Make net use v8 through gin

Make net use v8 through gin

- no longer try to use the default isolate (we want to remove it from v8)
- add the option to gin to manage an isolate in non-strict mode

BUG=359977
R=eroman@chromium.org,abarth@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=262559

[abarth-chromium, 2014-04-07 17:13:55]

LGTM

https://codereview.chromium.org/227233006/diff/20001/net/proxy/proxy_resolver...
File net/proxy/proxy_resolver_v8.cc (right):

https://codereview.chromium.org/227233006/diff/20001/net/proxy/proxy_resolver...
net/proxy/proxy_resolver_v8.cc:772: g_default_isolate_ =
Should we rename this variable?  It's now the proxy_resolver_isolate.

[jochen (gone - plz use gerrit), 2014-04-07 20:02:19]

On 2014/04/07 17:13:55, abarth wrote:
> LGTM
> 
>
https://codereview.chromium.org/227233006/diff/20001/net/proxy/proxy_resolver...
> File net/proxy/proxy_resolver_v8.cc (right):
> 
>
https://codereview.chromium.org/227233006/diff/20001/net/proxy/proxy_resolver...
> net/proxy/proxy_resolver_v8.cc:772: g_default_isolate_ =
> Should we rename this variable?  It's now the proxy_resolver_isolate.

done

[jochen (gone - plz use gerrit), 2014-04-08 17:30:25]

Eric, could you please review the changes under net/?

[eroman, 2014-04-08 19:47:24]

LGTM.

What is the memory impact during the transition? (i.e. is there still a default
isolate being statically initialized, and additionally the proxy resolver one?)

https://codereview.chromium.org/227233006/diff/40001/net/proxy/proxy_resolver...
File net/proxy/proxy_resolver_v8.cc (right):

https://codereview.chromium.org/227233006/diff/40001/net/proxy/proxy_resolver...
net/proxy/proxy_resolver_v8.cc:528: LOG(ERROR) << line_number << ": " <<
error_message;
Is this log addition intentional? Wander if it could lead to extra log spam. If
this is for debugging unittests I think it should be done by the
"js_bindings()->OnError()" instead.

[jochen (gone - plz use gerrit), 2014-04-08 20:43:39]

It's more or less one TLS key and an "v8::internal::Isolate" object that gets
wasted (note that the isolate object isn't initialized, so we don't allocate
memory for the heap or anything)

https://codereview.chromium.org/227233006/diff/40001/net/proxy/proxy_resolver...
File net/proxy/proxy_resolver_v8.cc (right):

https://codereview.chromium.org/227233006/diff/40001/net/proxy/proxy_resolver...
net/proxy/proxy_resolver_v8.cc:528: LOG(ERROR) << line_number << ": " <<
error_message;
On 2014/04/08 19:47:24, eroman wrote:
> Is this log addition intentional? Wander if it could lead to extra log spam.
If
> this is for debugging unittests I think it should be done by the
> "js_bindings()->OnError()" instead.

oops, that's left over from debugging.

[jochen (gone - plz use gerrit), 2014-04-08 20:44:04]

The CQ bit was checked by jochen@chromium.org

[commit-bot: I haz the power, 2014-04-08 20:44:24]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jochen@chromium.org/227233006/60001

[commit-bot: I haz the power, 2014-04-08 23:58:17]

Change committed as 262559