Weird graphics error with Radeon HD3200 & catalyst driver

lastchancetosee · 2009-05-03 07:50:46

Hi,
my Radeon HD3200 forced me, much to my chagrin, to use the newest Catalyst drivers. To my surprise installation worked flawlessly, however I get a weird graphics error:
Every few secons or so, a thin horizontal line flashes somewhere on the monitor, maybe 1-3 pixels high. Position and timing of this line is random and it is VERY annoying. The monitor is connected via DVI and gets sane values for sync/refresh/resolution (67kHz/60Hz/1280x1024).

So far I have noticed this only in X so I guess it's the driver, not the hardware.

What could this be?

[edit: I just had a whopping great glitch in the top quarter of my screen that shifted the whole picture below it visibly. I guess that happens with all of them but my eyes where to slow :-) ]
[edit2: The above is not the only problem :-( : Video performance is crappy. Jerky, sometimes with graphics errors as well.]

xorg.conf (mainly copied over from my previous box):

Section "ServerLayout"
    Identifier     "X.org Configured"
    Screen      0  "Screen0" 0 0
#    InputDevice    "Mouse0" "CorePointer"
#    InputDevice    "Keyboard0" "CoreKeyboard"
EndSection

#Section "Files"
#    ModulePath   "/usr/lib/xorg/modules"
#    FontPath     "/usr/share/fonts/misc"
#    FontPath     "/usr/share/fonts/100dpi:unscaled"
#    FontPath     "/usr/share/fonts/75dpi:unscaled"
#    FontPath     "/usr/share/fonts/TTF"
#    FontPath     "/usr/share/fonts/Type1"
#EndSection

Section "Module"
    Load  "dbe"
    Load  "dri"
    Load  "glx"
    Load  "dri2"
    Load  "drm"
    Load  "extmod"
    Load  "record"
EndSection

#Section "InputDevice"
#    Identifier  "Keyboard0"
#    Driver      "kbd"
#EndSection

#Section "InputDevice"
#    Identifier  "Mouse0"
#    Driver      "mouse"
#    Option        "Protocol" "auto"
#    Option        "Device" "/dev/input/mice"
#    Option        "ZAxisMapping" "4 5 6 7"
#EndSection

Section "Monitor"
    Identifier   "Monitor0"
    VendorName   "FUS"
    ModelName    "P19-2"
    Option        "DPMS"
#    HorizSync   30.0 - 82.0
#    VertRefresh 56.0 - 76.0
EndSection

Section "Device"
    Identifier  "Card0"
    Driver      "fglrx"
    VendorName  "ATI Technologies Inc"
    BoardName   "Radeon HD 3200 Graphics"
    Option        "XAANoOffscreenPixmaps" "true"
    Option        "TexturedXrender" "on"
    Option        "VideoOverlay" "on"
    Option        "OpenGLOverlay" "off"
    Option      "TexturedVideo" "on"
    Option      "TexturedVideoSync" "on"
    BusID       "PCI:1:5:0"
EndSection

Section "Screen"
    Identifier "Screen0"
    Device     "Card0"
    Monitor    "Monitor0"
    DefaultDepth     24
    SubSection "Display"
        Viewport   0 0
        Depth     24
        Modes    "1280x1024" "1024x768" "800x600"
    EndSubSection
EndSection

Section "DRI"
    Mode         0666
EndSection

Section "Extensions"
    Option        "Composite" "Enable"
EndSection

Xorg.0.log (only errors/warnings):

