Announcement

Collapse
No announcement yet.

AMD R600/700 DRM Interrupts Support Pushed

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

  • #61
    Originally posted by thalience View Post
    I'm not sure why, but I don't seem to be getting sync-to-vblank behavior with GL apps (glxgears still reports way more than 60fps). GL apps no-longer complain about a lack of interrupts, however.
    AIUI sync to vblank for gl apps doesn't work with DRI2 yet.

    Comment


    • #62
      Tried using http://kernel.ubuntu.com/~kernel-ppa...-next/current/ but and firmware in the right place but it will not load the irq microcode.
      Do I need some switch for it to load?

      Thanks

      Comment


      • #63
        I there any time frame for this patch to hit .33 git? I haven't been messing with custom kernels lately since ubuntu has been fairly close through the .32. But .33 will likely change that, since Ubuntu's release will likely target .32.

        If it's already in git, sorry. The logs are mainly large series of remote pulls which can be hard to track the specifics of.

        Comment


        • #64
          Dave sent the pull request to Linus earlier this week. If you want to track radeon development, follow Dave's tree:
          http://git.kernel.org/?p=linux/kerne....git;a=summary

          Comment


          • #65
            Cool, I hadn't checked drm-next in a few days. You've been busy Alex.

            BTW - drm-radeon-testing has a new commit porting HDMI support from radeonhd. Will this be the solution or just an interim fix?
            Last edited by rjwaldren; 12-11-2009, 12:35 PM.

            Comment


            • #66
              Originally posted by rjwaldren View Post
              Cool, I hadn't checked drm-next in a few days. You've been busy Alex.

              BTW - drm-radeon-testing has a new commit porting HDMI support from radeonhd. Will this be the solution or just an interim fix?
              Why should it be interim fix? It's real, great solution Of course not perfect yet, we don't have RV7x0 working but main part is working really nice

              Comment


              • #67
                Originally posted by Zajec View Post
                Why should it be interim fix? It's real, great solution Of course not perfect yet, we don't have RV7x0 working but main part is working really nice

                The reason I asked is that HDMI audio support in radeonhd was a rev-engineered solution. It has been reported here that Alex has some patches under IP review. I would expect that Alex has access to resources that RE might miss.

                The radeonHD solution does work, so it was just a question. Haven't tried the radeon patch yet. But I've been waiting on this feature anxiously as it completely removes my need for the fglrx driver. Now all of my use cases are at least supported, the performance and the move to mainline is all that I'm waiting on. That is coming faster than I ever would have guessed.

                Comment


                • #68
                  Originally posted by rjwaldren View Post
                  The reason I asked is that HDMI audio support in radeonhd was a rev-engineered solution. It has been reported here that Alex has some patches under IP review. I would expect that Alex has access to resources that RE might miss.
                  We do some single things incorrectly (Alex has that fixed locally) but generally most of the code is fine. If Alex will get permission to release his fixes, he will just fix some bits in current code and probably will give us real names of registers.

                  Comment


                  • #69
                    Originally posted by Zajec View Post
                    We do some single things incorrectly (Alex has that fixed locally) but generally most of the code is fine. If Alex will get permission to release his fixes, he will just fix some bits in current code and probably will give us real names of registers.
                    Please don't take my comment as a dig at the ported code, but I know that not all chips are supported yet and figured the fix will have to come from the IP release. Hopefully this will shorten the to time to getting the IP side in the tree and expedite the rv700+ bits.

                    Comment


                    • #70
                      Originally posted by rjwaldren View Post
                      Please don't take my comment as a dig at the ported code, but I know that not all chips are supported yet and figured the fix will have to come from the IP release. Hopefully this will shorten the to time to getting the IP side in the tree and expedite the rv700+ bits.
                      Yes, I am aware of that, thanks I'll try this anyway for fun

                      Comment

                      Working...
                      X