Move check for suspended displays out of common path

Move check for suspended displays out of common path

Checking if the displays are suspended in
DisplayConfigurator::RunPendingConfiguration is exposing a race
condition in the interaction between chrome and powerd.  The original
reason for adding the check was to ensure that chrome did not attempt to
perform any modesets while the chrome os device was in a dark resume.
Since this really only happens when the display configuration changes,
we can move the check into OnConfigurationChanged.  Also, to deal with
the possbility that the configuration timer was started right before the
system suspended, stop the timer in SuspendDisplays.  The
DisplayConfigurator will force a probe and re-configuration on resume
anyway so stopping the timer at suspend time will not really make a
difference.

BUG=chrome-os-partner:36723, 459783
Committed: https://crrev.com/797288279c906fc0b8d89aec55bad18ad975f187
Cr-Commit-Position: refs/heads/master@{#316918}

[Chirantan Ekbote, 2015-02-18 21:04:58]

chirantan@chromium.org changed reviewers:
+ derat@chromium.org, dnicoara@chromium.org

[Chirantan Ekbote, 2015-02-18 21:04:59]

Please take a look.  This change will block OnConfigurationChanged events from
doing anything with the displays suspended but will not block SetDisplayPower or
SetDisplayMode from doing so.

[Daniel Erat, 2015-02-18 21:12:38]

looks fine, but can you also update display_configurator_unittest.cc to cover
the change?

[Chirantan Ekbote, 2015-02-18 21:23:30]

On 2015/02/18 21:12:38, Daniel Erat wrote:
> looks fine, but can you also update display_configurator_unittest.cc to cover
> the change?

I knew I forgot something :-/  New patch set has the updated unittests.

[dnicoara, 2015-02-18 21:27:27]

lgtm

[Daniel Erat, 2015-02-18 21:28:50]

can you add a call to SetDisplayMode() while "suspended" to make sure it goes
through?

[Chirantan Ekbote, 2015-02-18 22:49:26]

New patchsets have been uploaded after l-g-t-m from dnicoara@chromium.org

[Chirantan Ekbote, 2015-02-18 22:50:59]

On 2015/02/18 21:28:50, Daniel Erat wrote:
> can you add a call to SetDisplayMode() while "suspended" to make sure it goes
> through?

Done.

[Daniel Erat, 2015-02-18 22:51:47]

lgtm

thanks!

[Chirantan Ekbote, 2015-02-18 22:57:15]

The CQ bit was checked by chirantan@chromium.org

[commit-bot: I haz the power, 2015-02-18 22:58:08]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/932203002/40001

[commit-bot: I haz the power, 2015-02-18 23:41:26]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-02-18 23:42:22]

Patchset 3 (id:??) landed as
https://crrev.com/797288279c906fc0b8d89aec55bad18ad975f187
Cr-Commit-Position: refs/heads/master@{#316918}