(WW) Open ACPI failed (/var/run/acpid.socket) (No such file or directory)
(WW) Falling back to old probe method for fglrx
(WW) This ATI Proprietary Linux Driver does not guarantee support of video driver ABI higher than 2.0
(WW) Video driver ABI version of the X server is 5.0
(WW) fglrx: No matching Device section for instance (BusID PCI:0@0:17:0) found
(WW) fglrx: No matching Device section for instance (BusID PCI:0@0:18:0) found
(WW) fglrx: No matching Device section for instance (BusID PCI:0@0:18:1) found
(WW) fglrx: No matching Device section for instance (BusID PCI:0@0:18:2) found
(WW) fglrx: No matching Device section for instance (BusID PCI:0@0:19:0) found
(WW) fglrx: No matching Device section for instance (BusID PCI:0@0:19:1) found
(WW) fglrx: No matching Device section for instance (BusID PCI:0@0:19:2) found
(WW) fglrx: No matching Device section for instance (BusID PCI:0@0:20:0) found
(WW) fglrx: No matching Device section for instance (BusID PCI:0@0:20:1) found
(WW) fglrx: No matching Device section for instance (BusID PCI:0@0:20:2) found
(WW) fglrx: No matching Device section for instance (BusID PCI:0@0:20:3) found
(WW) fglrx: No matching Device section for instance (BusID PCI:0@0:20:4) found
(WW) fglrx: No matching Device section for instance (BusID PCI:0@0:20:5) found
(WW) fglrx: No matching Device section for instance (BusID PCI:0@1:5:1) found
(WW) fglrx(0): board is an unknown third party board, chipset is supported
(WW) fglrx(0): could not detect X server version (query_status=-1)
(WW) AIGLX: 3D driver claims to not support visual 0x23
(WW) AIGLX: 3D driver claims to not support visual 0x24
(WW) AIGLX: 3D driver claims to not support visual 0x25
(WW) AIGLX: 3D driver claims to not support visual 0x26
(WW) AIGLX: 3D driver claims to not support visual 0x27
(WW) AIGLX: 3D driver claims to not support visual 0x28
(WW) AIGLX: 3D driver claims to not support visual 0x29
(WW) AIGLX: 3D driver claims to not support visual 0x2a
(WW) AIGLX: 3D driver claims to not support visual 0x2b
(WW) AIGLX: 3D driver claims to not support visual 0x2c
(WW) AIGLX: 3D driver claims to not support visual 0x2d
(WW) AIGLX: 3D driver claims to not support visual 0x2e
(WW) AIGLX: 3D driver claims to not support visual 0x2f
(WW) AIGLX: 3D driver claims to not support visual 0x30
(WW) AIGLX: 3D driver claims to not support visual 0x31
(WW) AIGLX: 3D driver claims to not support visual 0x32
(WW) AIGLX: 3D driver claims to not support visual 0x33
(WW) AIGLX: 3D driver claims to not support visual 0x34
(WW) AIGLX: 3D driver claims to not support visual 0x35
(WW) AIGLX: 3D driver claims to not support visual 0x36
(WW) AIGLX: 3D driver claims to not support visual 0x37
(WW) AIGLX: 3D driver claims to not support visual 0x38
(WW) AIGLX: 3D driver claims to not support visual 0x39
(WW) AIGLX: 3D driver claims to not support visual 0x3a
(WW) AIGLX: 3D driver claims to not support visual 0x3b
(WW) AIGLX: 3D driver claims to not support visual 0x3c
(WW) AIGLX: 3D driver claims to not support visual 0x3d
(WW) AIGLX: 3D driver claims to not support visual 0x3e
(WW) AIGLX: 3D driver claims to not support visual 0x3f
(WW) AIGLX: 3D driver claims to not support visual 0x40
(WW) AIGLX: 3D driver claims to not support visual 0x41
(WW) AIGLX: 3D driver claims to not support visual 0x42
(WW) AIGLX: 3D driver claims to not support visual 0x43
(WW) AIGLX: 3D driver claims to not support visual 0x44
(WW) AIGLX: 3D driver claims to not support visual 0x45
(WW) AIGLX: 3D driver claims to not support visual 0x46
(WW) AIGLX: 3D driver claims to not support visual 0x47
(WW) AIGLX: 3D driver claims to not support visual 0x48
(WW) AIGLX: 3D driver claims to not support visual 0x49
(WW) AIGLX: 3D driver claims to not support visual 0x4a
(WW) AIGLX: 3D driver claims to not support visual 0x4b
(WW) AIGLX: 3D driver claims to not support visual 0x4c
(WW) AIGLX: 3D driver claims to not support visual 0x4d
(WW) AIGLX: 3D driver claims to not support visual 0x4e
(WW) AIGLX: 3D driver claims to not support visual 0x4f
(WW) AIGLX: 3D driver claims to not support visual 0x50
(WW) AIGLX: 3D driver claims to not support visual 0x51
(WW) AIGLX: 3D driver claims to not support visual 0x52
(WW) AIGLX: 3D driver claims to not support visual 0x53
(WW) AIGLX: 3D driver claims to not support visual 0x54
(WW) AIGLX: 3D driver claims to not support visual 0x55
(WW) AIGLX: 3D driver claims to not support visual 0x56
(WW) AIGLX: 3D driver claims to not support visual 0x57
(WW) AIGLX: 3D driver claims to not support visual 0x58
(WW) AIGLX: 3D driver claims to not support visual 0x59
(WW) AIGLX: 3D driver claims to not support visual 0x5a
(WW) AIGLX: 3D driver claims to not support visual 0x5b
(WW) AIGLX: 3D driver claims to not support visual 0x5c
(WW) AIGLX: 3D driver claims to not support visual 0x5d
(WW) AIGLX: 3D driver claims to not support visual 0x5e
(WW) AIGLX: 3D driver claims to not support visual 0x5f
(WW) AIGLX: 3D driver claims to not support visual 0x60
(WW) AIGLX: 3D driver claims to not support visual 0x61
(WW) AIGLX: 3D driver claims to not support visual 0x62
(WW) AIGLX: 3D driver claims to not support visual 0x63
(WW) AIGLX: 3D driver claims to not support visual 0x64
(WW) AIGLX: 3D driver claims to not support visual 0x65
(WW) AIGLX: 3D driver claims to not support visual 0x66
(WW) AIGLX: 3D driver claims to not support visual 0x67
(WW) AIGLX: 3D driver claims to not support visual 0x68
(WW) AIGLX: 3D driver claims to not support visual 0x69
(WW) AIGLX: 3D driver claims to not support visual 0x6a
(WW) AIGLX: 3D driver claims to not support visual 0x6b
(WW) AIGLX: 3D driver claims to not support visual 0x6c
(WW) AIGLX: 3D driver claims to not support visual 0x6d
(WW) AIGLX: 3D driver claims to not support visual 0x6e
(WW) AIGLX: 3D driver claims to not support visual 0x6f
(WW) AIGLX: 3D driver claims to not support visual 0x70
(WW) AIGLX: 3D driver claims to not support visual 0x71
(WW) AIGLX: 3D driver claims to not support visual 0x72

