[Telemetry][Android] Wait for device under test to cool between pages.

[Telemetry][Android] Wait for device under test to cool between pages.

If the platform supports waiting for the device temperature to drop, wait.
Currently this only implements andorid.

BUG=chromium:669923
Review-Url: https://codereview.chromium.org/2541843007
Committed: https://chromium.googlesource.com/external/github.com/catapult-project/catapult/+/70f42a7c55ca69cdeb9aa6ec7e40ff3f155040b9

[rnephew (Reviews Here), 2016-12-01 18:38:00]

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

[rnephew (Reviews Here), 2016-12-01 18:38:00]

https://codereview.chromium.org/2541843007/diff/1/telemetry/telemetry/interna...
File telemetry/telemetry/internal/platform/platform_backend.py (right):

https://codereview.chromium.org/2541843007/diff/1/telemetry/telemetry/interna...
telemetry/telemetry/internal/platform/platform_backend.py:304: pass
Debating if I should add a CanWaitForTemperature method that defaults to false,
having each platform that implements it override it to true, and having this
raise NotImplementedError, but that also seems like unnecessary boilerplate code
when we can just make in a NOP on platforms that do not support it yet (or
platforms that have no need to support it).

https://codereview.chromium.org/2541843007/diff/1/telemetry/telemetry/page/sh...
File telemetry/telemetry/page/shared_page_state.py (right):

https://codereview.chromium.org/2541843007/diff/1/telemetry/telemetry/page/sh...
telemetry/telemetry/page/shared_page_state.py:189: # Make sure device under test
is at a suitable temperature.
Decided to put it here, because doing it here makes it wait between pages. This
will make sure that on long running tests, such as system health, that we do not
start overheating partway through. I do expect this to increase the amount of
time single benchmarks take since it will cool between pages, but I do not
expect this to increase the amount of time a test run takes in total though,
since the android test runner already waits between benchmarks for the temp to
drop.

[nednguyen, 2016-12-01 19:01:46]

On 2016/12/01 18:38:00, rnephew (Reviews Here) wrote:
>
https://codereview.chromium.org/2541843007/diff/1/telemetry/telemetry/interna...
> File telemetry/telemetry/internal/platform/platform_backend.py (right):
> 
>
https://codereview.chromium.org/2541843007/diff/1/telemetry/telemetry/interna...
> telemetry/telemetry/internal/platform/platform_backend.py:304: pass
> Debating if I should add a CanWaitForTemperature method that defaults to
false,
> having each platform that implements it override it to true, and having this
> raise NotImplementedError, but that also seems like unnecessary boilerplate
code
> when we can just make in a NOP on platforms that do not support it yet (or
> platforms that have no need to support it).
> 
>
https://codereview.chromium.org/2541843007/diff/1/telemetry/telemetry/page/sh...
> File telemetry/telemetry/page/shared_page_state.py (right):
> 
>
https://codereview.chromium.org/2541843007/diff/1/telemetry/telemetry/page/sh...
> telemetry/telemetry/page/shared_page_state.py:189: # Make sure device under
test
> is at a suitable temperature.
> Decided to put it here, because doing it here makes it wait between pages.
This
> will make sure that on long running tests, such as system health, that we do
not
> start overheating partway through. I do expect this to increase the amount of
> time single benchmarks take since it will cool between pages, but I do not
> expect this to increase the amount of time a test run takes in total though,
> since the android test runner already waits between benchmarks for the temp to
> drop.

I think this really should be a commandline flag
(--wait-to-cool-down-between-runs) or a benchmark property
(wait_to_cool_down_between_runs). We don't want to enable this by default for
all benchmark since it can be expensive, and benchmarks like power probably
don't care much about the temperature.

[rnephew (Reviews Here), 2016-12-01 19:14:02]

