Announcement

Collapse
No announcement yet.

Updated and Optimized Ubuntu Free Graphics Drivers

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

  • Originally posted by reCAPTCHA View Post
    It didn't brake for me at all, and I was updating quite often. Yesterday 2 or three times actually. It happened to me that I was having problems, and no one else was reporting it. In such cases purging oibaf, and adding it again usually solves the issue.
    purging oibafs ppa and readding it didn't helped. i had several programms crashing, all segfault in mesa_dri_drivers.so
    and the mesa updates from today fixed it again. as i previously said, it may be hardware dependend, i.e. it may depend if you were having an intel, amd or nvidia graphics...

    but in my case it broke yesterday with only updating mesa and got resolved today with only updating mesa, both from oibaf.
    so my best bet is that the mesa git got a bug and the automated oibaf builds catched it and the next build got the fix.

    Comment


    • I have a MSI R7970 Lightning on Ubuntu 13.10 and it works well with the PPA, except for one thing: the fans seems to be stuck around 50-70% all the time, making my desktop too noisy to be usable.
      I enabled dpm but it changed nothing.

      Is there a setting somewhere that I should tweak or is it a bug?


      For the record, here is the sensor output:
      Code:
      $ sensors
      acpitz-virtual-0
      Adapter: Virtual device
      temp1:        +27.8C  (crit = +106.0C)
      temp2:        +29.8C  (crit = +106.0C)
      
      radeon-pci-0100
      Adapter: PCI adapter
      temp1:        +33.0C  
      
      coretemp-isa-0000
      Adapter: ISA adapter
      Physical id 0:  +34.0C  (high = +85.0C, crit = +105.0C)
      Core 0:         +32.0C  (high = +85.0C, crit = +105.0C)
      Core 1:         +34.0C  (high = +85.0C, crit = +105.0C)
      Core 2:         +30.0C  (high = +85.0C, crit = +105.0C)
      Core 3:         +32.0C  (high = +85.0C, crit = +105.0C)
      
      pkg-temp-0-virtual-0
      Adapter: Virtual device
      temp1:        +34.0C

      Comment


      • Originally posted by Malizor View Post
        I have a MSI R7970 Lightning on Ubuntu 13.10 and it works well with the PPA, except for one thing: the fans seems to be stuck around 50-70% all the time, making my desktop too noisy to be usable.
        I enabled dpm but it changed nothing.

        Is there a setting somewhere that I should tweak or is it a bug?


        For the record, here is the sensor output:
        Code:
        $ sensors
        acpitz-virtual-0
        Adapter: Virtual device
        temp1:        +27.8C  (crit = +106.0C)
        temp2:        +29.8C  (crit = +106.0C)
        
        radeon-pci-0100
        Adapter: PCI adapter
        temp1:        +33.0C  
        
        coretemp-isa-0000
        Adapter: ISA adapter
        Physical id 0:  +34.0C  (high = +85.0C, crit = +105.0C)
        Core 0:         +32.0C  (high = +85.0C, crit = +105.0C)
        Core 1:         +34.0C  (high = +85.0C, crit = +105.0C)
        Core 2:         +30.0C  (high = +85.0C, crit = +105.0C)
        Core 3:         +32.0C  (high = +85.0C, crit = +105.0C)
        
        pkg-temp-0-virtual-0
        Adapter: Virtual device
        temp1:        +34.0C
        IIRC this is a known bug, but I can't find the bug report... search for it in https://bugs.freedesktop.org/

        Comment


        • Originally posted by ultij View Post
          Hello oibaf, are you planning to enable openmax in your builds for testing?
          Thanks!
          OpenMAX is now enabled.

          Comment


          • Originally posted by oibaf View Post
            OpenMAX is now enabled.
            Thanks, I updated your mesa build, installed the libomxil-bellagio packages, but it doesn't seem to work with VLC because libomx_r600.so is missing. Any thoughts?

            Comment


            • Originally posted by ultij View Post
              Thanks, I updated your mesa build, installed the libomxil-bellagio packages, but it doesn't seem to work with VLC because libomx_r600.so is missing. Any thoughts?
              You should install the *g3dvl* package on 13.10 or the *vdpau* package (well, the packages with that in the name) on 14.04, then that file should be installed somewhere under /usr/lib (try searching it with find /usr/lib | grep libomx ).

