Issue 1104053006: [Storage] Blob Storage perf tests

nednguyen

nednguyen@google.com changed reviewers: + nednguyen@google.com

5 years, 7 months ago (2015-05-07 10:47:46 UTC) #1

nednguyen

https://codereview.chromium.org/1104053006/diff/100001/tools/perf/benchmarks/blob_storage.py File tools/perf/benchmarks/blob_storage.py (right): https://codereview.chromium.org/1104053006/diff/100001/tools/perf/benchmarks/blob_storage.py#newcode18 tools/perf/benchmarks/blob_storage.py:18: TOPLEVEL_CATEGORY) My guess is this category filter confused telemetry ...

5 years, 7 months ago (2015-05-07 10:51:06 UTC) #2

chromium-reviews

That worked, thanks! Next problem that I'm having: For some reason telemetry isn't reading some ...

5 years, 7 months ago (2015-05-07 20:59:20 UTC) #3

dmurph

On 2015/05/07 at 20:59:20, chromium-reviews wrote: > That worked, thanks! > > Next problem that ...

5 years, 7 months ago (2015-05-07 22:50:34 UTC) #4

nednguyen

https://codereview.chromium.org/1104053006/diff/120001/tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py File tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py (right): https://codereview.chromium.org/1104053006/diff/120001/tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py#newcode24 tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:24: self._AddResultsInternal(renderer_thread.parent.IterAllSlices(), Instead of IterAllSlices(), try IterAllEvents instead. Async event ...

5 years, 7 months ago (2015-05-08 01:35:12 UTC) #5

chromium-reviews

We can keep going on this. I think this is close to be land-able. It ...

5 years, 7 months ago (2015-05-08 01:38:27 UTC) #6

We can keep going on this. I think this is close to be land-able. It would
be great if you let others on the original thread know that we  have
resolved the issue and will keep the discussion in code review so they
don't spend more time there :)

On Thu, May 7, 2015, 3:50 PM  <dmurph@chromium.org> wrote:

> Reviewers: nednguyen (ooo til 05-07),
>
> Message:
> On 2015/05/07 at 20:59:20, chromium-reviews wrote:
> > That worked, thanks!
>
> > Next problem that I'm having:
>
> > For some reason telemetry isn't reading some of my traces.  Specifically,
> > the "BlobRequest" trace.  This is present in the 0.html file (as you can
> > see), but it doesn't show up in the trace events.  Does telemetry perhaps
> > not support the 'async' trace events in the timeline?
>
> > You can see these events definitely fall within the "Action_ReadBlob"
> > interaction below.
>
> > Thanks so much!
> > Daniel
>
> > On Thu, May 7, 2015 at 3:51 AM <nednguyen@google.com> wrote:
>
> > >
> > >
> > >
>
>
https://codereview.chromium.org/1104053006/diff/100001/tools/perf/benchmarks/...
> > > File tools/perf/benchmarks/blob_storage.py (right):
> > >
> > >
> > >
>
>
https://codereview.chromium.org/1104053006/diff/100001/tools/perf/benchmarks/...
> > > tools/perf/benchmarks/blob_storage.py:18
> > >
> <
>
https://codereview.chromium.org/1104053006/diff/100001/tools/perf/benchmarks/...
> >:
> > > TOPLEVEL_CATEGORY)
> > > My guess is this category filter confused telemetry trace parsing code
> > > :-)
> > >
> > > For a speculative workaround, can you change those two lines to:
> > > cat_filter =
> > > tracing_category_filter.TracingCategoryFilter.CreateNoOverheadFilter()
> > > cat_filter.AddIncludedCategory(TOPLEVEL_CATEGORY)
> > >
> > > and see if it works?
> > >
> > > https://codereview.chromium.org/1104053006/
> > >
>
> > To unsubscribe from this group and stop receiving emails from it, send an
> email to chromium-reviews+unsubscribe@chromium.org.
>
>
> Also, should I move this back to the main thread? Or should we keep
> discussion
> here now?
>
> Description:
> [Storage] Blob Storage perf tests
>
> BUG=
>
>
CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_perf_bisect;tryserver.chromium.perf:win_perf_bisect;tryserver.chromium.perf:android_nexus5_perf_bisect
>
> Please review this at https://codereview.chromium.org/1104053006/
>
> Base URL: https://chromium.googlesource.com/chromium/src.git@master
>
> Affected files (+276, -2 lines):
>    M storage/browser/blob/blob_url_request_job.cc
>    A tools/perf/benchmarks/blob_storage.py
>    A tools/perf/page_sets/blob/blob-workshop.html
>    A tools/perf/page_sets/blob_workshop.py
>    A tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py
>    M tools/telemetry/telemetry/web_perf/timeline_based_measurement.py
>
>
>

To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

dmurph

any more ideas? I printed out every event name that I get and no "BlobRequest" ...

5 years, 7 months ago (2015-05-08 18:59:23 UTC) #7

nednguyen

On 2015/05/08 18:59:23, dmurph wrote: > any more ideas? I printed out every event name ...

5 years, 7 months ago (2015-05-08 19:28:48 UTC) #8

nednguyen

On 2015/05/08 19:28:48, nednguyen wrote: > On 2015/05/08 18:59:23, dmurph wrote: > > any more ...

5 years, 7 months ago (2015-05-08 21:35:44 UTC) #9

On 2015/05/08 19:28:48, nednguyen wrote:
> On 2015/05/08 18:59:23, dmurph wrote:
> > any more ideas?  I printed out every event name that I get and no
> "BlobRequest"
> > events are in there.
> Then I think this is definitely a trace parsing issue. I will try to patch
this
> and run in on my machine.
> > 
> >
>
https://codereview.chromium.org/1104053006/diff/120001/tools/telemetry/teleme...
> > File tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py (right):
> > 
> >
>
https://codereview.chromium.org/1104053006/diff/120001/tools/telemetry/teleme...
> > tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:24:
> > self._AddResultsInternal(renderer_thread.parent.IterAllSlices(),
> > On 2015/05/08 01:35:12, nednguyen wrote:
> > > Instead of IterAllSlices(), try IterAllEvents instead. Async event is not
> > slice,
> > > I think
> > 
> > Hm, still doesn't work :'(

Hmh, when I run this on my local machine, it works. An example output:

./tools/perf/run_benchmark blob_storage.blob_storage --browser=release
[ RUN      ] blob-workshop.html
[       OK ] blob-workshop.html (11949 ms)
[ RUN      ] blob-workshop.html
[       OK ] blob-workshop.html (11530 ms)
[  PASSED  ] 2 tests.