On 2016/12/01 19:01:46, nednguyen wrote:
> On 2016/12/01 18:38:00, rnephew (Reviews Here) wrote:
> >
>
https://codereview.chromium.org/2541843007/diff/1/telemetry/telemetry/interna...
> > File telemetry/telemetry/internal/platform/platform_backend.py (right):
> > 
> >
>
https://codereview.chromium.org/2541843007/diff/1/telemetry/telemetry/interna...
> > telemetry/telemetry/internal/platform/platform_backend.py:304: pass
> > Debating if I should add a CanWaitForTemperature method that defaults to
> false,
> > having each platform that implements it override it to true, and having this
> > raise NotImplementedError, but that also seems like unnecessary boilerplate
> code
> > when we can just make in a NOP on platforms that do not support it yet (or
> > platforms that have no need to support it).
> > 
> >
>
https://codereview.chromium.org/2541843007/diff/1/telemetry/telemetry/page/sh...
> > File telemetry/telemetry/page/shared_page_state.py (right):
> > 
> >
>
https://codereview.chromium.org/2541843007/diff/1/telemetry/telemetry/page/sh...
> > telemetry/telemetry/page/shared_page_state.py:189: # Make sure device under
> test
> > is at a suitable temperature.
> > Decided to put it here, because doing it here makes it wait between pages.
> This
> > will make sure that on long running tests, such as system health, that we do
> not
> > start overheating partway through. I do expect this to increase the amount
of
> > time single benchmarks take since it will cool between pages, but I do not
> > expect this to increase the amount of time a test run takes in total though,
> > since the android test runner already waits between benchmarks for the temp
to
> > drop.
> 
> I think this really should be a commandline flag
> (--wait-to-cool-down-between-runs) or a benchmark property
> (wait_to_cool_down_between_runs). We don't want to enable this by default for
> all benchmark since it can be expensive, and benchmarks like power probably
> don't care much about the temperature.

It would help decrease noise though if we do it between each page always. And as
stated above, in theory it shouldn't increase run times very much since its just
moving where it waits for temperatures to drop. I'll look into adding it as a
flag though.

[nednguyen, 2016-12-01 19:21:41]

On 2016/12/01 19:14:02, rnephew (Reviews Here) wrote:
> On 2016/12/01 19:01:46, nednguyen wrote:
> > On 2016/12/01 18:38:00, rnephew (Reviews Here) wrote:
> > >
> >
>
https://codereview.chromium.org/2541843007/diff/1/telemetry/telemetry/interna...
> > > File telemetry/telemetry/internal/platform/platform_backend.py (right):
> > > 
> > >
> >
>
https://codereview.chromium.org/2541843007/diff/1/telemetry/telemetry/interna...
> > > telemetry/telemetry/internal/platform/platform_backend.py:304: pass
> > > Debating if I should add a CanWaitForTemperature method that defaults to
> > false,
> > > having each platform that implements it override it to true, and having
this
> > > raise NotImplementedError, but that also seems like unnecessary
boilerplate
> > code
> > > when we can just make in a NOP on platforms that do not support it yet (or
> > > platforms that have no need to support it).
> > > 
> > >
> >
>
https://codereview.chromium.org/2541843007/diff/1/telemetry/telemetry/page/sh...
> > > File telemetry/telemetry/page/shared_page_state.py (right):
> > > 
> > >
> >
>
https://codereview.chromium.org/2541843007/diff/1/telemetry/telemetry/page/sh...
> > > telemetry/telemetry/page/shared_page_state.py:189: # Make sure device
under
> > test
> > > is at a suitable temperature.
> > > Decided to put it here, because doing it here makes it wait between pages.
> > This
> > > will make sure that on long running tests, such as system health, that we
do
> > not
> > > start overheating partway through. I do expect this to increase the amount
> of
> > > time single benchmarks take since it will cool between pages, but I do not
> > > expect this to increase the amount of time a test run takes in total
though,
> > > since the android test runner already waits between benchmarks for the
temp
> to
> > > drop.
> > 
> > I think this really should be a commandline flag
> > (--wait-to-cool-down-between-runs) or a benchmark property
> > (wait_to_cool_down_between_runs). We don't want to enable this by default
for
> > all benchmark since it can be expensive, and benchmarks like power probably
> > don't care much about the temperature.
> 
> It would help decrease noise though if we do it between each page always. And
as
> stated above, in theory it shouldn't increase run times very much since its
just
> moving where it waits for temperatures to drop. I'll look into adding it as a
> flag though.

We currently also have _WaitForThermalThrottling if needed in
https://github.com/catapult-project/catapult/blob/master/telemetry/telemetry/....
If you work on this, let take this slow & do it right.

A few questions I have are:
1) How is the method you're implementing different from
_WaitForThermalThrottling? Can we merge the two to into 1 thing?
2) On a typical long running benchmark like system_health.mobile_memory, how
much time does WaitingForDeviceToCool between story add to the overall benchmark
run?
3) Can we also implement this on desktop. IIRC, Charlie was having some problem
with the power increasing across runs on Mac.

