On Tue, Jan 12, 2010 at 2:08 AM, Christoph Rissner <<a href="mailto:c.r@visotech.at">c.r@visotech.at</a>> wrote: <div class="gmail_quote"><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> I pulled the latest version, but being a git noob I'm not sure about that, is a "git pull" sufficient?</blockquote><div> Yep. And the output you copied shows you had the latest. </div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> I found the sysfs file in a different path, is that ok? # find /sys/ -name "*edid*" /sys/devices/pci0000:00/0000:00:1d.7/usb1/1-5/graphics/fb1/edid # cat `find /sys/ -name "*edid*"` # There is no output, I guess that's what Bernie expected. </blockquote></div> I'm surprised there's no /sys/class/graphics tree (some of my other scripts, e.g. performance monitoring, assume it). I know sysfs is a loose contract, but hopefully that'll be in a standardized place in the future. But I'm assuming the edid file was mapped properly (cat metrics_misc and other udlfb-specific sysfs files in the same directory to confirm that's really a directory mapped to the udlfb driver). So I'm at a loss for your particular problem. For all the combinations of DisplayLink devices and monitors I have, I'm getting a proper EDID back. Now, there are valid scenarios where you won't (e.g. monitors with no edid), but this isn't one of them. One thing I'd check - you said EDID read appeared to be working on Windows, but wasn't sure -- can you check the display control panel, and see if it's offering you modes that are greater than the monitor's capabilities -- that would be a sign that edid is not getting returned there, either. Perhaps if someone else is seeing a problem, then can add to the datapoints? Thanks, Bernie Thinking of Haiti today