Fix leak: properly destroy TimezoneResolver instance.

Fix leak: properly destroy TimezoneResolver instance.

BUG=416494
TEST=manual

Committed: https://crrev.com/be51cdcb69512a02300ee84f204cadbe0948e9a5
Cr-Commit-Position: refs/heads/master@{#315017}

[Alexander Alekseev, 2015-02-05 14:07:15]

alemate@chromium.org changed reviewers:
+ sky@chromium.org

[Alexander Alekseev, 2015-02-05 14:07:16]

Please review.

[sky, 2015-02-05 17:59:22]

https://codereview.chromium.org/903733002/diff/1/chrome/browser/browser_proce...
File chrome/browser/browser_process_platform_part_chromeos.cc (right):

https://codereview.chromium.org/903733002/diff/1/chrome/browser/browser_proce...
chrome/browser/browser_process_platform_part_chromeos.cc:129:
timezone_resolver_.reset();
timezone_resolver_ is a scoped_ptr, so why is this necessary? And since I'm
asking anyone looking at this code is going to wonder the same thing. In other
words you should add a comment here.

[Alexander Alekseev, 2015-02-05 18:37:19]

https://codereview.chromium.org/903733002/diff/1/chrome/browser/browser_proce...
File chrome/browser/browser_process_platform_part_chromeos.cc (right):

https://codereview.chromium.org/903733002/diff/1/chrome/browser/browser_proce...
chrome/browser/browser_process_platform_part_chromeos.cc:129:
timezone_resolver_.reset();
On 2015/02/05 17:59:22, sky wrote:
> timezone_resolver_ is a scoped_ptr, so why is this necessary? And since I'm
> asking anyone looking at this code is going to wonder the same thing. In other
> words you should add a comment here.

Done.

[sky, 2015-02-06 00:38:44]

LGTM

[Alexander Alekseev, 2015-02-06 12:33:33]

The CQ bit was checked by alemate@chromium.org

[commit-bot: I haz the power, 2015-02-06 12:34:21]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/903733002/40001

[commit-bot: I haz the power, 2015-02-06 12:58:44]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-02-06 12:59:53]

Patchset 3 (id:??) landed as
https://crrev.com/be51cdcb69512a02300ee84f204cadbe0948e9a5
Cr-Commit-Position: refs/heads/master@{#315017}

[sky, 2015-02-06 15:58:24]

On 2015/02/06 12:59:53, I haz the power (commit-bot) wrote:
> Patchset 3 (id:??) landed as
> https://crrev.com/be51cdcb69512a02300ee84f204cadbe0948e9a5
> Cr-Commit-Position: refs/heads/master@{#315017}

I have a followup question here. What is unique about TimeResolver as compared
to the other objects BrowserProcessPlatformPart creates? Why does it need to be
special cased but not the others?

[Alexander Alekseev, 2015-02-06 16:12:18]

On 2015/02/06 15:58:24, sky wrote:
> On 2015/02/06 12:59:53, I haz the power (commit-bot) wrote:
> > Patchset 3 (id:??) landed as
> > https://crrev.com/be51cdcb69512a02300ee84f204cadbe0948e9a5
> > Cr-Commit-Position: refs/heads/master@{#315017}
> 
> I have a followup question here. What is unique about TimeResolver as compared
> to the other objects BrowserProcessPlatformPart creates? Why does it need to
be
> special cased but not the others?

There is nothing special. There are two lifetime scenarios of members of
BrowserProcessPlatformPart.
Some of them are created/destroyed by designated methods:

- |chrome_user_manager_| is created/destroyed by InitializeChromeUserManager() /
DestroyChromeUserManager() .

- |device_disabling_manager_| is created/destroyed by
InitializeDeviceDisablingManager() / ShutdownDeviceDisablingManager() .

On the contrary, |profile_helper_| is created by
BrowserProcessPlatformPart::profile_helper(), and destroyed by
BrowserProcessPlatformPart::StartTearDown().

So all members of BrowserProcessPlatformPart are destroyed (long?) before
BrowserProcessPlatformPart destruction.
Probably my comment should also include |profile_helper_|, but I am not sure
that this is the exact reason for it to be destroyed on StartTearDown().