Add new Telemetry tests to get CPU and memory usage for idle and flinging two test scenarios. Mirro…

Add new Telemetry tests to get CPU and memory usage for idle and flinging two test scenarios. Mirroring case will be added later.

It uses Chrome extension APIs to get CPU/memory usage for tab, browser, gpu and media router extension four processes. The test extension in this cl is implemented for this purpose. It collects the CPU/memory usage and posts it back the current tab.

We will get CPU/memory data each time when the Task Manager updates its process statistics, and will miss some data if media router extension is suspended.Right now I just use a simple way to present the results, which uses the average CPU/memory usage, but I feel this may not be the best way to present the results, since we have lots of 0 in extension CPU usage, it will definitely lower the average value. I need to figure out a better way to present the final results.

BUG=594225
Committed: https://crrev.com/50154e04711d8fac820f4f2c08519180c1f585bb
Cr-Commit-Position: refs/heads/master@{#388393}

[Lei Lei, 2016-03-31 18:36:40]

leilei@chromium.org changed reviewers:
+ apacible@chromium.org, mfoltz@chromium.org

[mark a. foltz, 2016-04-04 20:25:07]

Mostly minor comments.  It will be important to discard data points when the
event page is suspended.  Won't it have zero memory/cpu when this is the case?

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
File chrome/test/media_router/BUILD.gn (right):

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
chrome/test/media_router/BUILD.gn:17: media_router_extension_resources = [
Instead of duplicating file lists in the .gyp and .GN, it is preferred to import
the file list from the .gypi and refer to it here (using rebase_path where
necessary).  See c/b/media/router/media_router.{gn,gypi} for example.

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
File chrome/test/media_router/media_router_tests.gypi (right):

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
chrome/test/media_router/media_router_tests.gypi:17:
'media_router_extension_resources': [
media_router_test_extension_resources

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
File chrome/test/media_router/telemetry/benchmarks/media_router_benchmark.py
(right):

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/media_router_benchmark.py:25:
options.disable_component_extensions_with_background_pages = False
Will running additional background pages affect test results?

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/media_router_benchmark.py:41:
'--enable-logging=stderr'
Will logging affect the test performance?

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
File
chrome/test/media_router/telemetry/benchmarks/media_router_cpu_memory_metric.py
(right):

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/media_router_cpu_memory_metric.py:31:
print 'results_json' + ': ' + str(results_json)
Is this for debugging?  If so put it behind e.g. logging.info() so logging
doesn't affect normal test performance.

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/media_router_cpu_memory_metric.py:36:
units = ''
What are the units for metrics other than privateMemory?

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/media_router_cpu_memory_metric.py:38:
for process in metric_results:
Is metric_results a dictionary? If so, I think this can be written

for (process, process_results) in perf_results[metric]

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/media_router_cpu_memory_metric.py:44:
units = 'MB'
Do you need to reset this if metric != privateMemory?

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
File chrome/test/media_router/telemetry/benchmarks/pagesets/common.js (right):

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/pagesets/common.js:19: if
(!chrome.cast || !chrome.cast.isAvailable) {
Use the API-provided callback to start initialization,
window['__onGCastApiAvailable']

See the Chrome Sender SDK documentation for Cast for details.

(Sorry no URL - copy-paste of URLs is broken for me on Mac Canary)

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/pagesets/common.js:20:
setTimeout(startSession, 1000);
startSession is no longer defined?

[Lei Lei, 2016-04-13 00:22:43]

If MR extension is suspended, we will not get cpu/memory usage from task
manager, since there is no process for MR extension. 

But even when MR extension is active, we do see zero value for cpu usage, the
following is the CPU usage for MR extension for flinging case:
"mr_extension":[0,0,17.892893141654056,96.89912800774393,35.02795230594014,5.992395649920251,1.0000220004840106,0,0,0,0,0,8.986995817052724,4.00038403686754,0,0.998874268699176,0.9998000399920016,0,0,0,0,0,0,0,4.997481269440202]

The cpu usage drops to 0 sometimes, zero values definitely lower the average.
But I am not sure if it is a good idea to discard all the zero values, since at
half of the time, cpu usage is zero. 

