Bug 485501

Summary: Unable to launch X with radeon
Product: [Fedora] Fedora Reporter: Jesse Keating <jkeating>
Component: kernelAssignee: Anaconda Maintenance Team <anaconda-maint-list>
Status: CLOSED RAWHIDE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: medium Docs Contact:
Priority: medium    
Version: rawhideCC: airlied, anaconda-maint-list, cje, dcantrell, hdegoede, kernel-maint, quintela, vinagr3
Target Milestone: ---   
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2009-02-19 09:02:00 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 476774    
Attachments:
Description Flags
xorg.log of X failing to find screens on my radeon x1950
none
xorg.log of exact same X-start, but with logfile on usb-stick, this succeeds!! ??
none
xorg.log of trying to start X with radeon modprobed first. none

Description Jesse Keating 2009-02-13 19:29:23 UTC
(EE) RADEON: Failed to laod module fb (module does not exist, 0)

Comment 1 Chris Lumens 2009-02-16 16:02:01 UTC
fbdev_drv.so definitely exists on the install media.  Assuming this is reproducable, can you please attach /tmp/X.log?

Comment 2 Chris Lumens 2009-02-17 14:59:15 UTC
*** Bug 485838 has been marked as a duplicate of this bug. ***

Comment 3 Chris Lumens 2009-02-17 14:59:59 UTC
Taking out of NEEDINFO.

Comment 4 Chris Lumens 2009-02-17 15:12:41 UTC
Sorry, it's not fbdev_drv.so that's the problem.  It's libfb.so.  And that still exists on this machine I've got booted up, so I wonder what the deal is.

Comment 5 Hans de Goede 2009-02-19 08:05:52 UTC
I can reproduce this. With some trickery I've finally managed to get a full xorg.log of the failure. Strange enough if I make xorg log to a mounted usb-stick, it no longer fails to find the screens, so there seems to be some timing issue involved here.

Will attach logs.

Comment 6 Hans de Goede 2009-02-19 08:06:43 UTC
Created attachment 332507 [details]
xorg.log of X failing to find screens on my radeon x1950

Comment 7 Hans de Goede 2009-02-19 08:07:27 UTC
Created attachment 332508 [details]
xorg.log of exact same X-start, but with logfile on usb-stick, this succeeds!! ??

Comment 8 Hans de Goede 2009-02-19 08:09:08 UTC
Note: running with "nomodeset" makes the problem go away.

Comment 9 Hans de Goede 2009-02-19 08:11:14 UTC
Diffing the 2 logs the following stands out:

@@ -725,21 +725,263 @@
 (++) RADEON(0): DPI set to (96, 96)
 (II) Loading sub module "fb"
 (II) LoadModule: "fb"
-(WW) Warning, couldn't open module fb
-(II) UnloadModule: "fb"
-(EE) RADEON: Failed to load module "fb" (module does not exist, 0)
-(II) UnloadModule: "radeon"
-(II) UnloadModule: "int10" 
-(II) Unloading /usr/lib/xorg/modules//libint10.so
-(II) UnloadModule: "vgahw"
-(II) Unloading /usr/lib/xorg/modules//libvgahw.so
-(EE) Screen(s) found, but none have a usable configuration.

<snip>
+(II) Loading /usr/lib/xorg/modules//libfb.so
+(II) Module fb: vendor="X.Org Foundation"
+       compiled for 1.5.99.902, module version = 1.0.0
+       ABI class: X.Org ANSI C Emulation, version 0.4
+(==) RADEON(0): Using gamma correction (1.0, 1.0, 1.0)

So in one case it is finding / loading the fb module and in the other case it is not ???

Comment 10 Hans de Goede 2009-02-19 08:28:28 UTC
I've been discussing this with airlied on irc, the problem is that radeon is getting loading by Xorg instead of earlier by udev, triggering a race.

However I've just done a test where I first manually modprobed radeon and then I get a different error. Attaching log of this attempt.

Comment 11 Hans de Goede 2009-02-19 08:29:10 UTC
Created attachment 332511 [details]
xorg.log of trying to start X with radeon modprobed first.

Comment 12 Hans de Goede 2009-02-19 08:57:19 UTC
Comment on attachment 332511 [details]
xorg.log of trying to start X with radeon modprobed first.

Ok, the third time I screwed up, I didn't have /sys mounted making X unhappy.

With that fixed, and modprobing radeon befor starting X things work.

So I'm reassignign this to kernel (radeon is missing module ID's) and xlosing (airlied just fixed the missing module ID's).