Announcement

Collapse
No announcement yet.

AMD 8.39.4 Driver Released -- 8.35 no longer does it better

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

  • #16
    Michael,

    for XVideo, i have:

    (II) Loading extension XVideo
    (II) Loading extension XVideo-MotionCompensation


    and nothing more than that. but that message indicates that it
    should work.


    for openGL, this may be something:

    (EE) AIGLX error: dlopen of /usr/lib/dri/fglrx_dri.so failed (/usr/lib/dri/fglrx_dri.so: undefined symbol: __glXFindDRIScreen)
    (EE) AIGLX: reverting to software rendering
    (II) Loading local sub module "GLcore"
    (II) LoadModule: "GLcore"
    (II) Loading /usr/lib/xorg/modules/extensions//libGLcore.so
    (II) Module GLcore: vendor="X.Org Foundation"
    compiled for 1.3.0, module version = 1.0.0
    ABI class: X.Org Server Extension, version 0.3
    (II) GLX: Initialized MESA-PROXY GL provider for screen 0


    ...


    (EE) DoSwapInterval: cx = 0xa33f620, GLX screen = 0x9585d20
    (EE) AIGLX: cx->pGlxScreen->swapInterval == NULL


    ---

    let me know if you'd like the entire log file.


    thanks,
    sam

    Comment


    • #17
      "Dazed and confused"???

      I have found this message in my kernel log:

      ACPI: PCI Interrupt 0000:01:00.0[A] -> GSI 16 (level, low) -> IRQ 16
      Uhhuh. NMI received for unknown reason b1 on CPU 0.
      You have some hardware problem, likely on the PCI bus.
      Dazed and confused, but trying to continue

      I haven't enabled NMI because it makes the laptop's ACPI video functionality hang, so I don't know where this NMI came from.

      Comment


      • #18
        I went back to the 8.38.6's I lost my monitors native resolution of 1280x1024 with the 8.39.4, the highest available was 1152x864 ?
        Those who would give up Essential Liberty to purchase a little Temporary Safety,deserve neither Liberty nor Safety.
        Ben Franklin 1755

        Comment


        • #19
          Originally posted by DarkFoss View Post
          I went back to the 8.38.6's I lost my monitors native resolution of 1280x1024 with the 8.39.4, the highest available was 1152x864 ?

          DarkFoss,

          i have my resolution at 1600x1200 with 8.39.4. is the proper Modes
          set in your xorg.conf? eg..


          Code:
          Section "Screen"
                  Identifier "aticonfig-Screen[0]"
                  Device     "aticonfig-Device[0]"
                  Monitor    "aticonfig-Monitor[0]"
                  DefaultDepth     24
                  SubSection "Display"
                          Viewport   0 0
                          Depth     24
                          Modes    "1600x1200"         <<-- my default
                  EndSubSection
          EndSection

          Comment


          • #20
            Driver works fine here on Fedora Rawhide with Kernel "2.6.21-1.3244.fc8".


            ...i can not test the recent kernel, 2.6.23-0.30.rc0.git6.fc8, because of a "kernel panic"

            Comment


            • #21
              Originally posted by chans_ph View Post
              DarkFoss,

              i have my resolution at 1600x1200 with 8.39.4. is the proper Modes
              set in your xorg.conf? eg..


              Code:
              Section "Screen"
                      Identifier "aticonfig-Screen[0]"
                      Device     "aticonfig-Device[0]"
                      Monitor    "aticonfig-Monitor[0]"
                      DefaultDepth     24
                      SubSection "Display"
                              Viewport   0 0
                              Depth     24
                              Modes    "1600x1200"         <<-- my default
                      EndSubSection
              EndSection
              Yep the modes are set in the xorg.. I've had it happen with the 8.31 and the 8.34 drivers too, I just lose the resolution. Usually I just revert and wait for the next driver and everything is fine again.
              Code:
              Section "Screen"
                  Identifier "screen1"
                  Device "device1"
                  Monitor "monitor1"
                  DefaultColorDepth 24
              ]Subsection "Display"
                      Depth 24
                      Modes "1280x1024" "1152x864" "1024x768" "832x624" "800x600" "640x480" "480x360" "320x240"
                  EndSubsection
              Those who would give up Essential Liberty to purchase a little Temporary Safety,deserve neither Liberty nor Safety.
              Ben Franklin 1755

              Comment


              • #22
                "Testing Use Only" Watermark information -- http://www.phoronix.com/?page=news_item&px=NTkwNg
                Michael Larabel
                http://www.michaellarabel.com/

                Comment


                • #23
                  Michael Help Me!!

                  I am using ATI 9550 card..I installed the 8.39.4 driver with Xorg 1.3.0.0-9 on my Fedora 7,when system begins to start xorg, my screen shows "No signal" then the screen turns to be black..
                  I checked the Xorg.0.log,nothing wrong (EE) shows in that log file..

                  I downgrade Xorg to version 1.2 and reinstalled the 8.39.4 driver ,problem's the same.

                  I uninstalled fglrx 8.39.4 and installed 8.38.7,then the driver works fine..

                  what is the problem??
                  Nothing goes wrong when the Xorg starts and loads the fglrx driver,but why there is no signal???

                  Comment


                  • #24
                    Good News, Bad News

                    The good news of this release is that the EDID hex dump I get in my xorg log when using 1.3 isn't just 00FF000...0 anymore which suggests they are making progress in fixing (what I believe to be) this problem with certain rigs getting blank screens.

