Hi After an 'apt-get dist-upgrade' in Debian Unstable, I can't use my dual screen setup anymore with my ATI FireGL V5600. I'm using the git/master for the ati driver. At startup, the screens are configured in "same-as" mode but if I try to set one screen at the right (or left) of the other, both screens are displaying the same area (the left or right screen). % xrandr --output DVI-1 --right-of DVI-0 % xrandr Screen 0: minimum 320 x 200, current 3840 x 1200, maximum 3840 x 1200 DVI-1 connected 1920x1200+1920+0 (normal left inverted right x axis y axis) 519mm x 324mm 1920x1200 60.0*+ 59.9 1600x1200 59.9 ... DVI-0 connected 1920x1200+1920+0 (normal left inverted right x axis y axis) 519mm x 324mm 1920x1200 60.0*+ 59.9 1600x1200 59.9 ... If I try to move the position of one screen, I get the same thing : % xrandr --output DVI-0 --pos 1920x0 % xrandr Screen 0: minimum 320 x 200, current 3840 x 1200, maximum 3840 x 1200 DVI-1 connected 1920x1200+1920+0 (normal left inverted right x axis y axis) 519mm x 324mm 1920x1200 60.0*+ 59.9 1600x1200 59.9 ... DVI-0 connected 1920x1200+1920+0 (normal left inverted right x axis y axis) 519mm x 324mm 1920x1200 60.0*+ 59.9 1600x1200 59.9 ... If I do 'xrandr --auto', the screens come back to the "same-as" mode. Thank you
Created attachment 16757 [details] [review] Xorg.log failing
This is probably clone support (driving multiple outputs with one crtc) which was added recently. Can you attach the output of 'xrandr --verbose'? The following should work: xrandr --output DVI-1 --right-of DVI-0 --crtc 1 Ideally the xserver would be smart enough to handle this...
Created attachment 16765 [details] [review] xrandr --verbose
(In reply to comment #2) > This is probably clone support (driving multiple outputs with one crtc) which > was added recently. Can you attach the output of 'xrandr --verbose'? Sure, done. > The following should work: > xrandr --output DVI-1 --right-of DVI-0 --crtc 1 It doesn't :( The DVI-0 becomes black and go into sleep mode, if I do xrandr, I get this : % xrandr Screen 0: minimum 320 x 200, current 1920 x 1200, maximum 3840 x 1200 DVI-1 connected (normal left inverted right x axis y axis) 1920x1200 60.0 + 59.9 1600x1200 59.9 1680x1050 60.0 1600x1000 59.9 1280x1024 84.8 75.0 59.9 1280x960 59.9 1152x864 75.0 1152x720 60.0 1024x768 75.1 60.0 832x624 74.6 800x600 75.0 60.3 640x480 75.0 60.0 720x400 70.1 DVI-0 connected 1920x1200+0+0 (normal left inverted right x axis y axis) 519mm x 324mm 1920x1200 60.0*+ 59.9 1600x1200 59.9 1680x1050 60.0 1600x1000 59.9 1280x1024 84.8 75.0 59.9 1280x960 59.9 1152x864 75.0 1152x720 60.0 1024x768 75.1 60.0 832x624 74.6 800x600 75.0 60.3 640x480 75.0 60.0 720x400 70.1 Thank you!
Try the following: xrandr --output DVI-1 --mode 1920x1200 --crtc 1 xrandr --output DVI-1 --right-of DVI-0
Created attachment 16778 [details] Xorg.log with logverbose 7
(In reply to comment #5) > Try the following: > xrandr --output DVI-1 --mode 1920x1200 --crtc 1 > xrandr --output DVI-1 --right-of DVI-0 One screen becomes black, the other (DVI-1) is correct. I'm attaching to this bug the Xorg.0.log with -logverbose 7 It seems that I'm not alone to have this problem : http://article.gmane.org/gmane.comp.freedesktop.xorg.drivers.radeonhd/3556 Thanks
Looks like a general RandR bug in newer Xservers(In reply to comment #2) > This is probably clone support (driving multiple outputs with one crtc) which > was added recently. Can you attach the output of 'xrandr --verbose'? The logfile posted to the radeonhd list (the radeonhd driver has the same issue) showed that two different CRTCs were attached to the two outputs. radeonhd always had clone support, so this is not related to the driver. It seems there is a general RandR regression in newer Xservers. I don't think I have tried this kind of setup with 1.4.90 or later recently. > The following should work: > xrandr --output DVI-1 --right-of DVI-0 --crtc 1 > > Ideally the xserver would be smart enough to handle this... You also have to use at least xrandr version 1.2.3. Former versions were very buggy WRT uncloning.
(In reply to comment #8) > You also have to use at least xrandr version 1.2.3. Former versions were very > buggy WRT uncloning. I tried xrandr using git/master (4dcc8ae1a6903434def1a2706f7c68ff9e2a17c4) and same thing :(
(In reply to comment #8) > It seems there is a general RandR regression in newer Xservers. I don't think I > have tried this kind of setup with 1.4.90 or later recently. Maybe an issue with 1.4.x xservers? Works fine here with xserver from git.
if this is still an issue with ati 6.12.4 or git master, please re-open.
closing due to lack of feedback.
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.