Stop deferring commits on body tag insertion if there are no pending stylesheet loads.

Stop deferring commits on body tag insertion if there are no pending stylesheet loads.

At that point we can be fairly sure we'll have something accurate to paint soon, and eliminates long delays on first paint for pages that don't otherwise force any layouts until much later (like gws).

The linked bug has performance numbers, but note that there's a possibility this may regress zero-latency (i.e. no netsim) page cycler times. The tradeoff is well worth it however.

R=jamesr@chromium.org
BUG=344537
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=167465

[oystein (OOO til 10th of July), 2014-02-11 22:11:11]

James: this is the thing we talked about the other day. So far I haven't seen
any regressions in my testing, but pmeenan@ is doing some more comprehensive
testing of this later today (Android and PC) so we'll know for sure.

[jamesr, 2014-02-11 22:13:39]

This seems reasonable, but we should have some perf data to justify it and be
able to re-evaluate in the future.  How does it behave on our page cyclers? 
Could you file a tracking bug to dump any results into / track any fallout we
observe?

[oystein (OOO til 10th of July), 2014-02-11 22:21:25]

On 2014/02/11 22:13:39, jamesr wrote:
> This seems reasonable, but we should have some perf data to justify it and be
> able to re-evaluate in the future.  How does it behave on our page cyclers? 
> Could you file a tracking bug to dump any results into / track any fallout we
> observe?

Sure thing, I'll file one once Pat's results are in and update this.

[oystein (OOO til 10th of July), 2014-02-18 17:39:36]

On 2014/02/11 22:21:25, Oystein wrote:
> On 2014/02/11 22:13:39, jamesr wrote:
> > This seems reasonable, but we should have some perf data to justify it and
be
> > able to re-evaluate in the future.  How does it behave on our page cyclers? 
> > Could you file a tracking bug to dump any results into / track any fallout
we
> > observe?
> 
> Sure thing, I'll file one once Pat's results are in and update this.

James: Bug filed with results here:
https://code.google.com/p/chromium/issues/detail?id=344537

[oystein (OOO til 10th of July), 2014-02-19 20:57:32]

On 2014/02/18 17:39:36, Oystein wrote:
> On 2014/02/11 22:21:25, Oystein wrote:
> > On 2014/02/11 22:13:39, jamesr wrote:
> > > This seems reasonable, but we should have some perf data to justify it and
> be
> > > able to re-evaluate in the future.  How does it behave on our page
cyclers? 
> > > Could you file a tracking bug to dump any results into / track any fallout
> we
> > > observe?
> > 
> > Sure thing, I'll file one once Pat's results are in and update this.
> 
> James: Bug filed with results here:
> https://code.google.com/p/chromium/issues/detail?id=344537

Ping :)

[jamesr, 2014-02-19 21:11:16]

Those look really good! lgtm and let's keep an eye in it.  Make sure
there's enough info on the bug for the next person who comes along to repro
your results.

To unsubscribe from this group and stop receiving emails from it, send an email
to blink-reviews+unsubscribe@chromium.org.

[oystein (OOO til 10th of July), 2014-02-19 21:45:46]

The CQ bit was checked by oysteine@chromium.org

[commit-bot: I haz the power, 2014-02-19 21:59:40]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/oysteine@chromium.org/157963002/30001

[commit-bot: I haz the power, 2014-02-19 23:05:18]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/oysteine@chromium.org/157963002/30001

[commit-bot: I haz the power, 2014-02-20 07:19:27]

Change committed as 167465