Intel graphics machines G45 and HD Graphics "has no monitor section" #87

Closed
davshao opened this Issue Feb 15, 2014 · 8 comments

Projects

None yet

3 participants

@davshao
davshao commented Feb 15, 2014

For two separate machines with Intel graphics, one G45 and the other HD Graphics, using latest 3.7-DEVELOPMENT and after building dports from scratch locally post February 4, 2014, the monitor does not appear to be detected. From the resulting /var/log/Xorg.0.log:

345.562 --- Start of built-in configuration ---
[ 345.562] Section "Device"
[ 345.562] Identifier "Builtin Default intel Device 0"
[ 345.562] Driver "intel"
[ 345.562] EndSection
[ 345.562] Section "Screen"
[ 345.562] Identifier "Builtin Default intel Screen 0"
[ 345.562] Device "Builtin Default intel Device 0"
[ 345.562] EndSection
[ 345.562] Section "Device"
[ 345.562] Identifier "Builtin Default vesa Device 0"
[ 345.562] Driver "vesa"
[ 345.562] EndSection
[ 345.562] Section "Screen"
[ 345.562] Identifier "Builtin Default vesa Screen 0"
[ 345.562] Device "Builtin Default vesa Device 0"
[ 345.562] EndSection
[ 345.562] Section "Device"
[ 345.562] Identifier "Builtin Default fbdev Device 0"
[ 345.562] Driver "fbdev"
[ 345.562] EndSection
[ 345.562] Section "Screen"
[ 345.562] Identifier "Builtin Default fbdev Screen 0"
[ 345.562] Device "Builtin Default fbdev Device 0"
[ 345.562] EndSection
[ 345.562] Section "ServerLayout"
[ 345.562] Identifier "Builtin Default Layout"
[ 345.562] Screen "Builtin Default intel Screen 0"
[ 345.562] Screen "Builtin Default vesa Screen 0"
[ 345.562] Screen "Builtin Default fbdev Screen 0"
[ 345.562] EndSection
345.562 --- End of built-in configuration ---
345.563 ServerLayout "Builtin Default Layout"
345.563 |-->Screen "Builtin Default intel Screen 0" (0)
345.563 | |-->Monitor ""
345.563 | |-->Device "Builtin Default intel Device 0"
345.563 No monitor specified for screen "Builtin Default intel Screen 0".
Using a default monitor configuration.
...
[ 348.485] drmOpenDevice: node name is /dev/dri/card0
[ 348.485] drmOpenDevice: open result is 10, (OK)
[ 348.485] drmOpenByBusid: Searching for BusID pci:0000:00:02.0
[ 348.486] drmOpenDevice: node name is /dev/dri/card0
[ 348.486] drmOpenDevice: open result is 10, (OK)
[ 348.486] drmOpenByBusid: drmOpenMinor returns 10
[ 348.486] drmOpenByBusid: drmGetBusid reports pci:0000:00:02.0
348.486 Falling back to old probe method for vesa
348.486 VGA arbiter: cannot open kernel arbiter, no multi-card support
348.487 intel(0): Creating default Display subsection in Screen section
"Builtin Default intel Screen 0" for depth/fbbpp 24/32
348.487 intel(0): Depth 24, (--) framebuffer bpp 32
348.487 intel(0): RGB weight 888
348.487 intel(0): Default visual is TrueColor
348.487 intel(0): Integrated Graphics Chipset: Intel(R) HD Graphics
348.487 intel(0): CPU: x86-64, sse2, sse3, ssse3, sse4.1, sse4.2
348.487 intel(0): Framebuffer tiled
348.488 intel(0): Pixmaps tiled
348.488 intel(0): "Tear free" disabled
348.488 intel(0): Forcing per-crtc-pixmaps? no
348.488 intel(0): Output LVDS1 has no monitor section
348.640 intel(0): Output VGA1 has no monitor section
348.640 intel(0): Output HDMI1 has no monitor section
348.641 intel(0): Output DP1 has no monitor section
348.641 intel(0): DPI set to (96, 96)
...
349.582 XINPUT: Adding extended input device "" (type: KEYBOARD, id 7)
354.827 intel(0): EDID vendor "HSD", prod id 1206
354.827 intel(0): Printing DDC gathered Modelines:
354.827 intel(0): Modeline "1366x768"x0.0 71.90 1366 1414 1480 1486 768 774 782 806 -hsync -vsync (48.4 kHz eP)

@jrmarino
Collaborator

Three days ago the driver for the latest-3.7 was changed to 2.21.15
Prior to that it was 2.21.9.
If you are referencing 4 February, you might be not be using the latest. It might be worth updating dports and checking again.

@ftigeot
Collaborator
ftigeot commented Feb 15, 2014

This is not a dports issue but most likely a shortcoming our drm/i915 kernel driver.
For some reason, Xorg is unable to detect/set a default mode on the output the screen is plugged in.
Running xrandr manually is a possible workaround.

Example for a 1280x1024 monitor:
xrandr --screen 1 --output HDMI1 --mode 1280x1024

@davshao
davshao commented Feb 17, 2014

I was able to successfully start xfce4 on both machines using the xrandr suggestion.

I put in my .xinitrc file

xrandr --screen 0 --output HDMI1 --mode 1920x1080 &

for the Intel HD Graphics machine connected via HDMI -> DVI converter to a Samsung monitor and

xrandr --output LVDS1 --mode 1366x768 &

for the laptop.

@ftigeot
Collaborator
ftigeot commented Feb 27, 2014

This upstream bug report is mostly about the same issue:
https://bugs.freedesktop.org/show_bug.cgi?id=66494

@ftigeot
Collaborator
ftigeot commented Mar 2, 2014

Adding
Option "ReprobeOutputs" "true"
in the Device section of xorg.conf as suggested in commit 1445a62da8a08acd8a732176d085fd098f68cec3 of xf86-video-intel confirmed as a correct workaround for this issue.

@ftigeot ftigeot pushed a commit to DragonFlyBSD/DeltaPorts that referenced this issue Mar 2, 2014
François Tigeot Tweak xf86-video-intel for DragonFly 3.7+
* xf86-video-intel 2.21.15+ expects the kernel to set optimal video modes
  on all video outputs

* This is not the case with the current DragonFly implementation. Force
  a reprobe of outputs during X startup

Bug reports:
DragonFlyBSD/DPorts#87
https://bugs.freedesktop.org/show_bug.cgi?id=66494
844b0cc
@jrmarino
Collaborator
jrmarino commented Mar 3, 2014

we are currently between runs on muscles so I'll try to regenerate this port specifically soon.

@jrmarino
Collaborator
jrmarino commented Mar 3, 2014

okay, I think this closes the issue.
0b8ec5a

@jrmarino jrmarino closed this Mar 3, 2014
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment