X aborts w/ backtrace: 'Mode pool is empty' and 'No valid modes found'

Bug #180960 reported by Tye
2
Affects Status Importance Assigned to Milestone
X.Org X server
Won't Fix
High
xserver-xorg-video-imstt (Ubuntu)
Won't Fix
Undecided
Unassigned

Bug Description

Binary package hint: xorg

Already reported upstream to Xorg as their Bugzilla # 13949 (http://bugs.freedesktop.org/show_bug.cgi?id=13949), apologies if that precludes submitting it here, just covering all relevant/helpful bases:

As of Ubuntu 7.10 "Gutsy", and also appearing very early-on in Debian "Lenny" (testing), X now fails to start up for me, whereas X had worked just fine on this same hardware in Debian "Etch" and also in Ubuntu "Edgy" and "Feisty". Seems to be something about Xorg has changed, as deployed in the affected releases. Towards the end of my Xorg.0.log there's a whole bunch of lines rejecting every possible resolution (even known-valid ones), like this:

---
(II) imstt(0): Not using default mode "1024x768" (bad mode clock/interlace/doublescan)
(II) imstt(0): Not using default mode "1024x768" (unknown reason)

[Repeat w/ a wide variety of other possible resolutions, all rejected]

(WW) imstt(0): Mode pool is empty
(EE) imstt(0): No valid modes found

Backtrace:
0: /usr/bin/X(xf86SigHandler+0x94) [0x10093f44]
1: [0x100344]
2: /usr/lib/xorg/modules/drivers//imstt_drv.so [0xf941520]
3: /usr/lib/xorg/modules/drivers//imstt_drv.so [0xf94186c]
4: /usr/bin/X(InitOutput+0xb08) [0x1006e2c8]
5: /usr/bin/X(main+0x294) [0x1002e064]
6: /lib/libc.so.6 [0xfc79380]
7: /lib/libc.so.6 [0xfc795c4]

Fatal server error:
Caught signal 11. Server aborting
---

I have tried various mods to my xorg.conf, all to no avail, including:

* Revised HorizSync and VertRefresh to match monitor's actual range;
* Used a tried'n'true custom xorg.conf saved from previous releases;
* Ran 'sudo dpkg-reconfigure xserver-xorg' to build a fresh xorg.conf;
* Add/omit a PCI BusID (I have only one vidcard, and Xorg had worked fine without any BusID in earlier releases);
* Changed: Option "UseFBDev" to "false" in the Device section;
* Added: Option "NoDDC" or "IgnoreEDID" to the Device section;
* Added: Option "Int10Backend" "x86emu" to the ServerLayout section;
* Added a Modeline (actually tried two slightly different modelines produced by different Modeline Generator sites);
* Omit xorg.conf entirely (by renaming).

Interestingly, the "failsafe X" used in in Ubuntu Gutsy, which tries to use the vesa driver, also fails to start X; the Xorg.0.log for that one shows modules being loaded then ends with this:

---
(II) VESA: driver for VESA chipsets: vesa
(II) Primary Device is: PCI 00:0e:0
(EE) No devices detected.

Fatal server error:
no screens found
---

Hardware is a Umax S900 dual-200MHz PowerPC 604e (PowerMac clone) with the stock ixMicro/IMS TwinTurbo 128+ (imstt) PCI video card.

Attachment of full Xorg.0.log, xorg.conf, outputs of uname -a, lspci -vvnn, and ddcprobe to follow (xresprobe returns all-null values, just as it had under earlier releases where X worked fine).

Revision history for this message
In , Tye (webmaster-htdoctor) wrote :

Created an attachment (id=13557)
Log generated w/ std. xorg.conf in use

Revision history for this message
In , Tye (webmaster-htdoctor) wrote :

Created an attachment (id=13559)
xorg.conf (w/ commented experimental entries included)

Revision history for this message
In , Tye (webmaster-htdoctor) wrote :

Created an attachment (id=13560)
Log generated by "failsafe X" vesa fallback process

Revision history for this message
In , Tye (webmaster-htdoctor) wrote :

Created an attachment (id=13561)
Log generated in absence of any xorg.conf

Revision history for this message
In , Tye (webmaster-htdoctor) wrote :

Created an attachment (id=13562)
Output of 'lspci -vvnn'

Revision history for this message
In , Tye (webmaster-htdoctor) wrote :

Created an attachment (id=13563)
Output of 'ddcprobe'

Revision history for this message
In , Tye (webmaster-htdoctor) wrote :

$ uname -a
Linux vesta 2.6.22-14-powerpc-smp #1 SMP Tue Dec 18 09:17:24 UTC 2007 ppc GNU/Linux

Revision history for this message
Tye (webmaster-htdoctor) wrote :

Binary package hint: xorg

Already reported upstream to Xorg as their Bug # 13949 (http://bugs.freedesktop.org/show_bug.cgi?id=13949), apologies if that precludes submitting it here, just covering all relevant/helpful bases:

As of Ubuntu 7.10 "Gutsy", and also appearing very early-on in Debian "Lenny" (testing), X now fails to start up for me, whereas X had worked just fine on this same hardware in Debian "Etch" and also in Ubuntu "Edgy" and "Feisty". Seems to be something about Xorg has changed, as deployed in the affected releases. Towards the end of my Xorg.0.log there's a whole bunch of lines rejecting every possible resolution (even known-valid ones), like this:

---
(II) imstt(0): Not using default mode "1024x768" (bad mode clock/interlace/doublescan)
(II) imstt(0): Not using default mode "1024x768" (unknown reason)

[Repeat w/ a wide variety of other possible resolutions, all rejected]

(WW) imstt(0): Mode pool is empty
(EE) imstt(0): No valid modes found

Backtrace:
0: /usr/bin/X(xf86SigHandler+0x94) [0x10093f44]
1: [0x100344]
2: /usr/lib/xorg/modules/drivers//imstt_drv.so [0xf941520]
3: /usr/lib/xorg/modules/drivers//imstt_drv.so [0xf94186c]
4: /usr/bin/X(InitOutput+0xb08) [0x1006e2c8]
5: /usr/bin/X(main+0x294) [0x1002e064]
6: /lib/libc.so.6 [0xfc79380]
7: /lib/libc.so.6 [0xfc795c4]

Fatal server error:
Caught signal 11. Server aborting
---

I have tried various mods to my xorg.conf, all to no avail, including:

* Revised HorizSync and VertRefresh to match monitor's actual range;
* Used a tried'n'true custom xorg.conf saved from previous releases;
* Ran 'sudo dpkg-reconfigure xserver-xorg' to build a fresh xorg.conf;
* Add/omit a PCI BusID (I have only one vidcard, and Xorg had worked fine without any BusID in earlier releases);
* Changed: Option "UseFBDev" to "false" in the Device section;
* Added: Option "NoDDC" or "IgnoreEDID" to the Device section;
* Added: Option "Int10Backend" "x86emu" to the ServerLayout section;
* Added a Modeline (actually tried two slightly different modelines produced by different Modeline Generator sites);
* Omit xorg.conf entirely (by renaming).

Interestingly, the "failsafe X" used in in Ubuntu Gutsy, which tries to use the vesa driver, also fails to start X; the Xorg.0.log for that one shows modules being loaded then ends with this:

---
(II) VESA: driver for VESA chipsets: vesa
(II) Primary Device is: PCI 00:0e:0
(EE) No devices detected.

Fatal server error:
no screens found
---

Hardware is a Umax S900 dual-200MHz PowerPC 604e (PowerMac clone) with the stock ixMicro/IMS TwinTurbo 128+ (imstt) PCI video card.

Attachment of full Xorg.0.log, xorg.conf, outputs of uname -a, lspci -vvnn, and ddcprobe to follow (xresprobe returns all-null values, just as it had under earlier releases where X worked fine).

Revision history for this message
Tye (webmaster-htdoctor) wrote :

$ uname -a
Linux vesta 2.6.22-14-powerpc-smp #1 SMP Tue Dec 18 09:17:24 UTC 2007 ppc GNU/Linux

$ sudo ddcprobe
oem: IMS TT (IBM)
memory: 4096kb
noedid

Revision history for this message
Tye (webmaster-htdoctor) wrote :
Revision history for this message
Tye (webmaster-htdoctor) wrote :
Revision history for this message
Tye (webmaster-htdoctor) wrote :
Revision history for this message
Tye (webmaster-htdoctor) wrote :

Oops, disregard the above log attachment; that one was before I realized the vesa driver didn't get installed in the first place (strange to have a fallback-to-vesa process without installing vesa!).

This one is the correct log produced by the fallback-X after I installed vesa manually.

Revision history for this message
Tye (webmaster-htdoctor) wrote :
Revision history for this message
Tye (webmaster-htdoctor) wrote :
description: updated
Revision history for this message
Timo Aaltonen (tjaalton) wrote :

Please try Hardy alpha2.

Changed in xserver-xorg-video-imstt:
status: New → Incomplete
Changed in xorg-server:
status: Unknown → Confirmed
Revision history for this message
Tye (webmaster-htdoctor) wrote :

No proper Alphas exist for my PowerPC port, but I've tried to install a few recent daily PPC builds from ISOs and also tried network installation directly from the online repositories (my preferred method, successful for previous releases). I can always get the base Hoary system installed just fine, but when it comes to installing 'ubuntu-desktop', there always seems to be one dependency package or another missing from the ISO and/or repository, which causes failure of the 'ubuntu-desktop' meta-package installation.

I only need to install enough over the base system to see if X will start for me. If someone could recommend a more specific package, or brief list of packages, that I can install piecemeal via aptitude just to test my X startup, I'd be glad to try that.

Revision history for this message
Tye (webmaster-htdoctor) wrote :

I managed to get the rest of Hardy installed from the online repositories over the weekend, complete and current as of yesterday (Sun27Jan08). Nothing has improved in re: this bug -- X still aborts at startup with the same error and backtrace:

===

(WW) imstt(0): Mode pool is empty
(EE) imstt(0): No valid modes found

Backtrace:
0: /usr/bin/X(xf86SigHandler+0x94) [0x100927c4]
1: [0x100344]
2: /usr/lib/xorg/modules/drivers//imstt_drv.so [0xf7c14ac]
3: /usr/lib/xorg/modules/drivers//imstt_drv.so [0xf7c1880]
4: /usr/bin/X(InitOutput+0xb9c) [0x100701dc]
5: /usr/bin/X(main+0x2e8) [0x1002bad8]
6: /lib/libc.so.6 [0xfb6ab10]
7: /lib/libc.so.6 [0xfb6acd0]

Fatal server error:
Caught signal 11. Server aborting

===

(Complete Xorg.0.log attached herein)

Revision history for this message
Timo Aaltonen (tjaalton) wrote :

Unfortunately, imstt has been deprecated upstream and it hasn't bee ported to xorg-server 1.5, which means that it doesn't even build on it and has been removed from the archive. Marking as won't fix.

Changed in xserver-xorg-video-imstt:
status: Incomplete → Won't Fix
Changed in xorg-server:
importance: Unknown → High
Changed in xorg-server:
importance: High → Unknown
status: Confirmed → Won't Fix
Changed in xorg-server:
importance: Unknown → High
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.