FS#21839 - [kernel26] Blank screen after kernel26 and xf86-video-intel updates

Attached to Project: Arch Linux
Opened by Jeremy LaCroix (jlacroix) - Friday, 26 November 2010, 17:43 GMT
Last edited by Vesa Kaihlavirta (vegai) - Sunday, 13 March 2011, 11:28 GMT
Task Type Bug Report
Category Packages: Core
Status Closed
Assigned To Tobias Powalowski (tpowa)
Jan de Groot (JGC)
Architecture All
Severity High
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 4
Private No

Details

Description:
I have a Dell Latitude E6410 laptop with intel graphics. The laptop worked fine up until yesterdays updates, which included kernel26 and xf86-video-intel, as well as mesa 7.9. However, when those updates occurred and I rebooted, I get nothing but a blank screen. The brightness controls work (you can see the blacklight getting brighter) but there is nothing actually on the screen.

When I start my laptop, I see the Dell logo, then the Grub menu. After a few seconds after the grub menu, the screen goes black. I can SSH into it from another computer, but X just does not work at all anymore and I cannot use my laptop until this is fixed.

Steps to reproduce:
Just start my laptop, it happens every time.
This task depends upon

Closed by  Vesa Kaihlavirta (vegai)
Sunday, 13 March 2011, 11:28 GMT
Reason for closing:  Fixed
Comment by Jeremy LaCroix (jlacroix) - Friday, 26 November 2010, 18:24 GMT
I tried running without KMS, and that lets me get to the KDM screen, but it will crash immediately right back to the KDM screen after putting in my password.
Comment by Jeremy LaCroix (jlacroix) - Friday, 26 November 2010, 21:10 GMT
Downgrading to kernel 2.6.35 fixes this for me, so this must be something in the new kernel.
Comment by Ionut Biru (wonder) - Friday, 26 November 2010, 21:19 GMT
paste lspci
Comment by Jeremy LaCroix (jlacroix) - Friday, 26 November 2010, 21:36 GMT
00:00.0 Host bridge: Intel Corporation Core Processor DRAM Controller (rev 02)
00:02.0 VGA compatible controller: Intel Corporation Core Processor Integrated Graphics Controller (rev 02)
00:19.0 Ethernet controller: Intel Corporation 82577LM Gigabit Network Connection (rev 05)
00:1a.0 USB Controller: Intel Corporation 5 Series/3400 Series Chipset USB2 Enhanced Host Controller (rev 05)
00:1b.0 Audio device: Intel Corporation 5 Series/3400 Series Chipset High Definition Audio (rev 05)
00:1c.0 PCI bridge: Intel Corporation 5 Series/3400 Series Chipset PCI Express Root Port 1 (rev 05)
00:1c.2 PCI bridge: Intel Corporation 5 Series/3400 Series Chipset PCI Express Root Port 3 (rev 05)
00:1c.3 PCI bridge: Intel Corporation 5 Series/3400 Series Chipset PCI Express Root Port 4 (rev 05)
00:1d.0 USB Controller: Intel Corporation 5 Series/3400 Series Chipset USB2 Enhanced Host Controller (rev 05)
00:1e.0 PCI bridge: Intel Corporation 82801 Mobile PCI Bridge (rev a5)
00:1f.0 ISA bridge: Intel Corporation Mobile 5 Series Chipset LPC Interface Controller (rev 05)
00:1f.2 SATA controller: Intel Corporation 5 Series/3400 Series Chipset 6 port SATA AHCI Controller (rev 05)
00:1f.3 SMBus: Intel Corporation 5 Series/3400 Series Chipset SMBus Controller (rev 05)
00:1f.6 Signal processing controller: Intel Corporation 5 Series/3400 Series Chipset Thermal Subsystem (rev 05)
01:00.0 Network controller: Atheros Communications Inc. AR928X Wireless Network Adapter (PCI-Express) (rev 01)
02:00.0 CardBus bridge: Ricoh Co Ltd Device e476 (rev 02)
02:00.1 SD Host controller: Ricoh Co Ltd Device e822 (rev 03)
02:00.4 FireWire (IEEE 1394): Ricoh Co Ltd Device e832 (rev 03)
3f:00.0 Host bridge: Intel Corporation Core Processor QuickPath Architecture Generic Non-core Registers (rev 02)
3f:00.1 Host bridge: Intel Corporation Core Processor QuickPath Architecture System Address Decoder (rev 02)
3f:02.0 Host bridge: Intel Corporation Core Processor QPI Link 0 (rev 02)
3f:02.1 Host bridge: Intel Corporation Core Processor QPI Physical 0 (rev 02)
3f:02.2 Host bridge: Intel Corporation Core Processor Reserved (rev 02)
3f:02.3 Host bridge: Intel Corporation Core Processor Reserved (rev 02)
Comment by Andreas Radke (AndyRTR) - Saturday, 27 November 2010, 10:33 GMT
the intel driver doesn't support UMS mode anymore. you must use KMS mode. then report full logs (dmesg/Xorg.0.log).
Comment by Jeremy LaCroix (jlacroix) - Saturday, 27 November 2010, 16:04 GMT
How do I change from UMS to KMS?
Comment by Karol Błażewicz (karol) - Saturday, 27 November 2010, 16:08 GMT Comment by Jeremy LaCroix (jlacroix) - Saturday, 27 November 2010, 16:27 GMT
I followed the directions in the link, and it did not make a difference.

