Description of problem: With 3.14.9-200 I am able to dock my Thinkpad T540p and use an external Dell monitor at 1920x1200. Since 3.15.3-200, I am only able to select 1024x768 Version-Release number of selected component (if applicable): kernel-3.15.3-200.fc20 How reproducible: Always Steps to Reproduce: 1. Boot into 3.15.3-200 2. Dock laptop and use external monitor over HDMI 3. Attempt to change resolution via GNOME Display properties, or arandr.
Created attachment 916173 [details] dmesg.diff dmesg diff of 3.14.9-200 and 3.15.3-200
Created attachment 916174 [details] Xorg.0.log.diff Xorg log diff
Created attachment 916199 [details] dmesg-3.14.9-200.fc20.x86_64 drm.debug=6 dmesg output of the working kernel
Created attachment 916200 [details] dmesg-3.15.3-200.fc20.x86_64 drm.debug=6 dmesg output of the latest kernel
Created attachment 916201 [details] Xorg.0.log.3-14.9-200.fc20.x86_64
Created attachment 916202 [details] Xorg.0.log.3.15.3-200.fc20.x86_64
Same issue on Thinkpad T440s' when using the docking station. No issue without the docking station.
Created attachment 917308 [details] Kernel log with working 3.14.9 kernel
Created attachment 917309 [details] Kernel log with failing 3.15.4 kernel
Created attachment 917310 [details] Xorg log with working 3.14.9 kernel
Created attachment 917311 [details] Xorg log with failing 3.15.4 kernel
Same issue on my Thinkpad T440s with Ultra Dock 3. Upgrading to 3.15.4-200.fc20.x86_64 also did not help. Log files attached. Some maybe relevant excerpts as follows: dmesg: [ 0.000000] Reserving Intel graphics stolen memory at 0xbda00000-0xbf9fffff [ 1.790036] fb: conflicting fb hw usage inteldrmfb vs EFI VGA - removing generic driver vs. [ 1.292979] fb: switching to inteldrmfb from EFI VGA [ 15.475616] SELinux: initialized (dev drm, type drm), not configured for labeling Xorg.log.0: [ 14.991] (II) intel(0): Modeline "2560x1440"x60.0 241.50 2560 2608 2640 2720 1440 1443 1448 1481 +hsync +vsync (88.8 kHz eP) vs. [ 18.639] (II) intel(0): Modeline "1024x768"x60.0 65.00 1024 1048 1184 1344 768 771 777 806 -hsync -vsync (48.4 kHz e) [ 14.992] (II) intel(0): Using spanning desktop for initial modes vs. [ 18.639] (II) intel(0): Using fuzzy aspect match for initial modes [ 14.992] (II) intel(0): Output DP2 using initial mode 2560x1440 +1920+0 vs. [ 18.639] (II) intel(0): Output DP2 using initial mode 1024x768 +0+0 [ 14.993] (II) intel(0): Allocated new frame buffer 4480x1440 stride 17920, tiled vs. [ 18.641] (II) intel(0): Allocated new frame buffer 1024x768 stride 4096, tiled [ 15.055] (II) intel(0): Setting screen physical size to 1185 x 381 vs. [ 19.809] (II) intel(0): Setting screen physical size to 270 x 203 missing [ 15.821] (II) intel(0): EDID vendor "LGD", prod id 1034 [ 15.821] (II) intel(0): Printing DDC gathered Modelines: [ 15.821] (II) intel(0): Modeline "1920x1080"x0.0 140.10 1920 1980 2016 2092 1080 1083 1088 1116 +hsync -vsync (67.0 kHz eP) [ 16.043] (II) intel(0): Allocated new frame buffer 2560x1440 stride 10240, tiled [ 17.595] (II) intel(0): EDID vendor "LGD", prod id 1034 [ 17.595] (II) intel(0): Printing DDC gathered Modelines: [ 17.595] (II) intel(0): Modeline "1920x1080"x0.0 140.10 1920 1980 2016 2092 1080 1083 1088 1116 +hsync -vsync (67.0 kHz eP) [ 27.653] (II) intel(0): Using hsync ranges from config file [ 27.653] (II) intel(0): Using vrefresh ranges from config file vs. [ 20.628] (II) intel(0): Using EDID range info for horizontal sync [ 20.628] (II) intel(0): Using EDID range info for vertical refresh Also quite stunning that booting takes almost 4 seconds longer!
Same issue on a ThinkPad X240 with UltraDock on an EIZO monitor. Works again after downgrading to kernel-3.14.8-200. Bug 1118402 and Bug 1118950 seem to be duplicates.
