Bug 638656 - undefined symbol: bgNoneRoot trying to run any app that uses pyxf86config
Summary: undefined symbol: bgNoneRoot trying to run any app that uses pyxf86config
Keywords:
Status: CLOSED NOTABUG
Alias: None
Product: Fedora
Classification: Fedora
Component: pyxf86config
Version: 14
Hardware: All
OS: Linux
low
high
Target Milestone: ---
Assignee: Adam Jackson
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2010-09-29 15:27 UTC by Adam Williamson
Modified: 2010-11-10 20:26 UTC (History)
15 users (show)

Fixed In Version:
Clone Of:
Environment:
Last Closed: 2010-11-10 16:55:31 UTC
Type: ---
Embargoed:


Attachments (Terms of Use)

Description Adam Williamson 2010-09-29 15:27:42 UTC
After building an X server with the fix for 623742 and rebuilding pyxf86config , I now hit this problem:

[root@adam pyxf86config (f14 *%)]# system-config-display 
Traceback (most recent call last):
  File "/usr/share/system-config-display/xconf.py", line 27, in <module>
    import xf86config
  File "/usr/lib64/python2.7/site-packages/xf86config.py", line 1, in <module>
    import ixf86config
ImportError: /usr/lib64/python2.7/site-packages/ixf86configmodule.so: undefined symbol: bgNoneRoot

ajax says "<ajax> adamw: undefined data symbols like that you can fix by just adding a C file to pyxf86config that defines them" but I'm afraid I don't know how to do that. :) There is a xf86config_ext.c file in pyxf86config that looks like it might be the place to do this, but i'm not sure what to put in the definition.

Comment 1 Adam Williamson 2010-09-29 15:28:46 UTC
Proposing as nth for F14, same rationale as 623742: "this affects rather a lot of things, including system-config-keyboard according to https://bugzilla.redhat.com/show_bug.cgi?id=634890 . It breaks system-config-display and the related tools in Fusion which implement various out-of-tree drivers, which is something a lot of users rely on."



-- 
Fedora Bugzappers volunteer triage team
https://fedoraproject.org/wiki/BugZappers

Comment 2 Adam Williamson 2010-10-01 21:28:29 UTC
This was discussed at the 2010-10-01 blocker / NTH review meeting, and accepted as a nice-to-have.

Comment 3 Sandro Bonazzola 2010-10-04 14:30:49 UTC
Resolving the missing bgNoneRoot discovering another missing symbol: SmartScheduleMaxSlice.


--- pyxf86config-0.3.37/xf86config_ext.c.orig	2010-10-04 16:26:04.978300002 +0200
+++ pyxf86config-0.3.37/xf86config_ext.c	2010-10-04 16:25:37.141300056 +0200
@@ -11,6 +11,8 @@

 extern void xf86freeFiles (XF86ConfFilesPtr p);

+Bool bgNoneRoot = FALSE;
+
 void
 xf86freeInput (XF86ConfInputPtr ptr)
 {

Comment 4 Sandro Bonazzola 2010-10-04 14:40:14 UTC
All the missing symbols seems to be caused by utils.c inclusion, proposed solution for bug #623742, that include opaque.h where all those symbols missing are declared extern.

Comment 5 Sandro Bonazzola 2010-10-19 15:02:50 UTC
Any news from xorg people about this?

Comment 6 Adam Williamson 2010-10-19 16:17:48 UTC
it's pretty much going to require pouring large quantities of beer down ajax to get him to care enough to fix this. if you know what you're doing, it'd *really* help if you could come up with a patch. Unfortunately, I can't.



-- 
Fedora Bugzappers volunteer triage team
https://fedoraproject.org/wiki/BugZappers

Comment 7 Adam Williamson 2010-11-01 23:01:59 UTC
Removing F14 NTH status as F14 is now out.

Remaining open nice-to-have issues do NOT automatically become nice-to-have issues for Fedora 15. If you believe a Fedora 14 issue which was accepted as
nice-to-have but not resolved in time for release should also qualify for
nice-to-have status for Fedora 15, please re-propose it as nice-to-have for
Fedora 15.



-- 
Fedora Bugzappers volunteer triage team
https://fedoraproject.org/wiki/BugZappers

Comment 8 Michael Monreal 2010-11-02 07:00:49 UTC
I see this on every boot on F14 from the init script that ships with RPMFusion's Nvidia driver package. This means installation of that driver is busted in F14 as generation of a xorg.conf file does not work at all :/

Comment 9 Adam Williamson 2010-11-02 18:25:02 UTC
yes, we're aware of that. fglrx and psb from Fusion are similarly affected.



-- 
Fedora Bugzappers volunteer triage team
https://fedoraproject.org/wiki/BugZappers

Comment 10 James Begley 2010-11-05 10:33:43 UTC
Since this is going to affect anyone who uses the nvidia / psb / fglrx / etc ... surely this is a candidate for the common bugs wiki page? http://fedoraproject.org/wiki/Common_F14_bugs

Comment 11 Adam Williamson 2010-11-05 22:46:22 UTC
I'm not sure whether it's best to document it there or at Fusion.



-- 
Fedora Bugzappers volunteer triage team
https://fedoraproject.org/wiki/BugZappers

Comment 12 James Begley 2010-11-10 09:34:08 UTC
(In reply to comment #11)
> I'm not sure whether it's best to document it there or at Fusion.

I would have thought it would be better to document it here, since 

a) Fusion doesnt have a 'common bugs' page, and
b) the bug is within RPMs distributed from Fedora (although it shows up when used through the Fusion packages)

Anyway, as mentioned in #624297, the koji build of pyxf86config http://koji.fedoraproject.org/koji/buildinfo?buildID=203948 has resolved this issue for me.

Comment 13 Adam Williamson 2010-11-10 16:55:10 UTC
(although it shows up when used through the Fusion packages)

that's the important point. most people don't dig down and figure out which level the failure is at, all they see is 'my nvidia don't work'.

but yeah, we fixed it now, so no biggie.



-- 
Fedora Bugzappers volunteer triage team
https://fedoraproject.org/wiki/BugZappers

Comment 14 Adam Williamson 2010-11-10 16:55:31 UTC
closing this one, as you don't hit this problem unless you use a fix for the other bug which we rejected.



-- 
Fedora Bugzappers volunteer triage team
https://fedoraproject.org/wiki/BugZappers

Comment 15 Nadav Har'El 2010-11-10 20:10:11 UTC
Adam, in comment 13 you said "we fixed it now", but in comment 14 you implied you didn't, so which is it?

Comment 16 Adam Williamson 2010-11-10 20:26:33 UTC
we fixed the initial bug in pyxf86config a different way. this bug only came up if you used our *initial* proposed fix for pyxf86config, which was different from the fix we've now pushed.



-- 
Fedora Bugzappers volunteer triage team
https://fedoraproject.org/wiki/BugZappers


Note You need to log in before you can comment on or make changes to this bug.