[Telemetry] Change rasterize_and_record_micro to use top_25_pages page set.

[Telemetry] Change rasterize_and_record_micro to use top_25_pages page set.

Add top_25_smooth.json as archive_data_file for top_25_smooth page set.
top_25_pages page set uses top_25.json.

BUG=418375
Committed: https://crrev.com/636ee5e280e63d5ad63ef0f5cf891840d0e2864a
Cr-Commit-Position: refs/heads/master@{#318711}

[nednguyen, 2015-03-02 02:17:43]

nednguyen@google.com changed reviewers:
+ dtu@chromium.org, sullivan@chromium.org

[sullivan, 2015-03-02 14:23:01]

https://codereview.chromium.org/966283002/diff/1/tools/perf/page_sets/top_25_...
File tools/perf/page_sets/top_25_pages.py (right):

https://codereview.chromium.org/966283002/diff/1/tools/perf/page_sets/top_25_...
tools/perf/page_sets/top_25_pages.py:12: """ Pages hand-picked for 2012 CrOS
scrolling tuning efforts. """
Does this page set scroll? If so, how? If not, is the comment right?

[nednguyen, 2015-03-02 15:35:54]

https://codereview.chromium.org/966283002/diff/1/tools/perf/page_sets/top_25_...
File tools/perf/page_sets/top_25_pages.py (right):

https://codereview.chromium.org/966283002/diff/1/tools/perf/page_sets/top_25_...
tools/perf/page_sets/top_25_pages.py:12: """ Pages hand-picked for 2012 CrOS
scrolling tuning efforts. """
On 2015/03/02 14:23:01, sullivan wrote:
> Does this page set scroll? If so, how? If not, is the comment right?

Good catch, it's not. The fact that I copy top_25_smooth -> top_25 this & the
comment doesn't reflect what those pages do reflects that the model of sharing
page_set across different benchmarks is a broken model. :-/

[sullivan, 2015-03-02 15:47:33]

On 2015/03/02 15:35:54, nednguyen wrote:
>
https://codereview.chromium.org/966283002/diff/1/tools/perf/page_sets/top_25_...
> File tools/perf/page_sets/top_25_pages.py (right):
> 
>
https://codereview.chromium.org/966283002/diff/1/tools/perf/page_sets/top_25_...
> tools/perf/page_sets/top_25_pages.py:12: """ Pages hand-picked for 2012 CrOS
> scrolling tuning efforts. """
> On 2015/03/02 14:23:01, sullivan wrote:
> > Does this page set scroll? If so, how? If not, is the comment right?
> 
> Good catch, it's not. The fact that I copy top_25_smooth -> top_25 this & the
> comment doesn't reflect what those pages do reflects that the model of sharing
> page_set across different benchmarks is a broken model. :-/

I definitely see the problems with sharing the page set and not realizing all
the implications. But as a benchmark author, I find it a really useful feature
to be able to write a benchmark on the exact same content as another benchmark.
It would be great if it were easy to share content but authors needed to be
clearer about page interactions.

[sullivan, 2015-03-02 15:47:38]

lgtm

[nednguyen, 2015-03-02 16:09:08]

The CQ bit was checked by nednguyen@google.com

[commit-bot: I haz the power, 2015-03-02 16:10:21]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/966283002/20001

[nednguyen, 2015-03-02 16:10:26]

On 2015/03/02 15:47:33, sullivan wrote:
> On 2015/03/02 15:35:54, nednguyen wrote:
> >
>
https://codereview.chromium.org/966283002/diff/1/tools/perf/page_sets/top_25_...
> > File tools/perf/page_sets/top_25_pages.py (right):
> > 
> >
>
https://codereview.chromium.org/966283002/diff/1/tools/perf/page_sets/top_25_...
> > tools/perf/page_sets/top_25_pages.py:12: """ Pages hand-picked for 2012 CrOS
> > scrolling tuning efforts. """
> > On 2015/03/02 14:23:01, sullivan wrote:
> > > Does this page set scroll? If so, how? If not, is the comment right?
> > 
> > Good catch, it's not. The fact that I copy top_25_smooth -> top_25 this &
the
> > comment doesn't reflect what those pages do reflects that the model of
sharing
> > page_set across different benchmarks is a broken model. :-/
> 
> I definitely see the problems with sharing the page set and not realizing all
> the implications. But as a benchmark author, I find it a really useful feature
> to be able to write a benchmark on the exact same content as another
benchmark.
> It would be great if it were easy to share content but authors needed to be
> clearer about page interactions.

Yeah, that's why we are moving toward "one page = 1 userflow" as in bug/418375.
:)

[commit-bot: I haz the power, 2015-03-02 17:04:25]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-03-02 17:05:04]

Patchset 2 (id:??) landed as
https://crrev.com/636ee5e280e63d5ad63ef0f5cf891840d0e2864a
Cr-Commit-Position: refs/heads/master@{#318711}