Bug 23782 - Black screen Powercolor Radeon HD 3650 Pro AGP 512 MB using radeonhd-git drivers
Summary: Black screen Powercolor Radeon HD 3650 Pro AGP 512 MB using radeonhd-git drivers
Status: RESOLVED NOTABUG
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/radeonhd (show other bugs)
Version: 7.4 (2008.09)
Hardware: x86 (IA32) Linux (All)
: medium major
Assignee: Luc Verhaegen
QA Contact: Xorg Project Team
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2009-09-07 09:32 UTC by Sebastian Kuzlak
Modified: 2009-09-09 11:50 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:


Attachments
xorg.conf (2.96 KB, text/plain)
2009-09-07 09:32 UTC, Sebastian Kuzlak
no flags Details
lspci -v (4.98 KB, text/plain)
2009-09-07 09:34 UTC, Sebastian Kuzlak
no flags Details
Xorg.0.log (148.00 KB, text/x-log)
2009-09-07 09:36 UTC, Sebastian Kuzlak
no flags Details
kernel.log (4.13 KB, text/plain)
2009-09-07 09:41 UTC, Sebastian Kuzlak
no flags Details
Xorg.0.log (147.48 KB, text/plain)
2009-09-07 09:46 UTC, Sebastian Kuzlak
no flags Details

Description Sebastian Kuzlak 2009-09-07 09:32:36 UTC
Created attachment 29315 [details]
xorg.conf

if i start X i have a black screen with only mouse cursor visible- i can move   mouse but thats all. I can get to VT but if i try to get back to X system hangs. The only way i can get to x start i have to set Option "DRI" "false" in xorg ( no accel :/ )
Comment 1 Sebastian Kuzlak 2009-09-07 09:34:54 UTC
Created attachment 29316 [details]
lspci -v
Comment 2 Sebastian Kuzlak 2009-09-07 09:36:00 UTC
Created attachment 29317 [details]
Xorg.0.log
Comment 3 Sebastian Kuzlak 2009-09-07 09:41:13 UTC
Created attachment 29318 [details]
kernel.log
Comment 4 Sebastian Kuzlak 2009-09-07 09:46:23 UTC
Created attachment 29319 [details]
Xorg.0.log

Ive noticed that xorg.0.log is overwritten when i try to get back from vt to X so this log is until i get black screen ( s*** dont know exactly how to write it so ill just give you steps i took)
1. loaded modules (it seams that the are not autoladed after i type startx)
2. startx
3. change to VT1
The first xorg.0.log i attached has extra step that is
4 get back to x (total system hang)

sorry if it seems so messy
Comment 5 Yang Zhao 2009-09-07 10:13:19 UTC
Is this a recent regression? If so, can you bisect to find out which commit causes it?  Your log shows you have changes in the source, so make sure you're working with a clean checkout.

And, just in case, remember that Xserver-1.6 starts to a black screen by default.
Comment 6 Sebastian Kuzlak 2009-09-07 11:19:24 UTC
(In reply to comment #5)
> Is this a recent regression? If so, can you bisect to find out which commit
> causes it?  

No, at least it does not work at all at my machine. I came back to linux after a while ( and with the "new" graphics card - previously i used gf7600 )and since ive installed linux ( 1,5-2 months back ) im not able to get it working. Cant even start x with catalyst....

> Your log shows you have changes in the source, so make sure you're
> working with a clean checkout.

Well ill try clear old files and rebuild it but i think it wont help. Basicly i rebuild drivers every commit i see in hope that ill get it finally running. 

> And, just in case, remember that Xserver-1.6 starts to a black screen by
> default.



Comment 7 Yang Zhao 2009-09-07 11:23:09 UTC
Does radeon (xf86-video-ati) work any better?
Comment 8 Sebastian Kuzlak 2009-09-07 11:56:19 UTC
(In reply to comment #7)
> Does radeon (xf86-video-ati) work any better?
> 

no :/ Same as with radeonhd :/ 

Ive cleaned sources and rebuild radeonhd and nothing changed. I can see Mouse cursor, i can move it. I even hear kde starting up.

Now im starting to think that it may not be drivers but kernel.

Sep  7 19:55:42 acid-archlinux kernel: [drm] Setting GART location based on new memory map
Sep  7 19:55:42 acid-archlinux kernel: [drm] Can't use AGP base @0xe0000000, won't fit

I dont know. Im really tired fighting with ati hardware.
Comment 9 Andy Furniss 2009-09-09 09:52:30 UTC
(In reply to comment #8)
> (In reply to comment #7)
> > Does radeon (xf86-video-ati) work any better?
> > 
> 
> no :/ Same as with radeonhd :/ 
> 
> Ive cleaned sources and rebuild radeonhd and nothing changed. I can see Mouse
> cursor, i can move it. I even hear kde starting up.
> 
> Now im starting to think that it may not be drivers but kernel.
> 
> Sep  7 19:55:42 acid-archlinux kernel: [drm] Setting GART location based on new
> memory map
> Sep  7 19:55:42 acid-archlinux kernel: [drm] Can't use AGP base @0xe0000000,
> won't fit
> 
> I dont know. Im really tired fighting with ati hardware.
> 

It may work if you set your AGP Aperture size in bios to 256.

I have to do this for my AGP R6xx card. 

If that doesn't work there are other things to try.

The crash on VT switch back to X has existed on AGP cards ever since R6xx got accel.
  
Comment 10 Sebastian Kuzlak 2009-09-09 10:47:02 UTC
(In reply to comment #9)
 
> It may work if you set your AGP Aperture size in bios to 256.
> 
> I have to do this for my AGP R6xx card. 
> 
> If that doesn't work there are other things to try.
> 
> The crash on VT switch back to X has existed on AGP cards ever since R6xx got
> accel.
> 
> 

Setting AGP Aperture size in bios to 256 helped. Now i'm able to start X with DRI enabled. Driver works pretty damn well ;-) Thank you very much for help.

VT still crashes when changing but...

Ok now i dont know what resolution type should i set. I think ill set it to  NOTABUG. If any one here has a diferent opintion about resolution type, well please tell :)
Comment 11 Andy Furniss 2009-09-09 11:50:15 UTC
(In reply to comment #10)

> 
> Setting AGP Aperture size in bios to 256 helped. Now i'm able to start X with
> DRI enabled. Driver works pretty damn well ;-) Thank you very much for help.

Glad it's working for you now.

For the future, when the distros pick up the 3D work you may (or may not) find that you need to also disable AGP write backs.

This is a radeon kernel module parameter eg.

modprobe radeon no_wb=1


Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.