Announcement

Collapse
No announcement yet.

NVIDIA's PRIME Helpers Are Ready For Linux 3.9

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

  • NVIDIA's PRIME Helpers Are Ready For Linux 3.9

    Phoronix: NVIDIA's PRIME Helpers Are Ready For Linux 3.9

    Aside from a lot of other exciting DRM driver happenings for the Linux 3.9 kernel, it looks like the DRM "PRIME Helpers" that were conceived by NVIDIA to help them support DMA_BUF in their binary driver will be merged...

    http://www.phoronix.com/vr.php?view=MTMwMjI

  • #2
    Finally. It's easy for Linus to say fuck you to Nvidia when the kernel developers works against them by being shitheads.
    Now Nvidia should be able to create optimus support.

    Comment


    • #3
      Nvidia rocks once again and Linus, Cox & GPL nazis can go and four letter word themselves.

      Comment


      • #4
        Originally posted by GT220 View Post
        Nvidia rocks once again and Linus, Cox & GPL nazis can go and four letter word themselves.
        True.






        ten chars rule bla bla phoronix

        Comment


        • #5
          Hopefully, nvidia already has most of thier optimus work done on thier end and once these helpers are in place, we'll have a beta release with support next day.

          Also, lets hope that Alan Cox doesn't go and threaten to sue everyone rampage again and "cox block" the helpers as well.

          Comment


          • #6
            Although I like having open-source infrastructures, when it comes to the end user it doesn't really matter that much whether their GPU driver is os or not. In my opinion the only way we could achieve this is if NVIDIA and other hardware producers would decide to contribute to os drivers themselves(tegra coms to mind); reverse engineering is remarkable but not really sustainable, especially for complex hardware such as GPUs.

            PS. I hope Alan Cox and friends won't throw a tantrum over this just for the sake of suing somebody, considering NVIDIA actually implemented this functionality themselves.

            Comment


            • #7
              Originally posted by vitiv View Post
              PS. I hope Alan Cox and friends won't throw a tantrum over this just for the sake of suing somebody, considering NVIDIA actually implemented this functionality themselves.
              Why would that be a bad thing?
              Actually, one should hope he does (in some way).
              That would probably clear up the legal situation once for all.

              Comment


              • #8
                Originally posted by GT220 View Post
                Nvidia rocks once again and Linus, Cox & GPL nazis can go and four letter word themselves.
                Let's pretend that they weren't and let Nvidia link a proprietary BLOb directly into the kerenel internals without making a fuss. A few years down the road some competing company that had contributed code into the drm or dma_buf sections of the kernel could sue NVIDIA and set them back financially and legally. In such a case almost noone would be willing to get close to the kernel. Companies are much more comfortable around Linux when violations of the license are first met with loud and clear warnings, instead of court action. You may not like the GPL, but Linux for better or worse is pretty much stuck with it. If you don't like it, go implement dma_buf on FreeBSD. (Oh that's right, It's several years away from the point where you could even start)

                Comment


                • #9
                  Does this have any negative effects compared to simply using DMA_BUFF?

                  Next goal. KMS and I will cream myself.

                  Comment


                  • #10
                    How does this help them accomplish Optimus?

                    I thought it was GPL issues that were the primary stumbling block.

                    Comment


                    • #11
                      Originally posted by johnc View Post
                      How does this help them accomplish Optimus?

                      I thought it was GPL issues that were the primary stumbling block.
                      They are, the PRIME helpers introduces non-GPL interfaces for them to use instead of the standard DMA_BUF ones this way they arent directly touching GPL code and have a non-gpl middle layer between the driver and dma_buf

                      Comment


                      • #12
                        Originally posted by Ericg View Post
                        They are, the PRIME helpers introduces non-GPL interfaces for them to use instead of the standard DMA_BUF ones this way they arent directly touching GPL code and have a non-gpl middle layer between the driver and dma_buf
                        Non-GPL code included to the vanilla Linux sources?
                        How would that work?

                        I think you mean it does not rely on Linux code that requires EXPORT_SYMBOL_GPL?

                        Comment


                        • #13
                          Yea, I don't get it either. Unless it adds a new syscall, how can that help?

                          This paragraph from the article, when trimmed to be proper, explains the original issue pretty well:
                          NVIDIA can't directly utilize the Linux kernel's DMA_BUF buffer sharing mechanism -- a zero-copy way to share buffers between different kernel drivers whether it be DRM or other sub-systems -- due to GPL-only kernel

                          Comment


                          • #14
                            Originally posted by entropy View Post
                            Non-GPL code included to the vanilla Linux sources?
                            How would that work?

                            I think you mean it does not rely on Linux code that requires EXPORT_SYMBOL_GPL?
                            2 Points...

                            1) The majority of the code is under the GPL, yes. But considering Linux does not require copyright assignment, I'm pretty sure you can include any code that is under any license that is GPL compatible.

                            2) Also yes, the new code would not use EXPORT_SYMBOL_GPL and would instead declare EXPORT_SYMBOL

                            Comment


                            • #15
                              Originally posted by Ericg View Post
                              2) Also yes, the new code would not use EXPORT_SYMBOL_GPL and would instead declare EXPORT_SYMBOL
                              Which makes zero difference, as the kernel as a whole is GPL.

                              Comment

                              Working...
                              X