Pages: [blob-workshop.html,blob-workshop.html]
RESULT Action_CreateBlob-blob_writes_2Bx200: blob-workshop.html=
[0.052000,0.052000,0.075000,0.019000,0.017000,0.017000,0.019000,0.020000,0.020000,0.019000,0.022000,0.020000,0.020000,0.019000,0.033000,0.036000,0.040000,0.044000,0.042000,0.042000,0.038000,0.042000,0.058000,0.032000,0.025000,0.039000,0.035000,0.031000,0.030000,0.030000,0.031000,0.014000,0.041000,0.031000,0.033000,0.023000,0.041000,0.042000,0.041000,0.041000,0.036000,0.035000,0.037000,0.042000,0.048000,0.033000,0.030000,0.031000,0.051000,0.028000,0.035000,0.032000,0.027000,0.027000,0.020000,0.015000,0.015000,0.016000,0.016000,0.015000,0.019000,0.014000,0.014000,0.018000,0.015000,0.015000,0.014000,0.026000,0.024000,0.024000,0.037000,0.036000,0.037000,0.037000,0.037000,0.026000,0.039000,0.036000,0.036000,0.037000,0.046000,0.038000,0.029000,0.045000,0.017000,0.021000,0.036000,0.029000,0.030000,0.028000,0.047000,0.031000,0.031000,0.030000,0.023000,0.014000,0.015000,0.014000,0.013000,0.015000,0.014000,0.013000,0.014000,0.013000,0.014000,0.014000,0.013000,0.014000,0.014000,0.037000,0.042000,0.051000,0.039000,0.042000,0.043000,0.016000,0.017000,0.016000,0.019000,0.016000,0.015000,0.038000,0.038000,0.016000,0.041000,0.032000,0.020000,0.017000,0.017000,0.016000,0.016000,0.017000,0.016000,0.016000,0.016000,0.024000,0.029000,0.032000,0.038000,0.034000,0.037000,0.031000,0.029000,0.033000,0.035000,0.032000,0.034000,0.033000,0.033000,0.032000,0.051000,0.041000,0.040000,0.042000,0.020000,0.014000,0.016000,0.015000,0.025000,0.034000,0.024000,0.015000,0.046000,0.038000,0.030000,0.053000,0.030000,0.031000,0.038000,0.040000,0.040000,0.041000,0.039000,0.038000,0.039000,0.038000,0.041000,0.039000,0.042000,0.042000,0.081000,0.044000,0.024000,0.029000,0.033000,0.023000,0.043000,0.034000,0.044000,0.028000,0.028000,0.033000,0.035000,0.027000,0.029000,0.051000,0.042000,0.044000,0.045000,0.044000]
ms
Avg Action_CreateBlob-blob_writes_2Bx200: 0.030445ms
Sd  Action_CreateBlob-blob_writes_2Bx200: 0.012047ms
RESULT Action_CreateBlob-blob_writes_2Bx200: blob-workshop.html=
[0.035000,0.022000,0.015000,0.028000,0.021000,0.014000,0.016000,0.021000,0.015000,0.014000,0.015000,0.015000,0.046000,0.039000,0.037000,0.040000,0.045000,0.039000,0.041000,0.038000,0.030000,0.020000,0.022000,0.029000,0.031000,0.038000,0.048000,0.051000,0.037000,0.040000,0.033000,0.037000,0.026000,0.034000,0.032000,0.039000,0.056000,0.066000,0.039000,0.037000,0.036000,0.033000,0.037000,0.041000,0.030000,0.031000,0.043000,0.035000,0.023000,0.033000,0.022000,0.029000,0.027000,0.022000,0.020000,0.023000,0.022000,0.021000,0.032000,0.023000,0.021000,0.014000,0.033000,0.030000,0.028000,0.030000,0.028000,0.028000,0.029000,0.027000,0.027000,0.035000,0.031000,0.034000,0.030000,0.017000,0.042000,0.049000,0.023000,0.025000,0.024000,0.019000,0.020000,0.029000,0.026000,0.018000,0.024000,0.017000,0.017000,0.019000,0.017000,0.023000,0.024000,0.020000,0.025000,0.026000,0.022000,0.025000,0.024000,0.024000,0.044000,0.041000,0.056000,0.085000,0.033000,0.030000,0.033000,0.027000,0.023000,0.029000,0.020000,0.017000,0.024000,0.019000,0.019000,0.046000,0.027000,0.025000,0.026000,0.025000,0.026000,0.039000,0.017000,0.015000,0.015000,0.014000,0.015000,0.015000,0.026000,0.035000,0.039000,0.030000,0.027000,0.028000,0.025000,0.038000,0.037000,0.035000,0.036000,0.028000,0.028000,0.035000,0.022000,0.020000,0.036000,0.034000,0.021000,0.022000,0.026000,0.024000,0.023000,0.024000,0.022000,0.047000,0.042000,0.044000,0.044000,0.042000,0.037000,0.038000,0.039000,0.041000,0.046000,0.042000,0.047000,0.045000,0.039000,0.051000,0.041000,0.048000,0.041000,0.021000,0.025000,0.021000,0.025000,0.022000,0.026000,0.026000,0.025000,0.025000,0.025000,0.020000,0.018000,0.021000,0.022000,0.022000,0.020000,0.019000,0.022000,0.021000,0.018000,0.021000,0.021000,0.039000,0.038000,0.036000,0.039000,0.037000,0.038000,0.031000]
ms
Avg Action_CreateBlob-blob_writes_2Bx200: 0.029760ms
Sd  Action_CreateBlob-blob_writes_2Bx200: 0.010594ms
*RESULT Action_CreateBlob-blob_writes_2Bx200:
Action_CreateBlob-blob_writes_2Bx200=
[0.052000,0.052000,0.075000,0.019000,0.017000,0.017000,0.019000,0.020000,0.020000,0.019000,0.022000,0.020000,0.020000,0.019000,0.033000,0.036000,0.040000,0.044000,0.042000,0.042000,0.038000,0.042000,0.058000,0.032000,0.025000,0.039000,0.035000,0.031000,0.030000,0.030000,0.031000,0.014000,0.041000,0.031000,0.033000,0.023000,0.041000,0.042000,0.041000,0.041000,0.036000,0.035000,0.037000,0.042000,0.048000,0.033000,0.030000,0.031000,0.051000,0.028000,0.035000,0.032000,0.027000,0.027000,0.020000,0.015000,0.015000,0.016000,0.016000,0.015000,0.019000,0.014000,0.014000,0.018000,0.015000,0.015000,0.014000,0.026000,0.024000,0.024000,0.037000,0.036000,0.037000,0.037000,0.037000,0.026000,0.039000,0.036000,0.036000,0.037000,0.046000,0.038000,0.029000,0.045000,0.017000,0.021000,0.036000,0.029000,0.030000,0.028000,0.047000,0.031000,0.031000,0.030000,0.023000,0.014000,0.015000,0.014000,0.013000,0.015000,0.014000,0.013000,0.014000,0.013000,0.014000,0.014000,0.013000,0.014000,0.014000,0.037000,0.042000,0.051000,0.039000,0.042000,0.043000,0.016000,0.017000,0.016000,0.019000,0.016000,0.015000,0.038000,0.038000,0.016000,0.041000,0.032000,0.020000,0.017000,0.017000,0.016000,0.016000,0.017000,0.016000,0.016000,0.016000,0.024000,0.029000,0.032000,0.038000,0.034000,0.037000,0.031000,0.029000,0.033000,0.035000,0.032000,0.034000,0.033000,0.033000,0.032000,0.051000,0.041000,0.040000,0.042000,0.020000,0.014000,0.016000,0.015000,0.025000,0.034000,0.024000,0.015000,0.046000,0.038000,0.030000,0.053000,0.030000,0.031000,0.038000,0.040000,0.040000,0.041000,0.039000,0.038000,0.039000,0.038000,0.041000,0.039000,0.042000,0.042000,0.081000,0.044000,0.024000,0.029000,0.033000,0.023000,0.043000,0.034000,0.044000,0.028000,0.028000,0.033000,0.035000,0.027000,0.029000,0.051000,0.042000,0.044000,0.045000,0.044000,0.035000,0.022000,0.015000,0.028000,0.021000,0.014000,0.016000,0.021000,0.015000,0.014000,0.015000,0.015000,0.046000,0.039000,0.037000,0.040000,0.045000,0.039000,0.041000,0.038000,0.030000,0.020000,0.022000,0.029000,0.031000,0.038000,0.048000,0.051000,0.037000,0.040000,0.033000,0.037000,0.026000,0.034000,0.032000,0.039000,0.056000,0.066000,0.039000,0.037000,0.036000,0.033000,0.037000,0.041000,0.030000,0.031000,0.043000,0.035000,0.023000,0.033000,0.022000,0.029000,0.027000,0.022000,0.020000,0.023000,0.022000,0.021000,0.032000,0.023000,0.021000,0.014000,0.033000,0.030000,0.028000,0.030000,0.028000,0.028000,0.029000,0.027000,0.027000,0.035000,0.031000,0.034000,0.030000,0.017000,0.042000,0.049000,0.023000,0.025000,0.024000,0.019000,0.020000,0.029000,0.026000,0.018000,0.024000,0.017000,0.017000,0.019000,0.017000,0.023000,0.024000,0.020000,0.025000,0.026000,0.022000,0.025000,0.024000,0.024000,0.044000,0.041000,0.056000,0.085000,0.033000,0.030000,0.033000,0.027000,0.023000,0.029000,0.020000,0.017000,0.024000,0.019000,0.019000,0.046000,0.027000,0.025000,0.026000,0.025000,0.026000,0.039000,0.017000,0.015000,0.015000,0.014000,0.015000,0.015000,0.026000,0.035000,0.039000,0.030000,0.027000,0.028000,0.025000,0.038000,0.037000,0.035000,0.036000,0.028000,0.028000,0.035000,0.022000,0.020000,0.036000,0.034000,0.021000,0.022000,0.026000,0.024000,0.023000,0.024000,0.022000,0.047000,0.042000,0.044000,0.044000,0.042000,0.037000,0.038000,0.039000,0.041000,0.046000,0.042000,0.047000,0.045000,0.039000,0.051000,0.041000,0.048000,0.041000,0.021000,0.025000,0.021000,0.025000,0.022000,0.026000,0.026000,0.025000,0.025000,0.025000,0.020000,0.018000,0.021000,0.022000,0.022000,0.020000,0.019000,0.022000,0.021000,0.018000,0.021000,0.021000,0.039000,0.038000,0.036000,0.039000,0.037000,0.038000,0.031000]
ms
Avg Action_CreateBlob-blob_writes_2Bx200: 0.030103ms
Sd  Action_CreateBlob-blob_writes_2Bx200: 0.011335ms
RESULT telemetry_page_measurement_results: num_failed= 0 count
RESULT telemetry_page_measurement_results: num_errored= 0 count

chromium-reviews

Right, there are no "blob_reads". Only the writes. If you print out every event, you ...

5 years, 7 months ago (2015-05-08 21:36:45 UTC) #10

Right, there are no "blob_reads".  Only the writes.

If you print out every event, you never see the "BlobRequest" event.
However, if you look at the trace, it's there.

On Fri, May 8, 2015 at 2:35 PM <nednguyen@google.com> wrote:

> On 2015/05/08 19:28:48, nednguyen wrote:
> > On 2015/05/08 18:59:23, dmurph wrote:
> > > any more ideas?  I printed out every event name that I get and no
> > "BlobRequest"
> > > events are in there.
> > Then I think this is definitely a trace parsing issue. I will try to
> patch
> this
> > and run in on my machine.
> > >
> > >
>
>
>
https://codereview.chromium.org/1104053006/diff/120001/tools/telemetry/teleme...
> > > File tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py
> > (right):
> > >
> > >
>
>
>
https://codereview.chromium.org/1104053006/diff/120001/tools/telemetry/teleme...
> > > tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:24:
> > > self._AddResultsInternal(renderer_thread.parent.IterAllSlices(),
> > > On 2015/05/08 01:35:12, nednguyen wrote:
> > > > Instead of IterAllSlices(), try IterAllEvents instead. Async event is
> > not
> > > slice,
> > > > I think
> > >
> > > Hm, still doesn't work :'(
>
> Hmh, when I run this on my local machine, it works. An example output:
>
> ./tools/perf/run_benchmark blob_storage.blob_storage --browser=release
> [ RUN      ] blob-workshop.html
> [       OK ] blob-workshop.html (11949 ms)
> [ RUN      ] blob-workshop.html
> [       OK ] blob-workshop.html (11530 ms)
> [  PASSED  ] 2 tests.
>
> Pages: [blob-workshop.html,blob-workshop.html]
> RESULT Action_CreateBlob-blob_writes_2Bx200: blob-workshop.html=
>
>
[0.052000,0.052000,0.075000,0.019000,0.017000,0.017000,0.019000,0.020000,0.020000,0.019000,0.022000,0.020000,0.020000,0.019000,0.033000,0.036000,0.040000,0.044000,0.042000,0.042000,0.038000,0.042000,0.058000,0.032000,0.025000,0.039000,0.035000,0.031000,0.030000,0.030000,0.031000,0.014000,0.041000,0.031000,0.033000,0.023000,0.041000,0.042000,0.041000,0.041000,0.036000,0.035000,0.037000,0.042000,0.048000,0.033000,0.030000,0.031000,0.051000,0.028000,0.035000,0.032000,0.027000,0.027000,0.020000,0.015000,0.015000,0.016000,0.016000,0.015000,0.019000,0.014000,0.014000,0.018000,0.015000,0.015000,0.014000,0.026000,0.024000,0.024000,0.037000,0.036000,0.037000,0.037000,0.037000,0.026000,0.039000,0.036000,0.036000,0.037000,0.046000,0.038000,0.029000,0.045000,0.017000,0.021000,0.036000,0.029000,0.030000,0.028000,0.047000,0.031000,0.031000,0.030000,0.023000,0.014000,0.015000,0.014000,0.013000,0.015000,0.014000,0.013000,0.014000,0.013000,0.014000,0.014000,0.013000,0.014000,0.014000,0.037000,0.042000,0.051000,0.039000,0.042000,0.043000,0.016000,0.017000,0.016000,0.019000,0.016000,0.015000,0.038000,0.038000,0.016000,0.041000,0.032000,0.020000,0.017000,0.017000,0.016000,0.016000,0.017000,0.016000,0.016000,0.016000,0.024000,0.029000,0.032000,0.038000,0.034000,0.037000,0.031000,0.029000,0.033000,0.035000,0.032000,0.034000,0.033000,0.033000,0.032000,0.051000,0.041000,0.040000,0.042000,0.020000,0.014000,0.016000,0.015000,0.025000,0.034000,0.024000,0.015000,0.046000,0.038000,0.030000,0.053000,0.030000,0.031000,0.038000,0.040000,0.040000,0.041000,0.039000,0.038000,0.039000,0.038000,0.041000,0.039000,0.042000,0.042000,0.081000,0.044000,0.024000,0.029000,0.033000,0.023000,0.043000,0.034000,0.044000,0.028000,0.028000,0.033000,0.035000,0.027000,0.029000,0.051000,0.042000,0.044000,0.045000,0.044000]
> ms
> Avg Action_CreateBlob-blob_writes_2Bx200: 0.030445ms
> Sd  Action_CreateBlob-blob_writes_2Bx200: 0.012047ms
> RESULT Action_CreateBlob-blob_writes_2Bx200: blob-workshop.html=
>
>
[0.035000,0.022000,0.015000,0.028000,0.021000,0.014000,0.016000,0.021000,0.015000,0.014000,0.015000,0.015000,0.046000,0.039000,0.037000,0.040000,0.045000,0.039000,0.041000,0.038000,0.030000,0.020000,0.022000,0.029000,0.031000,0.038000,0.048000,0.051000,0.037000,0.040000,0.033000,0.037000,0.026000,0.034000,0.032000,0.039000,0.056000,0.066000,0.039000,0.037000,0.036000,0.033000,0.037000,0.041000,0.030000,0.031000,0.043000,0.035000,0.023000,0.033000,0.022000,0.029000,0.027000,0.022000,0.020000,0.023000,0.022000,0.021000,0.032000,0.023000,0.021000,0.014000,0.033000,0.030000,0.028000,0.030000,0.028000,0.028000,0.029000,0.027000,0.027000,0.035000,0.031000,0.034000,0.030000,0.017000,0.042000,0.049000,0.023000,0.025000,0.024000,0.019000,0.020000,0.029000,0.026000,0.018000,0.024000,0.017000,0.017000,0.019000,0.017000,0.023000,0.024000,0.020000,0.025000,0.026000,0.022000,0.025000,0.024000,0.024000,0.044000,0.041000,0.056000,0.085000,0.033000,0.030000,0.033000,0.027000,0.023000,0.029000,0.020000,0.017000,0.024000,0.019000,0.019000,0.046000,0.027000,0.025000,0.026000,0.025000,0.026000,0.039000,0.017000,0.015000,0.015000,0.014000,0.015000,0.015000,0.026000,0.035000,0.039000,0.030000,0.027000,0.028000,0.025000,0.038000,0.037000,0.035000,0.036000,0.028000,0.028000,0.035000,0.022000,0.020000,0.036000,0.034000,0.021000,0.022000,0.026000,0.024000,0.023000,0.024000,0.022000,0.047000,0.042000,0.044000,0.044000,0.042000,0.037000,0.038000,0.039000,0.041000,0.046000,0.042000,0.047000,0.045000,0.039000,0.051000,0.041000,0.048000,0.041000,0.021000,0.025000,0.021000,0.025000,0.022000,0.026000,0.026000,0.025000,0.025000,0.025000,0.020000,0.018000,0.021000,0.022000,0.022000,0.020000,0.019000,0.022000,0.021000,0.018000,0.021000,0.021000,0.039000,0.038000,0.036000,0.039000,0.037000,0.038000,0.031000]
> ms
> Avg Action_CreateBlob-blob_writes_2Bx200: 0.029760ms
> Sd  Action_CreateBlob-blob_writes_2Bx200: 0.010594ms
> *RESULT Action_CreateBlob-blob_writes_2Bx200:
> Action_CreateBlob-blob_writes_2Bx200=
>
>
[0.052000,0.052000,0.075000,0.019000,0.017000,0.017000,0.019000,0.020000,0.020000,0.019000,0.022000,0.020000,0.020000,0.019000,0.033000,0.036000,0.040000,0.044000,0.042000,0.042000,0.038000,0.042000,0.058000,0.032000,0.025000,0.039000,0.035000,0.031000,0.030000,0.030000,0.031000,0.014000,0.041000,0.031000,0.033000,0.023000,0.041000,0.042000,0.041000,0.041000,0.036000,0.035000,0.037000,0.042000,0.048000,0.033000,0.030000,0.031000,0.051000,0.028000,0.035000,0.032000,0.027000,0.027000,0.020000,0.015000,0.015000,0.016000,0.016000,0.015000,0.019000,0.014000,0.014000,0.018000,0.015000,0.015000,0.014000,0.026000,0.024000,0.024000,0.037000,0.036000,0.037000,0.037000,0.037000,0.026000,0.039000,0.036000,0.036000,0.037000,0.046000,0.038000,0.029000,0.045000,0.017000,0.021000,0.036000,0.029000,0.030000,0.028000,0.047000,0.031000,0.031000,0.030000,0.023000,0.014000,0.015000,0.014000,0.013000,0.015000,0.014000,0.013000,0.014000,0.013000,0.014000,0.014000,0.013000,0.014000,0.014000,0.037000,0.042000,0.051000,0.039000,0.042000,0.043000,0.016000,0.017000,0.016000,0.019000,0.016000,0.015000,0.038000,0.038000,0.016000,0.041000,0.032000,0.020000,0.017000,0.017000,0.016000,0.016000,0.017000,0.016000,0.016000,0.016000,0.024000,0.029000,0.032000,0.038000,0.034000,0.037000,0.031000,0.029000,0.033000,0.035000,0.032000,0.034000,0.033000,0.033000,0.032000,0.051000,0.041000,0.040000,0.042000,0.020000,0.014000,0.016000,0.015000,0.025000,0.034000,0.024000,0.015000,0.046000,0.038000,0.030000,0.053000,0.030000,0.031000,0.038000,0.040000,0.040000,0.041000,0.039000,0.038000,0.039000,0.038000,0.041000,0.039000,0.042000,0.042000,0.081000,0.044000,0.024000,0.029000,0.033000,0.023000,0.043000,0.034000,0.044000,0.028000,0.028000,0.033000,0.035000,0.027000,0.029000,0.051000,0.042000,0.044000,0.045000,0.044000,0.035000,0.022000,0.015000,0.028000,0.021000,0.014000,0.016000,0.021000,0.015000,0.014000,0.015000,0.015000,0.046000,0.039000,0.037000,0.040000,0.045000,0.039000,0.041000,0.038000,0.030000,0.020000,0.022000,0.029000,0.031000,0.038000,0.048000,0.051000,0.037000,0.040000,0.033000,0.037000,0.026000,0.034000,0.032000,0.039000,0.056000,0.066000,0.039000,0.037000,0.036000,0.033000,0.037000,0.041000,0.030000,0.031000,0.043000,0.035000,0.023000,0.033000,0.022000,0.029000,0.027000,0.022000,0.020000,0.023000,0.022000,0.021000,0.032000,0.023000,0.021000,0.014000,0.033000,0.030000,0.028000,0.030000,0.028000,0.028000,0.029000,0.027000,0.027000,0.035000,0.031000,0.034000,0.030000,0.017000,0.042000,0.049000,0.023000,0.025000,0.024000,0.019000,0.020000,0.029000,0.026000,0.018000,0.024000,0.017000,0.017000,0.019000,0.017000,0.023000,0.024000,0.020000,0.025000,0.026000,0.022000,0.025000,0.024000,0.024000,0.044000,0.041000,0.056000,0.085000,0.033000,0.030000,0.033000,0.027000,0.023000,0.029000,0.020000,0.017000,0.024000,0.019000,0.019000,0.046000,0.027000,0.025000,0.026000,0.025000,0.026000,0.039000,0.017000,0.015000,0.015000,0.014000,0.015000,0.015000,0.026000,0.035000,0.039000,0.030000,0.027000,0.028000,0.025000,0.038000,0.037000,0.035000,0.036000,0.028000,0.028000,0.035000,0.022000,0.020000,0.036000,0.034000,0.021000,0.022000,0.026000,0.024000,0.023000,0.024000,0.022000,0.047000,0.042000,0.044000,0.044000,0.042000,0.037000,0.038000,0.039000,0.041000,0.046000,0.042000,0.047000,0.045000,0.039000,0.051000,0.041000,0.048000,0.041000,0.021000,0.025000,0.021000,0.025000,0.022000,0.026000,0.026000,0.025000,0.025000,0.025000,0.020000,0.018000,0.021000,0.022000,0.022000,0.020000,0.019000,0.022000,0.021000,0.018000,0.021000,0.021000,0.039000,0.038000,0.036000,0.039000,0.037000,0.038000,0.031000]
> ms
> Avg Action_CreateBlob-blob_writes_2Bx200: 0.030103ms
> Sd  Action_CreateBlob-blob_writes_2Bx200: 0.011335ms
> RESULT telemetry_page_measurement_results: num_failed= 0 count
> RESULT telemetry_page_measurement_results: num_errored= 0 count
>
>
> https://codereview.chromium.org/1104053006/
>

To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

dmurph

Whoops, replying with @chromium Right, there are no "blob_reads". Only the writes. If you print ...

5 years, 7 months ago (2015-05-08 21:38:01 UTC) #11

Whoops, replying with @chromium

Right, there are no "blob_reads".  Only the writes.

If you print out every event, you never see the "BlobRequest" event.
However, if you look at the trace, it's there.

(you can comment out all be the user story to make the run go faster)

Daniel
On Fri, May 8, 2015 at 2:35 PM <nednguyen@google.com> wrote:

> On 2015/05/08 19:28:48, nednguyen wrote:
> > On 2015/05/08 18:59:23, dmurph wrote:
> > > any more ideas?  I printed out every event name that I get and no
> > "BlobRequest"
> > > events are in there.
> > Then I think this is definitely a trace parsing issue. I will try to
> patch
> this
> > and run in on my machine.
> > >
> > >
>
>
>
https://codereview.chromium.org/1104053006/diff/120001/tools/telemetry/teleme...
> > > File tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py
> > (right):
> > >
> > >
>
>
>
https://codereview.chromium.org/1104053006/diff/120001/tools/telemetry/teleme...
> > > tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:24:
> > > self._AddResultsInternal(renderer_thread.parent.IterAllSlices(),
> > > On 2015/05/08 01:35:12, nednguyen wrote:
> > > > Instead of IterAllSlices(), try IterAllEvents instead. Async event is
> > not
> > > slice,
> > > > I think
> > >
> > > Hm, still doesn't work :'(
>
> Hmh, when I run this on my local machine, it works. An example output:
>
> ./tools/perf/run_benchmark blob_storage.blob_storage --browser=release
> [ RUN      ] blob-workshop.html
> [       OK ] blob-workshop.html (11949 ms)
> [ RUN      ] blob-workshop.html
> [       OK ] blob-workshop.html (11530 ms)
> [  PASSED  ] 2 tests.
>
> Pages: [blob-workshop.html,blob-workshop.html]
> RESULT Action_CreateBlob-blob_writes_2Bx200: blob-workshop.html=
>
>
[0.052000,0.052000,0.075000,0.019000,0.017000,0.017000,0.019000,0.020000,0.020000,0.019000,0.022000,0.020000,0.020000,0.019000,0.033000,0.036000,0.040000,0.044000,0.042000,0.042000,0.038000,0.042000,0.058000,0.032000,0.025000,0.039000,0.035000,0.031000,0.030000,0.030000,0.031000,0.014000,0.041000,0.031000,0.033000,0.023000,0.041000,0.042000,0.041000,0.041000,0.036000,0.035000,0.037000,0.042000,0.048000,0.033000,0.030000,0.031000,0.051000,0.028000,0.035000,0.032000,0.027000,0.027000,0.020000,0.015000,0.015000,0.016000,0.016000,0.015000,0.019000,0.014000,0.014000,0.018000,0.015000,0.015000,0.014000,0.026000,0.024000,0.024000,0.037000,0.036000,0.037000,0.037000,0.037000,0.026000,0.039000,0.036000,0.036000,0.037000,0.046000,0.038000,0.029000,0.045000,0.017000,0.021000,0.036000,0.029000,0.030000,0.028000,0.047000,0.031000,0.031000,0.030000,0.023000,0.014000,0.015000,0.014000,0.013000,0.015000,0.014000,0.013000,0.014000,0.013000,0.014000,0.014000,0.013000,0.014000,0.014000,0.037000,0.042000,0.051000,0.039000,0.042000,0.043000,0.016000,0.017000,0.016000,0.019000,0.016000,0.015000,0.038000,0.038000,0.016000,0.041000,0.032000,0.020000,0.017000,0.017000,0.016000,0.016000,0.017000,0.016000,0.016000,0.016000,0.024000,0.029000,0.032000,0.038000,0.034000,0.037000,0.031000,0.029000,0.033000,0.035000,0.032000,0.034000,0.033000,0.033000,0.032000,0.051000,0.041000,0.040000,0.042000,0.020000,0.014000,0.016000,0.015000,0.025000,0.034000,0.024000,0.015000,0.046000,0.038000,0.030000,0.053000,0.030000,0.031000,0.038000,0.040000,0.040000,0.041000,0.039000,0.038000,0.039000,0.038000,0.041000,0.039000,0.042000,0.042000,0.081000,0.044000,0.024000,0.029000,0.033000,0.023000,0.043000,0.034000,0.044000,0.028000,0.028000,0.033000,0.035000,0.027000,0.029000,0.051000,0.042000,0.044000,0.045000,0.044000]
> ms
> Avg Action_CreateBlob-blob_writes_2Bx200: 0.030445ms
> Sd  Action_CreateBlob-blob_writes_2Bx200: 0.012047ms
> RESULT Action_CreateBlob-blob_writes_2Bx200: blob-workshop.html=
>
>
[0.035000,0.022000,0.015000,0.028000,0.021000,0.014000,0.016000,0.021000,0.015000,0.014000,0.015000,0.015000,0.046000,0.039000,0.037000,0.040000,0.045000,0.039000,0.041000,0.038000,0.030000,0.020000,0.022000,0.029000,0.031000,0.038000,0.048000,0.051000,0.037000,0.040000,0.033000,0.037000,0.026000,0.034000,0.032000,0.039000,0.056000,0.066000,0.039000,0.037000,0.036000,0.033000,0.037000,0.041000,0.030000,0.031000,0.043000,0.035000,0.023000,0.033000,0.022000,0.029000,0.027000,0.022000,0.020000,0.023000,0.022000,0.021000,0.032000,0.023000,0.021000,0.014000,0.033000,0.030000,0.028000,0.030000,0.028000,0.028000,0.029000,0.027000,0.027000,0.035000,0.031000,0.034000,0.030000,0.017000,0.042000,0.049000,0.023000,0.025000,0.024000,0.019000,0.020000,0.029000,0.026000,0.018000,0.024000,0.017000,0.017000,0.019000,0.017000,0.023000,0.024000,0.020000,0.025000,0.026000,0.022000,0.025000,0.024000,0.024000,0.044000,0.041000,0.056000,0.085000,0.033000,0.030000,0.033000,0.027000,0.023000,0.029000,0.020000,0.017000,0.024000,0.019000,0.019000,0.046000,0.027000,0.025000,0.026000,0.025000,0.026000,0.039000,0.017000,0.015000,0.015000,0.014000,0.015000,0.015000,0.026000,0.035000,0.039000,0.030000,0.027000,0.028000,0.025000,0.038000,0.037000,0.035000,0.036000,0.028000,0.028000,0.035000,0.022000,0.020000,0.036000,0.034000,0.021000,0.022000,0.026000,0.024000,0.023000,0.024000,0.022000,0.047000,0.042000,0.044000,0.044000,0.042000,0.037000,0.038000,0.039000,0.041000,0.046000,0.042000,0.047000,0.045000,0.039000,0.051000,0.041000,0.048000,0.041000,0.021000,0.025000,0.021000,0.025000,0.022000,0.026000,0.026000,0.025000,0.025000,0.025000,0.020000,0.018000,0.021000,0.022000,0.022000,0.020000,0.019000,0.022000,0.021000,0.018000,0.021000,0.021000,0.039000,0.038000,0.036000,0.039000,0.037000,0.038000,0.031000]
> ms
> Avg Action_CreateBlob-blob_writes_2Bx200: 0.029760ms
> Sd  Action_CreateBlob-blob_writes_2Bx200: 0.010594ms
> *RESULT Action_CreateBlob-blob_writes_2Bx200:
> Action_CreateBlob-blob_writes_2Bx200=
>
>
[0.052000,0.052000,0.075000,0.019000,0.017000,0.017000,0.019000,0.020000,0.020000,0.019000,0.022000,0.020000,0.020000,0.019000,0.033000,0.036000,0.040000,0.044000,0.042000,0.042000,0.038000,0.042000,0.058000,0.032000,0.025000,0.039000,0.035000,0.031000,0.030000,0.030000,0.031000,0.014000,0.041000,0.031000,0.033000,0.023000,0.041000,0.042000,0.041000,0.041000,0.036000,0.035000,0.037000,0.042000,0.048000,0.033000,0.030000,0.031000,0.051000,0.028000,0.035000,0.032000,0.027000,0.027000,0.020000,0.015000,0.015000,0.016000,0.016000,0.015000,0.019000,0.014000,0.014000,0.018000,0.015000,0.015000,0.014000,0.026000,0.024000,0.024000,0.037000,0.036000,0.037000,0.037000,0.037000,0.026000,0.039000,0.036000,0.036000,0.037000,0.046000,0.038000,0.029000,0.045000,0.017000,0.021000,0.036000,0.029000,0.030000,0.028000,0.047000,0.031000,0.031000,0.030000,0.023000,0.014000,0.015000,0.014000,0.013000,0.015000,0.014000,0.013000,0.014000,0.013000,0.014000,0.014000,0.013000,0.014000,0.014000,0.037000,0.042000,0.051000,0.039000,0.042000,0.043000,0.016000,0.017000,0.016000,0.019000,0.016000,0.015000,0.038000,0.038000,0.016000,0.041000,0.032000,0.020000,0.017000,0.017000,0.016000,0.016000,0.017000,0.016000,0.016000,0.016000,0.024000,0.029000,0.032000,0.038000,0.034000,0.037000,0.031000,0.029000,0.033000,0.035000,0.032000,0.034000,0.033000,0.033000,0.032000,0.051000,0.041000,0.040000,0.042000,0.020000,0.014000,0.016000,0.015000,0.025000,0.034000,0.024000,0.015000,0.046000,0.038000,0.030000,0.053000,0.030000,0.031000,0.038000,0.040000,0.040000,0.041000,0.039000,0.038000,0.039000,0.038000,0.041000,0.039000,0.042000,0.042000,0.081000,0.044000,0.024000,0.029000,0.033000,0.023000,0.043000,0.034000,0.044000,0.028000,0.028000,0.033000,0.035000,0.027000,0.029000,0.051000,0.042000,0.044000,0.045000,0.044000,0.035000,0.022000,0.015000,0.028000,0.021000,0.014000,0.016000,0.021000,0.015000,0.014000,0.015000,0.015000,0.046000,0.039000,0.037000,0.040000,0.045000,0.039000,0.041000,0.038000,0.030000,0.020000,0.022000,0.029000,0.031000,0.038000,0.048000,0.051000,0.037000,0.040000,0.033000,0.037000,0.026000,0.034000,0.032000,0.039000,0.056000,0.066000,0.039000,0.037000,0.036000,0.033000,0.037000,0.041000,0.030000,0.031000,0.043000,0.035000,0.023000,0.033000,0.022000,0.029000,0.027000,0.022000,0.020000,0.023000,0.022000,0.021000,0.032000,0.023000,0.021000,0.014000,0.033000,0.030000,0.028000,0.030000,0.028000,0.028000,0.029000,0.027000,0.027000,0.035000,0.031000,0.034000,0.030000,0.017000,0.042000,0.049000,0.023000,0.025000,0.024000,0.019000,0.020000,0.029000,0.026000,0.018000,0.024000,0.017000,0.017000,0.019000,0.017000,0.023000,0.024000,0.020000,0.025000,0.026000,0.022000,0.025000,0.024000,0.024000,0.044000,0.041000,0.056000,0.085000,0.033000,0.030000,0.033000,0.027000,0.023000,0.029000,0.020000,0.017000,0.024000,0.019000,0.019000,0.046000,0.027000,0.025000,0.026000,0.025000,0.026000,0.039000,0.017000,0.015000,0.015000,0.014000,0.015000,0.015000,0.026000,0.035000,0.039000,0.030000,0.027000,0.028000,0.025000,0.038000,0.037000,0.035000,0.036000,0.028000,0.028000,0.035000,0.022000,0.020000,0.036000,0.034000,0.021000,0.022000,0.026000,0.024000,0.023000,0.024000,0.022000,0.047000,0.042000,0.044000,0.044000,0.042000,0.037000,0.038000,0.039000,0.041000,0.046000,0.042000,0.047000,0.045000,0.039000,0.051000,0.041000,0.048000,0.041000,0.021000,0.025000,0.021000,0.025000,0.022000,0.026000,0.026000,0.025000,0.025000,0.025000,0.020000,0.018000,0.021000,0.022000,0.022000,0.020000,0.019000,0.022000,0.021000,0.018000,0.021000,0.021000,0.039000,0.038000,0.036000,0.039000,0.037000,0.038000,0.031000]
> ms
> Avg Action_CreateBlob-blob_writes_2Bx200: 0.030103ms
> Sd  Action_CreateBlob-blob_writes_2Bx200: 0.011335ms
> RESULT telemetry_page_measurement_results: num_failed= 0 count
> RESULT telemetry_page_measurement_results: num_errored= 0 count
>
>
> https://codereview.chromium.org/1104053006/
>

To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

chromium-reviews

I actually don't find BlobRequest in the trace-viewer either, only Registry::RegisterBlob On Fri, May 8, ...

5 years, 7 months ago (2015-05-08 22:04:20 UTC) #12

I actually don't find BlobRequest in the trace-viewer either, only
 Registry::RegisterBlob

On Fri, May 8, 2015 at 2:38 PM Daniel Murphy <dmurph@chromium.org> wrote:

