Restored missing broker initialization.

Restored missing broker initialization.

BUG=134574
TEST=none

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=145335

[Vitaly Buka (NO REVIEWS), 2012-07-03 00:12:13]

https://chromiumcodereview.appspot.com/10690070/diff/2002/chrome/service/serv...
File chrome/service/service_main.cc (left):

https://chromiumcodereview.appspot.com/10690070/diff/2002/chrome/service/serv...
chrome/service/service_main.cc:47:
content::InitializeSandbox(parameters.sandbox_info);
ContentMain already call InitializeSandBox from main_runner->Initialize

[Albert Bodenhamer, 2012-07-03 00:20:54]

https://chromiumcodereview.appspot.com/10690070/diff/2002/content/common/sand...
File content/common/sandbox_init_win.cc (right):

https://chromiumcodereview.appspot.com/10690070/diff/2002/content/common/sand...
content/common/sandbox_init_win.cc:21: sandbox::BrokerServices* broker_services
= sandbox_info->broker_services;
I'm not sure it's correct to InitBrokerServices in all cases.  Maybe instead add
an additional case below?  It currently triggers on process_type = NaCl; should
we just also trigger on process_type = service?

[jschuh, 2012-07-03 03:22:35]

https://chromiumcodereview.appspot.com/10690070/diff/2002/content/common/sand...
File content/common/sandbox_init_win.cc (right):

https://chromiumcodereview.appspot.com/10690070/diff/2002/content/common/sand...
content/common/sandbox_init_win.cc:21: sandbox::BrokerServices* broker_services
= sandbox_info->broker_services;
On 2012/07/03 00:20:54, Albert Bodenhamer wrote:
> I'm not sure it's correct to InitBrokerServices in all cases.  Maybe instead
add
> an additional case below?  It currently triggers on process_type = NaCl;
should
> we just also trigger on process_type = service?

Albert is right. The correct thing to do is branch based on the process type,
not whether or not broker_services is NULL.

[Vitaly Buka (NO REVIEWS), 2012-07-03 05:01:09]

https://chromiumcodereview.appspot.com/10690070/diff/2002/content/common/sand...
File content/common/sandbox_init_win.cc (right):

https://chromiumcodereview.appspot.com/10690070/diff/2002/content/common/sand...
content/common/sandbox_init_win.cc:21: sandbox::BrokerServices* broker_services
= sandbox_info->broker_services;
On 2012/07/03 00:20:54, Albert Bodenhamer wrote:
> I'm not sure it's correct to InitBrokerServices in all cases.  Maybe instead
add
> an additional case below?  It currently triggers on process_type = NaCl;
should
> we just also trigger on process_type = service?

Done.

https://chromiumcodereview.appspot.com/10690070/diff/2002/content/common/sand...
content/common/sandbox_init_win.cc:21: sandbox::BrokerServices* broker_services
= sandbox_info->broker_services;
On 2012/07/03 00:20:54, Albert Bodenhamer wrote:
> I'm not sure it's correct to InitBrokerServices in all cases.  Maybe instead
add
> an additional case below?  It currently triggers on process_type = NaCl;
should
> we just also trigger on process_type = service?

Done.

https://chromiumcodereview.appspot.com/10690070/diff/2002/content/common/sand...
content/common/sandbox_init_win.cc:24: // process to swap its window station.
During this time all the UI will be
I moved this to separate if because Winstation is required by NaCl part

https://chromiumcodereview.appspot.com/10690070/diff/6/content/common/sandbox...
File content/common/sandbox_init_win.cc (right):

https://chromiumcodereview.appspot.com/10690070/diff/6/content/common/sandbox...
content/common/sandbox_init_win.cc:32: // process to swap its window station.
During this time all the UI will be
Moved comment back and removed unnecessary changes.

[jschuh, 2012-07-03 14:41:26]

lgtm

[Albert Bodenhamer, 2012-07-03 15:56:20]

lgtm

[commit-bot: I haz the power, 2012-07-03 17:01:40]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/vitalybuka@chromium.org/10690070/6

[commit-bot: I haz the power, 2012-07-03 18:24:26]

Change committed as 145335