[DevTools] Port forwarding doesn't keep USB connection

[DevTools] Port forwarding doesn't keep USB connection

That means that when chrome://inspect is closed device is released.

BUG=387067
Committed: https://crrev.com/851cfc61ae374c3ff134a00d279c583078c32fee
Cr-Commit-Position: refs/heads/master@{#300667}

[vkuzkokov, 2014-10-20 13:47:56]

vkuzkokov@chromium.org changed reviewers:
+ dgozman@chromium.org

[dgozman, 2014-10-21 11:54:02]

https://codereview.chromium.org/644963003/diff/1/chrome/browser/devtools/devi...
File chrome/browser/devtools/device/android_device_manager.cc (right):

https://codereview.chromium.org/644963003/diff/1/chrome/browser/devtools/devi...
chrome/browser/devtools/device/android_device_manager.cc:394:
socket->Terminate();
I think this should issue Delegate::OnSocketClosed, but it doesn't.

https://codereview.chromium.org/644963003/diff/1/chrome/browser/devtools/devi...
File chrome/browser/devtools/device/android_web_socket.cc (right):

https://codereview.chromium.org/644963003/diff/1/chrome/browser/devtools/devi...
chrome/browser/devtools/device/android_web_socket.cc:198:
device_->device_message_loop_->DeleteSoon(FROM_HERE, socket_impl_);
I was about to ask whether this message loop could die together with last
device. But it's actually a MessageLoopProxy. Perhaps, we should rename the
field.

https://codereview.chromium.org/644963003/diff/1/chrome/browser/devtools/devi...
File chrome/browser/devtools/device/port_forwarding_controller.cc (right):

https://codereview.chromium.org/644963003/diff/1/chrome/browser/devtools/devi...
chrome/browser/devtools/device/port_forwarding_controller.cc:321:
device_(device.get()),
I think you either need |device_| or |browser_|, but not both.

[vkuzkokov, 2014-10-21 12:45:32]

Patchset #2 (id:20001) has been deleted

[vkuzkokov, 2014-10-21 13:00:54]

https://codereview.chromium.org/644963003/diff/1/chrome/browser/devtools/devi...
File chrome/browser/devtools/device/android_device_manager.cc (right):

https://codereview.chromium.org/644963003/diff/1/chrome/browser/devtools/devi...
chrome/browser/devtools/device/android_device_manager.cc:394:
socket->Terminate();
On 2014/10/21 11:54:02, dgozman wrote:
> I think this should issue Delegate::OnSocketClosed, but it doesn't.

It should. Fixed.

https://codereview.chromium.org/644963003/diff/1/chrome/browser/devtools/devi...
File chrome/browser/devtools/device/android_web_socket.cc (right):

https://codereview.chromium.org/644963003/diff/1/chrome/browser/devtools/devi...
chrome/browser/devtools/device/android_web_socket.cc:198:
device_->device_message_loop_->DeleteSoon(FROM_HERE, socket_impl_);
On 2014/10/21 11:54:02, dgozman wrote:
> I was about to ask whether this message loop could die together with last
> device. But it's actually a MessageLoopProxy. Perhaps, we should rename the
> field.

Done.

https://codereview.chromium.org/644963003/diff/1/chrome/browser/devtools/devi...
File chrome/browser/devtools/device/port_forwarding_controller.cc (right):

https://codereview.chromium.org/644963003/diff/1/chrome/browser/devtools/devi...
chrome/browser/devtools/device/port_forwarding_controller.cc:321:
device_(device.get()),
On 2014/10/21 11:54:02, dgozman wrote:
> I think you either need |device_| or |browser_|, but not both.

We store browser_ so that we could display on chrome://inspect which one is
used. device_ is needed to open forwarding sockets.

[dgozman, 2014-10-21 16:22:34]

lgtm

[vkuzkokov, 2014-10-22 11:07:40]

The CQ bit was checked by vkuzkokov@chromium.org

[commit-bot: I haz the power, 2014-10-22 11:08:37]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/644963003/80001

[commit-bot: I haz the power, 2014-10-22 12:16:09]

Committed patchset #4 (id:80001)

[commit-bot: I haz the power, 2014-10-22 12:16:55]

Patchset 4 (id:??) landed as
https://crrev.com/851cfc61ae374c3ff134a00d279c583078c32fee
Cr-Commit-Position: refs/heads/master@{#300667}