On Fri, Sep 28, 2007 at 05:38:55PM +0200, Maarten Lankhorst wrote:
Luc Verhaegen schreef:
On Thu, Sep 27, 2007 at 08:02:09AM +0200, Maarten Lankhorst wrote:
I have tested it, however it fails to initialise properly. It seems to be that my panel is not detected or connected in software properly. I attached a Xorg log. Do you think you could take a look at it?
Cheers, Maarten
Please try the latest version of our driver. I tried, I found out the folowing:
1. There is a bug in rhd_driver.c
if (!Monitor && (Connector->Type == RHD_CONNECTOR_PANEL)) {
...
} else {
It should be something like this:
if (!Monitor) { if (Connector->Type == RHD_CONNECTOR_PANEL)
xf86DrvMsg(rhdPtr->scrnIndex, X_WARNING, "Unable to attach a"
" monitor to connector \"%s\"\n", Connector->Name);
Output->Active = FALSE;
} else {
2. Monitor is only initialised if (Connector->DDC) exists, my inspiron 6400 has been defined with RHD_DDC_NONE, so Connector->DDC is false and my panel will never be detected. I'm assuming there is missing some code that would get my EDID data despite that.. (I just don't know what EDID is :-/)
Cheers, Maarten
My code is correct for the current situation. We need a rhdMonitor for a panel, we do not, under any circumstances, want to set a mode on the panel which it cannot accept. So we need information about this panel. No information == no panel, it is that simple. For DVI or VGA attached monitors, it is ok to rely on configured information. No real damage can be done, and they all should support the modes for early 90s fishbowls anyway, which is the default monitor that is attached when no EDID data is found. I will write up the code to get the panel information out of atombios this weekend. Luc Verhaegen. SUSE/Novell X Driver Developer. -- To unsubscribe, e-mail: radeonhd+unsubscribe@opensuse.org For additional commands, e-mail: radeonhd+help@opensuse.org