Announcement

Collapse
No announcement yet.

RadeonSI Gallium3D Gets UBO/TBO Support, OpenGL 3.3

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

  • #31
    Originally posted by dffx View Post
    Hmmm, wonder what your mesa-demos package is. As far as I can tell, 8.1.0 is the latest stable release: http://cgit.freedesktop.org/mesa/demos

    Which is likely what your mesa-demo-x package is.

    What distro?
    OpenSUSE 13.1 (RC2) with pontostroy repositorys for kernel and mesa.

    Comment


    • #32
      Originally posted by bridgman View Post
      - the "firmware" debates related to using AtomBIOS -- low level firmware shared by VBIOS and gfx drivers -- not the on-GPU microcode

      - you might find starting with 5xx acceleration and moving forward works better than starting with SI/CI and moving backward, since each new programming guide assumes some familiarity with the previous generations rather than making each subsequent doc longer and harder to read
      the problem lies that most whiners believe the firmwares do something like:

      #if FGLRX
      //do gl stuff very very fast
      #else
      //do less gl stuff slower
      #endif signed by evil AMD

      instead of the reality, where microcode and firmwares are used to power on the blocks in the chip/set boot freqs/expose API interfaces/etc which end up been completely unrelated to speed or features as long as you get all microcodes ofc[r600g and radeonsi do have them all btw]

      another interesting fact is that even if AMD release the microcode/firmware they most likely will be useless since you don't have the chip blueprints electronically speaking and without it you just will see a bunch of lines of nemo/hexa code that makes no sense, hence a new cycle of whine will start but for blueprints because evil AMD nerf open drivers in the chip, etc repeat cycle.

      the only real reason that fglrx is faster is :

      r600g/radeonsi send too much redundant shader data to the GPU[vadim did an superb job with SB to help here for r600g] and the cpu related code have more latency, which translate the GPU is not feed fast enough and have to do more work per frame because the shader code is not optimal enough and thats it, there is no conspiration theory.

      Comment


      • #33
        Sorry for hijacking...

        Originally posted by HD7950 View Post
        Code:
        OpenGL vendor string: X.Org
        OpenGL renderer string: Gallium 0.4 on AMD TAHITI
        OpenGL core profile version string: 3.1 (Core Profile) Mesa 10.0.0-devel (git-86cdff5)
        OpenGL core profile shading language version string: 1.40
        OpenGL core profile context flags: (none)
        OpenGL core profile extensions:
        OpenGL version string: 3.0 Mesa 10.0.0-devel (git-86cdff5)
        OpenGL shading language version string: 1.30
        OpenGL context flags: (none)
        OpenGL extensions:
        At least, kwin now works with 3.1 opengl.
        Hmm, can someone explain why my mesa build shows this:
        Code:
         
        OpenGL vendor string: X.Org
        OpenGL renderer string: Gallium 0.4 on AMD TAHITI
        OpenGL version string: 2.1 Mesa 10.0.0-devel (git-86cdff5)
        OpenGL shading language version string: 1.30
        OpenGL extensions:
        configure command:
        Code:
        /autogen.sh --prefix=/opt/xorg --with-dri-drivers=radeon,r200 --with-gallium-drivers=r300,r600,radeonsi,swrast --with-egl-platforms=x11,drm --enable-gbm --enable-shared-glapi --enable-glx-tls --enable-driglx-direct --enable-r600-llvm-compiler --enable-xorg --enable-xa --enable-gallium-egl --enable-gallium-gbm --enable-texture-float

        Comment


        • #34
          Originally posted by fooza View Post
          Sorry for hijacking...



          Hmm, can someone explain why my mesa build shows this:
          Code:
           
          OpenGL vendor string: X.Org
          OpenGL renderer string: Gallium 0.4 on AMD TAHITI
          OpenGL version string: 2.1 Mesa 10.0.0-devel (git-86cdff5)
          OpenGL shading language version string: 1.30
          OpenGL extensions:
          configure command:
          Code:
          /autogen.sh --prefix=/opt/xorg --with-dri-drivers=radeon,r200 --with-gallium-drivers=r300,r600,radeonsi,swrast --with-egl-platforms=x11,drm --enable-gbm --enable-shared-glapi --enable-glx-tls --enable-driglx-direct --enable-r600-llvm-compiler --enable-xorg --enable-xa --enable-gallium-egl --enable-gallium-gbm --enable-texture-float
          kernel must be too old,which version you use?

          Comment


          • #35
            Originally posted by fooza View Post
            Hmm, can someone explain why my mesa build shows this:
            Code:
             
            OpenGL vendor string: X.Org
            OpenGL renderer string: Gallium 0.4 on AMD TAHITI
            OpenGL version string: 2.1 Mesa 10.0.0-devel (git-86cdff5)
            OpenGL shading language version string: 1.30
            OpenGL extensions:
            configure command:
            Code:
            /autogen.sh --prefix=/opt/xorg --with-dri-drivers=radeon,r200 --with-gallium-drivers=r300,r600,radeonsi,swrast --with-egl-platforms=x11,drm --enable-gbm --enable-shared-glapi --enable-glx-tls --enable-driglx-direct --enable-r600-llvm-compiler --enable-xorg --enable-xa --enable-gallium-egl --enable-gallium-gbm --enable-texture-float
            You need to install LLVM 3.4.

            Comment


            • #36
              My RadeonSI HD 7950 card still gets corrupted text and all Source games crash with open source drivers. So there's plenty of bugs to be had on RadeonSI.

              Comment


              • #37
                L4D2 on my hd7770 works ok but serious sam crash with llvm assertion bug. So Source seems ok.

                Comment


                • #38
                  Originally posted by bridgman View Post

                  - you might find starting with 5xx acceleration and moving forward works better than starting with SI/CI and moving backward, since each new programming guide assumes some familiarity with the previous generations rather than making each subsequent doc longer and harder to read

                  That's a great tip. Thanks.

                  Comment


                  • #39
                    Originally posted by jrch2k8 View Post
                    kernel must be too old,which version you use?
                    nope, 3.12.0

                    Originally posted by smitty3268 View Post
                    You need to install LLVM 3.4.
                    ok, thx. I'll try it.

                    Comment


                    • #40
                      Originally posted by frosth View Post
                      L4D2 on my hd7770 works ok but serious sam crash with llvm assertion bug. So Source seems ok.
                      on my HD7950 l4d2 runs also fast and fine even with full AA and AF. But sound is gone... worked before...
                      Dota2 even improved (no glitches in the menu anymore) but sound is also gone.
                      Sound for Mark of the ninja and everywhere else however works fine - whats going on? which logs can I provide?

                      Code:
                      grep EE /var/log/Xorg.0.log
                      [     7.719] Current Operating System: Linux linux.site 3.12.0-3-desktop #1 SMP PREEMPT Mon Nov 4 12:54:39 UTC 2013 (ec24572) x86_64
                              (WW) warning, (EE) error, (NI) not implemented, (??) unknown.
                      [     7.725] Initializing built-in extension MIT-SCREEN-SAVER
                      [     7.738] (EE) Failed to load module "fglrx" (module does not exist, 0)
                      [     7.740] (EE) Failed to load module "vesa" (module does not exist, 0)
                      dmesg shows no error

                      Comment

                      Working...
                      X