Remove pseudo-idle notification for main frame reload.

Remove pseudo-idle notification for main frame reload.

GC work after page reload introduces unnecessary pause and regresses Octane (see bug).

This CL might increase memory usage in the case when page is reloaded many times within a small timeframe. If there are such regressions, then they should be fixed by adjusting V8 GC heuristics and not sending pseudo-idle notification.

This CL does not regress SunSpider benchmark.

BUG=404020
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=181394

[ulan, 2014-09-04 11:17:34]

ulan@chromium.org changed reviewers:
+ jochen@chromium.org

[ulan, 2014-09-04 11:18:19]

PTAL, as discussed offline this shouldn't regress any benchmarks.

[jochen (gone - plz use gerrit), 2014-09-04 11:20:43]

can you please add a more elaborate description why we're doing this and why
it's safe?

https://codereview.chromium.org/544593002/diff/1/Source/bindings/core/v8/V8GC...
File Source/bindings/core/v8/V8GCForContextDispose.cpp (right):

https://codereview.chromium.org/544593002/diff/1/Source/bindings/core/v8/V8GC...
Source/bindings/core/v8/V8GCForContextDispose.cpp:50: if
(!m_pseudoIdleTimer.isActive())
why start the timer if it's for the main frame?

https://codereview.chromium.org/544593002/diff/1/Source/bindings/core/v8/V8GC...
Source/bindings/core/v8/V8GCForContextDispose.cpp:56: if
(m_pseudoIdleTimer.isActive()) {
same here

[ulan, 2014-09-04 11:41:25]

Updated description.

https://codereview.chromium.org/544593002/diff/1/Source/bindings/core/v8/V8GC...
File Source/bindings/core/v8/V8GCForContextDispose.cpp (right):

https://codereview.chromium.org/544593002/diff/1/Source/bindings/core/v8/V8GC...
Source/bindings/core/v8/V8GCForContextDispose.cpp:50: if
(!m_pseudoIdleTimer.isActive())
On 2014/09/04 11:20:42, jochen wrote:
> why start the timer if it's for the main frame?

Done.

https://codereview.chromium.org/544593002/diff/1/Source/bindings/core/v8/V8GC...
Source/bindings/core/v8/V8GCForContextDispose.cpp:56: if
(m_pseudoIdleTimer.isActive()) {
On 2014/09/04 11:20:42, jochen wrote:
> same here

Done.

[jochen (gone - plz use gerrit), 2014-09-04 11:48:13]

lgtm

[ulan, 2014-09-04 11:49:06]

The CQ bit was checked by ulan@chromium.org

[commit-bot: I haz the power, 2014-09-04 11:49:20]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/ulan@chromium.org/544593002/40001

[commit-bot: I haz the power, 2014-09-04 19:44:25]

Committed patchset #3 (id:40001) as 181394