> Whoops, replying with @chromium
>
> Right, there are no "blob_reads".  Only the writes.
>
> If you print out every event, you never see the "BlobRequest" event.
> However, if you look at the trace, it's there.
>
> (you can comment out all be the user story to make the run go faster)
>
> Daniel
> On Fri, May 8, 2015 at 2:35 PM <nednguyen@google.com> wrote:
>
>> On 2015/05/08 19:28:48, nednguyen wrote:
>> > On 2015/05/08 18:59:23, dmurph wrote:
>> > > any more ideas?  I printed out every event name that I get and no
>> > "BlobRequest"
>> > > events are in there.
>> > Then I think this is definitely a trace parsing issue. I will try to
>> patch
>> this
>> > and run in on my machine.
>> > >
>> > >
>>
>>
>>
https://codereview.chromium.org/1104053006/diff/120001/tools/telemetry/teleme...
>> > > File tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py
>> > (right):
>> > >
>> > >
>>
>>
>>
https://codereview.chromium.org/1104053006/diff/120001/tools/telemetry/teleme...
>> > > tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:24:
>> > > self._AddResultsInternal(renderer_thread.parent.IterAllSlices(),
>> > > On 2015/05/08 01:35:12, nednguyen wrote:
>> > > > Instead of IterAllSlices(), try IterAllEvents instead. Async event
>> is
>> > not
>> > > slice,
>> > > > I think
>> > >
>> > > Hm, still doesn't work :'(
>>
>> Hmh, when I run this on my local machine, it works. An example output:
>>
>> ./tools/perf/run_benchmark blob_storage.blob_storage --browser=release
>> [ RUN      ] blob-workshop.html
>> [       OK ] blob-workshop.html (11949 ms)
>> [ RUN      ] blob-workshop.html
>> [       OK ] blob-workshop.html (11530 ms)
>> [  PASSED  ] 2 tests.
>>
>> Pages: [blob-workshop.html,blob-workshop.html]
>> RESULT Action_CreateBlob-blob_writes_2Bx200: blob-workshop.html=
>>
>>
[0.052000,0.052000,0.075000,0.019000,0.017000,0.017000,0.019000,0.020000,0.020000,0.019000,0.022000,0.020000,0.020000,0.019000,0.033000,0.036000,0.040000,0.044000,0.042000,0.042000,0.038000,0.042000,0.058000,0.032000,0.025000,0.039000,0.035000,0.031000,0.030000,0.030000,0.031000,0.014000,0.041000,0.031000,0.033000,0.023000,0.041000,0.042000,0.041000,0.041000,0.036000,0.035000,0.037000,0.042000,0.048000,0.033000,0.030000,0.031000,0.051000,0.028000,0.035000,0.032000,0.027000,0.027000,0.020000,0.015000,0.015000,0.016000,0.016000,0.015000,0.019000,0.014000,0.014000,0.018000,0.015000,0.015000,0.014000,0.026000,0.024000,0.024000,0.037000,0.036000,0.037000,0.037000,0.037000,0.026000,0.039000,0.036000,0.036000,0.037000,0.046000,0.038000,0.029000,0.045000,0.017000,0.021000,0.036000,0.029000,0.030000,0.028000,0.047000,0.031000,0.031000,0.030000,0.023000,0.014000,0.015000,0.014000,0.013000,0.015000,0.014000,0.013000,0.014000,0.013000,0.014000,0.014000,0.013000,0.014000,0.014000,0.037000,0.042000,0.051000,0.039000,0.042000,0.043000,0.016000,0.017000,0.016000,0.019000,0.016000,0.015000,0.038000,0.038000,0.016000,0.041000,0.032000,0.020000,0.017000,0.017000,0.016000,0.016000,0.017000,0.016000,0.016000,0.016000,0.024000,0.029000,0.032000,0.038000,0.034000,0.037000,0.031000,0.029000,0.033000,0.035000,0.032000,0.034000,0.033000,0.033000,0.032000,0.051000,0.041000,0.040000,0.042000,0.020000,0.014000,0.016000,0.015000,0.025000,0.034000,0.024000,0.015000,0.046000,0.038000,0.030000,0.053000,0.030000,0.031000,0.038000,0.040000,0.040000,0.041000,0.039000,0.038000,0.039000,0.038000,0.041000,0.039000,0.042000,0.042000,0.081000,0.044000,0.024000,0.029000,0.033000,0.023000,0.043000,0.034000,0.044000,0.028000,0.028000,0.033000,0.035000,0.027000,0.029000,0.051000,0.042000,0.044000,0.045000,0.044000]
>> ms
>> Avg Action_CreateBlob-blob_writes_2Bx200: 0.030445ms
>> Sd  Action_CreateBlob-blob_writes_2Bx200: 0.012047ms
>> RESULT Action_CreateBlob-blob_writes_2Bx200: blob-workshop.html=
>>
>>
[0.035000,0.022000,0.015000,0.028000,0.021000,0.014000,0.016000,0.021000,0.015000,0.014000,0.015000,0.015000,0.046000,0.039000,0.037000,0.040000,0.045000,0.039000,0.041000,0.038000,0.030000,0.020000,0.022000,0.029000,0.031000,0.038000,0.048000,0.051000,0.037000,0.040000,0.033000,0.037000,0.026000,0.034000,0.032000,0.039000,0.056000,0.066000,0.039000,0.037000,0.036000,0.033000,0.037000,0.041000,0.030000,0.031000,0.043000,0.035000,0.023000,0.033000,0.022000,0.029000,0.027000,0.022000,0.020000,0.023000,0.022000,0.021000,0.032000,0.023000,0.021000,0.014000,0.033000,0.030000,0.028000,0.030000,0.028000,0.028000,0.029000,0.027000,0.027000,0.035000,0.031000,0.034000,0.030000,0.017000,0.042000,0.049000,0.023000,0.025000,0.024000,0.019000,0.020000,0.029000,0.026000,0.018000,0.024000,0.017000,0.017000,0.019000,0.017000,0.023000,0.024000,0.020000,0.025000,0.026000,0.022000,0.025000,0.024000,0.024000,0.044000,0.041000,0.056000,0.085000,0.033000,0.030000,0.033000,0.027000,0.023000,0.029000,0.020000,0.017000,0.024000,0.019000,0.019000,0.046000,0.027000,0.025000,0.026000,0.025000,0.026000,0.039000,0.017000,0.015000,0.015000,0.014000,0.015000,0.015000,0.026000,0.035000,0.039000,0.030000,0.027000,0.028000,0.025000,0.038000,0.037000,0.035000,0.036000,0.028000,0.028000,0.035000,0.022000,0.020000,0.036000,0.034000,0.021000,0.022000,0.026000,0.024000,0.023000,0.024000,0.022000,0.047000,0.042000,0.044000,0.044000,0.042000,0.037000,0.038000,0.039000,0.041000,0.046000,0.042000,0.047000,0.045000,0.039000,0.051000,0.041000,0.048000,0.041000,0.021000,0.025000,0.021000,0.025000,0.022000,0.026000,0.026000,0.025000,0.025000,0.025000,0.020000,0.018000,0.021000,0.022000,0.022000,0.020000,0.019000,0.022000,0.021000,0.018000,0.021000,0.021000,0.039000,0.038000,0.036000,0.039000,0.037000,0.038000,0.031000]
>> ms
>> Avg Action_CreateBlob-blob_writes_2Bx200: 0.029760ms
>> Sd  Action_CreateBlob-blob_writes_2Bx200: 0.010594ms
>> *RESULT Action_CreateBlob-blob_writes_2Bx200:
>> Action_CreateBlob-blob_writes_2Bx200=
>>
>>
[0.052000,0.052000,0.075000,0.019000,0.017000,0.017000,0.019000,0.020000,0.020000,0.019000,0.022000,0.020000,0.020000,0.019000,0.033000,0.036000,0.040000,0.044000,0.042000,0.042000,0.038000,0.042000,0.058000,0.032000,0.025000,0.039000,0.035000,0.031000,0.030000,0.030000,0.031000,0.014000,0.041000,0.031000,0.033000,0.023000,0.041000,0.042000,0.041000,0.041000,0.036000,0.035000,0.037000,0.042000,0.048000,0.033000,0.030000,0.031000,0.051000,0.028000,0.035000,0.032000,0.027000,0.027000,0.020000,0.015000,0.015000,0.016000,0.016000,0.015000,0.019000,0.014000,0.014000,0.018000,0.015000,0.015000,0.014000,0.026000,0.024000,0.024000,0.037000,0.036000,0.037000,0.037000,0.037000,0.026000,0.039000,0.036000,0.036000,0.037000,0.046000,0.038000,0.029000,0.045000,0.017000,0.021000,0.036000,0.029000,0.030000,0.028000,0.047000,0.031000,0.031000,0.030000,0.023000,0.014000,0.015000,0.014000,0.013000,0.015000,0.014000,0.013000,0.014000,0.013000,0.014000,0.014000,0.013000,0.014000,0.014000,0.037000,0.042000,0.051000,0.039000,0.042000,0.043000,0.016000,0.017000,0.016000,0.019000,0.016000,0.015000,0.038000,0.038000,0.016000,0.041000,0.032000,0.020000,0.017000,0.017000,0.016000,0.016000,0.017000,0.016000,0.016000,0.016000,0.024000,0.029000,0.032000,0.038000,0.034000,0.037000,0.031000,0.029000,0.033000,0.035000,0.032000,0.034000,0.033000,0.033000,0.032000,0.051000,0.041000,0.040000,0.042000,0.020000,0.014000,0.016000,0.015000,0.025000,0.034000,0.024000,0.015000,0.046000,0.038000,0.030000,0.053000,0.030000,0.031000,0.038000,0.040000,0.040000,0.041000,0.039000,0.038000,0.039000,0.038000,0.041000,0.039000,0.042000,0.042000,0.081000,0.044000,0.024000,0.029000,0.033000,0.023000,0.043000,0.034000,0.044000,0.028000,0.028000,0.033000,0.035000,0.027000,0.029000,0.051000,0.042000,0.044000,0.045000,0.044000,0.035000,0.022000,0.015000,0.028000,0.021000,0.014000,0.016000,0.021000,0.015000,0.014000,0.015000,0.015000,0.046000,0.039000,0.037000,0.040000,0.045000,0.039000,0.041000,0.038000,0.030000,0.020000,0.022000,0.029000,0.031000,0.038000,0.048000,0.051000,0.037000,0.040000,0.033000,0.037000,0.026000,0.034000,0.032000,0.039000,0.056000,0.066000,0.039000,0.037000,0.036000,0.033000,0.037000,0.041000,0.030000,0.031000,0.043000,0.035000,0.023000,0.033000,0.022000,0.029000,0.027000,0.022000,0.020000,0.023000,0.022000,0.021000,0.032000,0.023000,0.021000,0.014000,0.033000,0.030000,0.028000,0.030000,0.028000,0.028000,0.029000,0.027000,0.027000,0.035000,0.031000,0.034000,0.030000,0.017000,0.042000,0.049000,0.023000,0.025000,0.024000,0.019000,0.020000,0.029000,0.026000,0.018000,0.024000,0.017000,0.017000,0.019000,0.017000,0.023000,0.024000,0.020000,0.025000,0.026000,0.022000,0.025000,0.024000,0.024000,0.044000,0.041000,0.056000,0.085000,0.033000,0.030000,0.033000,0.027000,0.023000,0.029000,0.020000,0.017000,0.024000,0.019000,0.019000,0.046000,0.027000,0.025000,0.026000,0.025000,0.026000,0.039000,0.017000,0.015000,0.015000,0.014000,0.015000,0.015000,0.026000,0.035000,0.039000,0.030000,0.027000,0.028000,0.025000,0.038000,0.037000,0.035000,0.036000,0.028000,0.028000,0.035000,0.022000,0.020000,0.036000,0.034000,0.021000,0.022000,0.026000,0.024000,0.023000,0.024000,0.022000,0.047000,0.042000,0.044000,0.044000,0.042000,0.037000,0.038000,0.039000,0.041000,0.046000,0.042000,0.047000,0.045000,0.039000,0.051000,0.041000,0.048000,0.041000,0.021000,0.025000,0.021000,0.025000,0.022000,0.026000,0.026000,0.025000,0.025000,0.025000,0.020000,0.018000,0.021000,0.022000,0.022000,0.020000,0.019000,0.022000,0.021000,0.018000,0.021000,0.021000,0.039000,0.038000,0.036000,0.039000,0.037000,0.038000,0.031000]
>> ms
>> Avg Action_CreateBlob-blob_writes_2Bx200: 0.030103ms
>> Sd  Action_CreateBlob-blob_writes_2Bx200: 0.011335ms
>> RESULT telemetry_page_measurement_results: num_failed= 0 count
>> RESULT telemetry_page_measurement_results: num_errored= 0 count
>>
>>
>> https://codereview.chromium.org/1104053006/
>>
>

To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

chromium-reviews

Did you do a full build of Chrome? The patch fixes a bug where that ...

5 years, 7 months ago (2015-05-08 22:21:01 UTC) #13

Did you do a full build of Chrome?  The patch fixes a bug where that wasn't
showing up.  You should at least see BlobRequest::CountSize

On Fri, May 8, 2015 at 3:04 PM Ned <nednguyen@google.com> wrote:

> I actually don't find BlobRequest in the trace-viewer either, only
>  Registry::RegisterBlob
>
> On Fri, May 8, 2015 at 2:38 PM Daniel Murphy <dmurph@chromium.org> wrote:
>
>> Whoops, replying with @chromium
>>
>> Right, there are no "blob_reads".  Only the writes.
>>
>> If you print out every event, you never see the "BlobRequest" event.
>> However, if you look at the trace, it's there.
>>
>> (you can comment out all be the user story to make the run go faster)
>>
>> Daniel
>> On Fri, May 8, 2015 at 2:35 PM <nednguyen@google.com> wrote:
>>
>>> On 2015/05/08 19:28:48, nednguyen wrote:
>>> > On 2015/05/08 18:59:23, dmurph wrote:
>>> > > any more ideas?  I printed out every event name that I get and no
>>> > "BlobRequest"
>>> > > events are in there.
>>> > Then I think this is definitely a trace parsing issue. I will try to
>>> patch
>>> this
>>> > and run in on my machine.
>>> > >
>>> > >
>>>
>>>
>>>
https://codereview.chromium.org/1104053006/diff/120001/tools/telemetry/teleme...
>>> > > File tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py
>>> > (right):
>>> > >
>>> > >
>>>
>>>
>>>
https://codereview.chromium.org/1104053006/diff/120001/tools/telemetry/teleme...
>>> > > tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:24:
>>> > > self._AddResultsInternal(renderer_thread.parent.IterAllSlices(),
>>> > > On 2015/05/08 01:35:12, nednguyen wrote:
>>> > > > Instead of IterAllSlices(), try IterAllEvents instead. Async event
>>> is
>>> > not
>>> > > slice,
>>> > > > I think
>>> > >
>>> > > Hm, still doesn't work :'(
>>>
>>> Hmh, when I run this on my local machine, it works. An example output:
>>>
>>> ./tools/perf/run_benchmark blob_storage.blob_storage --browser=release
>>> [ RUN      ] blob-workshop.html
>>> [       OK ] blob-workshop.html (11949 ms)
>>> [ RUN      ] blob-workshop.html
>>> [       OK ] blob-workshop.html (11530 ms)
>>> [  PASSED  ] 2 tests.
>>>
>>> Pages: [blob-workshop.html,blob-workshop.html]
>>> RESULT Action_CreateBlob-blob_writes_2Bx200: blob-workshop.html=
>>>
>>>
[0.052000,0.052000,0.075000,0.019000,0.017000,0.017000,0.019000,0.020000,0.020000,0.019000,0.022000,0.020000,0.020000,0.019000,0.033000,0.036000,0.040000,0.044000,0.042000,0.042000,0.038000,0.042000,0.058000,0.032000,0.025000,0.039000,0.035000,0.031000,0.030000,0.030000,0.031000,0.014000,0.041000,0.031000,0.033000,0.023000,0.041000,0.042000,0.041000,0.041000,0.036000,0.035000,0.037000,0.042000,0.048000,0.033000,0.030000,0.031000,0.051000,0.028000,0.035000,0.032000,0.027000,0.027000,0.020000,0.015000,0.015000,0.016000,0.016000,0.015000,0.019000,0.014000,0.014000,0.018000,0.015000,0.015000,0.014000,0.026000,0.024000,0.024000,0.037000,0.036000,0.037000,0.037000,0.037000,0.026000,0.039000,0.036000,0.036000,0.037000,0.046000,0.038000,0.029000,0.045000,0.017000,0.021000,0.036000,0.029000,0.030000,0.028000,0.047000,0.031000,0.031000,0.030000,0.023000,0.014000,0.015000,0.014000,0.013000,0.015000,0.014000,0.013000,0.014000,0.013000,0.014000,0.014000,0.013000,0.014000,0.014000,0.037000,0.042000,0.051000,0.039000,0.042000,0.043000,0.016000,0.017000,0.016000,0.019000,0.016000,0.015000,0.038000,0.038000,0.016000,0.041000,0.032000,0.020000,0.017000,0.017000,0.016000,0.016000,0.017000,0.016000,0.016000,0.016000,0.024000,0.029000,0.032000,0.038000,0.034000,0.037000,0.031000,0.029000,0.033000,0.035000,0.032000,0.034000,0.033000,0.033000,0.032000,0.051000,0.041000,0.040000,0.042000,0.020000,0.014000,0.016000,0.015000,0.025000,0.034000,0.024000,0.015000,0.046000,0.038000,0.030000,0.053000,0.030000,0.031000,0.038000,0.040000,0.040000,0.041000,0.039000,0.038000,0.039000,0.038000,0.041000,0.039000,0.042000,0.042000,0.081000,0.044000,0.024000,0.029000,0.033000,0.023000,0.043000,0.034000,0.044000,0.028000,0.028000,0.033000,0.035000,0.027000,0.029000,0.051000,0.042000,0.044000,0.045000,0.044000]
>>> ms
>>> Avg Action_CreateBlob-blob_writes_2Bx200: 0.030445ms
>>> Sd  Action_CreateBlob-blob_writes_2Bx200: 0.012047ms
>>> RESULT Action_CreateBlob-blob_writes_2Bx200: blob-workshop.html=
>>>
>>>
[0.035000,0.022000,0.015000,0.028000,0.021000,0.014000,0.016000,0.021000,0.015000,0.014000,0.015000,0.015000,0.046000,0.039000,0.037000,0.040000,0.045000,0.039000,0.041000,0.038000,0.030000,0.020000,0.022000,0.029000,0.031000,0.038000,0.048000,0.051000,0.037000,0.040000,0.033000,0.037000,0.026000,0.034000,0.032000,0.039000,0.056000,0.066000,0.039000,0.037000,0.036000,0.033000,0.037000,0.041000,0.030000,0.031000,0.043000,0.035000,0.023000,0.033000,0.022000,0.029000,0.027000,0.022000,0.020000,0.023000,0.022000,0.021000,0.032000,0.023000,0.021000,0.014000,0.033000,0.030000,0.028000,0.030000,0.028000,0.028000,0.029000,0.027000,0.027000,0.035000,0.031000,0.034000,0.030000,0.017000,0.042000,0.049000,0.023000,0.025000,0.024000,0.019000,0.020000,0.029000,0.026000,0.018000,0.024000,0.017000,0.017000,0.019000,0.017000,0.023000,0.024000,0.020000,0.025000,0.026000,0.022000,0.025000,0.024000,0.024000,0.044000,0.041000,0.056000,0.085000,0.033000,0.030000,0.033000,0.027000,0.023000,0.029000,0.020000,0.017000,0.024000,0.019000,0.019000,0.046000,0.027000,0.025000,0.026000,0.025000,0.026000,0.039000,0.017000,0.015000,0.015000,0.014000,0.015000,0.015000,0.026000,0.035000,0.039000,0.030000,0.027000,0.028000,0.025000,0.038000,0.037000,0.035000,0.036000,0.028000,0.028000,0.035000,0.022000,0.020000,0.036000,0.034000,0.021000,0.022000,0.026000,0.024000,0.023000,0.024000,0.022000,0.047000,0.042000,0.044000,0.044000,0.042000,0.037000,0.038000,0.039000,0.041000,0.046000,0.042000,0.047000,0.045000,0.039000,0.051000,0.041000,0.048000,0.041000,0.021000,0.025000,0.021000,0.025000,0.022000,0.026000,0.026000,0.025000,0.025000,0.025000,0.020000,0.018000,0.021000,0.022000,0.022000,0.020000,0.019000,0.022000,0.021000,0.018000,0.021000,0.021000,0.039000,0.038000,0.036000,0.039000,0.037000,0.038000,0.031000]
>>> ms
>>> Avg Action_CreateBlob-blob_writes_2Bx200: 0.029760ms
>>> Sd  Action_CreateBlob-blob_writes_2Bx200: 0.010594ms
>>> *RESULT Action_CreateBlob-blob_writes_2Bx200:
>>> Action_CreateBlob-blob_writes_2Bx200=
>>>
>>>
[0.052000,0.052000,0.075000,0.019000,0.017000,0.017000,0.019000,0.020000,0.020000,0.019000,0.022000,0.020000,0.020000,0.019000,0.033000,0.036000,0.040000,0.044000,0.042000,0.042000,0.038000,0.042000,0.058000,0.032000,0.025000,0.039000,0.035000,0.031000,0.030000,0.030000,0.031000,0.014000,0.041000,0.031000,0.033000,0.023000,0.041000,0.042000,0.041000,0.041000,0.036000,0.035000,0.037000,0.042000,0.048000,0.033000,0.030000,0.031000,0.051000,0.028000,0.035000,0.032000,0.027000,0.027000,0.020000,0.015000,0.015000,0.016000,0.016000,0.015000,0.019000,0.014000,0.014000,0.018000,0.015000,0.015000,0.014000,0.026000,0.024000,0.024000,0.037000,0.036000,0.037000,0.037000,0.037000,0.026000,0.039000,0.036000,0.036000,0.037000,0.046000,0.038000,0.029000,0.045000,0.017000,0.021000,0.036000,0.029000,0.030000,0.028000,0.047000,0.031000,0.031000,0.030000,0.023000,0.014000,0.015000,0.014000,0.013000,0.015000,0.014000,0.013000,0.014000,0.013000,0.014000,0.014000,0.013000,0.014000,0.014000,0.037000,0.042000,0.051000,0.039000,0.042000,0.043000,0.016000,0.017000,0.016000,0.019000,0.016000,0.015000,0.038000,0.038000,0.016000,0.041000,0.032000,0.020000,0.017000,0.017000,0.016000,0.016000,0.017000,0.016000,0.016000,0.016000,0.024000,0.029000,0.032000,0.038000,0.034000,0.037000,0.031000,0.029000,0.033000,0.035000,0.032000,0.034000,0.033000,0.033000,0.032000,0.051000,0.041000,0.040000,0.042000,0.020000,0.014000,0.016000,0.015000,0.025000,0.034000,0.024000,0.015000,0.046000,0.038000,0.030000,0.053000,0.030000,0.031000,0.038000,0.040000,0.040000,0.041000,0.039000,0.038000,0.039000,0.038000,0.041000,0.039000,0.042000,0.042000,0.081000,0.044000,0.024000,0.029000,0.033000,0.023000,0.043000,0.034000,0.044000,0.028000,0.028000,0.033000,0.035000,0.027000,0.029000,0.051000,0.042000,0.044000,0.045000,0.044000,0.035000,0.022000,0.015000,0.028000,0.021000,0.014000,0.016000,0.021000,0.015000,0.014000,0.015000,0.015000,0.046000,0.039000,0.037000,0.040000,0.045000,0.039000,0.041000,0.038000,0.030000,0.020000,0.022000,0.029000,0.031000,0.038000,0.048000,0.051000,0.037000,0.040000,0.033000,0.037000,0.026000,0.034000,0.032000,0.039000,0.056000,0.066000,0.039000,0.037000,0.036000,0.033000,0.037000,0.041000,0.030000,0.031000,0.043000,0.035000,0.023000,0.033000,0.022000,0.029000,0.027000,0.022000,0.020000,0.023000,0.022000,0.021000,0.032000,0.023000,0.021000,0.014000,0.033000,0.030000,0.028000,0.030000,0.028000,0.028000,0.029000,0.027000,0.027000,0.035000,0.031000,0.034000,0.030000,0.017000,0.042000,0.049000,0.023000,0.025000,0.024000,0.019000,0.020000,0.029000,0.026000,0.018000,0.024000,0.017000,0.017000,0.019000,0.017000,0.023000,0.024000,0.020000,0.025000,0.026000,0.022000,0.025000,0.024000,0.024000,0.044000,0.041000,0.056000,0.085000,0.033000,0.030000,0.033000,0.027000,0.023000,0.029000,0.020000,0.017000,0.024000,0.019000,0.019000,0.046000,0.027000,0.025000,0.026000,0.025000,0.026000,0.039000,0.017000,0.015000,0.015000,0.014000,0.015000,0.015000,0.026000,0.035000,0.039000,0.030000,0.027000,0.028000,0.025000,0.038000,0.037000,0.035000,0.036000,0.028000,0.028000,0.035000,0.022000,0.020000,0.036000,0.034000,0.021000,0.022000,0.026000,0.024000,0.023000,0.024000,0.022000,0.047000,0.042000,0.044000,0.044000,0.042000,0.037000,0.038000,0.039000,0.041000,0.046000,0.042000,0.047000,0.045000,0.039000,0.051000,0.041000,0.048000,0.041000,0.021000,0.025000,0.021000,0.025000,0.022000,0.026000,0.026000,0.025000,0.025000,0.025000,0.020000,0.018000,0.021000,0.022000,0.022000,0.020000,0.019000,0.022000,0.021000,0.018000,0.021000,0.021000,0.039000,0.038000,0.036000,0.039000,0.037000,0.038000,0.031000]
>>> ms
>>> Avg Action_CreateBlob-blob_writes_2Bx200: 0.030103ms
>>> Sd  Action_CreateBlob-blob_writes_2Bx200: 0.011335ms
>>> RESULT telemetry_page_measurement_results: num_failed= 0 count
>>> RESULT telemetry_page_measurement_results: num_errored= 0 count
>>>
>>>
>>> https://codereview.chromium.org/1104053006/
>>>
>>

To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

nednguyen

I confirm that I see those events in trace.json. I will work on the trace ...