Last edited by lastchancetosee (2009-05-03 08:25:45)

vacant · 2009-05-03 10:42:51

My laptop has HD3200 graphics. I tried the catalyst driver briefly. The Xorg log warnings are just that. What alarmed me was the volume of messages going to errors.log e.g.

Apr 26 21:38:34 clod kernel: [fglrx:drm_vm_open] *ERROR* map not found -> inconsistent kernel data!!! vma_start:0x7faae2630000,handle:0x72a5d000

I play 720HD video to my HD TV via VGA and found the catalyst driver gave the odd "line" corruption.

I can play HD video perfectly using xf86-video-ati-git and drm-radeon-module-git-r6xx-r7xx provided the laptop display is turned off.

3D is still rubbish but I don't need that.

lastchancetosee · 2009-05-03 11:43:29

I've checked and I get those errors too. Holy crap, there's a lot of those! You don't happen to know what they signify?
But thanks for the alternative drivers. I don't need 3D right now either, so I'll try those and report back.

Pity. I used to prefer ATI to NVidia. Then I discovered Linux ...

[edit: Potentially stupid question: What do I need the drm-module for? ]

Last edited by lastchancetosee (2009-05-03 11:46:31)

vacant · 2009-05-03 12:34:29

Sorry, I had a quick google at the time but didn't come up with much. The catalyst driver did enable the laptop to run cool, under 45C for some of the time whereas with xorg radeon I'm running around 55C (trip point for first level fan is 45C)

drm-radeon-module replaces radeon.ko and dri.ko direct rendering management modules siupplied with the kernel26 package.

lastchancetosee · 2009-05-03 12:48:21

OK.

I installed both after removing the catalyst stuff (to get the original libgl.so back) and reinstalling libgl and xorg-server for good measure.
Unfortunately now my desktop is back to it's old pre-fglrx sluggish self. Video is unbearable.

Anything unusual I need to have in xorg.conf for this to work?