Can you make a design doc about this work & answering these questions?

[nednguyen, 2016-12-01 19:21:56]

Description was changed from

==========
[Telemetry][Android] Wait for device under test to cool between pages.

If the platform supports waiting for the device temperature to drop, wait.
Currently this only implements andorid.

BUG=chromium:669923
==========

to

==========
[Telemetry][Android] Wait for device under test to cool between pages.

If the platform supports waiting for the device temperature to drop, wait.
Currently this only implements andorid.

BUG=chromium:669923
==========

[nednguyen, 2016-12-01 19:21:56]

nednguyen@google.com changed reviewers:
+ charliea@chromium.org

[rnephew (Reviews Here), 2016-12-01 19:28:20]

On 2016/12/01 19:21:41, nednguyen wrote:
> On 2016/12/01 19:14:02, rnephew (Reviews Here) wrote:
> > On 2016/12/01 19:01:46, nednguyen wrote:
> > > On 2016/12/01 18:38:00, rnephew (Reviews Here) wrote:
> > > >
> > >
> >
>
https://codereview.chromium.org/2541843007/diff/1/telemetry/telemetry/interna...
> > > > File telemetry/telemetry/internal/platform/platform_backend.py (right):
> > > > 
> > > >
> > >
> >
>
https://codereview.chromium.org/2541843007/diff/1/telemetry/telemetry/interna...
> > > > telemetry/telemetry/internal/platform/platform_backend.py:304: pass
> > > > Debating if I should add a CanWaitForTemperature method that defaults to
> > > false,
> > > > having each platform that implements it override it to true, and having
> this
> > > > raise NotImplementedError, but that also seems like unnecessary
> boilerplate
> > > code
> > > > when we can just make in a NOP on platforms that do not support it yet
(or
> > > > platforms that have no need to support it).
> > > > 
> > > >
> > >
> >
>
https://codereview.chromium.org/2541843007/diff/1/telemetry/telemetry/page/sh...
> > > > File telemetry/telemetry/page/shared_page_state.py (right):
> > > > 
> > > >
> > >
> >
>
https://codereview.chromium.org/2541843007/diff/1/telemetry/telemetry/page/sh...
> > > > telemetry/telemetry/page/shared_page_state.py:189: # Make sure device
> under
> > > test
> > > > is at a suitable temperature.
> > > > Decided to put it here, because doing it here makes it wait between
pages.
> > > This
> > > > will make sure that on long running tests, such as system health, that
we
> do
> > > not
> > > > start overheating partway through. I do expect this to increase the
amount
> > of
> > > > time single benchmarks take since it will cool between pages, but I do
not
> > > > expect this to increase the amount of time a test run takes in total
> though,
> > > > since the android test runner already waits between benchmarks for the
> temp
> > to
> > > > drop.
> > > 
> > > I think this really should be a commandline flag
> > > (--wait-to-cool-down-between-runs) or a benchmark property
> > > (wait_to_cool_down_between_runs). We don't want to enable this by default
> for
> > > all benchmark since it can be expensive, and benchmarks like power
probably
> > > don't care much about the temperature.
> > 
> > It would help decrease noise though if we do it between each page always.
And
> as
> > stated above, in theory it shouldn't increase run times very much since its
> just
> > moving where it waits for temperatures to drop. I'll look into adding it as
a
> > flag though.
> 
> We currently also have _WaitForThermalThrottling if needed in
>
https://github.com/catapult-project/catapult/blob/master/telemetry/telemetry/....
> If you work on this, let take this slow & do it right.
> 
> A few questions I have are:
> 1) How is the method you're implementing different from
> _WaitForThermalThrottling? Can we merge the two to into 1 thing?
> 2) On a typical long running benchmark like system_health.mobile_memory, how
> much time does WaitingForDeviceToCool between story add to the overall
benchmark
> run?
> 3) Can we also implement this on desktop. IIRC, Charlie was having some
problem
> with the power increasing across runs on Mac.
> 
> Can you make a design doc about this work & answering these questions?

If we already have _WaitForThermalThrottling, then maybe thermal throttling
isn't the issue in crbug.com/669923 making this unnecessary? 

I dont know enough about mac hardware to implement it there atm. Since this (if
this is even what is causing it) is causing problems on bisects for android, I'd
prefer to roll it out on android first, then work on the other platforms so we
can get unstuck there.

I dont think this requires a full DD, but I'll make a 1-pager for it.

