8086:2a42 [Intel GM45] Irregular sync flashes (Needs i915.powersave quirk)

I am experiencing the exact same problem on my Lenovo Thinkpad W500. My video card is the ATI Mobility Radeon HD 3650 hardware, however. I am also getting white (frequent but not constant) horizontal lines appearing and disappearing randomly all over the display.

I'm guessing this could be a Linux Kernel (KMS) problem, or an X-org problem, or a video driver problem. I will hold off on attaching a massive amount of log data and hardware info someone with a little more knowledge in these areas can tell me what would be most useful. :)

I can provide a video of this problem if it would help troubleshoot.

Be aware, this *might* be a dupe of
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/538344

538344 suggests booting with "nomodeset".

Adding nomodeset to the kernel parameters seems to have fixed this for me.

Probably best not to mark this as a dupe of 538344 just yet since Russel reports he is seeing it with an Intel card...?

538344 does mention that kernel 2.6.34 pre rc1 seems to fix the problem. Maybe Russel could test with that kernel to see if the problem goes away?

I've just marked bug #534894 as a dupe of this one, since that too is filed against the intel package and describes infrequent flickering that is comprised of extremely brief corruption of part of the screen. I think we're also all running fairly recent Intel chips.
To quickly re-mention a comment I added to the dupe bug:

[09:18] < Sarvatt> tell them to boot with i915.powersave=0 and mention in the bug if that fixes it :)
[09:18] < Sarvatt> that was default in -15 but not anymore in -16

I have this same problem on my Lenovo Ideapad U350

I'm experiencing this issue on my Lenovo Thinkpad R400.

Russel, can you please run the command apport-collect 538648? it will automatically attach all of the logs we need to look into this to the bug.

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

Thank you for that. Now have you tried booting with i915.powersave=0 added to the kernel command line in grub (after it says quiet splash) to see if it fixes your issue or not? That will help narrowing down the problem.

I have also been having this issue on a Lenovo S10. I have attempted to resolve by adding in the requested options into the kernel boot however it still persists. Below is a link to a video if the issue. The screen will flicker, blackout, and tear. the video below shows the tearing.

http://www.youtube.com/download_my_video?v=DjgnH9oedY8

fwiw I've not noticed any more of these flashes for a few days, although I hadn't set the powersave option.

The flashes have been greatly reduced if not eliminated yes.

Anecdotally, there does seem to be less of this flickering now, but I still get it occassionally. I'll see if I can get some actual data and report back.

I am confirming I am seeing flickering again as well, sorry for the false alarm

Looks like I spoke too soon, it's still happening with all of today's updates and 2.6.32-18-generic

This appears to have gotten worse for me with today's updates and -18. I used to only see it a few times a week but now it seems to be happening a "few" times a day. Should we try the i915.powersave=0 thing?

Sorry for the anecdotal evidence but I have no idea how to troubleshoot or reproduce this.

Does booting to the earlier kernel make it occur less frequently?

Perhaps this is a kernel drm bug.

Bryce, frequency is not particularly easy to measure. Over the course of running the *same* kernel, I have experienced this happening every 2 or 3 minutes for an hour to not happening for a whole day.

Jason, since you are on -radeon rather than -intel, I'm thinking you should probably file a separate bug report. Often these kinds of issues are video-driver specific, and many times particularly narrow to specific models of hardware, so as a general rule we limit the scope of X.org bug reports to the original reporter's system and those people who have exactly the same model of hardware.

Bryce, I am intel, I dont know why you think I am radeon?

Confirmed, passing i915.powersave=0 fixes the problem on my Thinkpad x200s - though it sounds like it'll cause more drain on the battery.

Jason, aha there was another Jason reporting earlier with a radeon. I scanned up to look for the post where you summarized your hardware and driver and since you apparently never provided it (tsk tsk bad bug etiquette) my eye caught the earlier Jason's post.

Yes: could people who experience this problem please try passing i915.powersave=0 to the kernel command line. We've got one report of success with an x200s, but it would be useful if everyone with a piece of affected hardware could try it.

Passing this kernel option fixed the problem on Lenovo G550 with Intel video card.

Sarvatt requested a dmesg from me with drm.debug=0x04 on the kernel commandline. Attaching.

