Panda: suspend/resume not working

[Avik Sil]

Hwpack: hwpack_linaro-lt-panda-x11-base_20111023-0_armel_supported.tar.gz
Image: linaro-o-ubuntu-desktop-tar-20111023-0.tar.gz

Suspended the system from desktop menu. After that, resume is not working on pressing USB keyboard/mouse.

Even when suspended using "echo mem > /sys/power/state" the desktop is not restored on keyboard/mouse press.

[warmcat (andy-warmcat)]

Testing with tilt-tracking, which is tilt-linux-linaro-3.1 without the linux-linaro-3.1, mem suspend is working fine.

With tilt-linux-linaro-3.1 it seems to suspend to the point that CPU1 is disabled and interrupts are off, then die before interrupts are coming back on again.

I saw something similar a couple of weeks ago when debugging Android suspend, can't remember what the fix was atm.

[Abhishek Paliwal (abhishek-paliwal)]

Observed here as well:
https://android-build.linaro.org/builds/~linaro-android/tracking-panda/#build=55

[Abhishek Paliwal (abhishek-paliwal)]

Also observed on Staging Panda 11.11 RC:
https://android-build.linaro.org/builds/~linaro-android/staging-panda-11.11-release/#build=4

[vishal (vishalbhoj)]

Also oserved for 4460 on 11.11 RC :
https://android-build.linaro.org/builds/~linaro-android/staging-panda-11.11-release/#build=4

[vishal (vishalbhoj)]

Also observed on 4460
for https://android-build.linaro.org/builds/~linaro-android/tracking-panda-11.11-release/#build=4

[vishal (vishalbhoj)]

Also observed for 4460 on RC build 11.11:
https://android-build.linaro.org/builds/~linaro-android/staging-panda-11.11-release/#build=4

[Abhishek Paliwal (abhishek-paliwal)]

Also observed on tracking-panda 11.11 RC:
https://android-build.linaro.org/builds/~linaro-android/tracking-panda-11.11-release/#build=4

[vishal (vishalbhoj)]

Also observed for 4460 on RC build 11.11:
https://android-build.linaro.org/builds/~linaro-android/panda-11.11-release/#build=5

[John Rigby (jcrigby)]

same issue in 2011.11 hwpack: hwpack_linaro-lt-panda-x11-base_20111121-2_armel_supported.tar.gz

[Avik Sil (aviksil)]

same issue on panda 4430 for hwpack: hwpack_linaro-lt-panda-x11-base_20111122-1_armel_supported.tar.gz

[warmcat (andy-warmcat)]

Jassi found a workaround, I confirmed it, and pushed on tilt-3.1 and updated tilt-linux-linaro-3.1.

[Abhishek Paliwal (abhishek-paliwal)]

On release : https://android-build.linaro.org/builds/~linaro-android/panda/#build=407
Device goes in suspend after screen timeout(but screen update fails - 897551) and pressing F1 key brings up lock state screen.

[Abhishek Paliwal (abhishek-paliwal)]

Suspend resume not working on : https://android-build.linaro.org/builds/~linaro-android/panda/#build=418
Unable to resume.

[warmcat (andy-warmcat)]

That log is from tilt-android-tracking kernel, we know that's broken by something else and it's on our list.

tilt-android-3.1 based kernel should be workable as should tilt-linux-linaro-3.1 for mem suspend.

[Ricardo Salveti (rsalveti)]

Seems it's working better with current tilt-linux-linaro-3.1, but my HDMI screen didn't get back to life after the resume: http://paste.ubuntu.com/770832/

[warmcat (andy-warmcat)]

Yeah these are further runtime_pm issues between hdmi's runtime_pm and its parent dss runtime_pm. Will get to them.

[Avik Sil (aviksil)]

Suspend/resume not working with 11.12 RC Ubuntu LEB image

[warmcat (andy-warmcat)]

I fixed the runtime_pm issue, now suspend is working, the 1080p raster is back but the screen is sometimes in blanking. Removing HDMI for > 10s then re-inserting gets normal pixel data back. Working on solving that...

[warmcat (andy-warmcat)]

OK this last issue is fixed, mem suspend should be good in tilt-3.1, tilt-linux-linaro-3.1 and tilt-tracking.

