On 9/23/07, Mich Lanners <[EMAIL PROTECTED]> wrote: > Hi Alex, > > On 22 Sep, this message from Alex Deucher echoed through cyberspace: > > On 9/22/07, Alex Deucher <[EMAIL PROTECTED]> wrote: > > > >> Once we sort this out, I'll add a MacModel option for the mini. > > > > fixed in ati git master: > > Option "MacModel" "mini" > > > > the following option should also work with the current driver: > > Option "ConnectorTable" "4,1,1,3,0,1,2,6" > > OK, this works to detect the correct monitor on DVI-I, either with a DVI > cable or a VGA cable (plus adapter). So yes, this seems to fix the EDID > detection problem. > > In the version I'm trying (1:6.7.192-4), radeon detects two monitor > ports: DVI-I and S-Video. Is that the intended way? Previous versions > detected 3 ports
they detected 3 ports because that's the default layout when the driver doesn't know what connectors the card has. I don't know what's included in 6.7.192-4 exactly. > > I still have the problem that the version from unstable (1:6.7.192-4) > hangs just after validating the modes, eating all CPU. Is this still a problem when using the connectortable option or with newer versions? > > Unfortunately the ConnectorTable option is not supported in 1:6.6.193-3, > so I can't verify if the EDID fix makes radeon choose a correct mode in > the absence of UseFBDev option. Right. 6.6.x didn't have that option. there was no way to force the connector layout. fbdev sometimes did the right thing. It's one of the improvements in the new driver. > > So, anything about the hang that I can try? does it still hang with the connectortable option or with the latest version from ati git? Alex -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
