Bug 16226

Summary: xrandr - ATI FireGL T2 (Thinkpad T42p) - Problem with two external Monitors
Product: xorg Reporter: el_lobo <wolfgang.knabl>
Component: Driver/RadeonAssignee: xf86-video-ati maintainers <xorg-driver-ati>
Status: RESOLVED INVALID QA Contact: Xorg Project Team <xorg-team>
Severity: normal    
Priority: high    
Version: unspecified   
Hardware: x86 (IA32)   
OS: Linux (All)   
Whiteboard:
i915 platform: i915 features:
Attachments:
Description Flags
Scrambled output with dual monitors using compiz with FireGL T2 (M10) none

Description el_lobo 2008-06-04 02:31:51 UTC
The ATI FireGL T2 has two CRTC's. If I put the Thinkpad into its Dockingstation I can attach two external monitors (1 VGA, 1 DVI) - and I can use them both if I turn off the LVDS (laptop-Panel).

My Problem: Either the DVI-0 or the VGA-0 doesn't display a proper image (looks like a sync problem); depending which monitor is on CRTC 0. The one using CRTC 1 is allways fine.

Playing arround with modes did not resolve my problem so far.

(OS: Ubuntu 8.04 / ATI-driver / xrandr 1.2).
Comment 1 Alex Deucher 2010-07-15 07:39:10 UTC
Is this still an issue?
Comment 2 Alex Deucher 2010-10-19 16:28:34 UTC
Is this still an issue with KMS or a newer version of the driver?
Comment 3 el_lobo 2010-10-20 03:45:26 UTC
Since Ubuntu Lucid (10.04) w/ KMS the dualhead configuration works fine; except sometimes returning from screensaver leaves the DVI-connected monitor blank so it needs a manual xrandr to turn it on again.
Comment 4 marcnesium 2011-09-04 06:44:57 UTC
Created attachment 50887 [details]
Scrambled output with dual monitors using compiz with FireGL T2 (M10)
Comment 5 marcnesium 2011-09-04 06:46:39 UTC
I have very similar problem, except it is only an issue while using compiz. In detail:
I'm using a Thinkpad T42p with ATI FireGL T2 (M10) with Ubuntu 11.04.
Using metacity while having two monitors is working flawlessly.
But with enabled compiz (on laptop screen only is no problem) and then connecting a 2nd monitor (or putting laptop into docking station with connected lcd) both screens get scrambled (see attachment). Can't find any error messages though.

(I tried to disable KMS, but no use.)

Some more details:

$ uname -a
Linux T42p 2.6.38-11-generic #48-Ubuntu SMP Fri Jul 29 19:05:14 UTC 2011 i686 i686 i386 GNU/Linux  

$ lspci -nn | grep VGA
01:00.0 VGA compatible controller [0300]: ATI Technologies Inc M10 NT [FireGL Mobility T2] [1002:4e54] (rev 80)

$ sudo lshw -class Display
  *-display               
       description: VGA compatible controller
       product: M10 NT [FireGL Mobility T2]
       vendor: ATI Technologies Inc
       physical id: 0
       bus info: pci@0000:01:00.0
       version: 80
       width: 32 bits
       clock: 66MHz
       capabilities: agp agp-2.0 pm vga_controller bus_master cap_list rom
       configuration: driver=radeon latency=66 mingnt=8
       resources: irq:11 memory:e0000000-e7ffffff ioport:3000(size=256) memory:c0100000-c010ffff memory:c0120000-c013ffff