[Ricardo Salveti (rsalveti)]

Great, tested with current kernel available at the overlay (from tilt-linux-linaro-3.1) and it worked just fine:

Suspend:
[ 2581.049102] omapdss error: operation stopped when reading edid
[ 2581.059997] [drm:drm_edid_block_valid] *ERROR* EDID checksum is invalid, remainder is 131
[ 2581.073028] Raw EDID:
[ 2581.075866] 00 ff ff ff ff ff ff 00 1e 6d dd 56 01 01 01 01
[ 2581.082122] 03 14 01 03 80 31 20 78 0a ae c5 a2 57 4a 9c 25
[ 2581.091369] 12 50 54 a7 6b 80 b3 00 81 80 81 8f 71 4f 01 01
[ 2581.097656] 01 01 01 01 01 01 1a 36 80 a0 70 38 1f 40 30 20
[ 2581.105773] 35 00 fe 22 11 00 00 1e 02 3a 80 18 71 38 2d 40
[ 2581.112060] 53 2c 45 00 fe 22 11 00 00 1e 00 00 00 fd 00 38
[ 2581.118316] 3d 1e 53 0f 00 0a 20 20 20 20 20 20 00 00 00 fc
[ 2581.124633] 00 57 32 32 35 33 0a 20 20 20 20 20 20 20 01 00
[ 2598.570922] omap_device: smartreflex.2: new worst case deactivate latency 0: 61035
[ 2610.221435] wl1271: down
[ 2629.136566] [drm:drm_edid_block_valid] *ERROR* EDID checksum is invalid, remainder is 226
[ 2629.146026] Raw EDID:
[ 2629.151458] 00 ff ff ff ff ff ff 00 1e 6d dd 56 01 01 01 01
[ 2629.172912] 03 14 01 03 80 31 20 78 0a ae c5 a2 57 4a 9c 25
[ 2629.178985] 12 50 54 a7 6b 80 b3 00 81 80 81 8f 71 4f 01 01
[ 2629.201293] 01 01 01 01 01 01 1a 36 80 a0 70 38 1f 40 30 20
[ 2629.210968] 35 00 fe 22 11 00 00 1e 02 3a 80 18 71 38 2d 40
[ 2629.231109] 53 2c 45 00 fe 22 11 00 00 1e 00 00 00 fd 00 38
[ 2629.249603] 3d 1e 53 0f 00 0a 20 20 20 20 20 20 00 00 00 fc
[ 2629.266113] 00 57 32 32 35 33 0a 20 20 00 00 00 00 00 00 00
[ 2665.092407] EXT4-fs (mmcblk0p2): re-mounted. Opts: errors=remount-ro,commit=0
[ 2670.485076] init: anacron main process (3317) killed by TERM signal
[ 2673.852325] omaplfb: omaplfb_Ioctl: PVR Device 1: Leave VT
[ 2694.503997] PM: Syncing filesystems ... done.
[ 2694.536987] PM: Preparing system for mem sleep
[ 2694.542572] Freezing user space processes ... (elapsed 0.02 seconds) done.
[ 2694.573608] Freezing remaining freezable tasks ... (elapsed 0.02 seconds) done.
[ 2694.604949] PM: Entering mem sleep
[ 2694.608520] Suspending console(s) (use no_console_suspend to debug)
[ 2694.952148] PM: suspend of devices complete after 335.601 msecs
[ 2694.955902] omap_device: omapdrm_pvr.0: new worst case deactivate latency 0: 2807617
[ 2694.963989] omap_device: omapdss_dispc.-1: new worst case deactivate latency 0: 5371093
[ 2694.966705] PM: late suspend of devices complete after 14.556 msecs
[ 2694.966796] Disabling non-boot CPUs ...
[ 2694.968078] CPU1: shutdown

Resume:
[ 2716.047546] Enabling non-boot CPUs ...
[ 2716.069763] CPU1: Booted secondary processor
[ 2716.069763] CPU1: Unknown IPI message 0x1
[ 2716.070007] Calibrating delay loop (skipped) already calibrated this CPU
[ 2716.070770] Switched to NOHz mode on CPU #1
[ 2716.070831] thermal_cooling_dev_register:Found cpu cpu
[ 2716.070831] thermal_init_thermal_state:Not all components registered for cpu domain sensor (null), gov=ee34c660, cooling=c0af70c4
[ 2716.070922] CPU1 is up
[ 2716.225769] omap_device: omapdss_dss.-1: new worst case activate latency 0: 12725830
[ 2716.226501] omap_device: smartreflex.0: new worst case activate latency...

[vk (vijay-k86)]

thanks a lot :)

[Avik Sil (aviksil)]

With 12.01 RC image, on resume, getting back the serial console, but not getting the Unity panel, only the desktop background is displayed.