How about discussing how to present the summary results in perf dashboard in our
sync up meeting? And I can update the metrics later, it should be a small
change.

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
File chrome/test/media_router/BUILD.gn (right):

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
chrome/test/media_router/BUILD.gn:17: media_router_extension_resources = [
On 2016/04/04 20:25:06, mark a. foltz wrote:
> Instead of duplicating file lists in the .gyp and .GN, it is preferred to
import
> the file list from the .gypi and refer to it here (using rebase_path where
> necessary).  See c/b/media/router/media_router.{gn,gypi} for example.

Done.

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
File chrome/test/media_router/media_router_tests.gypi (right):

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
chrome/test/media_router/media_router_tests.gypi:17:
'media_router_extension_resources': [
On 2016/04/04 20:25:06, mark a. foltz wrote:
> media_router_test_extension_resources

Done.

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
File chrome/test/media_router/telemetry/benchmarks/media_router_benchmark.py
(right):

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/media_router_benchmark.py:25:
options.disable_component_extensions_with_background_pages = False
On 2016/04/04 20:25:06, mark a. foltz wrote:
> Will running additional background pages affect test results?

Yes, I removed it. I just found this flag is not required, I only need to remove
the flag --disable-background-networking to enable chrome.runtime.

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/media_router_benchmark.py:41:
'--enable-logging=stderr'
On 2016/04/04 20:25:06, mark a. foltz wrote:
> Will logging affect the test performance?

Yes, I removed it.

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
File
chrome/test/media_router/telemetry/benchmarks/media_router_cpu_memory_metric.py
(right):

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/media_router_cpu_memory_metric.py:31:
print 'results_json' + ': ' + str(results_json)
On 2016/04/04 20:25:06, mark a. foltz wrote:
> Is this for debugging?  If so put it behind e.g. logging.info() so logging
> doesn't affect normal test performance.

Done.

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/media_router_cpu_memory_metric.py:36:
units = ''
On 2016/04/04 20:25:06, mark a. foltz wrote:
> What are the units for metrics other than privateMemory?

Right now we only have two metrics, privateMemory and CPU. The units for
privateMemory is MB, the units for CPU is empty string.

I defined a constant for the units of each metric, which makes the code easier
to read.

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/media_router_cpu_memory_metric.py:38:
for process in metric_results:
On 2016/04/04 20:25:06, mark a. foltz wrote:
> Is metric_results a dictionary? If so, I think this can be written
> 
> for (process, process_results) in perf_results[metric]

Done.

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/media_router_cpu_memory_metric.py:44:
units = 'MB'
On 2016/04/04 20:25:06, mark a. foltz wrote:
> Do you need to reset this if metric != privateMemory?

Yes, the units is reset at the beginning of for loop, see line 36.

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
File chrome/test/media_router/telemetry/benchmarks/pagesets/common.js (right):

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/pagesets/common.js:19: if
(!chrome.cast || !chrome.cast.isAvailable) {
On 2016/04/04 20:25:07, mark a. foltz wrote:
> Use the API-provided callback to start initialization,
> window['__onGCastApiAvailable']
> 
> See the Chrome Sender SDK documentation for Cast for details.
> 
> (Sorry no URL - copy-paste of URLs is broken for me on Mac Canary)

Done.

https://codereview.chromium.org/1843063004/diff/40001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/pagesets/common.js:20:
setTimeout(startSession, 1000);
On 2016/04/04 20:25:06, mark a. foltz wrote:
> startSession is no longer defined?

Done.

[Lei Lei, 2016-04-13 00:25:26]

leilei@chromium.org changed reviewers:
+ brettw@chromium.org

[Lei Lei, 2016-04-13 00:25:27]

+brettw for .gn file review.