$ dmesg | grep drm
[    1.717623] [drm] Initialized drm 1.1.0 20060810
[    1.794946] [drm] radeon defaulting to kernel modesetting.
[    1.794958] [drm] radeon kernel modesetting enabled.
[    1.798665] [drm] initializing kernel modesetting (RV350 0x1002:0x4E54).
[    1.798710] [drm] register mmio base: 0xC0100000
[    1.798715] [drm] register mmio size: 65536
[    1.799049] [drm] Generation 2 PCI interface, using max accessible memory
[    1.799081] [drm] Supports vblank timestamp caching Rev 1 (10.10.2010).
[    1.799087] [drm] Driver supports precise vblank timestamp query.
[    1.799106] [drm] radeon: irq initialized.
[    1.800352] [drm] Detected VRAM RAM=128M, BAR=128M
[    1.800364] [drm] RAM width 128bits DDR
[    1.800505] [drm] radeon: 128M of VRAM memory ready
[    1.800511] [drm] radeon: 256M of GTT memory ready.
[    1.800554] [drm] radeon: 1 quad pipes, 1 Z pipes initialized.
[    1.802644] [drm] Loading R300 Microcode
[    1.806285] [drm] radeon: ring at 0x00000000D0001000
[    1.806314] [drm] ring test succeeded in 1 usecs
[    1.806541] [drm] radeon: ib pool ready.
[    1.806622] [drm] ib test succeeded in 0 usecs
[    1.807712] [drm] Panel ID String: 1600x1200               
[    1.807719] [drm] Panel Size 1600x1200
[    1.807826] [drm] No TV DAC info found in BIOS
[    1.807906] [drm] Radeon Display Connectors
[    1.807912] [drm] Connector 0:
[    1.807916] [drm]   VGA
[    1.807921] [drm]   DDC: 0x60 0x60 0x60 0x60 0x60 0x60 0x60 0x60
[    1.807927] [drm]   Encoders:
[    1.807931] [drm]     CRT1: INTERNAL_DAC1
[    1.807936] [drm] Connector 1:
[    1.807940] [drm]   DVI-D
[    1.807944] [drm]   HPD1
[    1.807948] [drm]   DDC: 0x64 0x64 0x64 0x64 0x64 0x64 0x64 0x64
[    1.807954] [drm]   Encoders:
[    1.807958] [drm]     DFP1: INTERNAL_TMDS1
[    1.807963] [drm] Connector 2:
[    1.807967] [drm]   LVDS
[    1.807970] [drm]   Encoders:
[    1.807974] [drm]     LCD1: INTERNAL_LVDS
[    1.807979] [drm] Connector 3:
[    1.807982] [drm]   S-video
[    1.807986] [drm]   Encoders:
[    1.807990] [drm]     TV1: INTERNAL_DAC2
[    1.867425] [drm] radeon: power management initialized
[    2.004770] [drm] fb mappable at 0xE0040000
[    2.004777] [drm] vram apper at 0xE0000000
[    2.004782] [drm] size 7680000
[    2.004786] [drm] fb depth is 24
[    2.004790] [drm]    pitch is 6400
[    2.004989] fbcon: radeondrmfb (fb0) is primary device
[    2.180898] fb0: radeondrmfb frame buffer device
[    2.181061] drm: registered panic notifier
[    2.181292] [drm] Initialized radeon 2.8.0 20080528 for 0000:01:00.0 on minor 0

Is it a bug? Can i investigate more? What can i try?
Thanks in advance!
Bye, Marc.
Comment 6 Alex Deucher 2011-09-05 19:19:23 UTC
(In reply to comment #5)
> I have very similar problem, except it is only an issue while using compiz. In
> detail:

This is a different issue.  Compiz should query the driver for the max texture size and not enable effects if it is smaller than the total desktop size.  If your desktop is larger than 2048 pixels is width or height (max texture size) you will get undefined behavior as the surface is larger than the hardware limits.
Comment 7 marcnesium 2011-09-07 00:30:53 UTC
(In reply to comment #6)
> (In reply to comment #5)
> > I have very similar problem, except it is only an issue while using compiz. In
> > detail:
> 
> This is a different issue.  Compiz should query the driver for the max texture
> size and not enable effects if it is smaller than the total desktop size.  If
> your desktop is larger than 2048 pixels is width or height (max texture size)
> you will get undefined behavior as the surface is larger than the hardware
> limits.

Thank you for clearing this. Yes my desktop is larger than 2048 pixel width (1600x1200 + 1280x1024). If this is a hardware limit, i guess there is nothing to do about it.

But it is still left, that in earlier version (i tried a 10.4 Ubuntu live CD) this query worked already and does not work atm.
Comment 8 Michel Dänzer 2011-09-07 03:00:40 UTC
(In reply to comment #7)
> But it is still left, that in earlier version (i tried a 10.4 Ubuntu live CD)
> this query worked already and does not work atm.

What do you mean by 'this query worked' in Ubuntu 10.4?

el_lobo: Can this bug be resolved per comment #3?
Comment 9 Christopher M. Penalver 2016-02-24 04:59:02 UTC
el_lobo, Ubuntu 10.04 Desktop reached EOL on May 9, 2013. For more on this, please see https://wiki.ubuntu.com/Releases .

If this is reproducible in a supported release, it will help immensely if you filed a new report with Ubuntu by ensuring you have the package xdiagnose installed, and that you click the Yes button for attaching additional debugging information running the following from a terminal:
ubuntu-bug xorg

Also, please feel free to subscribe me to it.

For more on why this is helpful, please see https://wiki.ubuntu.com/ReportingBugs.

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.