Errors in Xorg.0.log:

(EE) RADEON(0): [dri] RADEONDRIGetVersion failed because of a version mismatch.
[dri] radeon.o kernel module version is 8.60.3 but version 1.17.0 or newer is needed.
[dri] Disabling DRI.
(II) RADEON(0): using shadow framebuffer
[...]
(II) RADEON(0):   MC_AGP_LOCATION  : 0x003f0000
(==) RADEON(0): Backing store disabled
(WW) RADEON(0): Direct rendering disabled
(EE) RADEON(0): Acceleration initialization failed
(II) RADEON(0): Acceleration disabled

How exactly is 8.xx not newer than 1.xx?

vacant · 2009-05-03 15:17:12

Sounds like catalyst remnants.

In my xorg.conf I needed a modeline to get 1280x768 to appear as an option on ksettings->display for my TV. AccelMethod EXA is fine, other options don't make much difference AFAIK.

Section "Module"
        Load    "freetype"
EndSection

Section "Monitor"
        Identifier      "laptop"
        Option          "DPMS" "true"
EndSection

Section "Monitor"
        Identifier      "tv"
        Option          "DPMS" "true"
        Modeline        "1280x768_60.00"  80.14  1280 1344 1480 1680  768 769 772 795  -HSync +Vsync
EndSection

Section "Device"
        Identifier      "hd3200"
        Driver          "radeon"
        BusID           "PCI:1:5:0"
        Option          "Monitor-LVDS"          "laptop"
        Option          "Monitor-VGA-0"         "tv"
        Option          "AccelMethod"           "EXA"
        Option          "ClockGating"           "true"
        Option          "DynamicPM"             "true"
EndSection

Section "DRI"
        Mode            0666
EndSection

lastchancetosee · 2009-05-03 15:44:02

vacant wrote:

Sounds like catalyst remnants.

Ouch. That would not be good. I've so far un- and reinstalled mesa, libgl, xorg-server and of course xf86-video-ati-git and the drm-stuff. What else could catalyst have messed up?
AccelMethod → EXA is in my config as well right now, no change.

My xorg.conf right now (without commented stuff):

Section "Module"
    Load  "dbe"
    Load  "dri"
    Load  "glx"
    Load  "dri2"
    Load  "drm"
    Load  "extmod"
    Load  "record"
EndSection

Section "Monitor"
    Identifier   "Monitor0"
    VendorName   "FUS"
    ModelName    "P19-2"
    Option        "DPMS"
EndSection

Section "Device"
    Identifier  "Card0"
    Driver      "radeon"
    VendorName  "ATI Technologies Inc"
    BoardName   "Radeon HD 3200 Graphics"
    Option        "XAANoOffscreenPixmaps" "true"
    Option      "AccelMethod" "exa"
    BusID       "PCI:1:5:0"
    Option      "DRI" "on"
EndSection

Section "Screen"
    Identifier "Screen0"
    Device     "Card0"
    Monitor    "Monitor0"
    DefaultDepth     24
    SubSection "Display"
        Viewport   0 0
        Depth     24
        Modes    "1280x1024" "1024x768" "800x600"
    EndSubSection
EndSection

Section "DRI"
    Mode         0666
EndSection

Section "Extensions"
    Option        "Composite" "Enable"
EndSection

Section "ServerFlags"
    Option "DontZap" "False"
EndSection

vacant · 2009-05-03 16:01:57

$ modinfo radeon
filename:       /lib/modules/2.6.29-ARCH/updates/radeon.ko
license:        GPL and additional rights
description:    ATI Radeon
author:         Gareth Hughes, Keith Whitwell, others.
depends:        drm
vermagic:       2.6.29-ARCH SMP preempt mod_unload
parm:           no_wb:Disable AGP writeback for scratch registers

I assume you have the correct radeon.ko and drm.ko loaded (from the updates directory that drm-module radeon supplies).

lastchancetosee · 2009-05-03 17:15:52

$ modinfo radeon
filename:       /lib/modules/2.6.29-ARCH/updates/radeon.ko
license:        GPL and additional rights
description:    ATI Radeon
author:         Gareth Hughes, Keith Whitwell, others.
depends:        drm
vermagic:       2.6.29-ARCH SMP preempt mod_unload 
parm:           no_wb:Disable AGP writeback for scratch registers
 (int)

$ modinfo drm
filename:       /lib/modules/2.6.29-ARCH/updates/drm.ko
license:        GPL and additional rights
description:    DRM shared core routines
author:         Gareth Hughes, Leif Delgass, José Fonseca, Jon Smirl
depends:        
vermagic:       2.6.29-ARCH SMP preempt mod_unload 
parm:           debug:Enable debug output (int)

[edit: The original errors with catalyst are probably due to the driver not supporting kernel 2.6.29 properly. At least the phoronix-forums tell me so.]