I'm really confused about this, because I have been using KMS since before this latest kernel was released. It allows me full resolution in TTY's and I use that a lot. I don't know why it's reporting that I'm not using KMS, because I am, and have been.
   Xorg.0.log (26.8 KiB)
   dmesg (41.4 KiB)
Comment by Andreas Radke (AndyRTR) - Sunday, 28 November 2010, 10:43 GMT
please add "verbose" to your grub kernel append line to get usefull output. please use KMS late mode to avoid mkinitcpio issues.
Comment by Jeremy LaCroix (jlacroix) - Sunday, 28 November 2010, 15:57 GMT
Here is the updated dmesg file after running verbose. Please note that I already run late start kms, as I believe it is the default. (I created this system after kms went mandatory, and use whatever is the default now days).
   dmesg.txt (42.1 KiB)
Comment by Jeremy LaCroix (jlacroix) - Sunday, 28 November 2010, 20:38 GMT
I did some further experimenting today. I grabbed a spare hard drive, and installed Arch on it with the latest install media. I did not install X, any drivers, or applications. I only installed core. After installing, I did a -Syu and rebooted. Same problem.
Comment by maxexcloo (maxexcloo) - Monday, 29 November 2010, 11:20 GMT
I have the same issue.
I had to downgrade my kernel and intel driver to post this.

In my case however, the laptop would boot and then take maybe 45 minutes till X would start, then I would unlock it (I use openbox and xscreensaver set to lock) and then it would lag severely and nothing but moving the mouse would work.
Comment by maxexcloo (maxexcloo) - Monday, 29 November 2010, 11:23 GMT
Attached are log files from my current boot, I can test the updated packages in a day, I have work to do :)
Comment by Jeremy LaCroix (jlacroix) - Monday, 29 November 2010, 15:03 GMT
I also want to add that with using 2.6.35 and all of the latest Intel packages, KWin crashes almost every time I exit a full screen game.
Comment by Peinthor Rene (rp-) - Monday, 29 November 2010, 20:31 GMT
same problem here.

I also tried the latest kernel26-git, here I get a picture after KMS turned on, but it's flickering all the time. Not usable if you ask me.
Comment by Christian Neukirchen (chneukirchen) - Monday, 29 November 2010, 21:32 GMT
Same problem with dark screen here on a HP ProBook 5320m, Core i5 integrated graphics.

The screen goes black as soon as i915/intel_agp is loaded (either in initrd or userland). Booting with i915.modeset=0 works, but of course only with xf86-video-vesa (and wrong DPI).

Downgrading to kernel26-2.6.35.8-1 works, so this is a kernel26 issue, not xf86-video-intel.
Comment by Jeremy LaCroix (jlacroix) - Monday, 29 November 2010, 21:33 GMT
I agree about it being a kernel bug. I'm the one that filed this bug initially but I don't know how to remove xf86-video-intel from the description. :(
Comment by awhan patnaik (awhan) - Tuesday, 30 November 2010, 17:25 GMT
am on ibm thinkpad r51. during the bootup when the udev events are being processed, screen goes black and keyboard goes dead, i have to hard shutdown. downgrading to kernel26-2.6.35.8-1 works.
Comment by Peinthor Rene (rp-) - Tuesday, 30 November 2010, 18:29 GMT
Alright! After a few hours building kernels and bisecting, for me it all comes to this commit:
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=7643a7fa16edf180d593f705f4fa5930c40e8d2d

I created a patch to revert this commit and build a working 2.6.36 kernel.
My hardware is a HP ProBook 5320m, I don't know if the same patch will work for you as I read some bug reports that it's often very hardware specific, as there are lot of reports with blank screen after KMS init.