[mark a. foltz, 2016-04-13 20:20:03]

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
File
chrome/test/media_router/telemetry/benchmarks/media_router_cpu_memory_metric.py
(right):

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/media_router_cpu_memory_metric.py:15:
'cpu': ''}
Can you define units for CPU?  Is it a utilization percentage or fraction?

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/media_router_cpu_memory_metric.py:37:
if results_json:
Slight preference for early return, to reduce indentation.

if not results_json:
  return

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/media_router_cpu_memory_metric.py:40:
units = UNITS.get(metric)
This can be put into L47.

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/media_router_cpu_memory_metric.py:42:
if len(process_results) > 0:
if not process_results:
  continue

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
File chrome/test/media_router/telemetry/benchmarks/pagesets/basic_test.html
(right):

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/pagesets/basic_test.html:7:
<script type="text/javascript"
src="https://www.gstatic.com/cv/js/sender/v1/cast_sender.js"></script>
Can this be put on one line?

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
File chrome/test/media_router/telemetry/benchmarks/pagesets/common.js (right):

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/pagesets/common.js:35:
console.info('Recevier listener: ' + JSON.stringify(availability));
Consider using the console format specifier %O for objects, instead of
stringifying them; this is really a personal preference though.

https://developer.chrome.com/devtools/docs/console-api#consolelogobject-object

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/pagesets/common.js:64: * Loads the
specific video on Chromecast.
Mention that the URL must point to a video/mp4 file.

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
File chrome/test/media_router/telemetry/extension/background.html (right):

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
chrome/test/media_router/telemetry/extension/background.html:7: <script
src="script.js"></script>
You don't need an HTML file if your background page only has scripts, instead
use the scripts property in the manifest.

https://developer.chrome.com/extensions/background_pages

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
File chrome/test/media_router/telemetry/extension/script.js (right):

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
chrome/test/media_router/telemetry/extension/script.js:59: console.log('message:
' + JSON.stringify(message));
Is all of this console logging necessary?  It will reflect in the test results
if time is being measured.

[mark a. foltz, 2016-04-13 20:21:12]

It sounds like the CPU and memory measurements should work out as long as the
samples are discarded when the extension process is missing.

[Lei Lei, 2016-04-14 01:55:13]

PTAL.

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
File
chrome/test/media_router/telemetry/benchmarks/media_router_cpu_memory_metric.py
(right):

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/media_router_cpu_memory_metric.py:15:
'cpu': ''}
On 2016/04/13 20:20:03, mark a. foltz wrote:
> Can you define units for CPU?  Is it a utilization percentage or fraction?

It is a utilization percentage, I renamed the metric name to cpu_utilization and
added '%' as units.

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/media_router_cpu_memory_metric.py:37:
if results_json:
On 2016/04/13 20:20:03, mark a. foltz wrote:
> Slight preference for early return, to reduce indentation.
> 
> if not results_json:
>   return
> 

Done.

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/media_router_cpu_memory_metric.py:40:
units = UNITS.get(metric)
On 2016/04/13 20:20:03, mark a. foltz wrote:
> This can be put into L47.

Done.

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/media_router_cpu_memory_metric.py:42:
if len(process_results) > 0:
On 2016/04/13 20:20:03, mark a. foltz wrote:
> if not process_results:
>   continue

Done.

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
File chrome/test/media_router/telemetry/benchmarks/pagesets/basic_test.html
(right):

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/pagesets/basic_test.html:7:
<script type="text/javascript"
src="https://www.gstatic.com/cv/js/sender/v1/cast_sender.js"></script>
On 2016/04/13 20:20:03, mark a. foltz wrote:
> Can this be put on one line?

This line exceeds 100 characters limitation, changed it to multiple lines.