5 years, 7 months ago (2015-05-11 16:05:53 UTC) #14

nednguyen

On 2015/05/11 16:05:53, nednguyen (slow review) wrote: > I confirm that I see those events ...

5 years, 7 months ago (2015-05-11 16:34:51 UTC) #15

dmurph

On 2015/05/11 at 16:34:51, nednguyen wrote: > On 2015/05/11 16:05:53, nednguyen (slow review) wrote: > ...

5 years, 7 months ago (2015-05-14 03:55:13 UTC) #16

nednguyen

On 2015/05/14 03:55:13, dmurph wrote: > On 2015/05/11 at 16:34:51, nednguyen wrote: > > On ...

5 years, 7 months ago (2015-05-14 16:45:11 UTC) #17

dmurph

On 2015/05/14 at 16:45:11, nednguyen wrote: > On 2015/05/14 03:55:13, dmurph wrote: > > On ...

5 years, 7 months ago (2015-05-18 22:28:56 UTC) #18

nednguyen

https://codereview.chromium.org/1104053006/diff/160001/tools/perf/benchmarks/blob_storage.py File tools/perf/benchmarks/blob_storage.py (right): https://codereview.chromium.org/1104053006/diff/160001/tools/perf/benchmarks/blob_storage.py#newcode11 tools/perf/benchmarks/blob_storage.py:11: nit: add another blank line https://codereview.chromium.org/1104053006/diff/160001/tools/perf/page_sets/blob_workshop.py File tools/perf/page_sets/blob_workshop.py (right): ...

5 years, 7 months ago (2015-05-18 22:55:54 UTC) #19

nednguyen

https://codereview.chromium.org/1104053006/diff/160001/tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py File tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py (right): https://codereview.chromium.org/1104053006/diff/160001/tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py#newcode35 tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:35: self._AddResultsInternal(events, interactions, results) On 2015/05/18 22:55:54, nednguyen wrote: > ...

5 years, 7 months ago (2015-05-19 00:34:16 UTC) #20

nednguyen

On 2015/05/19 00:34:16, nednguyen wrote: > https://codereview.chromium.org/1104053006/diff/160001/tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py > File tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py (right): > > https://codereview.chromium.org/1104053006/diff/160001/tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py#newcode35 > ...

5 years, 7 months ago (2015-05-19 15:52:53 UTC) #21

dmurph

It's working yay! Bug here: https://code.google.com/p/chromium/issues/detail?id=489799 +kinuko for storage/browser/blob/blob_url_request_job.cc https://codereview.chromium.org/1104053006/diff/160001/tools/perf/benchmarks/blob_storage.py File tools/perf/benchmarks/blob_storage.py (right): https://codereview.chromium.org/1104053006/diff/160001/tools/perf/benchmarks/blob_storage.py#newcode11 tools/perf/benchmarks/blob_storage.py:11: ...

5 years, 7 months ago (2015-05-21 00:21:36 UTC) #22

nednguyen

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

5 years, 7 months ago (2015-05-22 15:35:40 UTC) #23

eakuefner

looking good; some notes, mostly style stuff. https://codereview.chromium.org/1104053006/diff/180001/tools/perf/benchmarks/blob_storage.py File tools/perf/benchmarks/blob_storage.py (right): https://codereview.chromium.org/1104053006/diff/180001/tools/perf/benchmarks/blob_storage.py#newcode10 tools/perf/benchmarks/blob_storage.py:10: TOPLEVEL_CATEGORY = ...

5 years, 7 months ago (2015-05-22 18:33:19 UTC) #25

dmurph

Done w/ a couple questions https://codereview.chromium.org/1104053006/diff/180001/tools/perf/benchmarks/blob_storage.py File tools/perf/benchmarks/blob_storage.py (right): https://codereview.chromium.org/1104053006/diff/180001/tools/perf/benchmarks/blob_storage.py#newcode10 tools/perf/benchmarks/blob_storage.py:10: TOPLEVEL_CATEGORY = 'Blob' On ...

5 years, 7 months ago (2015-05-22 19:06:21 UTC) #26

dmurph

Whoops, I forgot to upload. Also, I changed to the lambda for the event checking.

5 years, 7 months ago (2015-05-22 22:30:48 UTC) #27

dmurph

dmurph@chromium.org changed reviewers: + kinuko@chromium.org

5 years, 7 months ago (2015-05-22 23:02:32 UTC) #28

dmurph

+kinuko for storage/browser/blob/blob_url_request_job.cc (and fixed tests)

5 years, 7 months ago (2015-05-22 23:02:32 UTC) #29

kinuko

On 2015/05/22 23:02:32, dmurph wrote: > +kinuko for storage/browser/blob/blob_url_request_job.cc lgtm for blob_url_request_job.cc > (and fixed ...

5 years, 7 months ago (2015-05-25 08:33:16 UTC) #30

nednguyen

https://codereview.chromium.org/1104053006/diff/220001/tools/perf/benchmarks/blob_storage.py File tools/perf/benchmarks/blob_storage.py (right): https://codereview.chromium.org/1104053006/diff/220001/tools/perf/benchmarks/blob_storage.py#newcode34 tools/perf/benchmarks/blob_storage.py:34: return (value.name.startswith('Action_CreateBlob') or I thought we want to filter ...

5 years, 6 months ago (2015-05-28 05:10:12 UTC) #32

https://codereview.chromium.org/1104053006/diff/220001/tools/perf/benchmarks/...
File tools/perf/benchmarks/blob_storage.py (right):

https://codereview.chromium.org/1104053006/diff/220001/tools/perf/benchmarks/...
tools/perf/benchmarks/blob_storage.py:34: return
(value.name.startswith('Action_CreateBlob') or
I thought we want to filter on "blob_writes", "blob_write_reads" & "blob_reads"
?

https://codereview.chromium.org/1104053006/diff/220001/tools/telemetry/teleme...
File tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py (right):

https://codereview.chromium.org/1104053006/diff/220001/tools/telemetry/teleme...
tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:56:
write_then_reads.append(interaction.end - interaction.start)
This is the wrong way of using timeline interaction record.

The idea of timeline interaction record is to specify a time range in which we
want to compute metrics on.

"Why compute metric on a specific time range instead of the whole browsing
section's life time?", you may ask. This is mainly to reduce noise.  From our
experience, we observe many situation in which some metrics X is unstable during
a specific time period of the browser, i.e: smoothness metrics are noisy because
of some time gap when users intentionally pause, thread time metrics are noisy
because of the idle time waiting for the page to load, etc.

Hence, there is no semantic built-in in interaction to guarantee it always
contain these "write_then_read" actions.

https://codereview.chromium.org/1104053006/diff/220001/tools/telemetry/teleme...
tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:63:
writes.append(duration)
You can use e.thread_duration to make sure that your metric is not sensitive to
the OS activities. Caveat is that thread_time info is not available on win, so
you need s.t like an if statement to fallback to use walltime for duration on
win platform.

https://codereview.chromium.org/1104053006/diff/220001/tools/telemetry/teleme...
tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:92: min_read_start =
min(event.start, min_read_start)
why can we just use reads.append(e.thread_duration)?

eakuefner

lgtm https://codereview.chromium.org/1104053006/diff/180001/tools/telemetry/telemetry/web_perf/metrics/blob_timeline_unittest.py File tools/telemetry/telemetry/web_perf/metrics/blob_timeline_unittest.py (right): https://codereview.chromium.org/1104053006/diff/180001/tools/telemetry/telemetry/web_perf/metrics/blob_timeline_unittest.py#newcode20 tools/telemetry/telemetry/web_perf/metrics/blob_timeline_unittest.py:20: blob_timeline.BlobTimelineMetric()._AddWriteResultsInternal( On 2015/05/22 19:06:21, dmurph wrote: > On ...

5 years, 6 months ago (2015-05-28 20:36:56 UTC) #33

dmurph

Switched to using thread_duration, using an accumulation of read times for blob read, and fixed ...

5 years, 6 months ago (2015-05-29 05:05:47 UTC) #35

Switched to using thread_duration, using an accumulation of read times for blob
read, and fixed the test.

https://codereview.chromium.org/1104053006/diff/220001/tools/perf/benchmarks/...
File tools/perf/benchmarks/blob_storage.py (right):

https://codereview.chromium.org/1104053006/diff/220001/tools/perf/benchmarks/...
tools/perf/benchmarks/blob_storage.py:34: return
(value.name.startswith('Action_CreateBlob') or
On 2015/05/28 at 05:10:12, nednguyen wrote:
> I thought we want to filter on "blob_writes", "blob_write_reads" &
"blob_reads" ?

The name contains the interaction and the result name.  Switched to checking for
the result name.

https://codereview.chromium.org/1104053006/diff/220001/tools/telemetry/teleme...
File tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py (right):

https://codereview.chromium.org/1104053006/diff/220001/tools/telemetry/teleme...
tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:44:
event_predicate=(lambda e: e.name == READ_EVENT_NAME)):
On 2015/05/28 at 20:36:56, eakuefner wrote:
> this looks fine, but let it be known that _if_ you want to factor this out,
@staticmethod will work, but @classmethod won't, because staticmethod creates a
true one-argument function whereas classmethod takes the class as an implicit
first parameter (this distinction exists in python but not c++ because classes
are objects in python).
> 
> whatever you choose to do from here is fine with me.

Switched to static methods

https://codereview.chromium.org/1104053006/diff/220001/tools/telemetry/teleme...
tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:56:
write_then_reads.append(interaction.end - interaction.start)
On 2015/05/28 at 05:10:12, nednguyen wrote:
> This is the wrong way of using timeline interaction record.
> 
> The idea of timeline interaction record is to specify a time range in which we
want to compute metrics on.
> 
> "Why compute metric on a specific time range instead of the whole browsing
section's life time?", you may ask. This is mainly to reduce noise.  From our
experience, we observe many situation in which some metrics X is unstable during
a specific time period of the browser, i.e: smoothness metrics are noisy because
of some time gap when users intentionally pause, thread time metrics are noisy
because of the idle time waiting for the page to load, etc.
> 
> Hence, there is no semantic built-in in interaction to guarantee it always
contain these "write_then_read" actions.

Based on external discussion, switched to using the read stats

https://codereview.chromium.org/1104053006/diff/220001/tools/telemetry/teleme...
tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:63:
writes.append(duration)
On 2015/05/28 at 05:10:12, nednguyen wrote:
> You can use e.thread_duration to make sure that your metric is not sensitive
to the OS activities. Caveat is that thread_time info is not available on win,
so you need s.t like an if statement to fallback to use walltime for duration on
win platform.

Done.

https://codereview.chromium.org/1104053006/diff/220001/tools/telemetry/teleme...
tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:92: min_read_start =
min(event.start, min_read_start)
On 2015/05/28 at 05:10:12, nednguyen wrote:
> why can we just use reads.append(e.thread_duration)?

Done.

nednguyen

https://codereview.chromium.org/1104053006/diff/240001/tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py File tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py (right): https://codereview.chromium.org/1104053006/diff/240001/tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py#newcode103 tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:103: not self.IsWriteReadInteraction(interaction)): Again, you cannot make any assumption about ...

5 years, 6 months ago (2015-05-29 17:57:53 UTC) #36

dmurph

But this is how I'm computing my metrics. How will I know if my 'read' ...

5 years, 6 months ago (2015-05-29 18:44:54 UTC) #37

dmurph

On 2015/05/29 at 18:44:54, dmurph wrote: > But this is how I'm computing my metrics. ...