Last edited by lastchancetosee (2009-05-03 17:49:08)

pseudonomous · 2009-05-03 20:10:32

I additionally suffer from these exact errors, with the exception that I couldn't get the catalyst driver to work at all. I'm now using the "ati-git" drivers for Xorg, but I had the same problem with the drivers in "Extra".

lastchancetosee · 2009-05-03 21:48:08

The graphic error problem? I didn't notice that with the driver from extra, but then I almost immediately switched to the catalyst one in search for better video.
So far with the -git one I don't notice these errors, but I still would like to get a bit more performance out of them :-) .

pseudonomous · 2009-05-03 22:53:19

I had the same error messages for the catalyst drivers and my display would lock after I'd start X.

Then I moved to the ati drivers, after which I could start X, but during startup I would get errors like:

(EE) RADEON(0): [dri] RADEONDRIGetVersion failed to open the DRM
(EE) RADEON(0): Acceleration initialization failed

when I would try and startx from another machine w/ ssh I could see it complain about something like:

[dri] radeon.o kernel module version is 8.60.3 but version 1.17.0 or newer is needed.
[dri] Disabling DRI.

I'm not 100% sure those numbers aren't off by a minor version, number, but they're in the ballpark. I got the same error messages on my console for both the ati and ati-git drivers. I also think I get some graphic curroption at boot-time becuase the arch-logo looks kind of funny (the colors aren't rendered right).

If run "glxinfo" it tells me I"m using direct rendering, but I don't think I believe it,

lastchancetosee · 2009-05-04 09:29:29

I just tested it with an old arch32 install that never saw the catalyst-drivers and I still have the same problem, only now the error is, as pseudonomas reported

(EE) RADEON(0): [dri] RADEONDRIGetVersion failed to open the DRM

and not the one with 8.xx < 1.xx. So I'd say that the latter is some remnant of catalyst which unfortunately doesn't help since it does not work without those remnants either.
glxinfo still reports that DRI is enabled although that is obviously not true.

@vacant: What kernel are you running? Do you have i686 or x86_64?

I also think I get some graphic curroption at boot-time becuase the arch-logo looks kind of funny (the colors aren't rendered right).

I have that with the 32bit install but not the new 64bit one.

I'm now officially out of ideas. Time to buy a NVidia card, I guess :-( .

vacant · 2009-05-04 09:37:27

lastchancetosee wrote:

@vacant: What kernel are you running? Do you have i686 or x86_64?

standard x64 kernel

$ uname -a
Linux clod 2.6.29-ARCH #1 SMP PREEMPT Wed Apr 29 15:36:46 CEST 2009 x86_64 AMD Athlon(tm)X2 DualCore QL-62 AuthenticAMD GNU/Linux
$ pacman -Qs kernel26
local/kernel26 2.6.29.2-1 (base)
    The Linux Kernel and modules

lastchancetosee · 2009-05-04 10:03:21

Same here.
Hmpf. Maybe you're just lucky :-) .

lastchancetosee · 2009-05-04 19:18:05

Where might I get the standard 2.6.28-ARCH kernel? I'd like to see if my problems go away with fglrx & 2.6.28, but I only ever had the 29er kernel on this box so I can't go back.

[edit: On second thought, maybe it's time I get into ABS and have a look at the kernel-PKGBUILD from it ...]