Also attaching a more verbose drm debugging output

And the output of intel_reg_dumper

I just had a glitch before rebooting to clear the drm debugging. I grabbed dmesg as quickly as I could

and intel_gpu_dumper output just in case it's useful.

that last dmesg dump was too brief to contain anything useful. Instead here's my /var/log/syslog with drm_ioctl grepped out (because with those log entries the log is >1GB.)

My Xorg logfile for that boot.

I've forwarded this upstream with Chris Jones' info.

https://bugs.freedesktop.org/show_bug.cgi?id=27589

I can confirm this happens also on my Thinkpad T500 (2082-7SG) using the integrated graphic card (Intel 4500MHD), I have not tested with the discrete graphic card (Radeon Mobilty 3650HD). I'm using Lucid x64 with all the latest updates.
Should I add some informations about my configuration on the upstream bug report ? Or is not necessary ?

Comfirm also on Asus UL20A with Intel chipset using Lucid beta 2.

The Acer Extensa 5630Z is also affected by this bug. The i915.powersave=0 workaround solves the problem.

 VGA compatible controller [0300]: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller [8086:2a42] (rev 07)
 Subsystem: Acer Incorporated [ALI] Device [1025:013c]

I'm having random flashes as well on an ASUS UL30A-X5 (intel 4500MHD). And I can confirm it's not limited to Compiz. I get the same flicker with desktop effects disabled.

Same problem here and with intel GMA4500HD.

I'll try the workaround.

same here - with or without compiz, is happening often when I use openoffice

I've been running Lucid alpha for a while now, and I'm 99% certain that this only started happening on my Thinkpad T500 since kernel package 2.6.32-16.

I also affected by this bug with latest Lucid updates. I have Ubuntu Desktop 64bit on a Lenovo Thinkpad T500.
However, I did follow Chris Jones advice (Comment #5) and it is working. I did not have this sync flashes today during the whole time.

This affects me on a Dell Inspiron 1545 with an Intel GM45 Chipset running Xubuntu 10.04. THe problem persists running either Compiz or Xfwm4.

+1 confirmed on Dell inspiron 1545 with Intel GM45 on gnome. I had no problems using karmic. the problems started after upgrade.

Like PK, also seeing this flickering with Dell Inspiron 1545 after Karmic upgrade.

Sorry, meant to say upgrade from Karmic to Lucid ...

+1 confirmed this just like the others on a Intel GM45 (Dell Inspiron 1545) but it seems to be independent on usage of compiz

The problem is still present with the latest kernel update: 2.6.32-22
Albeit, it seems less frequent, but still too much frequent.

confirmed on a dell vostro 1220 under latest kernel.

+1 confirmed on Dell Inspiron 1545 with Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07) on gnome. No problems using karmic. Random flickering on the bottom quarter of screen started after upgrade.

I can confirm the bug on my Lucid installation on my Sony Vaio Z21MN with the Intel graphics card (more details below). The workaround (i915.powersave=0) seems to work (after 2 days of use). I hope in a real fix.

00:01.0 PCI bridge: Intel Corporation Mobile 4 Series Chipset PCI Express Graphics Port (rev ff)
00:02.0 VGA compatible controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07)

I am also affected by the flashes on the screen on my Dell Latitude Z600 since I upgraded to Lucid 64bit. Here are the details of my video card :

00:02.0 VGA compatible controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07)
00:02.1 Display controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07)

00:02.0 0300: 8086:2a42 (rev 07)
                       Subsystem: 1028:02c8
00:02.1 0380: 8086:2a43 (rev 07)
                       Subsystem: 1028:02c8

The workaround i915.powersave=0 fixes the issue.

The workaround i915.powersave=0 fixes the issue for me too.

Hello, here is the first nvidia user with the same problem. I have an nvidia quadro fx 1700 with enabled adaptive clocking. Every time the device switches clock frequency there is a flickering on the screen. You can reproduce it using nvidia-settings -> powermizer -> Powermizer Settings and change preferred mode.

I don't think it is a driver issue, because the same driver with the same settings worked in 9.10 without problems.

I am testing the workaround as well.

