[15:27] jbarnes: any ideas on what's happening here? (bug #431812) [15:27] Launchpad bug 431812 in sysvinit "fbcon loading a mystery (screen powers off)" [High,Confirmed] https://launchpad.net/bugs/431812 [15:27] AFAIK fbcon is loaded when checking for KMS in the -intel driver [15:30] tseliot: Bug 397839 is not fixed even after applying the fixes Richard mentions in his blog... :( any ideas? [15:30] Launchpad bug 397839 in xorg-server "Screen randomly goes off in karmic" [Critical,Fix released] https://launchpad.net/bugs/397839 [15:31] this is in an ATI X1400 [15:31] let me check [15:31] see my last comment... gpm does not give any output , and i'm not sure how to debug it! :( [15:33] it seems to happen much rare than it used to , but still is a problem and the notification area keeps displaying the session is active icon [15:38] mac_v: it might be worth notifying hughsie about the problem [15:39] in the blog? ... yeah, i was thinking about it , but wanted to add a comment after i had some debug info [15:39] mac_v: or on irc.gnome.org (in #gnome-hackers) [15:43] hmm... ok , so if i mention the lp bug , they would have a look? [just asking , since i'm not sure how upstream irc guys ask for stuff ] ;) [15:55] mac_v: yes, if the report contains information which can help upstream [15:55] (in general) [15:55] ah ok , thanks [16:16] tseliot: yeah sounds like you need to load i915 and fbcon in your initrd instead [16:19] jbarnes: ah, right, let me try [16:37] jbarnes: I added i915 and fbcon to initramfs, rebuilt the initrd but I can still reproduce the problem [16:40] what kernel commit are you at? [16:40] apw: ^^ [16:59] jbarnes: in the meantime I have noticed that I get a few errors when fbcon and i915 are in the initramfs: http://paste.ubuntu.com/273572 [17:02] tseliot: yeah that looks like a pretty broken config... X isn't even detecting drm [17:02] loading those modules early shouldn't cause that though [17:03] jbarnes: but this means that the device wasn't even created (in time?) [drm] Failed to open DRM device for : No such file or directory [17:03] yeah something bad happened [17:04] if /dev/dri doesn't exist that would explain it [17:04] 3d acceleration was broken [17:05] so it could be [17:05] let me check [17:08] jbarnes: /dev/dri doesn't exist [17:12] jbarnes: also, if I pass i915.modeset=1 grub complains that it's an unknown option [17:12] i915.modeset:1 that is [17:12] sounds like some funkiness with your initrd [17:12] udev will create the nodes after drm/i915 load [17:13] or it should [17:13] so either i915 isn't loading or udev is failing somehow [17:13] tseliot: shouldn't drm be in initrd as well? i915 depends on it. [17:13] albert23: I tried putting the dependencies in the initrd too [17:15] loading only fbcon in the initrd seems to give me KMS back :-) === hal_line_ is now known as halfline [19:56] the 'unknown option i915.modeset' is expected, when i915 isn't builtin [20:04] would anyone have any idea of what could cause bug #429277 ? I would have tought that this is a gnome-settings-daemon issue again, but seb128 claims that this was fixed already. [20:04] Launchpad bug 429277 in xserver-xorg-input-mouse "left-handed mouse no longer working" [Undecided,Invalid] https://launchpad.net/bugs/429277 [20:06] so you filed a bug against a driver that you're not using instead. nice. [20:06] excuse-me? [20:06] you're not using xserver-xorg-input-mouse [20:07] ah? [20:07] … [20:08] so what driver does a mouse use these days? [20:08] is -mouse used at all, even? [20:10] or does it all goe through that new unified input device driver? [20:10] Q-FUNK: -evdev is normally used [20:10] albert23: interesting. so why is -mouse still shipped then? does it still have some limitted usage cases? [20:11] Q-FUNK: I think -vmmouse still needs -mouse [20:11] sorry if I'm not up-to-date on general X server issues, in this case, on the input modules.