Last edited by lastchancetosee (2009-05-04 19:48:24)

greenfish · 2009-05-05 16:30:04

oh my .... please this is really starting to anoy me. Compiz with bordes/wobbling doesn't work at all and the worst part is that dmesg is cluttered with

[fglrx:firegl_find_any_map] *ERROR* Invalid map handle!<3>[fglrx:drm_vm_open] *ERROR* map not found -> inconsistent kernel data!!! vma_start:0x7fef339ea000,handle:0xc29ec000
[fglrx:firegl_find_any_map] *ERROR* Invalid map handle!<3>[fglrx:drm_vm_close] *ERROR* map not found -> inconsistent kernel data!!! vma_start:0x7fef339ea000,handle:0xc29ec000
[fglrx:firegl_find_any_map] *ERROR* Invalid map handle!<3>[fglrx:drm_vm_open] *ERROR* map not found -> inconsistent kernel data!!! vma_start:0x7fef335b6000,handle:0xc5e76000
[fglrx:firegl_find_any_map] *ERROR* Invalid map handle!<3>[fglrx:drm_vm_open] *ERROR* map not found -> inconsistent kernel data!!! vma_start:0x7fef339ea000,handle:0xc29ec000

(won't paste the entire content in here) check out my paste if you wish: EDIT: I can't paste all those lines WOW! "time out while sending data" just imagine i have 800 lines of the same crap ^^

It's really starting to eat up my log, is there a way for me to hide this annoying message? Also everytime this error spams my dmesg, it's impossible to read everything else (I like to view dmesg on a daily basis).

any ideas?

I installed archlinux64 (ati radeonhd 3400) bit a few days ago, and i've had more issues than fun to be honest.

Last edited by greenfish (2009-05-05 16:35:11)

lastchancetosee · 2009-05-05 17:25:44

The problem, apparently, is this:

- the open source driver doesn't yet support R6xx/R7xx ATI cards.
- the catalyst driver doesn't support the 2.6.29 kernel (since they only support ubuntu and ubuntu 9.04 is still on kernel 2.6.28)
- I have no idea why the git-versions of the open source driver don't work for me while they work for vacant.

I just got the old kernel from the Arch Rollback Machine (see here, fifth paragraph) and will today try if this solves the problem with the catalyst driver. I'll report back.

lastchancetosee · 2009-05-05 18:42:22

OK, installed the kernel, changed the catalyst PKGBUILD (it patches the 2.6.29-kernel, so it really should work, but nevermind, it took that out), installed catalyst → didn't help.

The glitches are still there, the video performance ist a bit better, I imagine, the errors in error.log are gone, but these have appeared:

May  5 20:27:15 kakapo kernel: [fglrx:firegl_init_module] *ERROR* firegl_stub_register failed
May  5 20:36:35 kakapo kernel: [fglrx:firegl_lock] *ERROR* Process 6955 is using illegal context 0x00000004
May  5 20:37:10 kakapo kernel: [fglrx:firegl_lock] *ERROR* Process 7216 is using illegal context 0x00000005

The process is VLC.

Last edited by lastchancetosee (2009-05-05 19:15:41)

greenfish · 2009-05-05 19:43:11

lastchancetosee wrote:

The problem, apparently, is this:
- the open source driver doesn't yet support R6xx/R7xx ATI cards.
- the catalyst driver doesn't support the 2.6.29 kernel (since they only support ubuntu and ubuntu 9.04 is still on kernel 2.6.28)
- I have no idea why the git-versions of the open source driver don't work for me while they work for vacant.
I just got the old kernel from the Arch Rollback Machine (see here, fifth paragraph) and will today try if this solves the problem with the catalyst driver. I'll report back.

OS drivers can't say I favor any of them compared to proprietary drivers, goes for ATI & nvidia, but they are obviously doing a great job. ATI is working with Canonical?. If that's the case then it's gonna take a long time before we see any updates compared to our current situation.

OK, installed the kernel, changed the catalyst PKGBUILD (it patches the 2.6.29-kernel, so it really should work, but nevermind, it took that out), installed catalyst → didn't help.
The glitches are still there, the video performance ist a bit better, I imagine, the errors in error.log are gone

Why not downgrade the catalyst drivers instead of the kernel (so many new updates arrived this time) ?.

Are you running x86_64/compiz no issues?

Please post your updates as soon as possible, thank you!

