[telemetry] Add run_benchmark benchmarks for everything that runs on the bots.

[telemetry] Add run_benchmark benchmarks for everything that runs on the bots.


BUG=237412
TEST=None.
R=tonyg@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=212642

[dtu, 2013-07-18 01:11:41]

I went through
https://code.google.com/p/chromium/codesearch#chromium/tools/build/scripts/ma...
and converted every entry into a benchmark. Some have long or weird names.

[tonyg, 2013-07-18 01:39:00]

https://codereview.chromium.org/19721002/diff/1/tools/perf/benchmarks/blink_p...
File tools/perf/benchmarks/blink_perf.py (right):

https://codereview.chromium.org/19721002/diff/1/tools/perf/benchmarks/blink_p...
tools/perf/benchmarks/blink_perf.py:19: def _AddPage(path):
I assume all this is duplicated with blink_perf.py? If so, can we avoid the
duplication?

https://codereview.chromium.org/19721002/diff/1/tools/perf/benchmarks/dromaeo.py
File tools/perf/benchmarks/dromaeo.py (right):

https://codereview.chromium.org/19721002/diff/1/tools/perf/benchmarks/dromaeo...
tools/perf/benchmarks/dromaeo.py:9: class DromaeoDomcoreattr(test.Test):
Let's capitalize the last word:
DromaeoDomcoreAttr
DromaeoDomcoreModify
...
DromaeoJslibAttrJquery
DromaeoJslibAttrPrototype
...

https://codereview.chromium.org/19721002/diff/1/tools/perf/benchmarks/image_d...
File tools/perf/benchmarks/image_decoding.py (right):

https://codereview.chromium.org/19721002/diff/1/tools/perf/benchmarks/image_d...
tools/perf/benchmarks/image_decoding.py:11: page_set =
'page_sets/image_decoding_measurement.json'
Is this a good place for a TODO about the renaming?

https://codereview.chromium.org/19721002/diff/1/tools/perf/benchmarks/page_cy...
File tools/perf/benchmarks/page_cycler.py (right):

https://codereview.chromium.org/19721002/diff/1/tools/perf/benchmarks/page_cy...
tools/perf/benchmarks/page_cycler.py:9: class PageCyclerMoz(test.Test):
Can we alphabetize the cyclers in this file?

[dtu, 2013-07-18 02:17:05]

https://codereview.chromium.org/19721002/diff/1/tools/perf/benchmarks/blink_p...
File tools/perf/benchmarks/blink_perf.py (right):

https://codereview.chromium.org/19721002/diff/1/tools/perf/benchmarks/blink_p...
tools/perf/benchmarks/blink_perf.py:19: def _AddPage(path):
On 2013/07/18 01:39:00, tonyg wrote:
> I assume all this is duplicated with blink_perf.py? If so, can we avoid the
> duplication?

Done.

https://codereview.chromium.org/19721002/diff/1/tools/perf/benchmarks/dromaeo.py
File tools/perf/benchmarks/dromaeo.py (right):

https://codereview.chromium.org/19721002/diff/1/tools/perf/benchmarks/dromaeo...
tools/perf/benchmarks/dromaeo.py:9: class DromaeoDomcoreattr(test.Test):
On 2013/07/18 01:39:00, tonyg wrote:
> Let's capitalize the last word:
> DromaeoDomcoreAttr
> DromaeoDomcoreModify
> ...
> DromaeoJslibAttrJquery
> DromaeoJslibAttrPrototype
> ...

Done.

https://codereview.chromium.org/19721002/diff/1/tools/perf/benchmarks/image_d...
File tools/perf/benchmarks/image_decoding.py (right):

https://codereview.chromium.org/19721002/diff/1/tools/perf/benchmarks/image_d...
tools/perf/benchmarks/image_decoding.py:11: page_set =
'page_sets/image_decoding_measurement.json'
On 2013/07/18 01:39:00, tonyg wrote:
> Is this a good place for a TODO about the renaming?

Done.

https://codereview.chromium.org/19721002/diff/1/tools/perf/benchmarks/page_cy...
File tools/perf/benchmarks/page_cycler.py (right):

https://codereview.chromium.org/19721002/diff/1/tools/perf/benchmarks/page_cy...
tools/perf/benchmarks/page_cycler.py:9: class PageCyclerMoz(test.Test):
On 2013/07/18 01:39:00, tonyg wrote:
> Can we alphabetize the cyclers in this file?

Done.

