Announcement

Collapse
No announcement yet.

Recent linux-2.6 Git pull KMS won't recognize monitor 2

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

  • Recent linux-2.6 Git pull KMS won't recognize monitor 2

    I'm at work right now, so I am not able to access my machine at home to grab logs. I'm mostly just wondering if anyone else has run into this issue, or if someone knows if this is already fixed...

    I have a Radeon 4770 with two connected monitors. My secondary monitor is hooked up via a DVI->VGA adapter, and my primary monitor is directly connected over DVI.

    Up until I upgraded to 2.6.32-rc7 this weekend (git pull of Linus' tree), I was able to use KMS to initialize both monitors. Up until the X login, the monitors were just clones of each other, but after Gnome is started, they become separate desktops.

    Then this weekend I updated my kernel to 2.6.32-rc7 (from rc5). When I rebooted, my secondary display would no longer display anything, and when I log into X, it stays blank. In the cases that it doesn't stay blank, there's what looks like vertical monochrome columns which seem to be caused by sync issues between the monitor and video card. So it seems like KMS isn't detecting/initializing the second monitor.

    So my first thought was to reboot to rc5, since I had also upgraded my X.org ATI drivers (xorg-edgers ppa). This worked fine, so I figured something in the kernel was messed up. I set radeon.modeset=0 in my kernel boot parameters, and everything is displayed/detected correctly.

    Anyone else run into this, or heard about it? I'll try to grab some dmesg logs when I get home, but that'll be in about 10.5 hours.

    In the meantime, I'll be randomly browsing drm-next diffs that seem applicable.

  • #2
    I reported this bug a week or so back when I found the bug in drm-next.


    Dave Airlie fixed it in drm-next, but it hasn't been merged into Linus's tree yet. Maybe -rc8 will have it?

    Comment


    • #3
      Originally posted by pvtcupcakes View Post
      I reported this bug a week or so back when I found the bug in drm-next.


      Dave Airlie fixed it in drm-next, but it hasn't been merged into Linus's tree yet. Maybe -rc8 will have it?
      I'll probably pull a copy of drm-next and try it out, then verify the fix whenever linus does his next pull of the drm code.

      Thanks for the heads-up.

      Comment

      Working...
      X