Comment 5 for bug 375545
Revision history for this message
| Geir Ove Myhr (gomyhr) wrote Re: [i945] shuttle X50/intel 945 "No outputs definitely connected" | #5 |
© 2004
Canonical Ltd.
•
Terms of use
•
Data privacy
•
Contact Launchpad Support
•
Blog
•
Careers
•
System status
•
0e1f616
(Get the code!)
You're right, this is not about the xorg.conf. For some reason, the driver only checks the VGA connection, and does not find anything there. Normally, built in monitors are connected via an LVDS connection. In this case, the chipset is a 945G (not 945GM - which is the mobile version normally used in laptops). I haven't seen a computer with a non-mobile intel chipset and and a built in monitor before[*].
I wonder if the monitor is connected using LVDS, but since the driver sees that it is a non-mobile chipset, it simply assumes there is nothing there (I'm not even sure if the non-mobile chipsets have LVDS possibility, but let's assume it has for now). In this case, the telling the driver explicitly to use LVDS may help. If this is the case, we may quirk the driver to do this for this particular computer.
First, I would like you to add a minimal xorg.conf by running `dpkg-reconfigure -phigh xserver-xorg` and then add the line
Option "ModeDebug" "true"
to the "Device" section (or you can get a ready-made file from http://
This will add a lot of extra information to Xorg.0.log the next time you start. Please upload this Xorg.0.log.
Next, I would like to know if the log from the failsafe session tells us anything interesting. By adding the line
Driver "vesa"
to the "Device" section of xorg.log, that should use the generic vesa driver which is the same as what the failsafe option does. Please also upload the log for this case.
Finally, try to enable the LVDS explicitly in your xorg.conf. I'm not very familiar with the details for how to do this, but I guess following the "Ignore LVDS Quirk" at
https:/
should give the basic syntax, except that instead of
Option "Ignore" "True"
you should use
Option "Ignore" "False"
As I said, I'm not sure this is the right way to do it but hey, it just might work. If it does, it should be possible to add a quirk to the driver.
[*] Usually the problem is the other way around: A mobile chipset is used in a PC without a built-in monitor on the LVDS, and the driver implicitly assumes that all mobile chipsets have something attached to their LVDS.