521382

              Comment


              • #8
                Originally posted by MostAwesomeDude View Post
                No idea what VIA's doing, and Google is coming up empty-handed, suggesting I'm not the only one in the dark here.
                It would appear they're doing as little as possible while trying to maintain the illusion that they actually care.

                Comment


                • #9
                  Note: When I mentioned Google, I meant that I searched for a fair while for any open-source plans VIA might have published, and didn't find anything. Google and Xorg are kind of orthogonal.

                  Comment


                  • #10
                    Originally posted by MostAwesomeDude View Post
                    No idea what VIA's doing, and Google is coming up empty-handed, suggesting I'm not the only one in the dark here.
                    To propose a non-conspiracy suggestion here, I would consider it not impossible that VIA are in the same boat as everyone else! They may not really know what their intentions are. Afterall, their gfx side must be tiny compared to Intels, AMD/ATIs, nVidias... They can hardly hope to take anything and try and set the priority, so they are basically muddling through and winging it. Afterall, if we can hardly command much investment from AMD or nVidia, what chance have we with VIA who are a fraction of the size?

                    Comment

                    Working...
                    X