              I have no idea how can it be found by VLC.

              Comment


              • Over the last 2 weeks stability has decreased. Im using a cayman gpu. ATI 6950 2GB

                Mar 9 01:05:42 paul-desktop kernel: [50793.291773] radeon 0000:01:00.0: GPU lockup CP stall for more than 10000msec
                Mar 9 01:05:42 paul-desktop kernel: [50793.291783] radeon 0000:01:00.0: GPU lockup (waiting for 0x0000000000766abd last fence id 0x0000000000766abc)
                Mar 9 01:05:42 paul-desktop kernel: [50793.291865] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -16!
                Mar 9 01:05:42 paul-desktop kernel: [50793.292101] radeon 0000:01:00.0: couldn't schedule ib
                Mar 9 01:05:42 paul-desktop kernel: [50793.292107] [drm:radeon_cs_ib_chunk] *ERROR* Failed to schedule IB !
                Mar 9 01:05:54 paul-desktop kernel: [50805.286585] show_signal_msg: 162 callbacks suppressed

                This seems to kill the browser immediately as well and in turn it kills the x session and everything running under it. I can wake the system up and i cant get to the vterms.

                Mar 9 01:05:54 paul-desktop kernel: [50805.286597] Watchdog[3379]: segfault at 0 ip 00007ffc79ace51e sp 00007ffc6a95e650 error 6 in chrome[7ffc76050000+5ddb000]


                Under wine i am getting this again

                fixme:d3d:resource_check_usage Unhandled usage flags 0x8.


                Lastly - often when waking up after the screen saver the colour palette is messed up. looks like a few bits stuck some place. To fix this i need to let lock the screen and wake it up a few times until the color is correct.

                Comment


                • There was a change to radeon which caused lockups and was then reverted. Force an update:
                  sudo apt-get update
                  sudo apt-get upgrade

                  Comment


                  • Is anyone successfully running the Portal 2 beta using this PPA? I can't get it running on nvidia, I'll try AMD later, no Intel to test myself though.

                    Comment


                    • I for got to add - this is on fully patched 13.10 ubuntu. I had waited a 2 weeks incase a regression was found and removed. I have been through several cycles of updates and restarts but the problem persists.


                      Originally posted by paulf View Post
                      Over the last 2 weeks stability has decreased. Im using a cayman gpu. ATI 6950 2GB

                      Mar 9 01:05:42 paul-desktop kernel: [50793.291773] radeon 0000:01:00.0: GPU lockup CP stall for more than 10000msec
                      Mar 9 01:05:42 paul-desktop kernel: [50793.291783] radeon 0000:01:00.0: GPU lockup (waiting for 0x0000000000766abd last fence id 0x0000000000766abc)
                      Mar 9 01:05:42 paul-desktop kernel: [50793.291865] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -16!
                      Mar 9 01:05:42 paul-desktop kernel: [50793.292101] radeon 0000:01:00.0: couldn't schedule ib
                      Mar 9 01:05:42 paul-desktop kernel: [50793.292107] [drm:radeon_cs_ib_chunk] *ERROR* Failed to schedule IB !
                      Mar 9 01:05:54 paul-desktop kernel: [50805.286585] show_signal_msg: 162 callbacks suppressed

                      This seems to kill the browser immediately as well and in turn it kills the x session and everything running under it. I can wake the system up and i cant get to the vterms.

                      Mar 9 01:05:54 paul-desktop kernel: [50805.286597] Watchdog[3379]: segfault at 0 ip 00007ffc79ace51e sp 00007ffc6a95e650 error 6 in chrome[7ffc76050000+5ddb000]


                      Under wine i am getting this again

                      fixme:d3d:resource_check_usage Unhandled usage flags 0x8.


                      Lastly - often when waking up after the screen saver the colour palette is messed up. looks like a few bits stuck some place. To fix this i need to let lock the screen and wake it up a few times until the color is correct.

                      Comment

                      Working...
                      X