Consolidate RecordBrowserStartupTime() into one location

Consolidate RecordBrowserStartupTime() into one location

* Create a CurrentProcessInfo class as a central storage place for information about the current process.  For now store the process creation time there.
* Move RecordBrowserStartupTime() int browser_main.cc.

The motivation for this is that we will want to add some more UMA measurements which require knowledge of the process startup time.

BUG=None
Test=Chrome should continue to compile on all platforms.
TEST=


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=142984

[Avi (use Gerrit), 2012-06-15 21:13:27]

lgtm

[jeremy, 2012-06-15 21:15:30]

sky: owner review for chrome/browser
mark: owner review for base/

Thanks!

[Mark Mentovai, 2012-06-15 21:18:57]

https://chromiumcodereview.appspot.com/10561009/diff/1/base/process_info_mac.cc
File base/process_info_mac.cc (right):

https://chromiumcodereview.appspot.com/10561009/diff/1/base/process_info_mac....
base/process_info_mac.cc:36: return &process_creation_time;
Oh no! This is no good at all.

https://chromiumcodereview.appspot.com/10561009/diff/1/base/process_info_win.cc
File base/process_info_win.cc (right):

https://chromiumcodereview.appspot.com/10561009/diff/1/base/process_info_win....
base/process_info_win.cc:29: return &process_creation_time;
Once again. Looks like you chose a poor interface for what you’re trying to do.

[jeremy, 2012-06-19 15:40:34]

mark: Ready for another look.

[jeremy, 2012-06-19 15:44:20]

https://chromiumcodereview.appspot.com/10561009/diff/6001/base/process_info_w...
File base/process_info_win.cc (right):

https://chromiumcodereview.appspot.com/10561009/diff/6001/base/process_info_w...
base/process_info_win.cc:22: &ignore);
Looks like this function can fail too, so I need to make this match the Mac
version...

[Mark Mentovai, 2012-06-19 15:47:08]

https://chromiumcodereview.appspot.com/10561009/diff/6001/base/process_info_m...
File base/process_info_mac.cc (right):

https://chromiumcodereview.appspot.com/10561009/diff/6001/base/process_info_m...
base/process_info_mac.cc:31: *process_creation_time =
Can you clean this up and get rid of the second static by doing:

  Time* process_creation_time = new Time();
  *process_creation_time = FromTimeVal(…);
  return process_creation_time;

https://chromiumcodereview.appspot.com/10561009/diff/6001/base/process_info_m...
base/process_info_mac.cc:32:
base::Time::FromTimeVal(proc->kp_proc.p_un.__p_starttime);
Is the base:: qualifier necessary, since you’re using base::Time above on line
17?

https://chromiumcodereview.appspot.com/10561009/diff/6001/base/process_info_m...
base/process_info_mac.cc:43: static bool success =
ProcessCreationTimeInternal(&process_creation_time);
These are the two statics I’m talking about. You ought to be able to do this
with one.

[jeremy, 2012-06-19 16:03:59]

All fixed, ready for another look.

[Mark Mentovai, 2012-06-19 16:08:00]

https://chromiumcodereview.appspot.com/10561009/diff/19001/base/process_info_...
File base/process_info_mac.cc (right):

https://chromiumcodereview.appspot.com/10561009/diff/19001/base/process_info_...
base/process_info_mac.cc:40: static Time process_creation_time =
ProcessCreationTimeInternal();
Did “new Time()” and using the Time* interface to the internal function not
work?

This (pointerless interface) leaves open the possibility of the static
destructor. That’s no good.

It also doesn’t match the documented behavior of returning NULL on error. It
instead returns Time(0). NULL is better.

https://chromiumcodereview.appspot.com/10561009/diff/19001/base/process_info_...
File base/process_info_win.cc (right):

https://chromiumcodereview.appspot.com/10561009/diff/19001/base/process_info_...
base/process_info_win.cc:34: static Time process_creation_time =
ProcessCreationTimeInternal();
As in the Mac file.

[jeremy, 2012-06-19 16:29:48]

All fixed, ready for another look

[Mark Mentovai, 2012-06-19 16:32:44]

LGTM

[commit-bot: I haz the power, 2012-06-19 16:33:56]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jeremy@chromium.org/10561009/7013

[commit-bot: I haz the power, 2012-06-19 17:43:38]

Change committed as 142984