                    The bad news is I still get a blank screen with 1.3. I'll go ahead and give some more technical details to make up for my ranting below.

                    As of 1.3 there's no longer a separate DDC module for display detection. It's now built-in to the server itself. It seems (judging from my 1.1.1 log) fglrx likes being able to load the ddc module and probe the display.

                    Without a DesktopSetup line in my xorg.conf I get a message that two displays were detected, forcing clone mode. If I force single with the aforementioned line it gives some other error I don't have handy. I've only got one display connected and regardless of DesktopSetup and ForceMonitors it still chokes right after this stage.

                    These companies (AMD and Nvidia) are complete morons in the grand scheme of things, for their behavior over the drivers for their hardware in linux. There's two reasons:

                    1. This hardware is useless without proficient drivers behind it. I could be handed a graphics card capable of running a complete model of the visible universe at 1,000,000 fps but without a driver to use it, it might as well be a box of animal crackers. (Actually, the animal crackers would be better, you can eat those)

                    2. This is a community that is driven by the efforts of its members. That's the idea behind linux, that collectively working together can build software solutions that enable amazing things.

                    Who knows how many more people would get sucked in to making linux games and doing graphics work on the platform if it had proper support from that industry?

                    For now, at least, it looks like we won't find out. That's the shame of it because there are plenty of people who try to give detailed reports of their rigs, configs, and results and we don't have any idea if the developers are even listening or what the real problems holding things up are.

                    We do know they didn't take the finished driver and manually install it the way many of us do. That I can't explain. I understand that mistakes happen in building drivers or any type of software. Last semester I wrote a program in assembly that was wrong for exactly one input in 4,294,967,296. It happens.

                    But as soon as I was aware of the error I was able to fix it. That's the virtue of open source that is one of the driving forces that brings people to linux. The fact that we aren't crippled into waiting for a monthly driver to see if the man behind the curtain has sprinkled the right amount of fairy dust in with the beanstalk beans to make the damn thing work. That, if there's a problem, anyone with the know-how and tenacity can fix it and we can all benefit from it.

                    Comment


                    • #25
                      Then go and work as a fglrx project leader. As far as I know, they are currently looking for one.

                      Comment


                      • #26
                        Well, it looks like 8.39.4 is being recalled basically... AMD has asked it be pulled from repos.

                        So if you haven't downloaded the driver yet and rely upon distribution-specific packaging, wait and use the Avivo driver or the 8.38.7 driver.
                        Michael Larabel
                        http://www.michaellarabel.com/

                        Comment


                        • #27
                          Originally posted by Michael View Post
                          Well, it looks like 8.39.4 is being recalled basically... AMD has asked it be pulled from repos.

                          So if you haven't downloaded the driver yet and rely upon distribution-specific packaging, wait and use the Avivo driver or the 8.38.7 driver.
                          Has been recalled?? Too bad...

                          Any particular reason why this happened?

                          Comment


                          • #28
                            Originally posted by lenrek View Post
                            Has been recalled?? Too bad...

                            Any particular reason why this happened?
                            The watermark problem. AMD is still figuring out how this can be fixed.
                            Michael Larabel
                            http://www.michaellarabel.com/

                            Comment


                            • #29
                              Originally posted by d2kx View Post
                              Then go and work as a fglrx project leader. As far as I know, they are currently looking for one.
                              Well... Join them does not imply will be a project leader. Even if one landed him/her-self as a project leader, one may not be able to lead as one wants.

                              I am, surprise by this comment, there is no project leader? This is not a very good sign, it seems, these days, every time I read about ATI for Linux, it is always bad news.

                              Anyway, my company (due to my recommendation), had ordered 4 laptops (3 more are coming), all running Intel (CPU + Graphics).

                              Comment


                              • #30
                                Well now.... and who says it's working on FC7?
                                i get exact same behavior
                                ....
                                ABI class: X.Org XInput driver, version 0.6
                                (II) Primary Device is: PCI 01:00:0
                                (II) ATI Proprietary Linux Driver Version Identifier:8.39.4
                                (II) ATI Proprietary Linux Driver Release Identifier: UNSUPPORTED-8.393.1
                                (II) ATI Proprietary Linux Driver Build Date: Jul 18 2007 14:41:23
                                ....
                                Backtrace:
                                0: X(xf86SigHandler+0x6d) [0x48b99d]
                                1: /lib64/libc.so.6 [0x2aaaab60e630]
                                2: /usr/lib64/xorg/modules/drivers//fglrx_drv.so(atiddxSave64BitBAR+0x2d) [0x2aaaacd4cb8d]
                                3: /usr/lib64/xorg/modules/drivers//fglrx_drv.so(atiddxProbeMain+0x37f) [0x2aaaacd57a7f]
                                4: X(InitOutput+0x6e7) [0x463707]
                                5: X(main+0x275) [0x434625]
                                6: /lib64/libc.so.6(__libc_start_main+0xf4) [0x2aaaab5fbab4]
                                7: X(FontFileCompleteXLFD+0x229) [0x433ad9]

                                Fatal server error:
                                Caught signal 11. Server aborting


                                anyone can help?
                                ARCH: x86_64
                                kernel: 2.6.21

                                Comment

                                Working...
                                X