Announcement

Collapse
No announcement yet.

AMD Sends Out Linux Kernel Driver Support For Navi 12 GPUs

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

  • AMD Sends Out Linux Kernel Driver Support For Navi 12 GPUs

    Phoronix: AMD Sends Out Linux Kernel Driver Support For Navi 12 GPUs

    While we've already seen the RADV Vulkan driver land their slated support for Navi 12 GPUs on top of the recently launched Radeon RX 5700 "Navi 10" graphics cards, today is the first time we're seeing patches from AMD to wire in the support to the AMDGPU DRM Linux kernel driver for this next iteration of Navi...

    http://www.phoronix.com/scan.php?pag...Kernel-Navi-12

  • #2
    Another... VCN... revision?
    When Navi 12 hardware comes out can somebody from AMD point out the differences between this and VCN 2.0?

    Comment


    • #3
      Originally posted by tildearrow View Post
      Another... VCN... revision?
      When Navi 12 hardware comes out can somebody from AMD point out the differences between this and VCN 2.0?
      One item that might be different is regarding the power gating for VCN on Navi 12. There was a patch mentioning explicitly the VCN power gating for Navi 12 but didn't have a chance yet to compare to see if there is indeed no power gating for VCN on Navi 10.
      Michael Larabel
      http://www.michaellarabel.com/

      Comment


      • #4
        Navi RDNA ISA documentation published: https://gpuopen.com/compute-product/...-architecture/

        Comment


        • #5
          Originally posted by xxmitsu View Post
          Navi RDNA ISA documentation published: https://gpuopen.com/compute-product/...-architecture/
          Let's grab it before it gets removed! I remember back in 2017 when the Vega docs were published, and by September-October of that year I couldn't find them anymore.

          Comment


          • #6
            Originally posted by tildearrow View Post

            Let's grab it before it gets removed! I remember back in 2017 when the Vega docs were published, and by September-October of that year I couldn't find them anymore.
            They are still there...

            BTW: Wayback machine is your friend in those cases...

            Comment


            • #7
              Originally posted by xxmitsu View Post
              Navi RDNA ISA documentation published: https://gpuopen.com/compute-product/...-architecture/
              Thanks for the link. We can deduce a lot of things from the ISA spec. I'm particularly happy with a number of aspects of the new architecture for compute operations:
              . Maximum number of registers accessible by a workgroup seems to increase from 1024 (4 (queues) * 64 (wavefront) * 256 (max number of registers)) to 2048 ( 4 (SIMDs) * 32 (Wavefront) * 256 (max number of registers) * 4 (several wavefront per SIMD. A max of 1024 registers per SIMD)). I hope there is no restrictions on workgroups that reduce this number.
              . Lower LDS latency (1 cycle if good access pattern !)
              . More LDS accessible per workgroup: 64kb (was 16kb previously if my memory is right)

              Comment


              • #8
                Originally posted by mannerov View Post
                I'm particularly happy with a number of aspects of the new architecture for compute operation
                Now all we need are drivers that support compute. So far the OpenCL support is pretty broken, at launch few benchmarks even ran and performance was no better than Polaris.
                The AMDs open source ROCm driver has no support at all for Navi and a github issue asking about this never got a single reply from AMD devs.

                Navi might look good for compute but AMD seems to have designed it aggressively for gaming, leaving compute as an afterthought. This isn't a bad idea as it has allowed them to get much better performance in gaming and catch up to Nvidia. AMD has said that Vega (and GCN) will continue as compute focused parts.
                Arcturus (gfx908) is the next of these to be released and it already has support in ROCm, yet there is nothing from the devs mentioning Navi.

                Comment


                • #9
                  Originally posted by Madgemade View Post
                  The AMDs open source ROCm driver has no support at all for Navi and a github issue asking about this never got a single reply from AMD devs.
                  ...
                  AMD has said that Vega (and GCN) will continue as compute focused parts.Arcturus (gfx908) is the next of these to be released and it already has support in ROCm, yet there is nothing from the devs mentioning Navi.
                  The github thread went off the rails before anyone from AMD could respond, and the question has been pretty much overtaken by events if you look at the upstream commit history:

                  https://cgit.freedesktop.org/~agd5f/...aging-drm-next

                  https://git.kernel.org/pub/scm/linux...drm/amd/amdkfd

                  Compute-oriented cards are still top priority but we want to be able to run the same code on all hardware as much as possible (at least for Linux... Windows compute runs over PAL).
                  Last edited by bridgman; 08-02-2019, 04:46 PM.

                  Comment


                  • #10
                    Originally posted by bridgman View Post
                    The github thread went off the rails before anyone from AMD could respond, and the question has been pretty much overtaken by events if you look at the upstream commit history
                    The thread you mention did go off topic. But I don't think it's too much to ask for a GPU to have fully working support when it's launched. Even if it doesn't, then it really isn't to much effort for someone at AMD to add a paragraph to a readme explaining the current situation. Commit history is not something anyone would make a purchasing decision based on.
                    The thing that winds me up is that AMD compute performance is great and offers better value for money than Nvidia but nobody at AMD seems to know or care. ROCm gets no publicity at all whatsoever (outside of very specialist events). A quick browse around Reddit and the web generally and you only ever see Nvidia=CUDA and AMD=OpenCL, for most ROCm is unknown and the idea that CUDA can be run on AMD after a few tweaks is seen as madness.

                    If AMD actually funded ROCm properly then people would get to know about it and use it. The devs might get time to fix the numerous bugs it has (or at least time to respond to issues). Most of the time issues go completely unanswered by them. The documentation might even be expanded so that it covers more than a tiny fraction of what Nvidia cover.

                    Some of this (like documentation) can be helped by the community, but I'm not going my spend time on something that won't ever been seen, or will soon be deprecated (HC anyone?).

                    The problem at AMD seems to be a corporate one. Correct me if I'm wrong but it looks a lot like someone decided that trying to seriously compete with Nvidia was too hard and didn't meet their cost analysis. So instead a small underfunded effort would be make (ROCm) to move into the machine learning side (good buzzword for shareholders). It honestly feels like the compute performance of AMDs GPUs is wasted on AMD themselves. AMD is and has always been (since GCN) more competitive at compute than gaming but nothing is done about this. This RDNA GCN split seems good but I'm worried that the extra costs of maintaining this is going to slow or stall AMD's compute development.

                    Comment

                    Working...
                    X