[Bug 586053] New: [NoDDC/NoEDID][iG31] X -configure locks up system requiring hard reset
http://bugzilla.novell.com/show_bug.cgi?id=586053 http://bugzilla.novell.com/show_bug.cgi?id=586053#c0 Summary: [NoDDC/NoEDID][iG31] X -configure locks up system requiring hard reset Classification: openSUSE Product: openSUSE 11.3 Version: Factory Platform: PC OS/Version: Other Status: NEW Severity: Normal Priority: P5 - None Component: X.Org AssignedTo: bnc-team-xorg-bugs@forge.provo.novell.com ReportedBy: mrmazda@earthlink.net QAContact: xorg-maintainer-bugs@forge.provo.novell.com Found By: Community User Blocker: --- Display has no usable DDC or EDID. 00:00.0 Host bridge: Intel Corporation 82G33/G31/P35/P31 Express DRAM Controller (rev 10) 00:02.0 VGA compatible controller: Intel Corporation 82G33/G31 Express Integrated Graphics Controller (rev 10) running kernel is 2.6.33-rc7-3-default (M2) due to bug 586044 sax2 -p finds Chip: 0 is -> Intel G33 00:02:0 0x8086 0x29c2 PCI Intel sax2 & sax2 -c0 run before initial attempt to startx or runlevel 5 aborted X -configure following sax2 failure locked up the machine startx without xorg.conf brings up working 1024x768 No evidence in YaST2 of any means to configure my normally used 1600x1200 Krandrtray won't let me chose any resolution above 1024x768 sax2 after exiting a 1024x768 X session works to allow specifying 2048x1536@60hz display @ 1600x1200, but then startx exits: (EE) intel(0): No Kernel modesetting driver detected. (EE) Screen(s) found, but none have a usable configuration Fatal server error: no screens found Another attempt at 'X -configure' locked up the machine again, so I'm stuck at a miserably low 1024x768 resolution. Bug 586044 and 583424 are same system, with attached installation logs. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
http://bugzilla.novell.com/show_bug.cgi?id=586053
http://bugzilla.novell.com/show_bug.cgi?id=586053#c1
--- Comment #1 from Felix Miata
http://bugzilla.novell.com/show_bug.cgi?id=586053
http://bugzilla.novell.com/show_bug.cgi?id=586053#c2
Stefan Dirsch
Display has no usable DDC or EDID.
Ancient monitor? Broken VGA cable or still using BNC cables? KVM switch part of the game?
00:00.0 Host bridge: Intel Corporation 82G33/G31/P35/P31 Express DRAM Controller (rev 10) 00:02.0 VGA compatible controller: Intel Corporation 82G33/G31 Express Integrated Graphics Controller (rev 10) running kernel is 2.6.33-rc7-3-default (M2) due to bug 586044 sax2 -p finds Chip: 0 is -> Intel G33 00:02:0 0x8086 0x29c2 PCI Intel
Running default kernel 2.6.33 shouldn't be an issue here. Graphics hardware is Intel G33.
sax2 & sax2 -c0 run before initial attempt to startx or runlevel 5 aborted
/var/log/SaX.log ?
X -configure following sax2 failure locked up the machine
I don't think that's still supported. Often conflicting or broken drivers are tried to be loaded.
startx without xorg.conf brings up working 1024x768
1024x768 is the default resolution when DDC/EDID information is not available.
No evidence in YaST2 of any means to configure my normally used 1600x1200
See http://lists.opensuse.org/opensuse-factory/2009-12/msg00017.html
Krandrtray won't let me chose any resolution above 1024x768
Probably krandrtry can't add new modes. Should be possible by using xrandr, even higher than the initial one.
sax2 after exiting a 1024x768 X session works to allow specifying 2048x1536@60hz display @ 1600x1200, but then startx exits: (EE) intel(0): No Kernel modesetting driver detected. (EE) Screen(s) found, but none have a usable configuration Fatal server error: no screens found
Makes no sense. SaX2 itself starts an intel-based driver, on which the configuration interface is running. The error message means that KMS (Kernel mode setting) is not active, i.e. intel-agp and i915 kernel modules have not been loaded. intel meanwhile only supports KMS. So why should have been SaX2 able to start before? Anyway, I suggest to use xlog2xconf to create an xorg.conf and adjust this one to your needs. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
http://bugzilla.novell.com/show_bug.cgi?id=586053
http://bugzilla.novell.com/show_bug.cgi?id=586053#c3
--- Comment #3 from Felix Miata
sax2 & sax2 -c0 run before initial attempt to startx or runlevel 5 aborted
/var/log/SaX.log ?
