Bug 33331

Summary: [radeon 5450] screen damage + lack of 2D accelleration.
Product: xorg Reporter: Pawel Sikora <pawel_sikora>
Component: Driver/RadeonAssignee: xf86-video-ati maintainers <xorg-driver-ati>
Status: RESOLVED FIXED QA Contact: Xorg Project Team <xorg-team>
Severity: normal    
Priority: medium    
Version: unspecified   
Hardware: x86-64 (AMD64)   
OS: Linux (All)   
Whiteboard:
i915 platform: i915 features:
Attachments:
Description Flags
initial kde4 dual screen shot.
none
screen damage. step 1.
none
screen damage. step 2.
none
screen damage. step 3.
none
xserver log.
none
installed packages.
none
initial kde screen for radeon driver compiled from git sources.
none
xorg.log for radeon-git driver installed in /opt/xorg. none

Description Pawel Sikora 2011-01-21 04:56:01 UTC
hi,

i've recently bought an "asus EAH5450 silent" radeon-hd for my
lcd (1600x1050) and tv (1920x1080) connected as one virtual desktop.

01:00.0 VGA compatible controller: ATI Technologies Inc Device 68f9 (prog-if 00 [VGA controller])
        Subsystem: ASUSTeK Computer Inc. Device 0386
        Flags: bus master, fast devsel, latency 0, IRQ 43
        Memory at d0000000 (64-bit, prefetchable) [size=256M]
        Memory at fe8c0000 (64-bit, non-prefetchable) [size=128K]
        I/O ports at b000 [size=256]
        Expansion ROM at fe8a0000 [disabled] [size=128K]
        Capabilities: [50] Power Management version 3
        Capabilities: [58] Express Legacy Endpoint, MSI 00
        Capabilities: [a0] MSI: Enable+ Count=1/1 Maskable- 64bit+
        Capabilities: [100] Vendor Specific Information: ID=0001 Rev=1 Len=010 <?>
        Capabilities: [150] Advanced Error Reporting
        Kernel driver in use: radeon

i've configured the xorg as usual for dual screen setup:

Section "ServerLayout"
        Identifier     "Simple Layout"
        Screen      0  "Screen 1" 0 0
EndSection

Section "Files"
        FontPath     "/usr/share/fonts/misc"
        FontPath     "/usr/share/fonts/TTF"
        FontPath     "/usr/share/fonts/OTF"
        FontPath     "/usr/share/fonts/Type1"
        FontPath     "/usr/share/fonts/100dpi"
        FontPath     "/usr/share/fonts/75dpi"
EndSection

Section "Module"
        Load  "dbe"
        Load  "extmod"
        Load  "freetype"
        Load  "dri"
        Load  "glx"
EndSection

Section "Monitor"
        Identifier   "VGA-0"
        Option       "Primary" "true"
        Option       "PreferredMode" "1600x1050"
        Option       "LeftOf" "DVI-0"
EndSection

Section "Monitor"
        Identifier   "DVI-0"
        Option       "PreferredMode" "1920x1080"
        Option       "Position" "1600 0"
EndSection

Section "Device"
        Identifier  "ATI"
        Driver      "radeon"
        Option      "AccelMethod" "EXA"
EndSection

Section "Screen"
        Identifier "Screen 1"
        Device     "ATI"
        Monitor    "VGA-0"
        DefaultDepth     24
        SubSection "Display"
                Viewport   0 0
                Depth     24
                Modes    "1680x1050"
        EndSubSection
EndSection

Section "DRI"
        Group        "video"
        Mode         0660
EndSection

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

modprobe radeon kernel (2.6.37) module:

(...)
[  194.094552] [drm] Initialized drm 1.1.0 20060810
[  194.153135] [drm] radeon kernel modesetting enabled.
[  194.153203] radeon 0000:01:00.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16
[  194.153208] radeon 0000:01:00.0: setting latency timer to 64
[  194.155864] [drm] initializing kernel modesetting (CEDAR 0x1002:0x68F9).
[  194.156106] [drm] register mmio base: 0xFE8C0000
[  194.156108] [drm] register mmio size: 131072
[  194.156306] ATOM BIOS: 68F9.12.18.0.1.AS02
[  194.156319] radeon 0000:01:00.0: GPU softreset
[  194.156322] radeon 0000:01:00.0:   GRBM_STATUS=0x00003828
[  194.156324] radeon 0000:01:00.0:   GRBM_STATUS_SE0=0x00000007
[  194.156326] radeon 0000:01:00.0:   GRBM_STATUS_SE1=0x00000007
[  194.156328] radeon 0000:01:00.0:   SRBM_STATUS=0x200000C0
[  194.156388] radeon 0000:01:00.0:   GRBM_SOFT_RESET=0x00007F6B
[  194.156491] radeon 0000:01:00.0:   GRBM_STATUS=0x00003828
[  194.156493] radeon 0000:01:00.0:   GRBM_STATUS_SE0=0x00000007
[  194.156495] radeon 0000:01:00.0:   GRBM_STATUS_SE1=0x00000007
[  194.156498] radeon 0000:01:00.0:   SRBM_STATUS=0x200000C0
[  194.157581] radeon 0000:01:00.0: VRAM: 1024M 0x0000000000000000 - 0x000000003FFFFFFF (1024M used)
[  194.157583] radeon 0000:01:00.0: GTT: 512M 0x0000000040000000 - 0x000000005FFFFFFF
[  194.157592] mtrr: type mismatch for d0000000,10000000 old: write-back new: write-combining
[  194.157594] [drm] Detected VRAM RAM=1024M, BAR=256M
[  194.157596] [drm] RAM width 64bits DDR
[  194.157663] [TTM] Zone  kernel: Available graphics memory: 4097868 kiB.
[  194.157665] [TTM] Zone   dma32: Available graphics memory: 2097152 kiB.
[  194.157667] [TTM] Initializing pool allocator.
[  194.157694] [drm] radeon: 1024M of VRAM memory ready
[  194.157696] [drm] radeon: 512M of GTT memory ready.
[  194.157756] radeon 0000:01:00.0: irq 43 for MSI/MSI-X
[  194.157761] radeon 0000:01:00.0: radeon: using MSI.
[  194.157805] [drm] radeon: irq initialized.
[  194.157807] [drm] GART: num cpu pages 131072, num gpu pages 131072
[  194.158514] [drm] Loading CEDAR Microcode
[  194.241829] radeon 0000:01:00.0: WB enabled
[  194.258060] [drm] ring test succeeded in 1 usecs
[  194.258156] [drm] radeon: ib pool ready.
[  194.258270] [drm] ib test succeeded in 0 usecs
[  194.258281] failed to evaluate ATIF got AE_BAD_PARAMETER
[  194.260586] [drm] Radeon Display Connectors
[  194.260588] [drm] Connector 0:
[  194.260589] [drm]   HDMI-A
[  194.260590] [drm]   HPD1
[  194.260592] [drm]   DDC: 0x6440 0x6440 0x6444 0x6444 0x6448 0x6448 0x644c 0x644c
[  194.260593] [drm]   Encoders:
[  194.260594] [drm]     DFP1: INTERNAL_UNIPHY1
[  194.260596] [drm] Connector 1:
[  194.260597] [drm]   DVI-I
[  194.260598] [drm]   HPD4
[  194.260599] [drm]   DDC: 0x6460 0x6460 0x6464 0x6464 0x6468 0x6468 0x646c 0x646c
[  194.260601] [drm]   Encoders:
[  194.260602] [drm]     DFP2: INTERNAL_UNIPHY
[  194.260603] [drm]     CRT1: INTERNAL_KLDSCP_DAC1
[  194.260604] [drm] Connector 2:
[  194.260605] [drm]   VGA
[  194.260607] [drm]   DDC: 0x6430 0x6430 0x6434 0x6434 0x6438 0x6438 0x643c 0x643c
[  194.260608] [drm]   Encoders:
[  194.260609] [drm]     CRT2: INTERNAL_KLDSCP_DAC2
[  194.418508] [drm] Internal thermal controller without fan control
[  194.418589] [drm] radeon: power management initialized
[  194.592031] [drm] fb mappable at 0xD0141000
[  194.592033] [drm] vram apper at 0xD0000000
[  194.592034] [drm] size 8294400
[  194.592035] [drm] fb depth is 24
[  194.592036] [drm]    pitch is 7680
[  194.596862] [drm] force priority to high
[  194.796656] [drm] force priority to high
[  194.803525] [drm] force priority to high
[  194.803529] [drm] force priority to high
[  194.809357] Console: switching to colour frame buffer device 210x65
[  194.813269] fb0: radeondrmfb frame buffer device
[  194.813271] drm: registered panic notifier
[  194.813280] [drm] Initialized radeon 2.7.0 20080528 for 0000:01:00.0 on minor 0
[  226.760193] [drm] force priority to high
[  226.760197] [drm] force priority to high
[  226.780190] [drm] force priority to high
[  226.780193] [drm] force priority to high

