System Stutter When HDMI/Displayport Monitor Is Connected

[tdeering]

With a fresh install of Ubuntu 11.10 x64, the system has many "microfreezes" when I connect an external monitor via HDMI. These last maybe 10ms and occur about once per second. As I move my cursor around the screen, it stops motion during these microfreezes, causing a very irritating stutter effect. This only happens when using HDMI- when connected via VGA to an external monitor, the system behaves smoothly as expected. Not reproducible in Raring.

WORKAROUND: In Quantal use kernel boot parameter drm_kms_helper.poll=0.

ProblemType: Bug
DistroRelease: Ubuntu 11.10
Package: xorg 1:7.6+7ubuntu7.1
Uname: Linux 3.3.0-030300rc7-generic x86_64
.tmp.unity.support.test.0:

ApportVersion: 1.23-0ubuntu4
Architecture: amd64
CompizPlugins: [core,bailer,detection,composite,opengl,compiztoolbox,decor,grid,move,vpswitch,place,mousepoll,resize,wall,regex,session,snap,gnomecompat,imgpng,animation,expo,workarounds,unitymtgrabhandles,ezoom,fade,scale,unityshell]
CompositorRunning: compiz
Date: Sat Mar 24 18:30:09 2012
DistUpgraded: Fresh install
DistroCodename: oneiric
DistroVariant: ubuntu
ExtraDebuggingInterest: Yes, if not too technical
GraphicsCard:
 Intel Corporation 2nd Generation Core Processor Family Integrated Graphics Controller [8086:0116] (rev 09) (prog-if 00 [VGA controller])
   Subsystem: Device [1b0a:2080]
InstallationMedia: Ubuntu 11.10 "Oneiric Ocelot" - Release amd64 (20111012)
MachineType: PEGATRON CORPORATION A15
ProcEnviron:
 PATH=(cususername, no user)
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-3.3.0-030300rc7-generic root=UUID=a19e39dd-b510-4f9d-93f5-de6d40df10f1 ro quiet splash vt.handoff=7
SourcePackage: xorg
Symptom: display
Title: Xorg freeze
UpgradeStatus: No upgrade log present (probably fresh install)
dmi.bios.date: 06/17/2011
dmi.bios.vendor: American Megatrends Inc.
dmi.bios.version: 211
dmi.board.asset.tag: ATN12345678901234567
dmi.board.name: A15
dmi.board.vendor: PEGATRON CORPORATION
dmi.board.version: 1.0
dmi.chassis.asset.tag: No Asset Tag
dmi.chassis.type: 9
dmi.chassis.vendor: PEGATRON Computer Inc.
dmi.chassis.version: 1.0
dmi.modalias: dmi:bvnAmericanMegatrendsInc.:bvr211:bd06/17/2011:svnPEGATRONCORPORATION:pnA15:pvr1.0:rvnPEGATRONCORPORATION:rnA15:rvr1.0:cvnPEGATRONComputerInc.:ct9:cvr1.0:
dmi.product.name: A15
dmi.product.version: 1.0
dmi.sys.vendor: PEGATRON CORPORATION
version.compiz: compiz 1:0.9.6+bzr20110929-0ubuntu6.1
version.ia32-libs: ia32-libs N/A
version.libdrm2: libdrm2 2.4.26-1ubuntu1
version.libgl1-mesa-dri: libgl1-mesa-dri 7.11-0ubuntu3
version.libgl1-mesa-dri-experimental: libgl1-mesa-dri-experimental N/A
version.libgl1-mesa-glx: libgl1-mesa-glx 7.11-0ubuntu3
version.xserver-xorg: xserver-xorg 1:7.6+7ubuntu7.1
version.xserver-xorg-input-evdev: xserver-xorg-input-evdev 1:2.6.0-1ubuntu13
version.xserver-xorg-video-ati: xserver-xorg-video-ati 1:6.14.99~git20110811.g93fc084-0ubuntu1
version.xserver-xorg-video-intel: xserver-xorg-video-intel 2:2.15.901-1ubuntu2.1
version.xserver-xorg-video-nouveau: xserver-xorg-video-nouveau 1:0.0.16+git20110411+8378443-1

[tdeering (tomdeering7)]

Bump. This bug is still out there, making HDMI utterly unusable with Ubuntu on my laptop!

[tdeering (tomdeering7)]

This is still a problem with Precise (12.04).

[tdeering (tomdeering7)]

Just purchased a System76 Gazelle Professional laptop, and it has the same problem with 12.04. It seems this bug cuts across hardware configurations.

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[Chris Wilson (ickle)]

This one is confusing. There is a known (and fixed) stall with a periodicity of about 10s. But a stall every 1s, indicates something peculiar to your hardware, possibly something like an interrupt storm. Does anything show up in perf top or vmstat or etc...?

[tdeering (tomdeering7)]

FYI, Ubuntu 12.10 has the same issue with my monitor + HDMI. The freezes seem to be about 100ms in duration, and occur regularly every 1s. It's incredibly irritating.

Output of vmstat:

procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu----
 r b swpd free buff cache si so bi bo in cs us sy id wa
 0 0 0 13850796 84292 1289120 0 0 250 25 92 403 2 1 97 0

I installed perf, but I'm not exactly sure what to look for (nothing stuck out). Chris, can you let me know what else I can provide to be more helpful?

[Jeff Marcom (jeffmarcom)]

This occurs when going through display port as well.

[Chris Wilson (ickle)]

I think drm.debug=6 would help most, if there is a gfx related event it should show up there.

[Chris Wilson (ickle)]

I mean for example 'echo 6 > /sys/modules/drm/parameters/debug ; sleep 10; dmesg'

[tdeering (tomdeering7)]

Here's something: The stutter occurs when using HDMI with my primary monitor, but not with my HDTV.

Monitor: Acer S231HLbid http://www.newegg.com/Product/Product.aspx?Item=N82E16824009255
TV: Samsung LN37A550 http://www.amazon.com/Samsung-LN37A550-37-Inch-1080p-HDTV/dp/B00141AZR2

The root cause must lay in the difference between these two products. There is no "stutter" with either product using Windows or Mac. So it's strictly a Ubuntu thing, rather than a flaw with the monitor's hardware.

I think if we want to understand the problem, it's worth trying to pin down exactly what is different between these two displays.

[tdeering (tomdeering7)]

I've attached the manuals for the two displays. The relevant pages seem to be:

Samsung Manual: p44 (labeled 42), discusses HDMI in CEA mode for 1080p
Acer Manual: p12, standard timing table

My best guess for what is happening is that Ubuntu's display output timing over HDMI is slightly out of sync with what the Acer monitor expects, and hence some kind of "re-sync" process is happening about once a second. Or the timing of the laptop's built-in display is slightly different than that of the external monitor, causing the noveau driver to freak out about once a second as it tries to reconcile this.

In my Xorg log file, I see a lot of stuff like this:

[ 1878.231] (II) intel(0): EDID vendor "LGD", prod id 489
[ 1878.231] (II) intel(0): Printing DDC gathered Modelines:
[ 1878.231] (II) intel(0): Modeline "1920x1080"x0.0 138.50 1920 1968 2000 2080 1080 1083 1088 1111 -hsync -vsync (66.6 kHz eP)
[ 1879.796] (II) intel(0): Allocated new frame buffer 3840x1080 stride 15360, tiled
[ 2099.091] (II) intel(0): EDID vendor "LGD", prod id 489
[ 2099.091] (II) intel(0): Printing DDC gathered Modelines:
[ 2099.091] (II) intel(0): Modeline "1920x1080"x0.0 138.50 1920 1968 2000 2080 1080 1083 1088 1111 -hsync -vsync (66.6 kHz eP)
[ 2111.153] (II) intel(0): EDID vendor "LGD", prod id 489
[ 2111.153] (II) intel(0): Printing DDC gathered Modelines:
[ 2111.154] (II) intel(0): Modeline "1920x1080"x0.0 138.50 1920 1968 2000 2080 1080 1083 1088 1111 -hsync -vsync (66.6 kHz eP)
[ 2124.123] (II) intel(0): EDID vendor "LGD", prod id 489
[ 2124.123] (II) intel(0): Printing DDC gathered Modelines:

But if I look at the monitor's own menu, I see that the reported refresh rates are H:68KHz, V:60Hz The differences here seem like a likely culprit to me.

[tdeering (tomdeering7)]

Here's even more information. The output of 'xrandr --prop' with VGA and with HDMI connected

[Yang Kun (YK) (dayangkuncn)]

@Robert, Hi Robert, could you please take a look at this bug ? thanks

[Yang Kun (YK) (dayangkuncn)]

@Robert, hi robert, do you have any update ?

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[Robert Hooker (sarvatt)]

Would it be possible to get the info requested in comment #10 please? Does booting with drm_kms_helper.poll=0 added to the kernel command line help? That is the fix for the known stall referenced in comment #6. Is there a known good kernel that does work, and if not would it be possible to try later kernels to see if they do work please? Seeing if it works properly on http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.8.8-raring/ for instance would be a helpful start.

[tdeering (tomdeering7)]

Yep. I will try to get to this today.

Tom Deering | PhD Student | ISU ECpE Software Systems | [phone]

On Thu, Apr 18, 2013 at 7:56 AM, Robert Hooker <email address hidden> wrote:

> Would it be possible to get the info requested in comment #10 please?
> Does booting with drm_kms_helper.poll=0 added to the kernel command line
> help? That is the fix for the known stall referenced in comment #6. Is
> there a known good kernel that does work, and if not would it be
> possible to try later kernels to see if they do work please? Seeing if
> it works properly on http://kernel.ubuntu.com/~kernel-
> ppa/mainline/v3.8.8-raring/ for instance would be a helpful start.
>
> ** Changed in: linux (Ubuntu)
> Status: Confirmed => Incomplete
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/964231
>
> Title:
> System Stutter When HDMI Monitor Is Connected
>
> Status in “linux” package in Ubuntu:
> Incomplete
> Status in “linux” source package in Precise:
> Confirmed
> Status in “linux” source package in Quantal:
> Confirmed
>
> Bug description:
> With a fresh install of Ubuntu 11.10 x64, the system has many
> "microfreezes" when I connect an external monitor via HDMI. These last
> maybe 10ms and occur about once per second. As I move my cursor around
> the screen, it stops motion during these microfreezes, causing a very
> irritating stutter effect. This only happens when using HDMI- when
> connected via VGA to an external monitor, the system behaves smoothly
> as expected.
>
> I started a thread about this a long time ago, but I never did find an
> answer: http://ubuntuforums.org/showthread.php?t=1817910
>
> ProblemType: Bug
> DistroRelease: Ubuntu 11.10
> Package: xorg 1:7.6+7ubuntu7.1
> Uname: Linux 3.3.0-030300rc7-generic x86_64
> .tmp.unity.support.test.0:
>
> ApportVersion: 1.23-0ubuntu4
> Architecture: amd64
> CompizPlugins:
> [core,bailer,detection,composite,opengl,compiztoolbox,decor,grid,move,vpswitch,place,mousepoll,resize,wall,regex,session,snap,gnomecompat,imgpng,animation,expo,workarounds,unitymtgrabhandles,ezoom,fade,scale,unityshell]
> CompositorRunning: compiz
> Date: Sat Mar 24 18:30:09 2012
> DistUpgraded: Fresh install
> DistroCodename: oneiric
> DistroVariant: ubuntu
> ExtraDebuggingInterest: Yes, if not too technical
> GraphicsCard:
> Intel Corporation 2nd Generation Core Processor Family Integrated
> Graphics Controller [8086:0116] (rev 09) (prog-if 00 [VGA controller])
> Subsystem: Device [1b0a:2080]
> InstallationMedia: Ubuntu 11.10 "Oneiric Ocelot" - Release amd64
> (20111012)
> MachineType: PEGATRON CORPORATION A15
> ProcEnviron:
> PATH=(cususername, no user)
> LANG=en_US.UTF-8
> SHELL=/bin/bash
> ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-3.3.0-030300rc7-generic
> root=UUID=a19e39dd-b510-4f9d-93f5-de6d40df10f1 ro quiet splash vt.handoff=7
> SourcePackage: xorg
> Symptom: display
> Title: Xorg freeze
> UpgradeStatus: No upgrade log present (probably fresh install)
> dmi.bios.date: 06/17/2011
> dmi.bios.vendor: American Megatrends Inc.
> dmi.bios.versi...

[tdeering (tomdeering7)]

I will check these two things later today.

[tdeering (tomdeering7)]

@Robert I'm thrilled to say that *both* solutions worked!

First, "drm_kms_helper.poll=0" with the current Quantal kernel made the stutter problem go away.

Second, I grabbed the current daily build of Raring from http://cdimage.ubuntu.com/daily-live/current/, created a bootable USB, booted from it, and plugged in the monitor via HDMI. Again, no stutter problem when booting from Raring. Just to be sure it wasn't something special with booting from a live USB, I created a Quantal live USB, and the problem again manifested itself.

Thanks so much! :) Sorry I hadn't gotten around to trying the suggested from comment #10 before today. If I can provide anything else, let me know.

[Robert Hooker (sarvatt)]

Jeff: Is there any chance you could file a new bug on an affected cert machine with ubuntu-bug xorg so I can see the specific monitor info for what you are hitting it on? Also, can you include the info in comment #10 on the stock kernel, and also try a 3.9 based mainline kernel from http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.9-rc8-raring/ which contains the kms locking rework which should have fixed this to see if it does? I am unable to reproduce this on one of the cert machines tagged in this bug (v490u) because it is very monitor specific and all of mine are unaffected.

[tdeering (tomdeering7)]

FYI, the monitor with which I experienced this issue is an Acer Acer S231HLbid. See comment #11.

[James M. Leddy (jm-leddy)]

Jeff, have you had a chance to try the kernel from comment #23?

[Jeff Marcom (jeffmarcom)]

Okay, so...

I tried v3.9-rc8 and the latest in mainline and couldn't repro the issue.

Also, I cannot reproduce this problem with an ASUS monitor I have, only an acer model number: S231HL.

The issue, when reproducible seems to get progressively worse after the machine has resumed from suspend.

The kernel I reported this on was 3.5.0.23, but is also seen with 3.5.0.28.

[Brad Figg (brad-figg)]

This change was made by a bot.

[Robert Hooker (sarvatt)]

Jeff, how about the current raring kernel? If you're on precise then just sudo apt-get install linux-generic-lts-raring. Can you please get the info in comment 10 also so we can see what the actual problem is and dig into it more?

echo 6 > /sys/modules/drm/parameters/debug ; sleep 20; dmesg

[Jeff Marcom (jeffmarcom)]

the above notes were from precise 3.5.0-27. I tried the latest raring 3.8.0-22 and could not reproduce the issue even after suspending the machine 10 different times.

[Brad Figg (brad-figg)]

This change was made by a bot.

[Jeff Marcom (jeffmarcom)]

Tried updating BIOS version to G3ET93WW, and this actually disables the display port for L530 systems.

[James M. Leddy (jm-leddy)]

Hi Jeff,

Because we have a new kernel that works, we would have to bisect to figure out which commit fixes the issue. I'll chat more with Robert and figure out a way forward.

[Robert Hooker (sarvatt)]

Ok, I've started a reverse bisect on this to find where it was fixed (hence why the good and bad results below are inverted) and it will take approximately 10 steps. Can you try this kernel first and see if it works properly?

http://kernel.ubuntu.com/~sarvatt/lp964231/bisect1/

git bisect start 'drivers/gpu/drm'
# good: [28a33cbc24e4256c143dce96c7d93bf423229f92] Linux 3.5
git bisect good 28a33cbc24e4256c143dce96c7d93bf423229f92
# bad: [19f949f52599ba7c3f67a5897ac6be14bfcb1200] Linux 3.8
git bisect bad 19f949f52599ba7c3f67a5897ac6be14bfcb1200

[Jeff Marcom (jeffmarcom)]

Okay, so I have a bit more information. Most importantly this isn't limited to Lenovo hardware. Please see the following attachment for a matrix of kernel, laptops, and monitors tested.

[Robert Hooker (sarvatt)]

Thanks for that, I can see on there that it worked so I prepared the next step (again the good and bad markings are reversed since we're bisecting backwards to find the fix not the regression)

http://kernel.ubuntu.com/~sarvatt/lp964231/bisect2/

# good: [28a33cbc24e4256c143dce96c7d93bf423229f92] Linux 3.5
git bisect good 28a33cbc24e4256c143dce96c7d93bf423229f92
# bad: [19f949f52599ba7c3f67a5897ac6be14bfcb1200] Linux 3.8
git bisect bad 19f949f52599ba7c3f67a5897ac6be14bfcb1200
# bad: [a5a0fc67435599d9d787a8d7153967a70fed968e] Merge branch 'exynos-drm-next' of git://git.infradead.org/users/kmpark/linux-samsung into drm-next
git bisect bad a5a0fc67435599d9d787a8d7153967a70fed968e

[Jeff Marcom (jeffmarcom)]

Updated csv attached.

[Jeff Marcom (jeffmarcom)]

^^ No stuttering effect, but desktop is never shown on external monitor. I can tell a signal is being sent via the light indicator on the Acer monitor, but no image just a black screen.

[Robert Hooker (sarvatt)]

Thanks, we're only testing the stuttering here, other problems are going to pop up so just ignore them until we hit the final step :) Also there is no need to test multiple systems or anything, any of those that fail + the Acer S231HL would be fine. Next kernel

http://kernel.ubuntu.com/~sarvatt/lp964231/bisect3/

git bisect start 'drivers/gpu/drm'
# good: [28a33cbc24e4256c143dce96c7d93bf423229f92] Linux 3.5
git bisect good 28a33cbc24e4256c143dce96c7d93bf423229f92
# bad: [19f949f52599ba7c3f67a5897ac6be14bfcb1200] Linux 3.8
git bisect bad 19f949f52599ba7c3f67a5897ac6be14bfcb1200
# bad: [a5a0fc67435599d9d787a8d7153967a70fed968e] Merge branch 'exynos-drm-next' of git://git.infradead.org/users/kmpark/linux-samsung into drm-next
git bisect bad a5a0fc67435599d9d787a8d7153967a70fed968e
# bad: [ea9d758d6ddb9f4bb9639619100743e8f5fa85a0] drm/i915: push commit_output_state past the crtc/encoder preparing
git bisect bad ea9d758d6ddb9f4bb9639619100743e8f5fa85a0

[Jeff Marcom (jeffmarcom)]

This one looks good. No stuttering, and the external display works fine!!

[Jeff Marcom (jeffmarcom)]

a side note that the above kernel causes the system to not boot properly sometimes. In that the system halts before the unity desktop is displayed, cannot switch tty, no mouse of keyboard control.

[Robert Hooker (sarvatt)]

New one up at http://kernel.ubuntu.com/~sarvatt/lp964231/bisect4/

git bisect start 'drivers/gpu/drm'
# good: [28a33cbc24e4256c143dce96c7d93bf423229f92] Linux 3.5
git bisect good 28a33cbc24e4256c143dce96c7d93bf423229f92
# bad: [19f949f52599ba7c3f67a5897ac6be14bfcb1200] Linux 3.8
git bisect bad 19f949f52599ba7c3f67a5897ac6be14bfcb1200
# bad: [a5a0fc67435599d9d787a8d7153967a70fed968e] Merge branch 'exynos-drm-next' of git://git.infradead.org/users/kmpark/linux-samsung into drm-next
git bisect bad a5a0fc67435599d9d787a8d7153967a70fed968e
# bad: [ea9d758d6ddb9f4bb9639619100743e8f5fa85a0] drm/i915: push commit_output_state past the crtc/encoder preparing
git bisect bad ea9d758d6ddb9f4bb9639619100743e8f5fa85a0
# bad: [eeef9b3874d756405ab8f71b4012a2e7ce31458e] drm/i915: Add -EIO to the list of known errors for __wait_seqno
git bisect bad eeef9b3874d756405ab8f71b4012a2e7ce31458e

[Jeff Marcom (jeffmarcom)]

I am unable to verify if the kernel posted above fixes the display issue. The system cannot resume from suspend properly, it halts, no mouse/keyboard functionality and cannot switch tty. Plugging in external display does nothing. System is completely unresponsive.

[Robert Hooker (sarvatt)]

Thanks, skipped that kernel in the bisect then, next step:

http://kernel.ubuntu.com/~sarvatt/lp964231/bisect5/

git bisect start 'drivers/gpu/drm'
# good: [28a33cbc24e4256c143dce96c7d93bf423229f92] Linux 3.5
git bisect good 28a33cbc24e4256c143dce96c7d93bf423229f92
# bad: [19f949f52599ba7c3f67a5897ac6be14bfcb1200] Linux 3.8
git bisect bad 19f949f52599ba7c3f67a5897ac6be14bfcb1200
# bad: [a5a0fc67435599d9d787a8d7153967a70fed968e] Merge branch 'exynos-drm-next' of git://git.infradead.org/users/kmpark/linux-samsung into drm-next
git bisect bad a5a0fc67435599d9d787a8d7153967a70fed968e
# bad: [ea9d758d6ddb9f4bb9639619100743e8f5fa85a0] drm/i915: push commit_output_state past the crtc/encoder preparing
git bisect bad ea9d758d6ddb9f4bb9639619100743e8f5fa85a0
# bad: [eeef9b3874d756405ab8f71b4012a2e7ce31458e] drm/i915: Add -EIO to the list of known errors for __wait_seqno
git bisect bad eeef9b3874d756405ab8f71b4012a2e7ce31458e
# skip: [74da01dcfbb6300d758490d5d4efa1314c0e4e8b] drm/radeon: move r100_enable_bm to a more logic place
git bisect skip 74da01dcfbb6300d758490d5d4efa1314c0e4e8b

[Jeff Marcom (jeffmarcom)]

This doesn't resume properly from suspend either.

[Robert Hooker (sarvatt)]

Ok marked that one as skip too, just out of curiousity, is resuming properly from suspend a symptom or something? We're only testing the stuttering with this. Next kernel

http://kernel.ubuntu.com/~sarvatt/lp964231/bisect6/

git bisect start 'drivers/gpu/drm'
# good: [28a33cbc24e4256c143dce96c7d93bf423229f92] Linux 3.5
git bisect good 28a33cbc24e4256c143dce96c7d93bf423229f92
# bad: [19f949f52599ba7c3f67a5897ac6be14bfcb1200] Linux 3.8
git bisect bad 19f949f52599ba7c3f67a5897ac6be14bfcb1200
# bad: [a5a0fc67435599d9d787a8d7153967a70fed968e] Merge branch 'exynos-drm-next' of git://git.infradead.org/users/kmpark/linux-samsung into drm-next
git bisect bad a5a0fc67435599d9d787a8d7153967a70fed968e
# bad: [ea9d758d6ddb9f4bb9639619100743e8f5fa85a0] drm/i915: push commit_output_state past the crtc/encoder preparing
git bisect bad ea9d758d6ddb9f4bb9639619100743e8f5fa85a0
# bad: [eeef9b3874d756405ab8f71b4012a2e7ce31458e] drm/i915: Add -EIO to the list of known errors for __wait_seqno
git bisect bad eeef9b3874d756405ab8f71b4012a2e7ce31458e
# skip: [74da01dcfbb6300d758490d5d4efa1314c0e4e8b] drm/radeon: move r100_enable_bm to a more logic place
git bisect skip 74da01dcfbb6300d758490d5d4efa1314c0e4e8b
# skip: [e486fad9136dce21f5ba4322ae0454d23805d342] drm/i915: fix PIPE_WM_LINETIME definition
git bisect skip e486fad9136dce21f5ba4322ae0454d23805d342

[Jeff Marcom (jeffmarcom)]

The performance degradation is seen and worsened after the machine has resumed from suspend. It's often you won't see the issue reported UNLESS you've resumed from suspend.

[Jeff Marcom (jeffmarcom)]

Cool ! So, that one seems to resume from suspend okay, AND the external display works properly.

Progress!!

[Robert Hooker (sarvatt)]

Nice, that cut 2 more steps out of the bisect and we're down to 5 max left.

http://kernel.ubuntu.com/~sarvatt/lp964231/bisect7/

git bisect start 'drivers/gpu/drm'
# good: [28a33cbc24e4256c143dce96c7d93bf423229f92] Linux 3.5
git bisect good 28a33cbc24e4256c143dce96c7d93bf423229f92
# bad: [19f949f52599ba7c3f67a5897ac6be14bfcb1200] Linux 3.8
git bisect bad 19f949f52599ba7c3f67a5897ac6be14bfcb1200
# bad: [a5a0fc67435599d9d787a8d7153967a70fed968e] Merge branch 'exynos-drm-next' of git://git.infradead.org/users/kmpark/linux-samsung into drm-next
git bisect bad a5a0fc67435599d9d787a8d7153967a70fed968e
# bad: [ea9d758d6ddb9f4bb9639619100743e8f5fa85a0] drm/i915: push commit_output_state past the crtc/encoder preparing
git bisect bad ea9d758d6ddb9f4bb9639619100743e8f5fa85a0
# bad: [eeef9b3874d756405ab8f71b4012a2e7ce31458e] drm/i915: Add -EIO to the list of known errors for __wait_seqno
git bisect bad eeef9b3874d756405ab8f71b4012a2e7ce31458e
# skip: [74da01dcfbb6300d758490d5d4efa1314c0e4e8b] drm/radeon: move r100_enable_bm to a more logic place
git bisect skip 74da01dcfbb6300d758490d5d4efa1314c0e4e8b
# skip: [e486fad9136dce21f5ba4322ae0454d23805d342] drm/i915: fix PIPE_WM_LINETIME definition
git bisect skip e486fad9136dce21f5ba4322ae0454d23805d342
# bad: [c20dc3698dc7ecf053e2bf77299ae5982c0c2c45] drm/radeon: fix & improve ih ring handling v3
git bisect bad c20dc3698dc7ecf053e2bf77299ae5982c0c2c45

[Jeff Marcom (jeffmarcom)]

awesome, still looks okay with your rc4

[Robert Hooker (sarvatt)]

4 steps to go

http://kernel.ubuntu.com/~sarvatt/lp964231/bisect8/

git bisect start 'drivers/gpu/drm'
# good: [28a33cbc24e4256c143dce96c7d93bf423229f92] Linux 3.5
git bisect good 28a33cbc24e4256c143dce96c7d93bf423229f92
# bad: [19f949f52599ba7c3f67a5897ac6be14bfcb1200] Linux 3.8
git bisect bad 19f949f52599ba7c3f67a5897ac6be14bfcb1200
# bad: [a5a0fc67435599d9d787a8d7153967a70fed968e] Merge branch 'exynos-drm-next' of git://git.infradead.org/users/kmpark/linux-samsung into drm-next
git bisect bad a5a0fc67435599d9d787a8d7153967a70fed968e
# bad: [ea9d758d6ddb9f4bb9639619100743e8f5fa85a0] drm/i915: push commit_output_state past the crtc/encoder preparing
git bisect bad ea9d758d6ddb9f4bb9639619100743e8f5fa85a0
# bad: [eeef9b3874d756405ab8f71b4012a2e7ce31458e] drm/i915: Add -EIO to the list of known errors for __wait_seqno
git bisect bad eeef9b3874d756405ab8f71b4012a2e7ce31458e
# skip: [74da01dcfbb6300d758490d5d4efa1314c0e4e8b] drm/radeon: move r100_enable_bm to a more logic place
git bisect skip 74da01dcfbb6300d758490d5d4efa1314c0e4e8b
# skip: [e486fad9136dce21f5ba4322ae0454d23805d342] drm/i915: fix PIPE_WM_LINETIME definition
git bisect skip e486fad9136dce21f5ba4322ae0454d23805d342
# bad: [c20dc3698dc7ecf053e2bf77299ae5982c0c2c45] drm/radeon: fix & improve ih ring handling v3
git bisect bad c20dc3698dc7ecf053e2bf77299ae5982c0c2c45
# bad: [9d9740f099f2eaf309c4c9cbc0d732507140db28] drm/i915: add some barriers when changing DIPs
git bisect bad 9d9740f099f2eaf309c4c9cbc0d732507140db28

[Jeff Marcom (jeffmarcom)]

Hmm, the bug is present in the above kernel.

[Robert Hooker (sarvatt)]

3 more steps.

http://kernel.ubuntu.com/~sarvatt/lp964231/bisect9/

git bisect start 'drivers/gpu/drm'
# good: [28a33cbc24e4256c143dce96c7d93bf423229f92] Linux 3.5
git bisect good 28a33cbc24e4256c143dce96c7d93bf423229f92
# bad: [19f949f52599ba7c3f67a5897ac6be14bfcb1200] Linux 3.8
git bisect bad 19f949f52599ba7c3f67a5897ac6be14bfcb1200
# bad: [a5a0fc67435599d9d787a8d7153967a70fed968e] Merge branch 'exynos-drm-next' of git://git.infradead.org/users/kmpark/linux-samsung into drm-next
git bisect bad a5a0fc67435599d9d787a8d7153967a70fed968e
# bad: [ea9d758d6ddb9f4bb9639619100743e8f5fa85a0] drm/i915: push commit_output_state past the crtc/encoder preparing
git bisect bad ea9d758d6ddb9f4bb9639619100743e8f5fa85a0
# bad: [eeef9b3874d756405ab8f71b4012a2e7ce31458e] drm/i915: Add -EIO to the list of known errors for __wait_seqno
git bisect bad eeef9b3874d756405ab8f71b4012a2e7ce31458e
# skip: [74da01dcfbb6300d758490d5d4efa1314c0e4e8b] drm/radeon: move r100_enable_bm to a more logic place
git bisect skip 74da01dcfbb6300d758490d5d4efa1314c0e4e8b
# skip: [e486fad9136dce21f5ba4322ae0454d23805d342] drm/i915: fix PIPE_WM_LINETIME definition
git bisect skip e486fad9136dce21f5ba4322ae0454d23805d342
# bad: [c20dc3698dc7ecf053e2bf77299ae5982c0c2c45] drm/radeon: fix & improve ih ring handling v3
git bisect bad c20dc3698dc7ecf053e2bf77299ae5982c0c2c45
# bad: [9d9740f099f2eaf309c4c9cbc0d732507140db28] drm/i915: add some barriers when changing DIPs
git bisect bad 9d9740f099f2eaf309c4c9cbc0d732507140db28
# good: [199b2bc25ba587f666a712e9d8475d691d9cec4c] drm/i915: s/i915_wait_request/i915_wait_seqno/g
git bisect good 199b2bc25ba587f666a712e9d8475d691d9cec4c

[Jeff Marcom (jeffmarcom)]

Sorry for the delay on this. That kernel worked perfectly!

[Robert Hooker (sarvatt)]

Thanks! Next one up at http://kernel.ubuntu.com/~sarvatt/lp964231/bisect10/

git bisect start 'drivers/gpu/drm'
# good: [28a33cbc24e4256c143dce96c7d93bf423229f92] Linux 3.5
git bisect good 28a33cbc24e4256c143dce96c7d93bf423229f92
# bad: [19f949f52599ba7c3f67a5897ac6be14bfcb1200] Linux 3.8
git bisect bad 19f949f52599ba7c3f67a5897ac6be14bfcb1200
# bad: [a5a0fc67435599d9d787a8d7153967a70fed968e] Merge branch 'exynos-drm-next' of git://git.infradead.org/users/kmpark/linux-samsung into drm-next
git bisect bad a5a0fc67435599d9d787a8d7153967a70fed968e
# bad: [ea9d758d6ddb9f4bb9639619100743e8f5fa85a0] drm/i915: push commit_output_state past the crtc/encoder preparing
git bisect bad ea9d758d6ddb9f4bb9639619100743e8f5fa85a0
# bad: [eeef9b3874d756405ab8f71b4012a2e7ce31458e] drm/i915: Add -EIO to the list of known errors for __wait_seqno
git bisect bad eeef9b3874d756405ab8f71b4012a2e7ce31458e
# skip: [74da01dcfbb6300d758490d5d4efa1314c0e4e8b] drm/radeon: move r100_enable_bm to a more logic place
git bisect skip 74da01dcfbb6300d758490d5d4efa1314c0e4e8b
# skip: [e486fad9136dce21f5ba4322ae0454d23805d342] drm/i915: fix PIPE_WM_LINETIME definition
git bisect skip e486fad9136dce21f5ba4322ae0454d23805d342
# bad: [c20dc3698dc7ecf053e2bf77299ae5982c0c2c45] drm/radeon: fix & improve ih ring handling v3
git bisect bad c20dc3698dc7ecf053e2bf77299ae5982c0c2c45
# bad: [9d9740f099f2eaf309c4c9cbc0d732507140db28] drm/i915: add some barriers when changing DIPs
git bisect bad 9d9740f099f2eaf309c4c9cbc0d732507140db28
# good: [199b2bc25ba587f666a712e9d8475d691d9cec4c] drm/i915: s/i915_wait_request/i915_wait_seqno/g
git bisect good 199b2bc25ba587f666a712e9d8475d691d9cec4c
sarvatt@tangerine:~/linux-2.6$ git bisect bad
Bisecting: 2 revisions left to test after this (roughly 2 steps)
[0c14c7f957e70c3cb100e3fd6553b0ebea557571] drm/i915: properly alternate between DVI and HDMI

[Jeff Marcom (jeffmarcom)]

k, that one failed ^^

[penalvch (penalvch)]

tdeering, would you need a backport to a release prior to Raring, or may this be marked Status Invalid?

Declining the linux (Ubuntu) task as this is not reproducible in Raring for the original reporter as per https://bugs.launchpad.net/ubuntu/+source/linux/+bug/964231/comments/22 .

[tdeering (tomdeering7)]

It's fine to mark this as invalid, since it cannot be reproduced as of Raring.