Announcement

Collapse
No announcement yet.

System lockup with catalyst 8.01

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

  • System lockup with catalyst 8.01

    My system gets frozen whenever I try to log out from gnome, restart, shutdown, etc., even when I press Ctrl+Alt+Backspace to restart X. Switching to console works, but I can't issue a gdm restart in console either. Seems like the driver is refusing to reinitialize itself.

    Anyone has experienced this?

  • #2
    Originally posted by jasonxh View Post
    My system gets frozen whenever I try to log out from gnome, restart, shutdown, etc., even when I press Ctrl+Alt+Backspace to restart X. Switching to console works, but I can't issue a gdm restart in console either. Seems like the driver is refusing to reinitialize itself.

    Anyone has experienced this?
    BTW, I'm on Ubuntu feisty, with custom kernel 2.6.23.13 and Xorg 7.2.

    Comment


    • #3
      There are a couple of reports of this on the 8.01 thread. For some people, disabling the overlay options seems to make a difference :

      Option "OpenGLOverlay" "off"
      Option "VideoOverlay" "off"
      Which graphics card are you using ?

      Comment


      • #4
        Originally posted by bridgman View Post
        There are a couple of reports of this on the 8.01 thread. For some people, disabling the overlay options seems to make a difference :



        Which graphics card are you using ?
        Thanks for the hint. I'll try meddling with the overlay options tonight. I'm using mobility X1300, on a thinkpad T60.

        Comment


        • #5
          bridgman, it doesn't solve my problem at all. Still freezes my system. What a joke is ATI making at its customers!

          What is this NEW code base driver truly for? With this laggy 2d and bugs after bugs and absolutely nothing to do with the word "stable", who's gonna enjoy the "increased" 3d fps???

          Comment


          • #6
            Hi jason;

            Sorry this didn't work. I'll see if we have been able to reproduce the problem in house yet.

            The challenge here is that a lot of the worst issues don't show up on any of our in house systems or even in the beta test program, so without driving to your house, finding a clue on the forum threads, or getting a lot of info via Bugzilla we don't have a way to reproduce and fix the bug OR to stop it from happening in the future.

            I'm not sure if there is a Bugzilla ticket open for this one, but if not please file one and include hw info, logs & conf. That can really help our chances of configuring a system close enough to yours that we can see the problem and fix it.

            We'll go through this more in the Q&A, but the "new code base" is the 3d part and the low level, ASIC-specific bits of the X ("2d") driver. The rest of the X driver is *not* new code, but the dev team is now able to spend a lot more time on the "old code" so we do seem to be picking off more issues every month.
            Last edited by bridgman; 01-19-2008, 06:48 PM.

            Comment


            • #7
              Bugzilla issues

              Originally posted by bridgman View Post
              I'm not sure if there is a Bugzilla ticket open for this one, but if not please file one and include hw info, logs & conf.
              I suppose you are talking about http://ati.cchtml.com/ , right?

              While reporting the "failed to initialize ASIC" issue I am experiencing, I noticed that version 8.452.1 is not in the bugzilla pull-down menus yet - 7.12 is the latest version listed.

              The ATI Linux Feedback program is even stuck at version 8.433 (Catalyst 7-11)...

              Maybe you could trigger some updating of these tools...?

              Regards,
              knutbert

              Comment


              • #8
                i have the lockups too, and i'm using a hd2600xt by gigabyte on a p35 motherboard by msi using gutsy.
                probably this helps finding the issue.

                Comment


                • #9
                  Hi bridgman, I wasn't aware that you are from AMD/ATI, so forgive me if anything I said sounds improper.

                  Anyway it's nice to see that someone from inside is taking complaints from customers and trying to resolve problems.

                  Comment


                  • #10
                    I'd just like to add that for several releases of fglrx right back to the start of the "new style driver" at least, if I stop gdm with "/etc/init.d/gdm stop", the system will very often freeze if I then try to switch to the virtual console on which X was running - hard reset needed. (2.6.18 kernel)
                    On the other hand generally speaking if I kill X from within it, rather than using the gdm service stop command from the console, the VT gets freed properly and nothing seems to break.

                    Comment

                    Working...
                    X