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
File chrome/test/media_router/telemetry/benchmarks/pagesets/common.js (right):

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/pagesets/common.js:35:
console.info('Recevier listener: ' + JSON.stringify(availability));
On 2016/04/13 20:20:03, mark a. foltz wrote:
> Consider using the console format specifier %O for objects, instead of
> stringifying them; this is really a personal preference though.
> 
> https://developer.chrome.com/devtools/docs/console-api#consolelogobject-object

Thanks for the suggestion, I didn't know I can format the object as this way. 

I tried it, it works well in console, and prints out the correct logs.
Unfortunately the log from console.log with any formatter doesn't work in
chrome_debug.log, it just prints out the format specifiers, e.g 'error %s'. I
need the chrome debug log for debugging, since Telemetry will pause the test if
I open the developer tools during the test. But I can get the whole chrome debug
log after the test finishes.

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
chrome/test/media_router/telemetry/benchmarks/pagesets/common.js:64: * Loads the
specific video on Chromecast.
On 2016/04/13 20:20:03, mark a. foltz wrote:
> Mention that the URL must point to a video/mp4 file.

Done.

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
File chrome/test/media_router/telemetry/extension/background.html (right):

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
chrome/test/media_router/telemetry/extension/background.html:7: <script
src="script.js"></script>
On 2016/04/13 20:20:03, mark a. foltz wrote:
> You don't need an HTML file if your background page only has scripts, instead
> use the scripts property in the manifest.
> 
> https://developer.chrome.com/extensions/background_pages

Thanks for the comment, I removed background HTML file.

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
File chrome/test/media_router/telemetry/extension/script.js (right):

https://codereview.chromium.org/1843063004/diff/80001/chrome/test/media_route...
chrome/test/media_router/telemetry/extension/script.js:59: console.log('message:
' + JSON.stringify(message));
On 2016/04/13 20:20:03, mark a. foltz wrote:
> Is all of this console logging necessary?  It will reflect in the test results
> if time is being measured.

I removed it.

[apacible, 2016-04-14 17:35:20]

lgtm

https://codereview.chromium.org/1843063004/diff/140001/chrome/test/media_rout...
File chrome/test/media_router/telemetry/benchmarks/pagesets/cpu_memory_script.js
(right):

https://codereview.chromium.org/1843063004/diff/140001/chrome/test/media_rout...
chrome/test/media_router/telemetry/benchmarks/pagesets/cpu_memory_script.js:45:
return;
nit: return not needed

[Lei Lei, 2016-04-14 20:05:46]

https://codereview.chromium.org/1843063004/diff/140001/chrome/test/media_rout...
File chrome/test/media_router/telemetry/benchmarks/pagesets/cpu_memory_script.js
(right):

https://codereview.chromium.org/1843063004/diff/140001/chrome/test/media_rout...
chrome/test/media_router/telemetry/benchmarks/pagesets/cpu_memory_script.js:45:
return;
On 2016/04/14 17:35:20, apacible wrote:
> nit: return not needed

Done.

[Lei Lei, 2016-04-18 18:15:03]

Friendly ping.

[brettw, 2016-04-18 23:08:25]

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
File chrome/test/media_router/BUILD.gn (right):

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
chrome/test/media_router/BUILD.gn:5: gypi_values =
exec_script("//build/gypi_to_gn.py",
I think it's better to copy these file names than to try to share then with GYP
through this elaborate mechanism. We only use this for large and
frequently-changing lists. Then you can delete the change in .gn.

[mark a. foltz, 2016-04-18 23:48:11]

LGTM with minor comments.

brettw@: I asked leilei@ to consolidate the .gn/.gyp file list.  Is there a best
practices document I can refer to in the future?