[rnephew (Reviews Here), 2016-12-02 15:54:57]

I added a command line flag and moved where it calls LetTemperatureCool to
reflect where the thermal throttling already is called.

[charliea (OOO until 10-5), 2016-12-02 16:03:12]

https://codereview.chromium.org/2541843007/diff/20001/telemetry/telemetry/int...
File telemetry/telemetry/internal/story_runner.py (right):

https://codereview.chromium.org/2541843007/diff/20001/telemetry/telemetry/int...
telemetry/telemetry/internal/story_runner.py:64: help='Temperature to wait for
between pages. In tenths of'
Can we accept and pass around the temperature value as actual degrees celsius,
and only convert it at the last minute? We can mention here that the precision
is limited to 1/10 of a degree C, but it seems like this would be a lot more
intuitive to users.

[rnephew (Reviews Here), 2016-12-12 16:58:58]

https://codereview.chromium.org/2541843007/diff/20001/telemetry/telemetry/int...
File telemetry/telemetry/internal/story_runner.py (right):

https://codereview.chromium.org/2541843007/diff/20001/telemetry/telemetry/int...
telemetry/telemetry/internal/story_runner.py:64: help='Temperature to wait for
between pages. In tenths of'
On 2016/12/02 16:03:12, charliea wrote:
> Can we accept and pass around the temperature value as actual degrees celsius,
> and only convert it at the last minute? We can mention here that the precision
> is limited to 1/10 of a degree C, but it seems like this would be a lot more
> intuitive to users.

Done.

[rnephew (Reviews Here), 2016-12-15 15:25:51]

Ping.

[nednguyen, 2016-12-15 15:38:11]

https://codereview.chromium.org/2541843007/diff/40001/telemetry/telemetry/int...
File telemetry/telemetry/internal/story_runner.py (right):

https://codereview.chromium.org/2541843007/diff/40001/telemetry/telemetry/int...
telemetry/telemetry/internal/story_runner.py:64: help='Temperature to wait for
between pages. In degrees C.')
I think we should not expose this. Instead, can you do a local run of system
health benchmark on a test phone & see whether this has big effect on runtime?

If not, we should just go ahead & replace _WaitForThermalThrottlingIfNeeded with
this new mechanism

[rnephew (Reviews Here), 2016-12-15 15:47:05]

On 2016/12/15 15:38:11, nednguyen wrote:
>
https://codereview.chromium.org/2541843007/diff/40001/telemetry/telemetry/int...
> File telemetry/telemetry/internal/story_runner.py (right):
> 
>
https://codereview.chromium.org/2541843007/diff/40001/telemetry/telemetry/int...
> telemetry/telemetry/internal/story_runner.py:64: help='Temperature to wait for
> between pages. In degrees C.')
> I think we should not expose this. Instead, can you do a local run of system
> health benchmark on a test phone & see whether this has big effect on runtime?
> 
> If not, we should just go ahead & replace _WaitForThermalThrottlingIfNeeded
with
> this new mechanism

I dont think any amount of local testing will adequately warm a device like it
would be in the labs. We typically do not see temperature issues out side of lab
conditions, which are nearly impossible to replicate in the lab. I also expect a
single benchmark run to take longer, its the overall telemetry run that should
not take longer.

Old:                        New:
[Test 1 start]          [Test 1 start]
  [page 1]                   [page 1]
  [page 2]                   [Cool device]
  [page 3]                  [page 2] 
[Cool device]            [Cool device]
[Cool device]            [Page 3]
[Cool device]            [Cool device]

That is about how I expect it to change. Moving all the cooling at the end of
the benchmark to the between pages. Its likely that we will see a small increase
because of newtons law of cooling, but probably not very high. I just dont think
we can adequately test it outside of the lab.

