When I start X server, it freezes, I only see black screen. When I log in from another computer, 'ps axu' gives me: USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND root 6608 99.9 0.3 63368 3492 tty7 Rs+ 10:59 8:40 /usr/bin/X -dpi 96 -auth /var/run/slim.auth vt07 When I run X server with VESA, everrything is OK. I recompiled all X drivers (xf86-*) after upgrading xorg-server. Maybe this has something to do with my xorg.conf, I don't know, but I don't think settings in xorg.conf should make X server use 100% CPU... Versions used: [ebuild R ] x11-drivers/xf86-video-vesa-1.3.0 [ebuild R ] x11-base/xorg-server-1.4-r1 [ebuild R ] x11-base/xorg-x11-7.3 Reproducible: Always
Created attachment 132157 [details] xorg.log with i810 driver
Created attachment 132158 [details] xorg.log with vesa driver
Created attachment 132159 [details] xorg.conf
Created attachment 132160 [details] emerge --info
Does it happen with xf86-video-i810 v1.7.4? My hangups went away when using it
(In reply to comment #5) > Does it happen with xf86-video-i810 v1.7.4? > > My hangups went away when using it > You are right, everything is OK with v1.7.4.
Then your problem might be related to the intel driver which most recent versions of i810 (2.x+) equates to - but I don't know..
Ping :) Any updates here? I've just pushed 2.2.1 to portage. Could you try it out? Thanks
Still no improvement, neither with stable nor with testing.
Michał, which xorg-server version are you using? Could you try 1.4.0.90? Could you also try commenting out the "xtrap" extension? Thanks
Created attachment 144787 [details] Xorg.log for xorg-server-1.4.0.90 and without xtrap My previous test was with newest xorg-server found in portage. I attach xorg.log without xtrap extension. Still no change. BTW, I'm using 1.7.4 now to start xorg. I'll gladly help You any way You want, just write :-). Below I attach my xorg.conf (almost no change)
Created attachment 144788 [details] xorg.conf
Just for the sake of it : 1) In the Monitor section of your xorg.conf, could you : - comment out the Monitor-VGA, the HorizSync, the VertRefresh, the Modeline and the PreferredMode lines. - add Option "DPMS" to that section 2) In the Screen section, could you comment out all the Display SubSections. With both these changes, the Intel driver should be able to autoconfigure your screen. I have no idea if that will actually help, but everything leads me to believe that those lines are no longer needed anyway, so you might as well try without them. If that doesn't work, the only solution is to rebuild xorg with debug symbols and attach gdb from a remote shell to the Xorg process to see what piece of code it's looping over. Thanks
(In reply to comment #13) > Just for the sake of it : > > 1) In the Monitor section of your xorg.conf, could you : > - comment out the Monitor-VGA, the HorizSync, the VertRefresh, the Modeline > and the PreferredMode lines. > - add Option "DPMS" to that section > > 2) In the Screen section, could you comment out all the Display SubSections. > > With both these changes, the Intel driver should be able to autoconfigure your > screen. I have no idea if that will actually help, but everything leads me to > believe that those lines are no longer needed anyway, so you might as well try > without them. > > Thanks > Thanks for response. Well, it works :-). A least xorg-server starts. Great :-). However, I only got 1024x768, but my LCD's native resolution is 1280x1024. It looks similar to bug 192066. $ xrandr Screen 0: minimum 320 x 200, current 1024 x 768, maximum 1024 x 1024 VGA connected 1024x768+0+0 (normal left inverted right x axis y axis) 0mm x 0mm 1024x768 60.0* 800x600 60.3 640x480 59.9 Seems like I will have to try workaround mentioned in bug 192066. Anyway, I still think that wrong configuration should never lock X server...
Workaround fixed resolution problem (Monitor-VGA in Device Section and PreferredResolution in Monitor Section), so this bug report may be marked as resolved (invalid because of configuration problem?) Thanks for help :-).
(In reply to comment #15) > Workaround fixed resolution problem (Monitor-VGA in Device Section and > PreferredResolution in Monitor Section), so this bug report may be marked as > resolved (invalid because of configuration problem?) You've just found a really nasty bug, if I could kindly ask you to file a bug upstream. This is really something that needs to be fixed upstream. If you do open, assign it to the Intel/2D driver component, they'll figure out if it's their fault or not. Please paste the url here, so that I can track upstream's work. @x11, CCing you because it might interest you for other cases with other drivers. > Thanks for help :-). You're welcome, I'm glad to see it fixed
This has been already fixed. https://bugs.freedesktop.org/show_bug.cgi?id=12676
Great :) I'm not sure if it's worth a backport if you can workaround it for now... Donnie, up to you. Thanks
I'm planning to add the just-released 1.5 snapshot (1.4.99.x), so keep an eye out for that.
It's available in the x11 overlay, which you can get from layman. You'll need to add a few entries to package.unmask to get it going. Test it out if you can, and let us know how it works.
Created attachment 147189 [details] xorg.log Sorry for long delay, I haven't got much time now. Masked version doesn't work for me (xorg fails to start). Xorg.log is weird IMO (I have rebuilt all drivers), but I can't investigate this failure more now. I'll wait until new version hits the tree and then report :-).
Michał, Please don't hesitate to reopen this bug with more rencent versions of xorg-server and xf86-video-intel. Thanks