Would be nice if someone else could test this patch.
Comment by Jeremy LaCroix (jlacroix) - Tuesday, 30 November 2010, 18:35 GMT
I would be more than happy to try the patch, but I'm a newbie when it comes to patching and I don't know how. I'll see if one of my colleagues is available tonight to help me patch it.
Comment by Andreas Radke (AndyRTR) - Tuesday, 30 November 2010, 20:17 GMT
Rene, please report it upstream!
Comment by Peinthor Rene (rp-) - Tuesday, 30 November 2010, 20:30 GMT Comment by Peinthor Rene (rp-) - Tuesday, 30 November 2010, 21:55 GMT
I created an AUR package which builds the kernel 2.6.36.1 with this patch.
http://aur.archlinux.org/packages.php?ID=44045

the package isn't tested to much so don't expect to much.
Comment by Jeremy LaCroix (jlacroix) - Wednesday, 01 December 2010, 00:39 GMT
The patch you posted works, but has an undesired effect.

First of all, I am typing this from the patched kernel that Rene posted. This is more than I was ever able to do before. I am able to log in and use my machine. In addition, my card reader works now, and it didn't work in 2.6.35.

On the downside, tty's don't work. When I boot up, most of the boot process is a blank black screen, but the difference now is that X starts and I can use my laptop but tty's don't work.

Good job though! All we have to do is restore tty access and I'm good as gold.
Comment by Peinthor Rene (rp-) - Wednesday, 01 December 2010, 06:51 GMT
I don't know if it will help but did you try to load the i915 early, during the bootstrap?
Instructions how to do this is here: https://wiki.archlinux.org/index.php/Intel#KMS_.28Kernel_Mode_Setting.29
Comment by Jeremy LaCroix (jlacroix) - Wednesday, 01 December 2010, 13:18 GMT
That's actually what i was already doing. It doesn't seem to make a difference.
Comment by Christian Neukirchen (chneukirchen) - Wednesday, 01 December 2010, 21:08 GMT
kernel26-intelfix appears to work properly on this HP ProBook 5320m.

00:00.0 Host bridge: Intel Corporation Core Processor DRAM Controller (rev 02)
00:02.0 VGA compatible controller: Intel Corporation Core Processor Integrated Graphics Controller (rev 02)
00:16.0 Communication controller: Intel Corporation 5 Series/3400 Series Chipset HECI Controller (rev 06)
00:1a.0 USB Controller: Intel Corporation 5 Series/3400 Series Chipset USB2 Enhanced Host Controller (rev 05)
00:1b.0 Audio device: Intel Corporation 5 Series/3400 Series Chipset High Definition Audio (rev 05)
00:1c.0 PCI bridge: Intel Corporation 5 Series/3400 Series Chipset PCI Express Root Port 1 (rev 05)
00:1c.3 PCI bridge: Intel Corporation 5 Series/3400 Series Chipset PCI Express Root Port 4 (rev 05)
00:1c.5 PCI bridge: Intel Corporation 5 Series/3400 Series Chipset PCI Express Root Port 6 (rev 05)
00:1d.0 USB Controller: Intel Corporation 5 Series/3400 Series Chipset USB2 Enhanced Host Controller (rev 05)
00:1e.0 PCI bridge: Intel Corporation 82801 Mobile PCI Bridge (rev a5)
00:1f.0 ISA bridge: Intel Corporation Mobile 5 Series Chipset LPC Interface Controller (rev 05)
00:1f.2 SATA controller: Intel Corporation 5 Series/3400 Series Chipset 6 port SATA AHCI Controller (rev 05)
00:1f.6 Signal processing controller: Intel Corporation 5 Series/3400 Series Chipset Thermal Subsystem (rev 05)
02:00.0 Network controller: Broadcom Corporation Device 4727 (rev 01)
03:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller (rev 03)
ff:00.0 Host bridge: Intel Corporation Core Processor QuickPath Architecture Generic Non-core Registers (rev 02)
ff:00.1 Host bridge: Intel Corporation Core Processor QuickPath Architecture System Address Decoder (rev 02)
ff:02.0 Host bridge: Intel Corporation Core Processor QPI Link 0 (rev 02)
ff:02.1 Host bridge: Intel Corporation Core Processor QPI Physical 0 (rev 02)
ff:02.2 Host bridge: Intel Corporation Core Processor Reserved (rev 02)
ff:02.3 Host bridge: Intel Corporation Core Processor Reserved (rev 02)
Comment by Jeremy LaCroix (jlacroix) - Wednesday, 01 December 2010, 21:11 GMT
What are the chances of this patch making it in the actual kernel? Other than not being able to access tty's the patch works wonders, and makes an unusable system usable.
Comment by Christian Neukirchen (chneukirchen) - Wednesday, 01 December 2010, 22:31 GMT
On above HP ProBook, I can access the TTYs with kernel26-intelfix as usual (as with 2.6.35).

