Improve multi-process debugging.

Improve multi-process debugging.

1) --wait-for-debugger alone makes all child processes wait for a debugger
2) --wait-for-debugger=app1,app2... make only these apps wait for a debugger
3) the app name is added to the command line to make it easier to attach debugger after the process starts
4) MessageBox is used on Windows like chrome

BUG=478251
Committed: https://crrev.com/b41c6df16de616cbbb6d2a436f513458c0d22c38
Cr-Commit-Position: refs/heads/master@{#326676}

[jam, 2015-04-23 21:23:24]

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

[sky, 2015-04-23 21:48:54]

LGTM

https://codereview.chromium.org/1107633002/diff/1/mojo/shell/child_process.cc
File mojo/shell/child_process.cc (right):

https://codereview.chromium.org/1107633002/diff/1/mojo/shell/child_process.cc...
mojo/shell/child_process.cc:292: << " waiting for GDB.";
Can you add the process id to this output.

https://codereview.chromium.org/1107633002/diff/1/mojo/shell/child_process_ho...
File mojo/shell/child_process_host.cc (right):

https://codereview.chromium.org/1107633002/diff/1/mojo/shell/child_process_ho...
mojo/shell/child_process_host.cc:107:
child_command_line.AppendSwitchASCII(switches::kApp, name_);
Should you only do this if name_ is not empty?

https://codereview.chromium.org/1107633002/diff/1/mojo/shell/child_process_ho...
File mojo/shell/child_process_host.h (right):

https://codereview.chromium.org/1107633002/diff/1/mojo/shell/child_process_ho...
mojo/shell/child_process_host.h:69: std::string name_;
nit: const

[jam, 2015-04-23 22:18:44]

https://codereview.chromium.org/1107633002/diff/1/mojo/shell/child_process.cc
File mojo/shell/child_process.cc (right):

https://codereview.chromium.org/1107633002/diff/1/mojo/shell/child_process.cc...
mojo/shell/child_process.cc:292: << " waiting for GDB.";
On 2015/04/23 21:48:54, sky wrote:
> Can you add the process id to this output.

Done.

https://codereview.chromium.org/1107633002/diff/1/mojo/shell/child_process_ho...
File mojo/shell/child_process_host.cc (right):

https://codereview.chromium.org/1107633002/diff/1/mojo/shell/child_process_ho...
mojo/shell/child_process_host.cc:107:
child_command_line.AppendSwitchASCII(switches::kApp, name_);
On 2015/04/23 21:48:54, sky wrote:
> Should you only do this if name_ is not empty?

I don't expect name_ to be empty. i added a dcheck

https://codereview.chromium.org/1107633002/diff/1/mojo/shell/child_process_ho...
File mojo/shell/child_process_host.h (right):

https://codereview.chromium.org/1107633002/diff/1/mojo/shell/child_process_ho...
mojo/shell/child_process_host.h:69: std::string name_;
On 2015/04/23 21:48:54, sky wrote:
> nit: const

Done.

[jam, 2015-04-23 22:18:48]

The CQ bit was checked by jam@chromium.org

[jam, 2015-04-23 22:18:48]

The patchset sent to the CQ was uploaded after l-g-t-m from sky@chromium.org
Link to the patchset: https://codereview.chromium.org/1107633002/#ps20001
(title: "review comments")

[commit-bot: I haz the power, 2015-04-23 22:19:29]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1107633002/20001

[commit-bot: I haz the power, 2015-04-23 22:42:44]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-04-23 22:43:50]

Patchset 2 (id:??) landed as
https://crrev.com/b41c6df16de616cbbb6d2a436f513458c0d22c38
Cr-Commit-Position: refs/heads/master@{#326676}