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
File chrome/test/media_router/telemetry/benchmarks/media_router_benchmark.py
(right):

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
chrome/test/media_router/telemetry/benchmarks/media_router_benchmark.py:20:
page_set = media_router_perf_pages.MediaRouterDailogPageSet
typo in MediaRouterDialogPageSet

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
File chrome/test/media_router/telemetry/benchmarks/pagesets/basic_test.html
(right):

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
chrome/test/media_router/telemetry/benchmarks/pagesets/basic_test.html:12:
<button id="start_session_button" onclick="startFlingingSession()">
Are these tests started manually (by clicking a button) or is this just a
debugging page?

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
File chrome/test/media_router/telemetry/benchmarks/pagesets/common.js (right):

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
chrome/test/media_router/telemetry/benchmarks/pagesets/common.js:35:
console.info('Recevier listener: ' + JSON.stringify(availability));
Typo in Receiver

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
chrome/test/media_router/telemetry/benchmarks/pagesets/common.js:54:
function(session) { // Request session successful callback
Nit: two spaces before // comments here and elsewhere

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
File chrome/test/media_router/telemetry/extension/README.md (right):

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
chrome/test/media_router/telemetry/extension/README.md:1: <!-- Copyright 2016
The Chromium Authors. All rights reserved.
The copyright isn't required for documentation files.

[brettw, 2016-04-19 19:32:18]

On 2016/04/18 23:48:11, mark a. foltz wrote:
> LGTM with minor comments.
> 
> brettw@: I asked leilei@ to consolidate the .gn/.gyp file list.  Is there a
best
> practices document I can refer to in the future?

I thought there was but I don't see one at the moment. I'll add something in the
gypi_to_gn script comments.

[brettw, 2016-04-19 20:21:00]

On 2016/04/19 19:32:18, brettw wrote:
> On 2016/04/18 23:48:11, mark a. foltz wrote:
> > LGTM with minor comments.
> > 
> > brettw@: I asked leilei@ to consolidate the .gn/.gyp file list.  Is there a
> best
> > practices document I can refer to in the future?
> 
> I thought there was but I don't see one at the moment. I'll add something in
the
> gypi_to_gn script comments.

I did https://codereview.chromium.org/1905433002/ for the documentation.

[Lei Lei, 2016-04-19 21:56:02]

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
File chrome/test/media_router/BUILD.gn (right):

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
chrome/test/media_router/BUILD.gn:5: gypi_values =
exec_script("//build/gypi_to_gn.py",
On 2016/04/18 23:08:25, brettw wrote:
> I think it's better to copy these file names than to try to share then with
GYP
> through this elaborate mechanism. We only use this for large and
> frequently-changing lists. Then you can delete the change in .gn.

Done.

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
File chrome/test/media_router/telemetry/benchmarks/media_router_benchmark.py
(right):

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
chrome/test/media_router/telemetry/benchmarks/media_router_benchmark.py:20:
page_set = media_router_perf_pages.MediaRouterDailogPageSet
On 2016/04/18 23:48:11, mark a. foltz wrote:
> typo in MediaRouterDialogPageSet

Done.

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
File chrome/test/media_router/telemetry/benchmarks/pagesets/basic_test.html
(right):

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
chrome/test/media_router/telemetry/benchmarks/pagesets/basic_test.html:12:
<button id="start_session_button" onclick="startFlingingSession()">
On 2016/04/18 23:48:11, mark a. foltz wrote:
> Are these tests started manually (by clicking a button) or is this just a
> debugging page?

We need to click the button to start a new session, but it doesn't require to do
it manually, I have code to click the button to start a new session in line 86
in media_router_perf_pages.py.
https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...

This is a workaround for user gesture check, otherwise it will be failed to
start a new session if presentation.requestSession is called directly in js
code.

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
File chrome/test/media_router/telemetry/benchmarks/pagesets/common.js (right):

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
chrome/test/media_router/telemetry/benchmarks/pagesets/common.js:35:
console.info('Recevier listener: ' + JSON.stringify(availability));
On 2016/04/18 23:48:11, mark a. foltz wrote:
> Typo in Receiver

Done.