For older Radeons, AIGLX has long been supported as part of X.Org's official Radeon driver. As of version 8.42 of fglrx (the proprietary driver from ATI), AIGLX is supported. There are reports of instability and problems from users using the latest Compiz and X.Org versions, which are directly related to the driver, and can be circumvented by disabling the "Composite" extension and using at least version 9.1 of the driver.[2] Some users have chosen this, or to not use ATI's proprietary driver due to instability with the drivers. Visual tearing has been noted by some users when using the restricted driver

Thats up on AIGLX wiki, why would anyone want to remove "composite"? Doing so means no compiz and you're basically removing a function just to get decent performance. I can vouch for piss poor performance running youtube in fullscreen (probably no hw acceleration at all) and severe tearing.

Everyone is paying the same amount of money for their GPU, only difference is our choice of operating system, using nix means 0 support from the vendor.... hahahh.....yeah

Catalyst 9.4     2009-04-17     X Server 1.6 support, Ubuntu 9.04 support. Dropped support for older GPUs (R300 to R500).
Catalyst 9.3     2009-03-27     improved OpenGL composite support, last release to support pre X2xxx (pre DirectX10) Cards
Catalyst 9.2     2009-02-20     Bug fixes only.

I didn't have any issues like this on my i686 system (few days ago before I upgraded). So either i used an old catalyst driver probably 9.2-3 or something else is wrong, 64bit kernel?

Last edited by greenfish (2009-05-05 19:52:03)

lastchancetosee · 2009-05-06 06:04:50

greenfish wrote:

Why not downgrade the catalyst drivers instead of the kernel (so many new updates arrived this time) ?.

That's the next thing I'm going to try.

Are you running x86_64/compiz no issues?

I don't run compiz, xcompmgr seems to work fine, but I only use it for shadows. My problem is just the aformentionend glitches in the picture (I think I saw them on console yesterday. Maybe I should install some windows and check wether this isn't broken hardware after all ...) and lousy video playback. Haven't tried anything else yet.

[...] and can be circumvented by disabling the "Composite" extension and using at least version 9.1 of the driver.[2]

I'll give that a try as well, if only for testing. Not really a great solution.

I didn't have any issues like this on my i686 system (few days ago before I upgraded). So either i used an old catalyst driver probably 9.2-3 or something else is wrong, 64bit kernel?

I didn't either, but then a week ago I still had a Radeon 9600 which is supported flawlessly by xf86-video-ati.
@i686: I still have my i686-System from that computer on my harddisk and have already played around with that a bit. There didn't seem to be much of a difference, but I'll do some further testing on that tonight. Now I'm off to work.

ToDo-list:
- disable Composite
- try older catalyst with old/new kernel
- try 32bit
(- get money)
(- buy NVidia card)

greenfish · 2009-05-06 10:21:58

quote @lastchancetosee

Hmm don't we need to downgrade the current kernel < 2.6.29 as well (kernel headers), no expert here just a theory?

I don't run compiz, xcompmgr seems to work fine, but I only use it for shadows. My problem is just the aformentionend glitches in the picture (I think I saw them on console yesterday. Maybe I should install some windows and check wether this isn't broken hardware after all ...) and lousy video playback. Haven't tried anything else yet.

"some windows" you mean the os? I've read a lot of threads since yesterday and were all having the exact same issues, glitches, no hw accelerating, poor performance etc etc.

@i686: I still have my i686-System from that computer on my harddisk and have already played around with that a bit. There didn't seem to be much of a difference, but I'll do some further testing on that tonight. Now I'm off to work.
ToDo-list:
- disable Composite
- try older catalyst with old/new kernel
- try 32bit
(- get money)
(- buy NVidia card)

Thanks for your update again sir, always easier to debug the more people we can get to try out various tests. I'm a bit jealous you have the option of buying a new GPU, since my main system is a laptop im out luck. The wiki/forum isnt clear if open source ati drivers supports 3D acceleration/effects or not, guess the only way is to try out my self when I get back from work.

Good luck sir!

ToDo-list:

-downgrading catalyst
-try open source driver xf86-video-radeonhd
-use a screwdriver

lastchancetosee · 2009-05-06 10:44:43

greenfish wrote:

"some windows" you mean the os? I've read a lot of threads since yesterday and were all having the exact same issues, glitches, no hw accelerating, poor performance etc etc.

Yes I meant that. XP or Vista (because for that I can get a 64bit version via MSDNAA). I don't think that it's some sort of hardware error, but just to make sure...

I'm a bit jealous you have the option of buying a new GPU, since my main system is a laptop im out luck.