[nednguyen, 2016-12-15 16:01:19]

On 2016/12/15 15:47:05, rnephew (Reviews Here) wrote:
> On 2016/12/15 15:38:11, nednguyen wrote:
> >
>
https://codereview.chromium.org/2541843007/diff/40001/telemetry/telemetry/int...
> > File telemetry/telemetry/internal/story_runner.py (right):
> > 
> >
>
https://codereview.chromium.org/2541843007/diff/40001/telemetry/telemetry/int...
> > telemetry/telemetry/internal/story_runner.py:64: help='Temperature to wait
for
> > between pages. In degrees C.')
> > I think we should not expose this. Instead, can you do a local run of system
> > health benchmark on a test phone & see whether this has big effect on
runtime?
> > 
> > If not, we should just go ahead & replace _WaitForThermalThrottlingIfNeeded
> with
> > this new mechanism
> 
> I dont think any amount of local testing will adequately warm a device like it
> would be in the labs. We typically do not see temperature issues out side of
lab
> conditions, which are nearly impossible to replicate in the lab. I also expect
a
> single benchmark run to take longer, its the overall telemetry run that should
> not take longer.
> 
> Old:                        New:
> [Test 1 start]          [Test 1 start]
>   [page 1]                   [page 1]
>   [page 2]                   [Cool device]
>   [page 3]                  [page 2] 
> [Cool device]            [Cool device]
> [Cool device]            [Page 3]
> [Cool device]            [Cool device]
> 
> That is about how I expect it to change. Moving all the cooling at the end of
> the benchmark to the between pages. Its likely that we will see a small
increase
> because of newtons law of cooling, but probably not very high. I just dont
think
> we can adequately test it outside of the lab.

I see. I think the next step then could be:
1) Remove the commandline flag. We should be the one who know best about which
temperature to use. In general, less public APIs will give you less headache in
the long run.
2) Go ahead with landing this patch, but taking it as an experiment. Add logging
to know whether the wait for temperature is actually triggered & how long it
takes.
3) Collect the data: 
i) Whether this helps fixing crbug.com/669923
ii) How much runtime increase does this contribute to the benchmark.
iii) Whether this makes other benchmarks more stable or less stable. --> Gives
this about 3 weeks to see if this CL got blamed by any bisect.

4) Assuming data in 3) is good, go ahead with migrate
_WaitForThermalThrottlingIfNeeded(state.platform) over to this new method of
waiting for device to cool
If the data in 3) is not good, we may need to revert this CL & decide what to do
next.

[rnephew (Reviews Here), 2016-12-15 16:29:01]

https://codereview.chromium.org/2541843007/diff/40001/telemetry/telemetry/int...
File telemetry/telemetry/internal/story_runner.py (right):

https://codereview.chromium.org/2541843007/diff/40001/telemetry/telemetry/int...
telemetry/telemetry/internal/story_runner.py:64: help='Temperature to wait for
between pages. In degrees C.')
On 2016/12/15 15:38:10, nednguyen wrote:
> I think we should not expose this. Instead, can you do a local run of system
> health benchmark on a test phone & see whether this has big effect on runtime?
> 
> If not, we should just go ahead & replace _WaitForThermalThrottlingIfNeeded
with
> this new mechanism

Got rid of the cmd line option.

[rnephew (Reviews Here), 2016-12-15 16:29:04]

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

[commit-bot: I haz the power, 2016-12-15 16:29:13]

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

[rnephew (Reviews Here), 2016-12-15 16:30:47]

On 2016/12/15 16:29:01, rnephew (Reviews Here) wrote:
>
https://codereview.chromium.org/2541843007/diff/40001/telemetry/telemetry/int...
> File telemetry/telemetry/internal/story_runner.py (right):
> 
>
https://codereview.chromium.org/2541843007/diff/40001/telemetry/telemetry/int...
> telemetry/telemetry/internal/story_runner.py:64: help='Temperature to wait for
> between pages. In degrees C.')
> On 2016/12/15 15:38:10, nednguyen wrote:
> > I think we should not expose this. Instead, can you do a local run of system
> > health benchmark on a test phone & see whether this has big effect on
runtime?
> > 
> > If not, we should just go ahead & replace _WaitForThermalThrottlingIfNeeded
> with
> > this new mechanism
> 
> Got rid of the cmd line option.

The devil method this uses already logs the temperature, so no extra logging
needed. We can tell by how many times the log message appears how long it
waited.

[commit-bot: I haz the power, 2016-12-15 16:39:37]

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

[commit-bot: I haz the power, 2016-12-15 16:39:38]

Dry run: Try jobs failed on following builders:
  Catapult Mac Tryserver on master.tryserver.client.catapult (JOB_FAILED,
https://build.chromium.org/p/tryserver.client.catapult/builders/Catapult%20Ma...)