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
chrome/test/media_router/telemetry/benchmarks/pagesets/common.js:54:
function(session) { // Request session successful callback
On 2016/04/18 23:48:11, mark a. foltz wrote:
> Nit: two spaces before // comments here and elsewhere

Done.

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
File chrome/test/media_router/telemetry/extension/README.md (right):

https://codereview.chromium.org/1843063004/diff/160001/chrome/test/media_rout...
chrome/test/media_router/telemetry/extension/README.md:1: <!-- Copyright 2016
The Chromium Authors. All rights reserved.
On 2016/04/18 23:48:11, mark a. foltz wrote:
> The copyright isn't required for documentation files.

Done.

[Lei Lei, 2016-04-19 21:57:18]

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

[commit-bot: I haz the power, 2016-04-19 21:57:47]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1843063004/180001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1843063004/180001

[commit-bot: I haz the power, 2016-04-19 22:19:31]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-04-19 22:19:33]

Dry run: Try jobs failed on following builders:
  linux_chromium_clobber_rel_ng on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[Lei Lei, 2016-04-19 23:46:37]

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

[commit-bot: I haz the power, 2016-04-19 23:47:14]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1843063004/200001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1843063004/200001

[commit-bot: I haz the power, 2016-04-20 01:02:37]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-04-20 01:02:38]

Dry run: This issue passed the CQ dry run.

[Lei Lei, 2016-04-20 01:08:51]

The CQ bit was checked by leilei@chromium.org

[Lei Lei, 2016-04-20 01:08:51]

The patchset sent to the CQ was uploaded after l-g-t-m from
apacible@chromium.org, mfoltz@chromium.org
Link to the patchset: https://codereview.chromium.org/1843063004/#ps200001
(title: "Fix GN build failure")

[commit-bot: I haz the power, 2016-04-20 01:09:16]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1843063004/200001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1843063004/200001

[commit-bot: I haz the power, 2016-04-20 01:14:54]

Committed patchset #11 (id:200001)

[commit-bot: I haz the power, 2016-04-22 19:19:34]

Description was changed from

==========
Add new Telemetry tests to get CPU and memory usage for idle and flinging two
test scenarios. Mirroring case will be added later.

It uses Chrome extension APIs to get CPU/memory usage for tab, browser, gpu and
media router extension four processes. The test extension in this cl is
implemented for this purpose. It collects the CPU/memory usage and posts it back
the current tab.

We will get CPU/memory data each time when the Task Manager updates its process
statistics, and will miss some data if media router extension is suspended.Right
now I just use a simple way to present the results, which uses the average
CPU/memory usage, but I feel this may not be the best way to present the
results, since we have lots of 0 in extension CPU usage, it will definitely
lower the average value. I need to figure out a better way to present the final
results.

BUG=594225
==========

to

==========
Add new Telemetry tests to get CPU and memory usage for idle and flinging two
test scenarios. Mirroring case will be added later.

It uses Chrome extension APIs to get CPU/memory usage for tab, browser, gpu and
media router extension four processes. The test extension in this cl is
implemented for this purpose. It collects the CPU/memory usage and posts it back
the current tab.

We will get CPU/memory data each time when the Task Manager updates its process
statistics, and will miss some data if media router extension is suspended.Right
now I just use a simple way to present the results, which uses the average
CPU/memory usage, but I feel this may not be the best way to present the
results, since we have lots of 0 in extension CPU usage, it will definitely
lower the average value. I need to figure out a better way to present the final
results.

BUG=594225

Committed: https://crrev.com/50154e04711d8fac820f4f2c08519180c1f585bb
Cr-Commit-Position: refs/heads/master@{#388393}
==========

[commit-bot: I haz the power, 2016-04-22 19:19:35]

Patchset 11 (id:??) landed as
https://crrev.com/50154e04711d8fac820f4f2c08519180c1f585bb
Cr-Commit-Position: refs/heads/master@{#388393}