[tonyg, 2013-07-18 02:32:39]

lgtm

https://codereview.chromium.org/19721002/diff/8001/tools/perf/benchmarks/page...
File tools/perf/benchmarks/page_cycler.py (right):

https://codereview.chromium.org/19721002/diff/8001/tools/perf/benchmarks/page...
tools/perf/benchmarks/page_cycler.py:9: class PageCyclerBloat(test.Test):
Oh, I remembered one more thing we could clean up here. The page cyclers hack
--pageset-repeat to be =10. We should specify the --pageset-repeat flag in each
benchmark in this file for now. Then after we switch over, we can remove the
hack from the measurement.

[dtu, 2013-07-18 20:39:43]

https://codereview.chromium.org/19721002/diff/8001/tools/perf/benchmarks/page...
File tools/perf/benchmarks/page_cycler.py (right):

https://codereview.chromium.org/19721002/diff/8001/tools/perf/benchmarks/page...
tools/perf/benchmarks/page_cycler.py:9: class PageCyclerBloat(test.Test):
On 2013/07/18 02:32:39, tonyg wrote:
> Oh, I remembered one more thing we could clean up here. The page cyclers hack
> --pageset-repeat to be =10. We should specify the --pageset-repeat flag in
each
> benchmark in this file for now. Then after we switch over, we can remove the
> hack from the measurement.

Done.

[tonyg, 2013-07-18 20:46:25]

On 2013/07/18 20:39:43, Dave Tu wrote:
>
https://codereview.chromium.org/19721002/diff/8001/tools/perf/benchmarks/page...
> File tools/perf/benchmarks/page_cycler.py (right):
> 
>
https://codereview.chromium.org/19721002/diff/8001/tools/perf/benchmarks/page...
> tools/perf/benchmarks/page_cycler.py:9: class PageCyclerBloat(test.Test):
> On 2013/07/18 02:32:39, tonyg wrote:
> > Oh, I remembered one more thing we could clean up here. The page cyclers
hack
> > --pageset-repeat to be =10. We should specify the --pageset-repeat flag in
> each
> > benchmark in this file for now. Then after we switch over, we can remove the
> > hack from the measurement.
> 
> Done.

Could you also move the netsim page cycler into page_cycler.py?

[dtu, 2013-07-18 21:23:02]

On 2013/07/18 20:46:25, tonyg wrote:
> On 2013/07/18 20:39:43, Dave Tu wrote:
> >
>
https://codereview.chromium.org/19721002/diff/8001/tools/perf/benchmarks/page...
> > File tools/perf/benchmarks/page_cycler.py (right):
> > 
> >
>
https://codereview.chromium.org/19721002/diff/8001/tools/perf/benchmarks/page...
> > tools/perf/benchmarks/page_cycler.py:9: class PageCyclerBloat(test.Test):
> > On 2013/07/18 02:32:39, tonyg wrote:
> > > Oh, I remembered one more thing we could clean up here. The page cyclers
> hack
> > > --pageset-repeat to be =10. We should specify the --pageset-repeat flag in
> > each
> > > benchmark in this file for now. Then after we switch over, we can remove
the
> > > hack from the measurement.
> > 
> > Done.
> 
> Could you also move the netsim page cycler into page_cycler.py?

Done.

[tonyg, 2013-07-18 21:32:33]

lgtm

https://codereview.chromium.org/19721002/diff/26001/tools/perf/benchmarks/pag...
File tools/perf/benchmarks/page_cycler.py (right):

https://codereview.chromium.org/19721002/diff/26001/tools/perf/benchmarks/pag...
tools/perf/benchmarks/page_cycler.py:9: class NetsimTop25(test.Test):
Let's rename it PageCyclerNetsimTop25 for consistency. Nothing depends on the
name yet.

[dtu, 2013-07-18 21:38:50]

https://codereview.chromium.org/19721002/diff/26001/tools/perf/benchmarks/pag...
File tools/perf/benchmarks/page_cycler.py (right):

https://codereview.chromium.org/19721002/diff/26001/tools/perf/benchmarks/pag...
tools/perf/benchmarks/page_cycler.py:9: class NetsimTop25(test.Test):
On 2013/07/18 21:32:34, tonyg wrote:
> Let's rename it PageCyclerNetsimTop25 for consistency. Nothing depends on the
> name yet.

Done.

[dtu, 2013-07-19 20:36:23]

Committed patchset #5 manually as r212642 (presubmit successful).