Bug 20679

Summary: [i915] blank screen with 2.6.29-rc kernels
Product: DRI Reporter: Sami Kerola <kerolasa>
Component: DRM/IntelAssignee: Jesse Barnes <jbarnes>
Status: CLOSED INVALID QA Contact:
Severity: normal    
Priority: medium Keywords: NEEDINFO, regression
Version: XOrg git   
Hardware: x86-64 (AMD64)   
OS: Linux (All)   
Whiteboard:
i915 platform: i915 features:
Attachments:
Description Flags
debug files: dmesg.txt lspci-vvv.txt Xorg.0.log xorg.conf xrandr-verbose.txt none

Description Sami Kerola 2009-03-16 01:36:11 UTC
Created attachment 23905 [details]
debug files: dmesg.txt lspci-vvv.txt Xorg.0.log xorg.conf xrandr-verbose.txt

System environment
------------------

-- chipset:

00:02.0 VGA compatible controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 0c) (prog-if 00 [VGA controller])
        Subsystem: Dell Latitude D630
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+
        Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 0
        Interrupt: pin A routed to IRQ 27
        Region 0: Memory at fea00000 (64-bit, non-prefetchable) [size=1M]
        Region 2: Memory at e0000000 (64-bit, prefetchable) [size=256M]
        Region 4: I/O ports at efe8 [size=8]
        Capabilities: [90] Message Signalled Interrupts: Mask- 64bit- Queue=0/0 Enable+
                Address: fee0300c  Data: 4191
        Capabilities: [d0] Power Management version 3
                Flags: PMEClk- DSI+ D1- D2- AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
                Status: D0 PME-Enable- DSel=0 DScale=0 PME-
                Bridge: PM- B3+
        Kernel modules: intelfb

00:02.1 Display controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 0c)
        Subsystem: Dell Unknown device 01f9
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
        Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 0
        Region 0: Memory at feb00000 (64-bit, non-prefetchable) [size=1M]
        Capabilities: [d0] Power Management version 3
                Flags: PMEClk- DSI+ D1- D2- AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
                Status: D0 PME-Enable- DSel=0 DScale=0 PME-
                Bridge: PM- B3+


-- system architecture: x86_64

-- xf86-video-intel/xserver/mesa/libdrm version:

Name        : mesa-dri-drivers             Relocations: (not relocatable)
Version     : 7.1                               Vendor: Fedora Project
Release     : 0.37.fc9                      Build Date: Fri 27 Jun 2008 10:24:46 PM CEST

X.Org X Server 1.5.2
Release Date: 10 October 2008
X Protocol Version 11, Revision 0
Build Operating System: Linux 2.6.18-92.1.10.el5 x86_64
Current Operating System: Linux lelux 2.6.29-rc7-00143-g99adcd9 #1 SMP Tue Mar 10 13:10:40 CET 2009 x86_64
Build Date: 03 November 2008  12:34:48AM
Build ID: xorg-x11-server 1.5.2-3.fc9

server glx version string: 1.2

Name        : libdrm                       Relocations: (not relocatable)
Version     : 2.4.0                             Vendor: Fedora Project
Release     : 0.13.fc9                      Build Date: Wed 18 Jun 2008 01:25:28 AM CEST

-- kernel version: 2.6.29-rc7-00143-g99adcd9

-- Linux distribution: Fedora release 9

-- Machine or mobo model: Dell Latitude D630

http://www.dell.com/content/products/productdetails.aspx/latit_d630?c=us&l=en&s=biz&cs=555

-- Display connector: docking connector, VGA,

Reproduce steps. Probability if not 100% reproducible
-----------------------------------------------------

Problems started with 2.6.29-rc* kernels. I am not sure which was the first time I saw the behavior first time but rc3 is quite close. My laptop is nearly always attached to docking station, so it took a while to figure out how bad this bug is.

Startup. Initially I noticed blank screen after boot. That was easily cured with switching to console and back to login page (ctrl+alt+f1 & ctrl+alt+f7).

Mode switching. When switching modes nearly 50% of times when I come back to X screen is blank. Retry works fine.

Deattach from dock while system running test. After this X was never blank, but console mode was always. The console mode was perfectly usable.

Reattach to dock. System freeze without stacktrace and the Magic SysRq keys did nothing. All of my reattachment tries has always caused this, and because behavior is so annoying I have done this only three times.

When I send email to Jessa Barnes he replied: "Sounds like a regression,
but it could be related to the new KMS bits..."
Comment 1 Sami Kerola 2009-03-17 04:25:36 UTC
(In reply to comment #0)

Results with 2.6.29-rc8-00124 are different.

> Mode switching. When switching modes nearly 50% of times when I come back to X
> screen is blank. Retry works fine.

100% of switches work.

> Deattach from dock while system running test. After this X was never blank, but
> console mode was always. The console mode was perfectly usable.

Same as reattach.

> Reattach to dock. System freeze without stacktrace and the Magic SysRq keys did
> nothing. All of my reattachment tries has always caused this, and because
> behavior is so annoying I have done this only three times.

Screen is not blank, it is frozen with what ever there happens to be. Input does nothing including SysRq stuff.
Comment 2 Dan Nicholson 2009-03-17 09:20:39 UTC
(In reply to comment #1)
> (In reply to comment #0)
> 
> > Reattach to dock. System freeze without stacktrace and the Magic SysRq keys did
> > nothing. All of my reattachment tries has always caused this, and because
> > behavior is so annoying I have done this only three times.
> 
> Screen is not blank, it is frozen with what ever there happens to be. Input
> does nothing including SysRq stuff.

I was having problems with docking on a different laptop (Thinkpad X60) that were because of acpi bugs. Can you check syslog to make sure this is drm causing the issue and not another subsystem? I was getting a flood of events that essentially made the system lock up.
Comment 3 Sami Kerola 2009-03-18 10:01:41 UTC
After recompiling with following as yes

CONFIG_PM_DEBUG
CONFIG_ACPI_DEBUG
CONFIG_PCI_DEBUG
CONFIG_DEBUG_DRIVER
CONFIG_DEBUG_KERNEL
CONFIG_DEBUG_BUGVERBOSE
CONFIG_DEBUG_INFO  

and making acpi to output more

echo 0x1F >/sys/module/acpi/parameters/debug_level
echo 0x1F >/sys/module/acpi/parameters/debug_layer

I still do not see anything in syslog and I am not sure what is causing the problem. Initial problems with blank screen seemed like graphics driver issue, but reattach to dock may very well be something else. Lets see will I be able to figure out what, if I do I let you know.
Comment 4 Gordon Jin 2009-07-09 00:11:41 UTC
Does this still exist in 2.6.30 or 2.6.31-rcX?
Comment 5 Sami Kerola 2009-07-12 02:45:06 UTC
My testing environment stopped working properly so I do not know. I hope to have time within week or two to get environment fixed so that I can say something more reliable.
Comment 6 Jesse Barnes 2009-08-31 10:26:07 UTC
Would be good if you could test a distro kernel w/KMS enabled too, or try a kernel config based on a distro config, that way you're likely to be trying well tested options.
Comment 7 Jesse Barnes 2009-10-05 10:39:51 UTC
Closing due to inactivity.  If you still see this problem (hopefully it's been fixed) it may be due to ACPI docking activity.  We have some hooks to use to reprogram the outputs in that case, so it may be easy to fix.
Comment 8 Jari Tahvanainen 2016-11-03 08:10:28 UTC
Closing resolved+invalid, no activity in 7 years.

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.