[Avik Sil (aviksil)]

The above observation is with 4430.

[warmcat (andy-warmcat)]

Didn't change anything here since it was last working. Are there any funnies in dmesg?

[Avik Sil (aviksil)]

On suspend:

[ 2051.478332] wl1271: down
[ 2053.661895] EXT4-fs (mmcblk0p2): re-mounted. Opts: errors=remount-ro,commit=0
[ 2054.696990] omaplfb: omaplfb_Ioctl: PVR Device 1: Leave VT
[ 2054.939666] PM: Syncing filesystems ... done.
[ 2054.968749] PM: Preparing system for mem sleep
[ 2054.973815] Freezing user space processes ... (elapsed 0.02 seconds) done.
[ 2055.004486] Freezing remaining freezable tasks ... (elapsed 0.02 seconds) done.
[ 2055.035675] PM: Entering mem sleep
[ 2055.039245] Suspending console(s) (use no_console_suspend to debug)
[ 2055.349029] PM: suspend of devices complete after 301.818 msecs
[ 2055.352752] omap_device: omapdrm_pvr.0: new worst case deactivate latency 0: 2655029
[ 2055.355468] omap_device: omap-mailbox.-1: new worst case deactivate latency 0: 2655029
[ 2055.363555] PM: late suspend of devices complete after 14.495 msecs
[ 2055.363647] Disabling non-boot CPUs ...
[ 2055.364868] CPU1: shutdown

On resume:

[ 4100.153625] Enabling non-boot CPUs ...
[ 4100.190093] CPU1: Booted secondary processor
[ 4100.190399] CPU1: Unknown IPI message 0x1
[ 4100.190429] Calibrating delay loop (skipped) already calibrated this CPU
[ 4100.191223] thermal_cooling_dev_register:Found cpu cpu
[ 4100.191223] thermal_init_thermal_state:Not all components registered for cpu domain sensor (null), gov=eeb88d20, cooling=c0af70c4
[ 4100.191223] CPU1 is up
[ 4100.194335] Switched to NOHz mode on CPU #1
[ 4100.346130] omap_device: omapdss_dss.-1: new worst case activate latency 0: 5584716
[ 4100.346923] omap_device: smartreflex.2: new worst case activate latency 0: 640869
[ 4100.347015] PM: early resume of devices complete after 155.731 msecs
[ 4100.759460] PM: resume of devices complete after 412.048 msecs
[ 4100.870178] PM: Finishing wakeup.
[ 4100.873779] Restarting tasks ... done.
[ 4101.095642] omaplfb: omaplfb_Ioctl: PVR Device 1: Enter VT
[ 4101.684356] EXT4-fs (mmcblk0p2): re-mounted. Opts: errors=remount-ro,commit=0
[ 4102.709991] wl1271: firmware booted (Rev 6.1.5.50.74)
[ 4102.738555] ADDRCONF(NETDEV_UP): wlan0: link is not ready
[ 4102.924072] ieee80211 phy0: device no longer idle - scanning
[ 4104.917266] smsc95xx 1-1.1:1.0: eth0: link up, 100Mbps, half-duplex, lpa 0x0080
[ 4105.598846] ieee80211 phy0: device now idle
[ 4113.422607] eth0: no IPv6 routers present
[ 4122.810516] ieee80211 phy0: device no longer idle - scanning
[ 4125.471954] ieee80211 phy0: device now idle
[ 4152.802093] ieee80211 phy0: device no longer idle - scanning
[ 4155.483062] ieee80211 phy0: device now idle
[ 4192.788787] ieee80211 phy0: device no longer idle - scanning
[ 4195.454528] ieee80211 phy0: device now idle

[warmcat (andy-warmcat)]

Looks normal afaict. Maybe there's something in the X log.

[Ricardo Salveti (rsalveti)]

On Tue, Jan 24, 2012 at 7:42 AM, warmcat <email address hidden> wrote:
> Looks normal afaict.  Maybe there's something in the X log.

Yeah, maybe unity 3d stopped working during resume, will try to
reproduce it here.

[Ricardo Salveti (rsalveti)]

Tried here and I can confirm that there are still a few issues with Unity 3D (probably an issue with Compiz). Besides that the kernel is behaving correctly, so it'll close this bug for now and retest with the new unity 3d version that will be integrated in the next few days.

[Milosz Wasilewski (mwasilew)]

This bug is pretty old. It wasn't fixed only in linaro-android project. The current bug for this issue is: https://bugs.launchpad.net/linaro-landing-team-ti/+bug/1033944. Closing.