Yeah, lucky me . More money I didn't want to spend... Unfortunately I bought this mainboard specifically so I could get rid of additional graphics cards. if all else fails I'll return my board and replace it with something with an NVidia chipset. But you're of course right, at least I HAVE that option.

The wiki/forum isnt clear if open source ati drivers supports 3D acceleration/effects

They don't, at least not for newer cards (R600/R700). They do, and do so flawlessly in my experience, with older cards (why oh why can't I plug my Radeon 9600 AGP into this board? Damn you, PCIe!). I think the radeonhd-driver's support for newer cards is even more experimental than video-ati, but no harm in trying.

btw., no need to call me sir, I'm not that old :-) ...

(Disabling Composite didn't change a thing.)

When you downgrade catalyst, you will probably need the old Xorg as well (1.5). You'll probably find that on the Rollback Machine.

greenfish · 2009-05-06 16:24:21

Yeah, lucky me . More money I didn't want to spend... Unfortunately I bought this mainboard specifically so I could get rid of additional graphics cards. if all else fails I'll return my board and replace it with something with an NVidia chipset. But you're of course right, at least I HAVE that option.

True that, buying a new product just because it's not supported isn't my idea of fun either. At least were NOT forced to go back to windows (last time that happened was in 2006 for me), if x fails there's always the open source drivers (no 3D).

They don't, at least not for newer cards (R600/R700). They do, and do so flawlessly in my experience, with older cards (why oh why can't I plug my Radeon 9600 AGP into this board? Damn you, PCIe!). I think the radeonhd-driver's support for newer cards is even more experimental than video-ati, but no harm in trying.
btw., no need to call me sir, I'm not that old :-) ...
(Disabling Composite didn't change a thing.)
When you downgrade catalyst, you will probably need the old Xorg as well (1.5). You'll probably find that on the Rollback Machine.

Yeah you're right, just tried both OS drivers, compiz breaks with a freeze/white screen so im back to "fglrx". Oh heheheh no sir then

Disabling the composite has no effect for me either, other then compiz breaks (no suprise). The very idea of going back to Xorg 1.5 is a nightmare for me, most of the hw i have on my sony works PERFECT on my current build, old one I had to manually correct things. Oh well i'm gonna accept no compiz-fusion for me now until I start downgrading, but that won't happen until i'm 100% done with my arch64 setup.

Hmm take a look at this thread I posted on phoronix board. look at stef telfords post, he might be onto something? http://www.phoronix.com/forums/showthre … #post73148

Have a good one!

lastchancetosee · 2009-05-08 12:09:07

I'm afraid I will be deserting you. I'll bring back my board and exchange it with one with NVidia-chipset.

In the meantime I tried the radeonhd-driver, with mixed results: It works far better with the RS780, but I couldn't get DRI to work even though others apparently did.
The whole problem seems to be strangely random.

I didn't have time yet to test the old xorg-server, but maybe I'll do that over the weekend before returning the board.

Arch Linux

#1 2009-05-03 07:50:46

Weird graphics error with Radeon HD3200 & catalyst driver

#2 2009-05-03 10:42:51

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#3 2009-05-03 11:43:29

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#4 2009-05-03 12:34:29

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#5 2009-05-03 12:48:21

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#6 2009-05-03 15:17:12

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#7 2009-05-03 15:44:02

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#8 2009-05-03 16:01:57

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#9 2009-05-03 17:15:52

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#10 2009-05-03 20:10:32

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#11 2009-05-03 21:48:08

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#12 2009-05-03 22:53:19

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#13 2009-05-04 09:29:29

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#14 2009-05-04 09:37:27

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#15 2009-05-04 10:03:21

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#16 2009-05-04 19:18:05

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#17 2009-05-05 16:30:04

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#18 2009-05-05 17:25:44

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#19 2009-05-05 18:42:22

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#20 2009-05-05 19:43:11

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#21 2009-05-06 06:04:50

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#22 2009-05-06 10:21:58

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#23 2009-05-06 10:44:43

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#24 2009-05-06 16:24:21

Re: Weird graphics error with Radeon HD3200 & catalyst driver

#25 2009-05-08 12:09:07

Re: Weird graphics error with Radeon HD3200 & catalyst driver

Board footer