[systemd-devel] Someone working on gui for multiseat?

[poma]

On 01/23/2013 02:55 AM, Lennart Poettering wrote:
> On Mon, 21.01.13 21:56, poma (pomidorabelisima at gmail.com) wrote:
> 
>> This is contradictory to what I have experienced with DisplayLink DL-165
>> based device[1].
>> *Without engaging multiseat(udev's rule)* picture is garbled on the
>> connected monitor, regardless of whether monitor(DVI) comes with the
>> correct EDID or monitor(VGA)[2] without/broken EDID.
>> EDID loading by /etc/modprobe.d/drm_kms_helper.conf:
>> options drm_kms_helper edid_firmware=edid/848-480[2]
>> doesn't work at all,
>> and even with working kernel boot parameter:
>> drm_kms_helper.edid_firmware=DVI-I-1:edid/848-480[2]
>> system boot stuck few minutes and picture is garbled as before.
>> [drm] Got external EDID base block and 0 extensions from "edid/848-480"
>> for connector "DVI-I-1"
>> [drm] write mode info 153
>> …
>> And when connected monitor get activated with e.g.
>> xfce4-display-settings(picture is still garbled),
>> then nouveau start spitting messages:
>> nouveau E[     DRM] fail set_domain
>> nouveau E[     DRM] validate gart_list
>> nouveau E[     DRM] validate: -22
>> nouveau E[     DRM] fail set_domain
>> …
>> with slowing down the system.
>>
>> *With engaging multiseat(udev's rule)*[3] monitor adapter[1] is powered
>> on, but connected monitor(VGA/DVI) stays powered off(NO SYNC).
>> If the system ever reaches gdm.
>> Major Bummer!
> 
> garbled output and issues in the kernel logs usually indicate a
> kernel/driver problem, not a systemd problem. Have you filed a bug in
> rhbz against the kernel? Make sure to CC airlied who write the DRM driver.
> 
> Lennart
> 

https://bugzilla.redhat.com/show_bug.cgi?id=904232
;)

Cheers,
poma