Yield HTMLDocumentParser after executing parser blocking script

Yield HTMLDocumentParser after executing parser blocking script

Example trace screenshot: https://drive.google.com/open?id=0B5o-0Snm1_rFUG1QYU9kR1RSVVk
Example trace: https://drive.google.com/open?id=0B5o-0Snm1_rFM3FmUGVJTE9NUTg

BUG=TBC

[kouhei (in TOK), 2016-07-11 07:33:23]

kouhei@chromium.org changed reviewers:
+ csharrison@chromium.org

[kouhei (in TOK), 2016-07-11 07:33:23]

wip

[kouhei (in TOK), 2016-07-11 07:33:28]

The CQ bit was checked by kouhei@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-07-11 07:33:34]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[kouhei (in TOK), 2016-07-11 08:13:14]

Description was changed from

==========
Yield HTMLDocumentParser after executing parser blocking script

BUG=TBC
==========

to

==========
Yield HTMLDocumentParser after executing parser blocking script

Example trace screenshot:
https://drive.google.com/open?id=0B5o-0Snm1_rFbTlKcjZkdGN3eEk
Example trace: https://drive.google.com/open?id=0B5o-0Snm1_rFM3FmUGVJTE9NUTg

BUG=TBC
==========

[kouhei (in TOK), 2016-07-11 08:13:40]

Description was changed from

==========
Yield HTMLDocumentParser after executing parser blocking script

Example trace screenshot:
https://drive.google.com/open?id=0B5o-0Snm1_rFbTlKcjZkdGN3eEk
Example trace: https://drive.google.com/open?id=0B5o-0Snm1_rFM3FmUGVJTE9NUTg

BUG=TBC
==========

to

==========
Yield HTMLDocumentParser after executing parser blocking script

Example trace screenshot:
https://drive.google.com/open?id=0B5o-0Snm1_rFUG1QYU9kR1RSVVk
Example trace: https://drive.google.com/open?id=0B5o-0Snm1_rFM3FmUGVJTE9NUTg

BUG=TBC
==========

[Charlie Harrison, 2016-07-11 12:39:58]

The idea sgtm. I'm nervous about regressions in the short term but in the long
term I think rescheduling parsing is the best approach.

[kouhei (in TOK), 2016-07-11 13:02:45]

On 2016/07/11 12:39:58, csharrison wrote:
> The idea sgtm. I'm nervous about regressions in the short term but in the long
> term I think rescheduling parsing is the best approach.

I'm surprised to see that this regresses both ttfmp/ttfcp. Let me take a closer
look tomorrow.
win_perf_bisect:
http://storage.googleapis.com/chromium-telemetry/html-results/results-2016-07...

[Charlie Harrison, 2016-07-11 13:24:27]

On 2016/07/11 13:02:45, kouhei wrote:
> On 2016/07/11 12:39:58, csharrison wrote:
> > The idea sgtm. I'm nervous about regressions in the short term but in the
long
> > term I think rescheduling parsing is the best approach.
> 
> I'm surprised to see that this regresses both ttfmp/ttfcp. Let me take a
closer
> look tomorrow.
> win_perf_bisect:
>
http://storage.googleapis.com/chromium-telemetry/html-results/results-2016-07...

I bet non-essential work is descheduling parsing. I'd love to see some trace
analysis of these regressions.

[Charlie Harrison, 2017-05-06 01:49:12]

Removing myself from listed reviewer to clean up my queue. Please add me back if
you still need a review.

[Charlie Harrison, 2017-05-06 01:54:48]

csharrison@chromium.org changed reviewers:
- csharrison@chromium.org