Forgot. :-p -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
http://bugzilla.novell.com/show_bug.cgi?id=586053
http://bugzilla.novell.com/show_bug.cgi?id=586053#c4
--- Comment #4 from Felix Miata
http://bugzilla.novell.com/show_bug.cgi?id=586053
http://bugzilla.novell.com/show_bug.cgi?id=586053#c5
Felix Miata
http://bugzilla.novell.com/show_bug.cgi?id=586053
http://bugzilla.novell.com/show_bug.cgi?id=586053#c6
Stefan Dirsch
http://bugzilla.novell.com/show_bug.cgi?id=586053
http://bugzilla.novell.com/show_bug.cgi?id=586053#c7
Felix Miata
(In reply to comment #1)
Display has no usable DDC or EDID.
Ancient monitor? Broken VGA cable or still using BNC cables? KVM switch part of the game?
Long ago I got a bunch of Dell/Trinitron P991s for free, about half of which for reasons unknown to me had non-working DDC/EDID. I kept three, and use primarily one with non-working in order to confirm the devs are not neglecting keeping X working as well as before for those with similarly handicapped displays.
00:00.0 Host bridge: Intel Corporation 82G33/G31/P35/P31 Express DRAM Controller (rev 10) 00:02.0 VGA compatible controller: Intel Corporation 82G33/G31 Express Integrated Graphics Controller (rev 10) running kernel is 2.6.33-rc7-3-default (M2) due to bug 586044 sax2 -p finds Chip: 0 is -> Intel G33 00:02:0 0x8086 0x29c2 PCI Intel
Running default kernel 2.6.33 shouldn't be an issue here. Graphics hardware is Intel G33.
According the the motherboard manufacturer the graphics hardware is G31, though I see according to Xorg.0.log output there doesn't seem to be any such device. Who am I to disagree with the people who made it? http://www.biostar.com.tw/app/en/mb/content.php?S_ID=355
X -configure following sax2 failure locked up the machine
I don't think that's still supported. Often conflicting or broken drivers are tried to be loaded.
It was precisely http://lists.opensuse.org/opensuse-factory/2009-12/msg00017.html and related SaX2 discussion following that brought X -configure to my attention. Though not as powerful as SaX2, it has worked on every attempt I've made to use it since, until this bug.
startx without xorg.conf brings up working 1024x768
1024x768 is the default resolution when DDC/EDID information is not available.
That constitutes an improvement over 800x600, which is what I remember has been the default for many moons since the jump from 640x480.
Krandrtray won't let me chose any resolution above 1024x768
Probably krandrtry can't add new modes. Should be possible by using xrandr, even higher than the initial one.
This is where my memory of my lost response attempt is worst. I now only remember being unable to tell from the xrandr man page how to add the modes that formerly were available automatically just from providing via xorg.conf HorizSync and VertRefresh ranges, in part by not finding any description of the "name" parameter required as input for --newmode, and in part by not seeing for --fb whether size means device pixels, pt, mm, em or something else.
sax2 after exiting a 1024x768 X session works to allow specifying 2048x1536@60hz display @ 1600x1200, but then startx exits: (EE) intel(0): No Kernel modesetting driver detected. (EE) Screen(s) found, but none have a usable configuration Fatal server error: no screens found
Makes no sense. SaX2 itself starts an intel-based driver, on which the configuration interface is running. The error message means that KMS (Kernel mode setting) is not active, i.e. intel-agp and i915 kernel modules have not been loaded. intel meanwhile only supports KMS. So why should have been SaX2 able to start before?
Considering that X will start with no xorg.conf at all, it certainly makes no sense to me.
Anyway, I suggest to use xlog2xconf to create an xorg.conf and adjust this one to your needs.