Maybe this could be relevant: I have *never* experienced this problem when the laptop has an external screen attached. It does not matter if I only use the external screen or both screens.

This seems to be a general problem with FBC on 8086:2a42 devices, has anyone experiencing this found a kernel that did not have this problem? Can anyone try the maverick kernel, or the most recent mainline kernel to see if the problems are still there before we look into globally disabling FBC or powersave options on this specific gpu for lucid?

Kernel 2.6.34-999-generic #201005121008 SMP Wed May 12 09:10:15 UTC 2010 x86_64 still shows the bug on:
Acer Extensa 5630Z
 VGA compatible controller [0300]: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller [8086:2a42] (rev 07)
 Subsystem: Acer Incorporated [ALI] Device [1025:013c]

I have a Dell Inspiron 1545 with the same video card (same lspci output) as Alexis de Lattre (#74) and the workaround has kept me flicker-free all morning.

Has everyone still posting that they have this problem tried the workaround? (See the bug description.)

Robert, I have been using the latest maverick kernel for about 24 hours now:

$ uname -r
2.6.34-2-generic

I have an Inspiron 1545 and the Intel GM45:

$ lspci | grep -i graphics
00:02.0 VGA compatible controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07)
00:02.1 Display controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07)

I was previously using the i915.powersave workaround. To test the newer kernel, I installed it, and disabled the i915.powersave workaround. To ensure that it was disabled, I rebooted into the lucid kernel again, and the flickering returned--about every few minutes, like before.

Now I've been booted into the latest maverick kernel for about 24 hours now without seeing any flickering. Given how frequently I saw the flickering before, I think that it is likely that the maverick kernel has resolved the issue. However, given the flickering's nondeterministic nature, I will continue using the maverick kernel and post back if the flickering returns.

Let me know if there is anything else I can do, short of using git-bisect, to help you track this down.

I ran Lucid alpha/daily installs since about January IIRC, and am 99% sure that the flicker problem first occurred with kernel image 2.6.32-16 (http://ubuntuforums.org/showpost.php?p=8942276&postcount=9). It's possible that the problem has been fixed upstream in 2.6.34 (or .33 even), but if you're looking to find the culprit, there will be far fewer differences between the two Ubuntu packages of 2.6.32-15 and 2.6.32-16, than a complete different kernel version.

According to comment 5, we already know why the -15 kernel had no flicker issues: it had i915.powersave=0 by default.

The -16 kernel backported the entire drm from 2.6.33 back to the 2.6.32 kernel, so your statement about it being less isn't true unfortunately

@Jeff250, @Robert Hooker: well that certainly explains the mystery.... for me at least anyway. I never realized that i915.powersave=0 had been default in -15 (and should have read the other 81 comments before mine)....

Can't see any flickering on a current ubuntu 10.10 (maverick). Which surprises me since I saw flickering with kernel 2.6.34-999 on lucid (see comment 79).
(All this with [8086:2a42].)

Ok since this is for sure a kernel problem I've opened a new task for it. Thank you for the confirmation that the problem is fixed in 2.6.34-rc7. It would be especially helpful if anyone could narrow down which 2.6.34-rc the problem was fixed by installing different mainline kernels until you find one where it is fixed. I would start at 2.6.34-rc1 and work your way up. Andy Whitcroft is building a test kernel that just disables FBC on this device as well, but if we could narrow down where it was fixed it would be a great help. Also, please do make sure that you are not using powersave=0 when you test the .34 kernels before saying it is fixed.

I have built some test kernels with FBC disabled for the GM45 hardware reported here. If those of you affected by this flickering issue could test these kernels and report back here that would be great. Please remember to test without the i915.powersave=0 override else the test is meaningless. The kernels are uploading now and will be at the URL below shortly:

    http://people.canonical.com/~apw/lp538648-lucid/

Thanks!

invalidating the xserver-xorg-video-intel target, it was just on here for easier tracking.

Andy, no flicker seen in the first 35 minutes of using your FBC disabled kernel on lucid.
dmesg |grep "Command line"
[ 0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-2.6.32-22-generic root=/dev/sda7
dmesg |grep "Linux version"
[ 0.000000] Linux version 2.6.32-22-generic (root@chloe) (gcc version 4.4.3 (Ubuntu 4.4.3-3ubuntu1) ) #34~lp538648v201005171834 SMP Mon May 17 17:55:25 UTC 2010 (Ubuntu 2.6.32-22.34~lp538648v201005171834-generic 2.6.32.11+drm33.2)

Robert, I do not quite follow your your comment #87 about flicker being fixed in 2.6.34-rc7 and trying the different 2.6.34-rc's. From my comment #79 the daily mainline kernel from May 12th compiled for lucid tried with lucid flickers. And that daily kernel is newer then rc7. And I just tried the rc7-lucid itself in lucid and it flickers. Jeff250's comment #81 sounds like lucid is flicker free with the maverick kernel (compiled for maverick)? And maverick (ubuntu 10.10) is also flicker free (my comment #86). So does that not mean it is how the kernel is compiled that makes the flicker go away?

Hello,

I also have this weird bug in my Sony Vaio VGN-FJ1S with an Intel 915 VGA. Strangely this only started yesterday I guess (!)

Well more important: I found a workaround for this: "Add to Panel..." > "CPU Frequency Scaling Monitor" and then I set it to work at 1.33 GHz (the max. for me it's 1.73 GHz) and it's ok, no more screen dim/flickering !

Nevertheless I hope this will be fixed soon.

Cheers,
JC

@JC -- could you test the kernels in comment #88 and see if those fix the issue for you. Do ensure you do not have the frequency fixed when testing. Please report back here.

@wolfgang -- not necessarily as the kernel has a lot of different patches applied in maverick which are not in -rc7. Some of those affect the specific option we are testing on here on some hardware.

@wolfgang -- indeed could you also test the kernels in comment #88 and report back.

@Andy

The kernel from comment #88 looks flicker free to me now for over 5 hours. I have already been testing it as indicated in my comment #90.
details to make sure:
uptime; dmesg |grep "Command line"; dmesg |grep "Linux version"
 16:01:03 up 5:20, 4 users, load average: 0.42, 0.97, 0.97
[ 0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-2.6.32-22-generic root=/dev/sda7
[ 0.000000] Linux version 2.6.32-22-generic (root@chloe) (gcc version 4.4.3 (Ubuntu 4.4.3-3ubuntu1) ) #34~lp538648v201005171834 SMP Mon May 17 17:55:25 UTC 2010 (Ubuntu 2.6.32-22.34~lp538648v201005171834-generic 2.6.32.11+drm33.2)

About your comment #94: Is there a solution in those maverick patches to rc7 or do they just work around the issue?

Andy, I have been using the kernel in comment 88 for about 6 hours now without flickering.

FBC patch works like a charm, thx to Andy Whitcroft for the compiled kernel!

@ Andy Whitcroft

I hadn't the time to try that kernel of yours but latter maybe this weekend I'll try then.

I have another *maybe interesting* fact to report: I installed KDE in this Ubuntu 10.04 and since then I have no screen glitches/flickering! Nothing at all. Even in Gnome. Maybe this problem/bug is related to the gnome-power-manager or something?

Notice I'm running '2.6.32-22-generic' kernel and this Ubuntu 10.04 was a fresh install. I have an Intel 915GM Chipset in this Sony laptop and X.Org X Server 1.7.6.

JC: 915GM wasn't affected by this, and we disabled FBC on 915GM/945GM a few months ago already.

I would beg to differ here, I have a compaq NC6220 with the intel 915GM and it flickers like crazy when under any CPU load. Tried the modeset

Fresh install here as well,

Tried the i915.powersave=0 thing, no joy

X.Org X Server 1.7.6

root@nc6220:/etc# cat lsb-release
DISTRIB_ID=Ubuntu
DISTRIB_RELEASE=10.04
DISTRIB_CODENAME=lucid
DISTRIB_DESCRIPTION="Ubuntu 10.04 LTS"
root@nc6220:/etc# uname -r
2.6.32-22-generic
root@nc6220:/etc# lspci -s 00:02.0
00:02.0 VGA compatible controller: Intel Corporation Mobile 915GM/GMS/910GML Express Graphics Controller (rev 03)

@Brian Threlkeld:
Are you sure that you are effectivelly using the suggested parameter? For example you could forgot the 'update-grub' command.

Check the effective use with: dmesg | grep 'Kernel command'

Yep, pretty familiar with how to do that, currently driving a project at my company where we are deploying 100's of ubuntu desktops...

But to answer your question Mario,

root@nc6220:/# dmesg | grep 'Kernel command'
[ 0.000000] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-2.6.32-22-generic root=UUID=56efd8c3-56e5-420b-b6be-14d7919778a6 ro quiet splash i915.powersave=0
root@nc6220:/#

Look ok?

Strange thing is (and this is a home laptop, not part of my work project) it runs fine after reboot, then after some time the flickering starts to come back when the CPU gets loaded up. I was also mainly addressing Robert Hooker's comment:

"JC: 915GM wasn't affected by this, and we disabled FBC on 915GM/945GM a few months ago already."

When in fact I have the subject card and current kernel and it isnt fixed.

Thanks

Brian

Andy no flicker Dell Inspiron 1545 laptop
I added then deleted the i915.powersave=0
from line
GRUB_CMDLINE_LINUX "i915.powersave=0 "
and loaded the new 64bit patched kernal located in your link and
no flicker. (the i915.powersave=0 fix worked for me) It has only been a day but i think its a winner it works.(little too early to tell though100 percent)
I don't know what a FBC is, buts thats fine by me lol.
P.S.
I started a blog a while back still going, same Issue http://ubuntuforums.org/showthread.php?t=1473580
I also posted a bug report so there is a duplicate out there.
Thanks

Update
Getting a flicker again not bad, one or two a day, so it was hard to know if this kernal thing
works or not, well I guess it doesn't for me.
tnx

I get this on my Dell Inspiron 1750 with Intel Graphics card.

I have the exact same symptoms as comment #72 on an HP dv4 laptop
>> Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07) >> on gnome. No problems using
 >> karmic. Random flickering on the bottom quarter of screen started after upgrade.

Two things I've observed:

- The flickering NEVER happens (on either screen) if I have my external screen plugged in.

- It seems to get much worse with the terminal open and on top.

the i915.powersave=0 thing didn't work for me.

Same problem in Acer Aspire 5738Z with Intel Graphics Media Accelerator. Does this problem happen in 32 bit pae enabled ubuntu? the option i915.powersave=0 is not an option for me due to battery life. Thanks.

Hi,

I tried installing Andy's kernels but the results are not very good. While I get about 2-3 flickers/day normally, the rate gets higher when I try to run a heavy application. If at the same time, I do the powersave thing, the flickers seem to go away.

Can anybody tell me if I am going to run into some loss in terms of hardware (like battery and CPU life time) if I do the powersave thing ? I don't mind using this workaround for the time being, if it doesnt cost me anything !!

I love my computer more than I hate the flickers though!!

Let me know if you need to have any information about my system.

Seeing flicker on maverick with ubuntu's mainline 2.6.35rc1 kernel. No flicker with stock maverick.

The kernels in #88 have worked perfectly for me, for about a week now. Thanks!

$ lspci -s 00:02.0
00:02.0 VGA compatible controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07)

+1 confirmed here. Dell Inspiron 1545, 00:02.0 VGA compatible controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07)
00:02.1 Display controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07).
Occurs in 10.04. Did not occur in 9.04. I am not interested in using either workaround. My thanks to those whom continue to research this.

+1 confirmed.

Asus UL80
00:02.0 VGA compatible controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07)
00:02.1 Display controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07)

Installed http://kernel.ubuntu.com/~kernel-ppa/mainline/v2.6.34-lucid/ for touchpad to work on multitap (horizontal scroll) and FN+F9. I didn't try the grub workaround, I feel there is a drawback to it (maybe its just me).

Screen flicker shows up once or twice when fresh from boot, but after that it won't happen for some hours.

Loading the kernel with i915.powersave=0 stops the flickering in my case. Next week I will try the new kernel in post #88.
I have an Intel 4500MHD.

+1 confirmed.

I'm on a Lenovo G550 and adding the i915.powersave=0 appears to have stopped the flickering.

Andy Whitcroft's test kernel works just fine on Lenovo T500 with Intel Corporation Mobile 4 Series.
thx for that!!! cause with i915.powersave=0 sleep/suspend does not work.

+1 for me as well... can we expect the fixes with Andy's test Kernel to be included in upcoming updates?

I have a 8086:2a42 video card and successfully used the i915.powersave=0 option to stop the sync flashes.

I installed Andy Whitcroft's kernel and removed the i915.powersave=0 option yesterday. Everything continue to work fine ; no sync flashes at all.

@Peter : it seems that FBC means FrameBuffer Compression...

What about including the #88 patch in the official kernel of Lucid (or in the proposed repository)? It should not affect the power consuming of the card.

Neither installing Andy Whitcroft's test kernel nor the option i915.powersave=0 helped with me (Lenovo Think Pad T400s). To get rid of the annoying flickering, I re-installed Karmic Koala. I hope the bug will be solved at least for the next distribution.

same here. neither of the workarounds work for me perfectly. kinda annoying
actually.

On Sat, Jul 3, 2010 at 3:05 PM, Martin Schraner <email address hidden>wrote:

> Neither installing Andy Whitcroft's test kernel nor the option
> i915.powersave=0 helped with me (Lenovo Think Pad T400s). To get rid of
> the annoying flickering, I re-installed Karmic Koala. I hope the bug
> will be solved at least for the next distribution.
>
> --
> [Intel GM45] Irregular sync flashes on 8086:2a42 (Needs i915.powersave
> quirk)
> https://bugs.launchpad.net/bugs/538648
> You received this bug notification because you are a direct subscriber
> of the bug.
>
> Status in X.org xf86-video-intel: Unknown
> Status in “linux” package in Ubuntu: Triaged
> Status in “xserver-xorg-video-intel” package in Ubuntu: Invalid
>
> Bug description:
> [Workaround 1]
> Install Andy Whitcroft's test kernel from
> http://people.canonical.com/~apw/lp538648-lucid/ (see comment #88).
>
> [Workaround 2]
> As workaround add the option i915.powersave=0 to the kernel command line in
> the grub config (for GRUB 2: edit /etc/default/grub and add
> i915.powersave=0 to GRUB_CMDLINE_LINUX, then run sudo update-grub; for GRUB
> 1: edit /boot/grub/menu.lst and add i915.powersave=0 to the end of the line
> beginning with # kopt=, then run sudo update-grub).
>
> [Report]
> Binary package hint: xserver-xorg
>
> When I first installed Lucid on my T500 I believe everything seemd to be
> fine with the display. Recently though the screen has started giving
> flashes -- as though there was some form of sync problem for the display.
> The flashes are of irregular frequency and occurrence. I haven't yet seen
> any pattern to the behaviour. They are however exceptionally annoying.
>
> I believe I am using the Intel chipset graphics device -- I am not
> actually aware of how to switch to using the alternate graphics hardware,
> but I am told it is not worth bothering with, so I haven't.
>
> I am using the Compiz window manager so I guess the problem could be there
> rather than with Xorg.
>
> There is no xorg.conf file so there is no overide of the default behaviour.
>
> I am not sure what programs to run to give useful output for this problem.
> I have attached the current Xorg log file.
> ---
> Architecture: amd64
> DistroRelease: Ubuntu 10.04
> DkmsStatus:
> virtualbox-ose, 3.1.4, 2.6.32-17-generic, x86_64: installed
> virtualbox-ose, 3.1.4, 2.6.32-16-generic, x86_64: installed
> virtualbox-ose-guest, 3.1.4, 2.6.32-17-generic, x86_64: installed
> virtualbox-ose-guest, 3.1.4, 2.6.32-16-generic, x86_64: installed
> InstallationMedia: Ubuntu 10.04 "Lucid Lynx" - Alpha amd64 (20100224.1)
> MachineType: LENOVO 20827SG
> Package: xserver-xorg-video-intel 2:2.9.1-1ubuntu16
> PackageArchitecture: amd64
> PccardctlIdent:
> Socket 0:
> no product info available
> PccardctlStatus:
> Socket 0:
> no card
> ProcCmdLine: BOOT_IMAGE=/boot/vmlinuz-2.6.32-17-generic
> root=UUID=d075e3b9-f26d-4896-9983-4afe962b115a ro quiet splash
> ProcVersionSignature: Ubuntu 2.6.32-17.26-generic 2.6.32.10+drm33.1
> Tags: lucid lucid
> Uname: Linux 2.6.32-17-generic x86_64
> UserGroups: adm admin cdrom dialout lpadmin ...

I have this same problem on my Lenovo G450.

It may be worth noting that the effect is far easier to see in lower surrounding light levels: in darker room light.

 i915.powersave=0 fixes it on my Lenovo Edge, batterie time seems to get down a lot is this possible?

powersave=0 works for me too on my T500. Regarding the battery drain: this is totally screwed up because of https://bugs.launchpad.net/ubuntu/+source/linux/+bug/524281 anyhow for me. I get good battery with the modified kernel from there.

This bug is absolutely annoying and implicitly also a security issue since I'm using apw's kernel mentioned as a workaround, so I'm not able to fetch security updates to the kernel. (workaround no. 2 is not an option since battery life is important to me.)

Is there a good reason for not including the patches from apw's kernel into the official ubuntu kernels to make the flicker disappear?

@Max good question, interestet in this too, nobody is stating if or when this will be included...

With the current updates, it seems to occur in pairs, twice in quick succession now.

I just switched to 32 bit ubuntu with pae kernel and the problem persists. I still don't know if the problem is from ubuntu itself or from this repository: https://launchpad.net/~ubuntu-x-swat/+archive/x-updates

Can anyone confirm this problem happening without this repository?

Thanks.

I experiense the same problem on my Asus X5dij laptop, which has also an integreted Intel 4 Series Express Video Card(Intel GMA 4500M) and Intel GL40 Chipset.
My whole screen flickers. It is really unpleasent for my eyes. I can see the horizontal refresh lines. So I've tried to create a xorg.conf and to add a Modeline, as I believed that the problem is due to wrong Horizontal sync value. But I had no success and the problem still exists.
I have also tried the kernel from the Comment #88 and to use i915.powersave=0, but neither solution was successful.
I hope that someone will find a solution to this problem.
Thanks.

I have HP2230s and i915.powersave=0 does not work for me.

I can confirm that this does NOT happen when you 2 displays working simultaneously, e.g. external display attached to HDMI working as primary display and laptop internal display working as secundary (extended) display.

However as soon as you de-attach external display, internal laptop display starts to flickers on random basis. What seems to make flicker more intensive is if you open terminal in full screen (black background color).

Adding i915.powersave=0, solves the issue on my laptop: HP, Intel GMA 4500M, Intel GL40.

Since the last kernel update (version 2.6.32-24.42), I haven't seen any flickers. I DON'T use i915.powersave=0.

I have a Dell Latitude Z600 laptop.
PCI ID : 8086:2a42 (rev 07)
Subsystem: 1028:02c8

Forget what I said in my previous comment ; I just saw my screen flickering :-) But I should say that it's much less frequent than before.

Upgraded to Maverick beta and haven't seen any flickers since then.

Weird, I had removed the i915.powersave=0 and for a long time now (maybe a month) I had almost no flickering (could count them on one hand).
However, today is hell. It is flickering every 3-4 minutes or so...
I've check back the history of recently installed software, and I only have Chromium and FF nightly + tzdata + gwibber updated, so all unrelated to the Video system...
Do you know when a fixed version of the video driver will be available for Lucid?

PS: while waiting, I will add again the kernel parameter workaround.

On Dell Latitude E4310 amd64 I have the same problem too
powersave=0 workaround does not work!
kernels 2.6.32.21 or 2.6.32 22 do not have the problem !
kernel 2.6.32.23, 2.6.32.24, 2.6.32.25 have the flicker problem !
so some update from 2.6.32 22 to 2.6.32.23 is probably buggy

Any chance of getting this fix backported to Lucid? We have a fair number of systems rolled out with this hardware, and trying to explain the grub workaround to ordinary users is painful...

AfC

I didn't investigate the exact state of the packages, but I observed an all-Intel Vaio running Maverick yesterday and it was showing the same flicker I saw in Lucid.

I upgraded to Maverick 5 days ago, and I haven't seen any sync flash since then. So I consider that the bug is fixed in Maverick. I have a 8086:2a42 video card on a Dell Latitude Z600.

Can someone summarize the actual real fix (not any workarounds) and/or highlight the changeset to whichever Debian/Ubuntu package to <email address hidden> so that Debian Squeeze can have the problem resolved as well?

@Russel,

You will find an up-to-date summary of the situation at the very beginning of the bug report (your bug report ! :-), in the 3 first paragraphs of the bug description.

Alexis,

On Mon, 2010-11-01 at 21:55 +0000, Alexis de Lattre wrote:
> @Russel,
>
> You will find an up-to-date summary of the situation at the very
> beginning of the bug report (your bug report ! :-), in the 3 first
> paragraphs of the bug description.

I think the problem is that when Julien looked at this bug report, it
wasn't clear which kernel version is the one that provides the fix so
that the Debian folk can investigate whether they can issue that kernel
for Squeeze.

> ** Description changed:
>
> [Status]
> Fixed in maverick
> Not fixed in lucid (bug still seen on 2010-09-22 with kernel 2.6.32-25)

Debian Squeeze is still running the 2.6.32-5-amd64 kernel, Maverick is
running 2.6.35-22. Does anyone know which changeset to which kernel
series actually cures the problem? Do we even know whether this is a
2.6.32, 2.6.33, 2.6.34 or 2.6.35 issue?

> [Workaround]
> As workaround add the option i915.powersave=0 to the kernel command line in the grub config (for GRUB 2: edit /etc/default/grub and add i915.powersave=0 to GRUB_CMDLINE_LINUX, then run sudo update-grub; for GRUB 1: edit /boot/grub/menu.lst and add i915.powersave=0 to the end of the line beginning with # kopt=, then run sudo update-grub).
>
> [Alternative workaround for experts]
> - Compile your kernel with Andy Whitcroft's patch from http://people.canonical.com/~apw/lp538648-lucid/ (see comment #88).
> + Compile your kernel with Andy Whitcroft's patch from http://people.canonical.com/~apw/lp538648-lucid/ (see comment #92).

I am not sure this is an option for the Debian folk, they need to run a
kernel from the official series.

Thanks.
--
Russel.
=============================================================================
Dr Russel Winder t: +44 20 7585 2200 voip: sip:<email address hidden>
41 Buckmaster Road m: +44 7770 465 077 xmpp: <email address hidden>
London SW11 1EN, UK w: www.russel.org.uk skype: russel_winder

Does anyone know when those comment numbers shifted? What once was comment #88 (Andy Whitcroft's patch) is now #92, breaking referrals. I filed bug 669722 about this.

Russel Winder, this bug was reported a while ago and there hasn't been any activity in it recently. We were wondering if this is still an issue? If so, could you please test for this with the latest development release of Ubuntu? ISO images are available from http://cdimage.ubuntu.com/daily-live/current/ .

If it remains an issue, could you please run the following command in the development release from a Terminal (Applications->Accessories->Terminal), as it will automatically gather and attach updated debug information to this report:

apport-collect -p linux <replace-with-bug-number>

Also, could you please test the latest upstream kernel available following https://wiki.ubuntu.com/KernelMainlineBuilds ? It will allow additional upstream developers to examine the issue. Please do not test the daily kernel folder, but the one all the way at the bottom. Once you've tested the upstream kernel, please comment on which kernel version specifically you tested. If this bug is fixed in the mainline kernel, please add the following tags:
kernel-fixed-upstream
kernel-fixed-upstream-VERSION-NUMBER

where VERSION-NUMBER is the version number of the kernel you tested. For example:
kernel-fixed-upstream-v3.12-rc2

This can be done by clicking on the yellow circle with a black pencil icon next to the word Tags located at the bottom of the bug description. As well, please remove the tag:
needs-upstream-testing

If the mainline kernel does not fix this bug, please add the following tags:
kernel-bug-exists-upstream
kernel-bug-exists-upstream-VERSION-NUMBER

As well, please remove the tag:
needs-upstream-testing

Once testing of the upstream kernel is complete, please mark this bug's Status as Confirmed. Please let us know your results. Thank you for your understanding.

This version is now outdated and no more supported