and finally started up kde4 gui (capture.jpeg) and tested following things...

1). basic Xv check:

$ xvinfo
X-Video Extension version 2.2
screen #0
 no adaptors present

$ xvattr
Found Xv 2.2

$ xdriinfo
Screen 0: not direct rendering capable.

afaics on the http://www.x.org/wiki/RadeonFeature there's dri/texturedxv/exa
support for 'evergreen' radeon family, so what is wrong with my card?

2). basic 'window drag&move test'.

please look at the 3 screenshots (*_213[0-2].jpg). with each few moves
of dragged window i'm observing more damages on screen. the one way to fix
the screen is window zooming to full screen and back.


i can provide more information/debug/test you need to solve this problems.
Comment 1 Pawel Sikora 2011-01-21 04:56:55 UTC
Created attachment 42262 [details]
initial kde4 dual screen shot.
Comment 2 Pawel Sikora 2011-01-21 04:58:00 UTC
Created attachment 42263 [details]
screen damage. step 1.
Comment 3 Pawel Sikora 2011-01-21 04:58:27 UTC
Created attachment 42264 [details]
screen damage. step 2.
Comment 4 Pawel Sikora 2011-01-21 04:58:52 UTC
Created attachment 42265 [details]
screen damage. step 3.
Comment 5 Pawel Sikora 2011-01-21 05:00:30 UTC
Created attachment 42266 [details]
xserver log.
Comment 6 Pawel Sikora 2011-01-21 05:01:31 UTC
Created attachment 42267 [details]
installed packages.
Comment 7 Pawel Sikora 2011-01-21 06:57:32 UTC
Created attachment 42270 [details]
initial kde screen for radeon driver compiled from git sources.
Comment 8 Pawel Sikora 2011-01-21 06:58:45 UTC
Created attachment 42271 [details] [review]
xorg.log for radeon-git driver installed in /opt/xorg.
Comment 9 Alex Deucher 2011-01-21 07:20:09 UTC
You are using UMS which does not support acceleration with evergreen cards.  Only KMS supports acceleration.  The display issues are likely caused by a mix of a UMS ddx (xf86-video-ati) with a KMS kernel; both drivers are trying to control the hardware.  Make sure you built your ddx with KMS support and that the drm is loaded before you start X.
Comment 10 Pawel Sikora 2011-01-21 07:52:49 UTC
(In reply to comment #9)
> You are using UMS which does not support acceleration with evergreen cards. 
> Only KMS supports acceleration.  The display issues are likely caused by a mix
> of a UMS ddx (xf86-video-ati) with a KMS kernel; both drivers are trying to
> control the hardware.  Make sure you built your ddx with KMS support and that
> the drm is loaded before you start X.

ahh, my fault. i've missed that ./configure doesn't detect KMS automatically
due to lack of some -devel packages in my system. in my case the lack
of xinerama devel stuff leads to pkg-config error for xorg-server.pc request.

with all necessary packages the radeon git driver seems to work fine
with my evergreen card. at least the aiglx, compositing, and textured xv
used by tvtime are fine.
Comment 11 Pawel Sikora 2011-01-21 07:53:55 UTC
current git driver works fine.

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.