[rnephew (Reviews Here), 2016-12-15 16:50:55]

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

[charliea (OOO until 10-5), 2016-12-15 16:50:57]

lgtm, but defer to Ned's lgtm ultimately

[commit-bot: I haz the power, 2016-12-15 16:50:57]

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

[commit-bot: I haz the power, 2016-12-15 17:01:27]

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

[commit-bot: I haz the power, 2016-12-15 17:01:28]

Dry run: Try jobs failed on following builders:
  Catapult Mac Tryserver on master.tryserver.client.catapult (JOB_FAILED,
https://build.chromium.org/p/tryserver.client.catapult/builders/Catapult%20Ma...)

[nednguyen, 2016-12-15 17:48:55]

lgtm

[nednguyen, 2016-12-15 17:49:39]

On 2016/12/15 17:48:55, nednguyen wrote:
> lgtm

+Juan & Hector since this may affect crbug.com/671156

[rnephew (Reviews Here), 2016-12-19 21:31:10]

On 2016/12/15 17:49:39, nednguyen wrote:
> On 2016/12/15 17:48:55, nednguyen wrote:
> > lgtm
> 
> +Juan & Hector since this may affect crbug.com/671156

Fixed unit tests and re. Landing.

[rnephew (Reviews Here), 2016-12-19 21:31:22]

On 2016/12/19 21:31:10, rnephew (Reviews Here) wrote:
> On 2016/12/15 17:49:39, nednguyen wrote:
> > On 2016/12/15 17:48:55, nednguyen wrote:
> > > lgtm
> > 
> > +Juan & Hector since this may affect crbug.com/671156
> 
> Fixed unit tests and re. Landing.

rebased*

[rnephew (Reviews Here), 2016-12-19 21:31:25]

On 2016/12/19 21:31:10, rnephew (Reviews Here) wrote:
> On 2016/12/15 17:49:39, nednguyen wrote:
> > On 2016/12/15 17:48:55, nednguyen wrote:
> > > lgtm
> > 
> > +Juan & Hector since this may affect crbug.com/671156
> 
> Fixed unit tests and re. Landing.

rebased*

[rnephew (Reviews Here), 2016-12-19 21:31:34]

The CQ bit was checked by rnephew@chromium.org

[rnephew (Reviews Here), 2016-12-19 21:31:35]

The patchset sent to the CQ was uploaded after l-g-t-m from
nednguyen@google.com, charliea@chromium.org
Link to the patchset: https://codereview.chromium.org/2541843007/#ps120001
(title: "rebase")

[commit-bot: I haz the power, 2016-12-19 21:31:38]

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

[commit-bot: I haz the power, 2016-12-19 21:43:02]

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

[commit-bot: I haz the power, 2016-12-19 21:43:03]

Try jobs failed on following builders:
  Catapult Mac Tryserver on master.tryserver.client.catapult (JOB_FAILED,
https://build.chromium.org/p/tryserver.client.catapult/builders/Catapult%20Ma...)

[rnephew (Reviews Here), 2016-12-19 22:43:00]

The CQ bit was checked by rnephew@chromium.org

[rnephew (Reviews Here), 2016-12-19 22:43:01]

The patchset sent to the CQ was uploaded after l-g-t-m from
nednguyen@google.com, charliea@chromium.org
Link to the patchset: https://codereview.chromium.org/2541843007/#ps140001
(title: "fix more tests")

[commit-bot: I haz the power, 2016-12-19 22:43:04]

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

[commit-bot: I haz the power, 2016-12-19 23:01:40]

CQ is committing da patch.

Bot data: {"patchset_id": 140001, "attempt_start_ts": 1482187380743900,
"parent_rev": "d77eaf7f69e860436f0245c10ea2c186937ab666", "commit_rev":
"70f42a7c55ca69cdeb9aa6ec7e40ff3f155040b9"}

[commit-bot: I haz the power, 2016-12-19 23:01:43]

Description was changed from

==========
[Telemetry][Android] Wait for device under test to cool between pages.

If the platform supports waiting for the device temperature to drop, wait.
Currently this only implements andorid.

BUG=chromium:669923
==========

to

==========
[Telemetry][Android] Wait for device under test to cool between pages.

If the platform supports waiting for the device temperature to drop, wait.
Currently this only implements andorid.

BUG=chromium:669923

Review-Url: https://codereview.chromium.org/2541843007
Committed:
https://chromium.googlesource.com/external/github.com/catapult-project/catapu...
==========

[commit-bot: I haz the power, 2016-12-19 23:01:44]

Committed patchset #8 (id:140001) as
https://chromium.googlesource.com/external/github.com/catapult-project/catapu...