*** Bug 1118950 has been marked as a duplicate of this bug. ***
I am also hitting this bug on my docked t440s. I tested following kernel: kernel-3.14.9-200.fc20.x86_64 - ok kernel-3.15.2-200.fc20.x86_64 - not ok kernel-3.15.3-200.fc20.x86_64 - not ok kernel-3.15.4-200.fc20.x86_64 - not ok kernel-3.16.0-0.rc4.git3.1.fc20.x86_64 - not ok Note that the last kernel is custom built. So it looks like this regression appeared in 3.15 and it is still in upstream kernel.
Created attachment 917655 [details] revert upstream commit patch just sent upstream.
please apply that fix to Fedora 3.15 or newer kernels.
*** Bug 1118402 has been marked as a duplicate of this bug. ***
Dave, can you give us link to mailing list or patchwork? I cannot find your post. Thanks.
(In reply to Jiri Pirko from comment #19) > Dave, can you give us link to mailing list or patchwork? I cannot find your > post. Thanks. http://www.spinics.net/lists/intel-gfx/msg49145.html
*** Bug 1119021 has been marked as a duplicate of this bug. ***
Added to F20-rawhide. Thanks.
Just tested 3.15.5-200.fc20.x86_64. Works fine. Thanks!
kernel-3.15.5-200.fc20 has been submitted as an update for Fedora 20. https://admin.fedoraproject.org/updates/kernel-3.15.5-200.fc20
*** Bug 1119864 has been marked as a duplicate of this bug. ***
kernel-3.15.5-200.fc20 has been pushed to the Fedora 20 stable repository. If problems still persist, please make note of it in this bug report.
The updated kernel does _not_ fix this problem on my T540p. Should I re-open the issue?
The update fixes the issue for me with my T540p + Dell monitor over HDMI from the dock.
kernel-3.15.5-200.fc20 also fixes the issue on my ThinkPad X240. Thank you for the quick fix.
Does it work for you with kernel-3.15.7-200.fc20? On the same setup (T440s + UltraDock + NEC monitor + Samsung monitor) When I use kernel-3.15.5-200.mst.fc20 I see in xrandr: Screen 0: minimum 320 x 200, current 3840 x 1080, maximum 8192 x 8192 eDP1 connected (normal left inverted right x axis y axis) 1920x1080 60.01 + ... DP1 disconnected (normal left inverted right x axis y axis) HDMI1 disconnected (normal left inverted right x axis y axis) DP2 disconnected (normal left inverted right x axis y axis) HDMI2 disconnected (normal left inverted right x axis y axis) DP3 disconnected (normal left inverted right x axis y axis) DP4 connected primary 1920x1080+0+0 (normal left inverted right x axis y axis) 510mm x 287mm 1920x1080 60.00*+ 50.00 59.94 ... DP5 connected 1920x1080+1920+0 (normal left inverted right x axis y axis) 477mm x 268mm 1920x1080 60.00*+ ... And all works as expected. However if I update to kernel-3.15.7-200.fc20 I'm getting mirrored/one huge screen again. xrandr there: Screen 0: minimum 320 x 200, current 3840 x 1080, maximum 8192 x 8192 eDP1 connected (normal left inverted right x axis y axis) 1920x1080 60.01 + ... DP1 disconnected (normal left inverted right x axis y axis) HDMI1 disconnected (normal left inverted right x axis y axis) DP2 connected 3840x1080+0+0 (normal left inverted right x axis y axis) 510mm x 287mm 1920x1080 60.00 + 3840x1080 60.00* ... HDMI2 disconnected (normal left inverted right x axis y axis)
For what it's worth, this did not fix the problem on my T540p. I've worked around it by attaching my external screen directly to the laptop and not the docking station.
(In reply to Ted Ross from comment #31) > For what it's worth, this did not fix the problem on my T540p. I've worked > around it by attaching my external screen directly to the laptop and not the > docking station. This bug is not about that, its about the connections external connections on the laptops not docks resolutions being dropped down to xvga res. The dock problem is something else. Which I see with my t540p as well. This kernel works for me. xrandr | grep 1080 Screen 0: minimum 320 x 200, current 3840 x 1080, maximum 8192 x 8192 1920x1080 60.01 + HDMI1 connected 1920x1080+1920+0 (normal left inverted right x axis y axis) 521mm x 293mm 1920x1080 60.00*+ 50.00 59.94 1920x1080i 60.00 50.00 59.94 DP2 connected 1920x1080+0+0 (normal left inverted right x axis y axis) 521mm x 293mm 1920x1080 60.00*+ ~]$ uname -a Linux WOPR2 3.15.6-200.fc20.x86_64 #1 SMP Fri Jul 18 02:36:27 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux