Performance degraded and video glitches with i915 after upgrade to utopic

[luca]

After upgrading from trusty to utopic, I noticed a performance degradation in the video performance. In particular the flash plugin video (like youtube) starts OK but it degrades after a couple of seconds when in full screen, looking as if it was doing something like 10~15 fps with videos that worked fine before. When switching to windowed more again, the video goes more smoothly but the seek bar blinks and other parts of the video present some glitches. Also when switching from X to a console, via Ctrl-Alt-F1 for example, it takes like 2 seconds to get the console, and I see the below message in the console. When switching back to X is fast as usual, but when switching back to the console, again 2 seconds delay and the message:
[ 184.436477] [drm:intel_enable_lvds] *ERROR* timed out waiting for panel to power on

Other weird looking messages are found in dmesg:
[ 4.673834] [drm] GMBUS [i915 gmbus panel] timed out, falling back to bit banging on pin 3
[...]
[ 4.732400] [drm] Wrong MCH_SSKPD value: 0x16040307
[ 4.732404] [drm] This can cause pipe underruns and display issues.
[ 4.732406] [drm] Please upgrade your BIOS to fix this.

WORKAROUND: Change intel_pscale cpufreq driver setting to performance from powersave.

ProblemType: Bug
DistroRelease: Ubuntu 14.10
Package: xorg 1:7.7+7ubuntu2
ProcVersionSignature: Ubuntu 3.16.0-25.33-generic 3.16.7
Uname: Linux 3.16.0-25-generic x86_64
.tmp.unity.support.test.0:

ApportVersion: 2.14.7-0ubuntu8
Architecture: amd64
CompizPlugins: [core,composite,opengl,compiztoolbox,decor,vpswitch,snap,mousepoll,resize,place,move,wall,grid,regex,imgpng,session,gnomecompat,animation,fade,workarounds,scale,expo,ezoom]
CompositorRunning: None
CurrentDesktop: Unity
Date: Sat Nov 15 22:52:27 2014
DistUpgraded: 2014-11-08 12:14:07,528 DEBUG enabling apt cron job
DistroCodename: utopic
DistroVariant: ubuntu
DkmsStatus: virtualbox, 4.3.18, 3.16.0-25-generic, x86_64: installed
ExtraDebuggingInterest: Yes
GraphicsCard:
 Intel Corporation 2nd Generation Core Processor Family Integrated Graphics Controller [8086:0116] (rev 09) (prog-if 00 [VGA controller])
   Subsystem: Toshiba America Info Systems Device [1179:0009]
InstallationDate: Installed on 2012-04-26 (933 days ago)
InstallationMedia: Ubuntu 12.04 LTS "Precise Pangolin" - Release amd64 (20120425)
MachineType: TOSHIBA SATELLITE Z830
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-3.16.0-25-generic root=UUID=36929bf3-a158-44d9-a80d-3adac2840fa8 ro quiet splash acpi_backlight=vendor pcie_aspm=force i915.enable_rc6=7 i915.enable_fbc=1 i915.lvds_downclock=1 vt.handoff=7
SourcePackage: xorg
Symptom: display
UpgradeStatus: Upgraded to utopic on 2014-11-08 (7 days ago)
dmi.bios.date: 04/18/2013
dmi.bios.vendor: TOSHIBA
dmi.bios.version: Version 1.80
dmi.board.asset.tag: 0000000000
dmi.board.name: Portable PC
dmi.board.vendor: TOSHIBA
dmi.board.version: Version A0
dmi.chassis.asset.tag: 0000000000
dmi.chassis.type: 10
dmi.chassis.vendor: TOSHIBA
dmi.chassis.version: Version 1.0
dmi.modalias: dmi:bvnTOSHIBA:bvrVersion1.80:bd04/18/2013:svnTOSHIBA:pnSATELLITEZ830:pvrPT22LE-00300GGR:rvnTOSHIBA:rnPortablePC:rvrVersionA0:cvnTOSHIBA:ct10:cvrVersion1.0:
dmi.product.name: SATELLITE Z830
dmi.product.version: PT22LE-00300GGR
dmi.sys.vendor: TOSHIBA
version.compiz: compiz 1:0.9.12+14.10.20140918-0ubuntu1
version.ia32-libs: ia32-libs N/A
version.libdrm2: libdrm2 2.4.56-1
version.libgl1-mesa-dri: libgl1-mesa-dri 10.3.0-0ubuntu3
version.libgl1-mesa-dri-experimental: libgl1-mesa-dri-experimental N/A
version.libgl1-mesa-glx: libgl1-mesa-glx 10.3.0-0ubuntu3
version.xserver-xorg-core: xserver-xorg-core 2:1.16.0-1ubuntu1
version.xserver-xorg-input-evdev: xserver-xorg-input-evdev 1:2.9.0-1ubuntu2
version.xserver-xorg-video-ati: xserver-xorg-video-ati 1:7.4.0-2ubuntu2
version.xserver-xorg-video-intel: xserver-xorg-video-intel 2:2.99.914-1~exp1ubuntu4.1
version.xserver-xorg-video-nouveau: xserver-xorg-video-nouveau 1:1.0.11-1ubuntu2
xserver.bootTime: Sat Nov 15 22:40:14 2014
xserver.configfile: default
xserver.errors:

xserver.logfile: /var/log/Xorg.0.log
xserver.outputs:

xserver.version: 2:1.16.0-1ubuntu1

[luca (llucax)]

I tried booting with kernel 3.13.11.11 (AFAIK trusty uses kernel 3.13 at least) form http://kernel.ubuntu.com/~kernel-ppa/mainline/ and the problems are the same, the only message I don't get in the console is "*ERROR* timed out waiting for panel to power on" but switching to the console is still slow.

Now I'm trying with kernel 3.17.1 and no difference either...

[Michael Picher (mpicher)]

I'm having similar speed issues and strange horizontal lines all over the screen randomly... they appear for a second and disappear.

I too have gone from Trusty to Utopic.

This page had some things to try https://wiki.archlinux.org/index.php/Intel_graphics#SNA_issues

Going to accelmode UXA seemed to fix the speed issues but I couldn't get the resolutions right and the video would freeze when trying to come back from sleep.

My system is a Samsung Series 5.

I'm not seeing the same i915 issue:
[ 0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-3.16.0-24-generic root=UUID=beb26549-7f0b-47e2-8a51-9b8e8d185d8c ro quiet splash pcie_aspm=force i915.i915_enable_rc6=1 i915.i915_enable_fbc=1 i915.lvds_downclock=1 vt.handoff=7
[ 0.000000] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-3.16.0-24-generic root=UUID=beb26549-7f0b-47e2-8a51-9b8e8d185d8c ro quiet splash pcie_aspm=force i915.i915_enable_rc6=1 i915.i915_enable_fbc=1 i915.lvds_downclock=1 vt.handoff=7
[ 18.669068] i915: unknown parameter 'i915_enable_rc6' ignored
[ 18.669073] i915: unknown parameter 'i915_enable_fbc' ignored
[ 18.691624] i915 0000:00:02.0: irq 43 for MSI/MSI-X
[ 19.321014] i915 0000:00:02.0: fb0: inteldrmfb frame buffer device
[ 19.321017] i915 0000:00:02.0: registered panic notifier
[ 19.348617] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0

[luca (llucax)]

To get rid off these 2 errors:
 [ 18.669068] i915: unknown parameter 'i915_enable_rc6' ignored
 [ 18.669073] i915: unknown parameter 'i915_enable_fbc' ignored

Just remove the "i915_" prefix, it was removed in recent kernels. Maybe when those two options are used again you can get results similar to the ones before the upgrade.

I'll try to change to UXA acceleration, thanks!

[luca (llucax)]

BTW, you can also click on "This bug also affects me" to give it more heat/visibility.

[penalvch (penalvch)]

luca, if you remove the non-default kernel parameters does this change anything:
acpi_backlight=vendor pcie_aspm=force i915.enable_rc6=7 i915.enable_fbc=1 i915.lvds_downclock=1

As well, could you please comment to why these are in use?

[luca (llucax)]

> luca, if you remove the non-default kernel parameters does this change anything:
> acpi_backlight=vendor pcie_aspm=force i915.enable_rc6=7 i915.enable_fbc=1 i915.lvds_downclock=1

Yes, it just makes it slower (although more consistently slow). Without
those kernel options, after a reboot, it takes around 8 ~ 10 minutes to
start being slow, with them the slowness is immediate.

> As well, could you please comment to why these are in use?

Hard to remember but fortunately I had some comments in my
/etc/default/grub :)

pcie_aspm=force
https://access.redhat.com/site/documentation/en-US/Red_Hat_Enterprise_Linux/6/html/Power_Management_Guide/ASPM.html
(forced aspm because it not automatically detected)

i915.*
http://www.phoronix.com/scan.php?page=article&item=intel_i915_power&num=1
       modinfo: i915_enable_rc6:Enable power-saving render C-state 6.
       Different stages can be selected via bitmask values (0 = disable;
       1 = enable rc6; 2 = enable deep rc6; 4 = enable deepest rc6).
       For example, 3 would enable rc6 and deep rc6, and 7 would enable
       everything. default: -1 (use per-chip default) (int)
(I enabled everything because it seemed to work well)

acpi_osi=Linux / acpi_backlight=vendor
http://www.linlap.com/toshiba_portege_z830-10f
(acpi_osi=Linux not used because it doesn't seem necessary with the hack
mentioned in
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/935778/comments/68)

I used to have issues in the past with the backlight after suspending,
but I got everything working fine using those options for a couple of
Ubuntu releases, until utopic :(

Thanks!

PS: I also tried switching to UXA acceleration method, but is not making
a difference.

[luca (llucax)]

OK, I just noticed I only have 2 CPU scaling governors now, so looking around I bumped into this:

http://askubuntu.com/questions/544266/why-are-missing-the-frequency-options-on-cpufreq-utils-indicator

I switched to "performance" (it was in "powersave") and now all seems to be working fine. It seems like it was just a problem with CPU scaling.

Nevertheless, this is a very nasty regression, and very hard to find out, as the only evident effect I've seen from it is this slowdown when watching video using the flash plugin (probably because the flash plugin sucks so much, because I could see videos, even in HD, with any other decent video player like mplayer or vlc).

So it seems like the intel_pscale cpufreq driver is kind of broken, the CPU frequency is in fact scaled (I have a small indicator for that and I see it increasing even with the "powersave" governor), but it doesn't seem to calculate correctly the CPU power needed.

Maybe as a workaround the "performance" governor should be used by default. Or maybe I should switch back to the old ACPI driver, as I'm also noticing that the CPU frequency is kept quite high, even when idling, so I'm afraid of the consequences of this in my battery life :-/

[luca (llucax)]

This seems like a related issue https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1188647

[penalvch (penalvch)]

luca, could you please test the latest upstream kernel available from the very top line at the top of the page (the release names are irrelevant for testing, and please do not test the daily folder) following https://wiki.ubuntu.com/KernelMainlineBuilds ? It will allow additional upstream developers to examine the issue.

If the test did not allow you to test to the issue (ex. you couldn't boot into the OS) please make a comment in your report about this, and continue to test the next most recent kernel version until you can test to the issue. 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 exactly shown as:
kernel-fixed-upstream-3.18-rc6

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.

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

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.

[luca (llucax)]

I'll try it as soon as I have some time.

[luca (llucax)]

After using this for a while, I noticed after some time the performance seems to be degraded again even when using the performance governor. Also, I noticed the battery life is reduced significantly, even when using the powersave governor (before it was about 3.5 hours, now is more like 2.5 hours).

I will try with the latest kernel and also I'll try using the old ACPI cpu freq scaling driver.

[luca (llucax)]

Hi, I've been using kernel 3.18.0-031800-lowlatency for a while now, and the performance using the powersave governor seems to be much better. I would say that problem is resolved, but the battery life isn't improved compared to how much it lasted when the old acpi driver was used. I'll test now for a while disabling the pstate driver and I will report back.

[Launchpad Janitor (janitor)]

[Expired for linux (Ubuntu) because there has been no activity for 60 days.]