That seems to be an unrelated regression.
Comment by mlmeola (mlmeola) - Thursday, 02 December 2010, 07:20 GMT
I have a seemingly similar problem (blank screen on lcd for intel graphics when kms kicks in on 2.6.36-line kernels), but 2.6.36.1 intelfix doesn't fix it.

I have a Sony Vaio Z series laptop (vpcz1) with switchable Intel Core i5 graphics/nvidia graphics set to work with static graphics switching. Booting with only the intel card enabled under 2.6.35.8 works fine. Under 2.6.36.1 from arch (or 2.6.36.1 from AUR's kernel-intelfix, and at least as far back as 2.6.36-rc1 from kernel.org) the embedded LCD screen goes black (no backlight) as soon as KMS starts. It doesn't change whether or not I have X enabled, the screen just stays black.
Comment by Peinthor Rene (rp-) - Thursday, 02 December 2010, 07:36 GMT
I can't really help you fix these problem as I can't reproduce this on my hardware, but you could also try to bisect the problem to specific commit. It is a bit of pain and takes time, but try to build 2.6.36-rc1, because between rc1 and rc2 there where some commits which might possible the reason.
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=shortlog;h=refs/tags/v2.6.36-rc2
Here the drm/i915 commits are the hot ones. if rc1 doesn't work for you, you would need to check even more between stable 2.6.35 and 36-rc1.
Comment by Jeremy LaCroix (jlacroix) - Tuesday, 07 December 2010, 16:40 GMT
Quick question. I am still unable to access tty's in the patched 2.6.36 kernel. I know that it's perceived that it is a separate issue, however I recently downgraded the Intel driver to the previous version and I still cannot access tty's. Is it possible it's related to this issue? Or should I file a separate bug report for it?
Comment by Jeremy LaCroix (jlacroix) - Saturday, 11 December 2010, 14:57 GMT
I tried 2.6.36.2 today, and the problem is still not solved.
Comment by Kimmo (Dehir) - Saturday, 18 December 2010, 22:29 GMT
Having similar type of problem with intel 855GME. The screen wents sometimes total blank and wont change to other terminals. It falls back to 1024x768 if i can get picture. Also finds only 1024x768 resolution with vesa driver which i kinda of odd. Not related to xf86-video-intel ?
Comment by Jeremy LaCroix (jlacroix) - Friday, 07 January 2011, 18:16 GMT
It looks like this bug was fixed upstream:
https://bugs.freedesktop.org/show_bug.cgi?id=29278

I know this is asking a lot, but can this fix be implemented into the Arch kernel? If not, is it possible for an AUR package to be made consistent with the patches found in that bug report? I would do it myself but I am not currently at that skill level.
Comment by maxexcloo (maxexcloo) - Wednesday, 02 February 2011, 06:01 GMT
I still have the issue with the latest kernel and intel packages. Anything I can do? I currently run an old kernel because I am forced to...
Comment by Peinthor Rene (rp-) - Wednesday, 02 February 2011, 07:30 GMT
With 2.6.37 I have a different issue. I can see a picture but have heavy flicker. I already reported this upstream and it was fixed in the drm-intel-staging branch, which I'm currently using (with 1 revert because of another black screen bug....)

Your only chance is to report the bug directly to the intel developers and pray that they will backport a fix for the 2.6.37 kernel.

https://bugs.freedesktop.org./

Loading...