This bug is my first exposure to the existence of this file. xlog2xconf has no manual entry, and no --help explanations for -e or -f or - options. I tried it on both i845G and iG31 systems, getting virtually useless 821 byte skeletons: Section "Device" Identifier "Builtin Default intel Device 0" Driver "intel" EndSection Section "Screen" Identifier "Builtin Default intel Screen 0" Device "Builtin Default intel Device 0" EndSection Section "Device" Identifier "Builtin Default fbdev Device 0" Driver "fbdev" EndSection Section "Screen" Identifier "Builtin Default fbdev Screen 0" Device "Builtin Default fbdev Device 0" EndSection Section "Device" Identifier "Builtin Default vesa Device 0" Driver "vesa" EndSection Section "Screen" Identifier "Builtin Default vesa Screen 0" Device "Builtin Default vesa Device 0" EndSection Section "ServerLayout" Identifier "Builtin Default Layout" Screen "Builtin Default intel Screen 0" Screen "Builtin Default fbdev Screen 0" Screen "Builtin Default vesa Screen 0" EndSection -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
http://bugzilla.novell.com/show_bug.cgi?id=586053
http://bugzilla.novell.com/show_bug.cgi?id=586053#c8
--- Comment #8 from Felix Miata
I don't think we have a solution for that issue (broken DDC/EDID). It should be exotic nowadays. Really.
I disagree. Just because the monitor reports some particular information doesn't mean what it reports will give the right results for a user's needs. Just because I have a working DDC/EDID CRT that supports 2048x1536 doesn't mean that's the resolution I want used. Has X -configure been announced upstream as deprecated or removed? If not, it should still work. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
http://bugzilla.novell.com/show_bug.cgi?id=586053
http://bugzilla.novell.com/show_bug.cgi?id=586053#c9
Stefan Dirsch
http://bugzilla.novell.com/show_bug.cgi?id=586053
http://bugzilla.novell.com/show_bug.cgi?id=586053#c10
--- Comment #10 from Felix Miata
The fact, that there are still monitors around without (working) DDC/EDID doesn't justify any amount of work trying to support also these corner cases.
Of itself, you're probably right. But as I already wrote, just because working DDC/EDID exists does not mean what results from it is right for the user. There needs to be understandable and readily discoverable instructions for what mere mortal users with bad DDC/EDID need to do to get the results they want.
How to add a new mode by using xrandr. Basically this is
xrandr --newmode <newmode> xrandr --addmode <output> <newmode> xrandr --output <outpu> --mode <newmode>
See "xrandr --help".
I saw all of that. It's too terse. What's the difference between add and new? Why would anyone need to specify an output whenever there is only one output? In the past people have been able to configure X without ever having to start any display manager. I don't see any way to do this now. Installation, where I normally first did it at fresh install, now skips the whole process, apparently assuming xorg.conf-free automatic configuration cannot fail.
Documentation about the usage of xlog2xconf can be found in the first lines of the script.
I had no idea it was a script.
It's main purpose is to create a xorg.conf, which reflects the automatic configuration and then can be adjusted to your needs. Hope that helps.
In my first tries, example of which ends comment 7, it essentially requires knowledge to build all video-related file components from scratch. Maybe it does what it needs to do with mga, nv or radeon, which I've not yet had time to try in M2 or M3, but it sure doesn't seem at all helpful with intel. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
http://bugzilla.novell.com/show_bug.cgi?id=586053
http://bugzilla.novell.com/show_bug.cgi?id=586053#c
Felix Miata
http://bugzilla.novell.com/show_bug.cgi?id=586053
http://bugzilla.novell.com/show_bug.cgi?id=586053#c11
Stefan Dirsch
http://bugzilla.novell.com/show_bug.cgi?id=586053
http://bugzilla.novell.com/show_bug.cgi?id=586053#c12
Felix Miata
http://bugzilla.novell.com/show_bug.cgi?id=586053
http://bugzilla.novell.com/show_bug.cgi?id=586053#c13
Stefan Dirsch
http://bugzilla.novell.com/show_bug.cgi?id=586053
http://bugzilla.novell.com/show_bug.cgi?id=586053#c14
Felix Miata
No idea, why you think that is critical.
Because I equate system lockup with crash/see whiteboard. From: https://bugzilla.novell.com/page.cgi?id=fields.html#importance "Critical Crash..."
It has never been recommended by us to use 'X -configure'.
As a standard upstream component, its recommendation would seem to be implied by deprecation of SaX2. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
http://bugzilla.novell.com/show_bug.cgi?id=586053
http://bugzilla.novell.com/show_bug.cgi?id=586053#c15
Stefan Dirsch
http://bugzilla.novell.com/show_bug.cgi?id=586053
http://bugzilla.novell.com/show_bug.cgi?id=586053#c16
--- Comment #16 from Felix Miata
participants (1)
-
bugzilla_noreply＠novell.com