Don't crash service process on start initializing V8

Don't crash service process on start initializing V8.

https://codereview.chromium.org/944913002 wrongly assumes that
non empty --type means a child process.

This is a quick fix for merging into Chrome 42. We need better solution for 43. Content does not know about "service" process and does not include "service" constant from chrome.

BUG=466917
Committed: https://crrev.com/1a16f9bf016a616d12be7003ebe84fb1aebc748e
Cr-Commit-Position: refs/heads/master@{#323758}

[Vitaly Buka (NO REVIEWS), 2015-04-03 07:04:05]

vitalybuka@chromium.org changed reviewers:
+ agl@chromium.org, jochen@chromium.org, rmcilroy@chromium.org

[Vitaly Buka (NO REVIEWS), 2015-04-03 07:06:01]

vitalybuka@chromium.org changed reviewers:
+ amineer@chromium.org

[Vitaly Buka (NO REVIEWS), 2015-04-03 07:06:01]

PTAL

[rmcilroy, 2015-04-03 08:01:25]

On 2015/04/03 07:06:01, Vitaly Buka wrote:
> PTAL

I added a comment to the bug. I don't think this will fix the original bug, it
will only allow the service process to be able to launch standalone (which is
should never need to do since it's always launched from the browser process,
which should be fine assuming it uses ChildProcessLauncher).

However, if this does fix the issue i'm fine with it landing as a temporary
measure, just be aware it may cause crashes in the service process during Chrome
updates as the original bug described.

[Vitaly Buka (NO REVIEWS), 2015-04-03 15:50:44]

On 2015/04/03 08:01:25, rmcilroy wrote:
> On 2015/04/03 07:06:01, Vitaly Buka wrote:
> > PTAL
> 
> I added a comment to the bug. I don't think this will fix the original bug, it
> will only allow the service process to be able to launch standalone (which is
> should never need to do since it's always launched from the browser process,
> which should be fine assuming it uses ChildProcessLauncher).
> 
> However, if this does fix the issue i'm fine with it landing as a temporary
> measure, just be aware it may cause crashes in the service process during
Chrome
> updates as the original bug described.

Probably service is running not as regular child process.

[Vitaly Buka (NO REVIEWS), 2015-04-03 15:57:05]

vitalybuka@chromium.org changed reviewers:
+ amineer1. avi@chromium.org
- amineer@chromium.org

[Vitaly Buka (NO REVIEWS), 2015-04-03 15:58:20]

On 2015/04/03 15:50:44, Vitaly Buka wrote:
> On 2015/04/03 08:01:25, rmcilroy wrote:
> > On 2015/04/03 07:06:01, Vitaly Buka wrote:
> > > PTAL
> > 
> > I added a comment to the bug. I don't think this will fix the original bug,
it
> > will only allow the service process to be able to launch standalone (which
is
> > should never need to do since it's always launched from the browser process,
> > which should be fine assuming it uses ChildProcessLauncher).
> > 
> > However, if this does fix the issue i'm fine with it landing as a temporary
> > measure, just be aware it may cause crashes in the service process during
> Chrome
> > updates as the original bug described.
> 
> Probably service is running not as regular child process.

And I am sure, it resolves the issue.

[Avi (use Gerrit), 2015-04-03 15:59:54]

avi@chromium.org changed reviewers:
+ avi@chromium.org

[Avi (use Gerrit), 2015-04-03 15:59:54]

LGTM stamp

[Vitaly Buka (NO REVIEWS), 2015-04-03 16:03:05]

The CQ bit was checked by vitalybuka@chromium.org

[commit-bot: I haz the power, 2015-04-03 16:03:16]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1055413002/1

[commit-bot: I haz the power, 2015-04-03 17:15:08]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2015-04-03 20:38:26]

Patchset 1 (id:??) landed as
https://crrev.com/1a16f9bf016a616d12be7003ebe84fb1aebc748e
Cr-Commit-Position: refs/heads/master@{#323758}