FS#20363 - [kernel26] [i915] laptop display blank with 2.6.34.2 and 2.6.35

Attached to Project: Arch Linux
Opened by Byron Clark (byronc) - Wednesday, 04 August 2010, 03:13 GMT
Last edited by Jan de Groot (JGC) - Wednesday, 09 March 2011, 22:47 GMT
Task Type Bug Report
Category Upstream Bugs
Status Closed
Assigned To Tobias Powalowski (tpowa)
Thomas Bächler (brain0)
Architecture x86_64
Severity High
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 2
Private No

Details

Description:

I have a laptop with this video card:

00:02.0 VGA compatible controller: Intel Corporation Core Processor Integrated Graphics Controller (rev 02) (prog-if 00 [VGA controller])
Subsystem: Dell Device 040a
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 44
Region 0: Memory at f0000000 (64-bit, non-prefetchable) [size=4M]
Region 2: Memory at e0000000 (64-bit, prefetchable) [size=256M]
Region 4: I/O ports at 70b0 [size=8]
Expansion ROM at <unassigned> [disabled]
Capabilities: [90] MSI: Enable+ Count=1/1 Maskable- 64bit-
Address: fee0f00c Data: 4181
Capabilities: [d0] Power Management version 2
Flags: PMEClk- DSI+ D1- D2- AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-
Capabilities: [a4] PCI Advanced Features
AFCap: TP+ FLR+
AFCtrl: FLR-
AFStatus: TP-
Kernel driver in use: i915
Kernel modules: i915

I use the method described at http://wiki.archlinux.org/index.php/Intel#KMS_.28Kernel_Mode_Setting.29 to load the driver early in the boot process.

As of kernel26-2.6.34.2-1, the screen is backlit, but blank when booted. After starting Xorg blind, running "xrandr --output DP1 --off; xrandr --output DP1 --auto" turns the backlight off and then on again, but nothing appears on the display.

When booting the laptop connected to a docking station with two external displays, the correct output appears on the external displays.

I updated to kernel26-2.6.35-1 from testing, but the problem was still present.
This task depends upon

Closed by  Jan de Groot (JGC)
Wednesday, 09 March 2011, 22:47 GMT
Reason for closing:  Fixed
Additional comments about closing:  2011-03-09: A task closure has been requested. Reason for request: I mistakenly requested closure of  FS#21839  which appears to be a duplicate of this bug. Upstream bug is closed and all the patches appear to have landed for 2.6.38. Additionally, the 2.6.37.2 kernel fixes the issue for me.
Comment by Byron Clark (byronc) - Wednesday, 04 August 2010, 03:21 GMT
Just to confirm, downgrading to kernel26-2.6.34.1-1 fixes the problem. The laptop panel actually displays something during boot.
Comment by Gerardo Exequiel Pozzi (djgera) - Wednesday, 04 August 2010, 03:25 GMT
Please report to upstream, this is the way to resolve this type of issues. Thanks.
Comment by Byron Clark (byronc) - Wednesday, 04 August 2010, 12:35 GMT
Reported upstream: https://bugzilla.kernel.org/show_bug.cgi?id=16514
Comment by Byron Clark (byronc) - Thursday, 05 August 2010, 01:43 GMT
Moved to a different tracker: https://bugs.freedesktop.org/show_bug.cgi?id=29278
Comment by Lucas Cano (canolucas) - Thursday, 05 August 2010, 10:43 GMT
this happened to me too, but it wsa solved after i did some things:
* updated the bios.
* tried the hotkey to turn the backlight on.
* removed the battery and inserted it again.
don't know which of those solved the issue though.
Comment by Byron Clark (byronc) - Thursday, 05 August 2010, 21:27 GMT
Updating to the latest bios for this machine didn't change anything.
Turning the backlight off and then on again didn't change anything.
Removing the battery and reinserting didn't change anything.
Comment by David C. Rankin (drankinatty) - Saturday, 07 August 2010, 18:05 GMT
Try kernel26-2.6.35-2 from testing. My laptop was hardlocking during boot and compiz was whitescreening when I could get it to boot. It wasn't until I tried kernel26-2.6.35-2 that I was able to boot consistently and compiz is back.
Comment by Thomas Bächler (brain0) - Saturday, 07 August 2010, 19:30 GMT
We also expect 2.6.34.3 and 2.6.35.1 next week.
Comment by Byron Clark (byronc) - Sunday, 08 August 2010, 02:17 GMT
@drankinatty - Aren't all your machines that exhibit that problem using radeon cards? This is specifically an intel problem.

Unfortunately, it doesn't look like a fix for this is present in the 2.6.34.3 and 2.6.35.1 kernels currently under review.
Comment by David C. Rankin (drankinatty) - Sunday, 08 August 2010, 21:08 GMT
Byron, You are right, all my laptops are radeon based (never again... nVidia only). The reason this caught my eye was the backlight on/off, display-nothing, but docked-OK is that they can all relate to the 'module load' stage where I'm failry convinced there is a generic issue that is causing display issues on a whole lot of laptops. A wild-a guess is that whatever lookup table that coordinates hardware id's with modules to load is messed up for the less common (i.e. laptop) configs.

I have been bitten by this before in non-radeon areas as well like atheros wireless, etc. The lspci data you posted shows an i915 module selected, but take a careful look at dmesg to see if that is actually what the hardware is initialized with and look for any other errors noted with DPMS, etc.

Other useful info can probably be gleaned from /var/log/Xorg.0.log to see if there is anything strange there.

My purpose was to advise that if anyone wants to see additional logs, etc. from someone else that has had a number of problems with the 2.6.33 -> 34 -> 35 transition, I'm more than happy to make them available for whatever they may be worth. If there is no interest, no problem.

Good luck with your intel issue, but keep an eye on whether it is actually a kenerl module load issue that shows up as an intel issue in your case.
Comment by Byron Clark (byronc) - Monday, 09 August 2010, 02:35 GMT
@drankinatty - thanks for the tips. This bug probably won't help you to find a generic problem though. It's been bisected (by me and others) to a specific commit in the i915 code. Check the linked upstream bugs if you're interested.
Comment by Jonathan Steel (jsteel) - Wednesday, 11 August 2010, 21:17 GMT
I get a blank display on my Asus Eee 1001P netbook unless I add "acpi_osi=Linux" to my kernel line in my GRUB menu.lst. It was fine before the most recent Kernel update.

For example:

# (0) Arch Linux
title Arch Linux
root (hd0,0)
kernel /vmlinuz26 root=/dev/mapper/root cryptdevice=/dev/sda2:root ro acpi_osi=Linux

Does that help?
Comment by Byron Clark (byronc) - Thursday, 12 August 2010, 02:48 GMT
@jsteel - thanks for the tip, unfortunately, this bug is specific to a different type of intel card than the one present in your netbook. Upstream (see the freedesktop.org bug) is working on it.
Comment by Karol Babioch (johnpatcher) - Monday, 14 February 2011, 17:35 GMT
Is this bug still an issue? I had the same problem with these kernels, but 2.6.36 does fix it for me. Maybe this one can be closed?
Comment by Byron Clark (byronc) - Monday, 14 February 2011, 17:55 GMT
2.6.36 is better, but doesn't entirely fix the issue. It looks like the real fix landed for 2.6.38 and possibly a future 2.6.37.y stable kernel.

Loading...