Control process backgrounding from within the child process on Windows.

Control process backgrounding from within the child process on Windows.

Windows has a fancy backgrounding mode which lowers IO priority.
Sadly it can only be set from within the process to be backgrounded.
So add a new ChildProcessMsg to tell child processes when they should
enter and leave background mode.

This reduces glitching during background tab load. This should only be
landed along with https://codereview.chromium.org/298253004/ which
removes backgrounding for active audio tabs.

I expect this will reduce CPU and I/O load significantly for background
processes which will lead to power savings.

This is only done on Windows since ChromeOS / Linux require privileges
only available to the browser process for setting background mode.

BUG=362294
TEST=Play audio. Load background tabs. Observe less/no glitching.
R=jam@chromium.org, luken@chromium.org, nasko@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=274071

[DaleCurtis, 2014-05-28 23:00:09]

luken: General review (per cpu@).
jam: content/ approval.
nasko: content/common/child_messages.h approval.

[nasko, 2014-05-28 23:08:37]

LGTM with a nit.

https://codereview.chromium.org/305533006/diff/30001/content/child/child_thre...
File content/child/child_thread.cc (right):

https://codereview.chromium.org/305533006/diff/30001/content/child/child_thre...
content/child/child_thread.cc:557:
base::Process::Current().SetProcessBackgrounded(background);
Out of curiosity, does that work from inside the sandbox?

https://codereview.chromium.org/305533006/diff/30001/content/common/child_pro...
File content/common/child_process_messages.h (right):

https://codereview.chromium.org/305533006/diff/30001/content/common/child_pro...
content/common/child_process_messages.h:117: bool /* should background? */)
nit: the comment is usually the parameter name in the handler method.

[DaleCurtis, 2014-05-29 00:06:59]

https://codereview.chromium.org/305533006/diff/30001/content/child/child_thre...
File content/child/child_thread.cc (right):

https://codereview.chromium.org/305533006/diff/30001/content/child/child_thre...
content/child/child_thread.cc:557:
base::Process::Current().SetProcessBackgrounded(background);
On 2014/05/28 23:08:37, nasko wrote:
> Out of curiosity, does that work from inside the sandbox?

Yup, I've confirmed the cpu, io priorities of the processes backgrounded have
gone from "6 (below normal), normal"  to "4 (idle), very low"

https://codereview.chromium.org/305533006/diff/30001/content/common/child_pro...
File content/common/child_process_messages.h (right):

https://codereview.chromium.org/305533006/diff/30001/content/common/child_pro...
content/common/child_process_messages.h:117: bool /* should background? */)
On 2014/05/28 23:08:37, nasko wrote:
> nit: the comment is usually the parameter name in the handler method.

Done.

[jam, 2014-05-29 17:28:58]

lgtm

[luken, 2014-05-30 17:57:22]

lgtm

[DaleCurtis, 2014-05-30 19:41:56]

The CQ bit was checked by dalecurtis@chromium.org

[commit-bot: I haz the power, 2014-05-30 19:43:08]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/dalecurtis@chromium.org/305533006/50001

[commit-bot: I haz the power, 2014-05-31 01:20:21]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  win_chromium_rel on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/win_chromium_rel/buil...)

[commit-bot: I haz the power, 2014-05-31 03:38:16]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-05-31 03:38:16]

Try jobs failed on following builders:
  win_chromium_rel on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/win_chromium_rel/buil...)

[DaleCurtis, 2014-05-31 04:05:26]

I had to disable the ChromeRenderProcessHostTest.Backgrounding test since there
doesn't appear to be a clean way to wait for renderers to process IPC messages. 
I'll send out a follow up CL to re-enable the test once I can do so in a
non-flaky manner.

[DaleCurtis, 2014-05-31 04:06:09]

The CQ bit was checked by dalecurtis@chromium.org

[commit-bot: I haz the power, 2014-05-31 04:08:29]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/dalecurtis@chromium.org/305533006/70001

[commit-bot: I haz the power, 2014-05-31 07:40:12]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  win_chromium_rel on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/win_chromium_rel/buil...)

[DaleCurtis, 2014-05-31 22:26:39]

Committed patchset #3 manually as r274071 (presubmit successful).