Can't get suspend/resume working on yocto 1.7

[ossian]

I'm currently using yocto 1.7 and am having some issues getting the cpu suspend/resume power feature to work.

When issuing the following two command at the command prompt:

echo +30 > /sys/class/rtc/rtc0/wakealarm

echo standby > /sys/power/state

I get the following out put:

[  104.593519] PM: Syncing filesystems ... done.

[  104.619884] Freezing user space processes ... (elapsed 0.01 seconds) done.

[  104.645795] Freezing remaining freezable tasks ... (elapsed 0.01 seconds) done.

[  104.773140] hub 2-0:1.0: port 1 nyet suspended

[  104.834071] PM: suspend of devices complete after 167.729 msecs

[  104.840003] PM: suspend devices took 0.170 seconds

[  104.845402] PM: late suspend of devices complete after 0.573 msecs

[  104.852308] PM: noirq suspend of devices complete after 0.717 msecs

[  104.858645] Disabling non-boot CPUs ...

[  104.863205] PM: noirq resume of devices complete after 0.330 msecs

[  104.869970] PM: early resume of devices complete after 0.396 msecs

After this the system is locked up.

I had experienced a similar issue with yocto 1.6, but when I run a master build from last October, 2014 I get the following output:

PM: Syncing filesystems ... done.

Freezing user space processes ... (elapsed 0.01 seconds) done.

Freezing remaining freezable tasks ... (elapsed 0.01 seconds) done.

PM: suspend of devices complete after 99.514 msecs

PM: suspend devices took 0.100 seconds

PM: late suspend of devices complete after 0.433 msecs

PM: noirq suspend of devices complete after 0.565 msecs

Disabling non-boot CPUs ...

CPU1: shutdown

Enabling non-boot CPUs ...

CPU1: Booted secondary processor

CPU1 is up

PM: noirq resume of devices complete after 0.311 msecs

PM: early resume of devices complete after 0.316 msecs

mxc_sdc_fb fb.26: 1920x1080 h_sync,r,l: 44,88,148  v_sync,l,u: 5,4,36 pixclock=148500000 Hz

PM: resume of devices complete after 190.285 msecs

PM: resume devices took 0.190 seconds

Restarting tasks ... done.

This suspend/resume works.

So, it looks like the feature was working at one point on the development branch.

I can see that in the distribution that works I get a CPU1: shutdown, I don't see this in the yocto 1.7 release.

Does anyone have any ideas what might be causing this issue, or what other steps I might take to track down what might be causing what looks like a lock on shutdown?

Has any one see a similar issue? Is suspend/resume working for others on yocto 1.7?

Any help would be much appreciated.

Thanks,

Ossian

[Yuri]

  From "Linux Reference Manual" of BSP documentation :

"... refer to our script in /unit_tests/suspend_random_auto.sh

or /unit_tests/suspend_quick_auto.sh".

Have a great day,
Yuri

-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------

[ossian · Accepted solution]

Yuri,

Thank you for the suggestion.

After additional investigation it turns out the issue I was having was caused by a device tree configuration issue.

If anyone else experiences a similar issue I would suggest reviewing your regulator settings in your device tree. In particular review the "pu-supply" settings.

Regards,

Ossian