Announcement

Collapse
No announcement yet.

NVIDIA Wants Feedback On Its Device Memory Allocator Project

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

  • #41
    Originally posted by schmidtbag View Post
    Interesting to me how Nvidia suddenly cares about the opinions of Linux users, and even Nouveau. Back in Linus' "fuck you Nvidia" days, the available technology and drivers at the time gave Nvidia enough of an edge where they could be their usual selves and just brush it off like "I don't care what any of you think, you're doing it my way because you don't have a better option". But now, AMD+Intel are almost fully caught-up with Nvidia (in terms of performance and features), and Wayland is starting to be more widely adopted. Ignoring hardware limitations, we're very close to a point where AMD will be a resounding better option for Linux (regardless of your principles) and where Nvidia will be holding back development to the point where they're resented rather than "begrudgingly accepted".
    Who does that remind me of... oh yeah, Shuttleworth and Mir. NVidia is bigger so it'll be harder for them to fail as spectacularly as Mir did, but they will feel the $$$ sting this time.

    Comment


    • #42
      I don't feel particularly bad for ignoring NVidia when it comes to GPU purchasing decisions; I tend to buy reasonably good hardware and use it on my Windows box first and then migrate it to my Linux boxes over time as support for them mature.

      With AMD, at least I know they have a reasonably sane FLOSS strategy and I'm all for supporting that, because it allows game and engine developers and other interested parties to dive in and help optimize FLOSS drivers for their particular platform to the benefit of everyone. NVidia? Big fat nope.

      Comment


      • #43
        NVIDIA
        What is this?

        Comment


        • #44
          Originally posted by DanL View Post
          Nvidia made EGLstreams and then admitted it wasn't ideal.
          I heard no such admission.

          Originally posted by DanL View Post
          The GBM devs admit their solution isn't ideal.
          Never heard that one as well.

          Originally posted by DanL View Post
          So the two camps look to make something better.
          Look at the commits. It's a two-man's project. Neither from the kernel camp.

          Originally posted by DanL View Post
          There's no guarantee it doesn't end up in squabbling and NIH fragmentation, but based on what we've seen so far, that's a big conclusion to jump to.
          Again, look at commit log and the mailing list. The lack of responses speaks volume.

          Comment


          • #45
            Originally posted by c117152 View Post
            I heard no such admission.
            James Jones (of Nvidia) said:
            "But viewed shortcomings of EGLStream are the open standard being implemented differently by vendors, there isn't cross-device support, it's based upon EGL, there is a lot of encapsulation, and the behavior can vary in some areas."

            GBM requires KMS and is Mesa-specific.

            Look at the commits. It's a two-man's project. Neither from the kernel camp.
            No one disputes that. Nvidia is looking to change it though, as I noted in my previous post (#37). Read the quote again. No, wait, I already told you to read it again. So you just want to argue for the sake of arguing or your reading comprehension is piss poor. Either way, quit wasting my time.

            Comment


            • #46
              Originally posted by DanL View Post
              James Jones (of Nvidia) said:
              "But viewed shortcomings of EGLStream are the open standard being implemented differently by vendors, there isn't cross-device support, it's based upon EGL, there is a lot of encapsulation, and the behavior can vary in some areas."
              Saying "viewed shortcomings" is saying those aren't shortcomings. Only that they're perceived as such. It's passive-aggressive legal talk for "I hear what you're saying and care about you. Now shut up and do as I say.".

              Originally posted by DanL View Post
              No one disputes that. Nvidia is looking to change it though, as I noted in my previous post (#37).
              And I'm saying I read it and they're not. All I'm seeing is them developing something after the fact and in github instead of the mailing list and doing it years after the original discussions took place even though they were directly asked for their opinion despite their previous and ongoing transgressions with regards to everything to do with kernel driver development.

              It's like you throwing an event and asking everyone for feedback and to RSVP, and everyone does it except some asshole. Then, two weeks before the event they call up saying the date and location doesn't work for them and offering you "help" by sending you and everyone else a list of possible alternative locations and dates while saying "I hope we can work together and reach a fair compromise". Oh, and this asshole did the exact same thing the last 3 events.

              Comment

              Working...
              X