5 years, 6 months ago (2015-06-01 19:22:18 UTC) #38

nednguyen

A bit of caveat: we will move the interaction record out of the metric name, ...

5 years, 6 months ago (2015-06-01 23:29:27 UTC) #39

A bit of caveat: we will move the interaction record out of the metric name, and
make each Value looks s.t like:

Value:
  Page: XYZ
  Interaction: ABC
  name: BlobRead   (<-- right now, this is ABC_BlobRead)
  value: 123
  units: ms

See https://code.google.com/p/chromium/issues/detail?id=461101

https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/teleme...
File tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py (right):

https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/teleme...
tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:74: values=writes,
Be aware that if the list is empty, this will cause exception. You may want to
do a check like:

if not writes:
  results.AddValue(list_of_scalar_values.ListOfScalarValues(        
page=results.current_page,
name='blob-writes',
units='ms',
values=None,
none_value_reason='No blob write events found')

https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/teleme...
tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:91:
reads.append(total_thread_time)
Why does this reads contains the list of total_thread_time by interaction?

If I read this correctly, the list contains
[total_thread_time_of_read_in_record_1,
total_thread_time_of_read_in_record_2,...]

Is this because you expect the record 1 to contain only read events of the first
read, record 2 to contain only read events of the 2nd read? If so, then note
that we can't make this assumption. I would just make it as straight forward as
output a total number of all read events or the list of all read events' latency
found in interaction records, then you can create interaction record with
different names in your pages. Doing so will make the
BlobTimelineMetric.AddResults called multiple times, which allow you to
differentiate the values of one interaction from others

https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/teleme...
tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:99:
description='List of durations of blob reading.'))
This also needs to deal with empty list

dmurph

On 2015/06/01 at 23:29:27, nednguyen wrote: > A bit of caveat: we will move the ...

5 years, 6 months ago (2015-06-01 23:42:48 UTC) #40

On 2015/06/01 at 23:29:27, nednguyen wrote:
> A bit of caveat: we will move the interaction record out of the metric name,
and make each Value looks s.t like:
> 
> Value:
>   Page: XYZ
>   Interaction: ABC
>   name: BlobRead   (<-- right now, this is ABC_BlobRead)
>   value: 123
>   units: ms
> 
> See https://code.google.com/p/chromium/issues/detail?id=461101
> 
>
https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/teleme...
> File tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py (right):
> 
>
https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/teleme...
> tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:74: values=writes,
> Be aware that if the list is empty, this will cause exception. You may want to
do a check like:
> 
> if not writes:
>   results.AddValue(list_of_scalar_values.ListOfScalarValues(        
> page=results.current_page,
> name='blob-writes',
> units='ms',
> values=None,
> none_value_reason='No blob write events found')
> 
>
https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/teleme...
> tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:91:
reads.append(total_thread_time)
> Why does this reads contains the list of total_thread_time by interaction?
> 
> If I read this correctly, the list contains
> [total_thread_time_of_read_in_record_1,
total_thread_time_of_read_in_record_2,...]
> 
> Is this because you expect the record 1 to contain only read events of the
first read, record 2 to contain only read events of the 2nd read? If so, then
note that we can't make this assumption. I would just make it as straight
forward as output a total number of all read events or the list of all read
events' latency found in interaction records, then you can create interaction
record with different names in your pages. Doing so will make the
BlobTimelineMetric.AddResults called multiple times, which allow you to
differentiate the values of one interaction from others
> 
>
https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/teleme...
> tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:99:
description='List of durations of blob reading.'))
> This also needs to deal with empty list

This means it looks like there will be 0 ways for me to report different metrics
from the same 'trace' event in one 'metric'.  How can I can do this?  As
discussed before:

I have 2 different situations where I'm measuring the read time.
 * Worst-case stale read, where it's immediately after write.
 * Warm read, where it's after write time.

eakuefner

On 2015/06/01 23:42:48, dmurph wrote: > On 2015/06/01 at 23:29:27, nednguyen wrote: > > A ...

5 years, 6 months ago (2015-06-02 00:04:57 UTC) #41

On 2015/06/01 23:42:48, dmurph wrote:
> On 2015/06/01 at 23:29:27, nednguyen wrote:
> > A bit of caveat: we will move the interaction record out of the metric name,
> and make each Value looks s.t like:
> > 
> > Value:
> >   Page: XYZ
> >   Interaction: ABC
> >   name: BlobRead   (<-- right now, this is ABC_BlobRead)
> >   value: 123
> >   units: ms
> > 
> > See https://code.google.com/p/chromium/issues/detail?id=461101
> > 
> >
>
https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/teleme...
> > File tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py (right):
> > 
> >
>
https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/teleme...
> > tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:74:
values=writes,
> > Be aware that if the list is empty, this will cause exception. You may want
to
> do a check like:
> > 
> > if not writes:
> >   results.AddValue(list_of_scalar_values.ListOfScalarValues(        
> > page=results.current_page,
> > name='blob-writes',
> > units='ms',
> > values=None,
> > none_value_reason='No blob write events found')
> > 
> >
>
https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/teleme...
> > tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:91:
> reads.append(total_thread_time)
> > Why does this reads contains the list of total_thread_time by interaction?
> > 
> > If I read this correctly, the list contains
> > [total_thread_time_of_read_in_record_1,
> total_thread_time_of_read_in_record_2,...]
> > 
> > Is this because you expect the record 1 to contain only read events of the
> first read, record 2 to contain only read events of the 2nd read? If so, then
> note that we can't make this assumption. I would just make it as straight
> forward as output a total number of all read events or the list of all read
> events' latency found in interaction records, then you can create interaction
> record with different names in your pages. Doing so will make the
> BlobTimelineMetric.AddResults called multiple times, which allow you to
> differentiate the values of one interaction from others
> > 
> >
>
https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/teleme...
> > tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:99:
> description='List of durations of blob reading.'))
> > This also needs to deal with empty list
> 
> This means it looks like there will be 0 ways for me to report different
metrics
> from the same 'trace' event in one 'metric'.  How can I can do this?  As
> discussed before:
> 
> I have 2 different situations where I'm measuring the read time.
>  * Worst-case stale read, where it's immediately after write.
>  * Warm read, where it's after write time.

Wait, I think Ned's caveat has generated a bit of confusion. To be clear, those
situations are contained within disjoint interaction records, right?

dmurph

Switched to UUID-based tracking for reading and expanded tests a bit. patch selfie: http://i.imgur.com/YiEGExB.gif https://codereview.chromium.org/1104053006/diff/240001/tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py ...

5 years, 6 months ago (2015-06-02 20:04:07 UTC) #42

Switched to UUID-based tracking for reading and expanded tests a bit.

patch selfie:
http://i.imgur.com/YiEGExB.gif

https://codereview.chromium.org/1104053006/diff/240001/tools/telemetry/teleme...
File tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py (right):

https://codereview.chromium.org/1104053006/diff/240001/tools/telemetry/teleme...
tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:103: not
self.IsWriteReadInteraction(interaction)):
On 2015/05/29 at 17:57:52, nednguyen wrote:
> Again, you cannot make any assumption about what interaction labels could be.
> 
> I think this is partly our fault that this AddResults(self, model,
renderer_thread, interactions, results) API give you the full interactions
instead of just the time range ~_~
> 
> I file a bug about that here:
https://code.google.com/p/chromium/issues/detail?id=493785

Done.

https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/teleme...
File tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py (right):

https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/teleme...
tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:74: values=writes,
On 2015/06/01 at 23:29:27, nednguyen wrote:
> Be aware that if the list is empty, this will cause exception. You may want to
do a check like:
> 
> if not writes:
>   results.AddValue(list_of_scalar_values.ListOfScalarValues(        
> page=results.current_page,
> name='blob-writes',
> units='ms',
> values=None,
> none_value_reason='No blob write events found')

See the line above it :)

https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/teleme...
tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:91:
reads.append(total_thread_time)
On 2015/06/01 at 23:29:27, nednguyen wrote:
> Why does this reads contains the list of total_thread_time by interaction?
> 
> If I read this correctly, the list contains
> [total_thread_time_of_read_in_record_1,
total_thread_time_of_read_in_record_2,...]
> 
> Is this because you expect the record 1 to contain only read events of the
first read, record 2 to contain only read events of the 2nd read? If so, then
note that we can't make this assumption. I would just make it as straight
forward as output a total number of all read events or the list of all read
events' latency found in interaction records, then you can create interaction
record with different names in your pages. Doing so will make the
BlobTimelineMetric.AddResults called multiple times, which allow you to
differentiate the values of one interaction from others

Switched to using UUIDs :)

https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/teleme...
tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:99:
description='List of durations of blob reading.'))
On 2015/06/01 at 23:29:27, nednguyen wrote:
> This also needs to deal with empty list

See the line above it :)

eakuefner

On 2015/06/02 at 20:04:07, dmurph wrote: > Switched to UUID-based tracking for reading and expanded ...

5 years, 6 months ago (2015-06-02 20:29:00 UTC) #43

On 2015/06/02 at 20:04:07, dmurph wrote:
> Switched to UUID-based tracking for reading and expanded tests a bit.
> 
> patch selfie:
> http://i.imgur.com/YiEGExB.gif

I suspect most people still don't know about this amazing feature :D
> 
>
https://codereview.chromium.org/1104053006/diff/240001/tools/telemetry/teleme...
> File tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py (right):
> 
>
https://codereview.chromium.org/1104053006/diff/240001/tools/telemetry/teleme...
> tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:103: not
self.IsWriteReadInteraction(interaction)):
> On 2015/05/29 at 17:57:52, nednguyen wrote:
> > Again, you cannot make any assumption about what interaction labels could
be.
> > 
> > I think this is partly our fault that this AddResults(self, model,
renderer_thread, interactions, results) API give you the full interactions
instead of just the time range ~_~
> > 
> > I file a bug about that here:
https://code.google.com/p/chromium/issues/detail?id=493785
> 
> Done.
> 
>
https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/teleme...
> File tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py (right):
> 
>
https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/teleme...
> tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:74: values=writes,
> On 2015/06/01 at 23:29:27, nednguyen wrote:
> > Be aware that if the list is empty, this will cause exception. You may want
to do a check like:
> > 
> > if not writes:
> >   results.AddValue(list_of_scalar_values.ListOfScalarValues(        
> > page=results.current_page,
> > name='blob-writes',
> > units='ms',
> > values=None,
> > none_value_reason='No blob write events found')
> 
> See the line above it :)
> 
>
https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/teleme...
> tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:91:
reads.append(total_thread_time)
> On 2015/06/01 at 23:29:27, nednguyen wrote:
> > Why does this reads contains the list of total_thread_time by interaction?
> > 
> > If I read this correctly, the list contains
> > [total_thread_time_of_read_in_record_1,
total_thread_time_of_read_in_record_2,...]
> > 
> > Is this because you expect the record 1 to contain only read events of the
first read, record 2 to contain only read events of the 2nd read? If so, then
note that we can't make this assumption. I would just make it as straight
forward as output a total number of all read events or the list of all read
events' latency found in interaction records, then you can create interaction
record with different names in your pages. Doing so will make the
BlobTimelineMetric.AddResults called multiple times, which allow you to
differentiate the values of one interaction from others
> 
> Switched to using UUIDs :)
> 
>
https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/teleme...
> tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:99:
description='List of durations of blob reading.'))
> On 2015/06/01 at 23:29:27, nednguyen wrote:
> > This also needs to deal with empty list
> 
> See the line above it :)

nednguyen

lgtm with nits https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py File tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py (right): https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py#newcode74 tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py:74: values=writes, On 2015/06/02 20:04:07, dmurph wrote: ...

5 years, 6 months ago (2015-06-02 21:06:58 UTC) #44

dmurph

On 2015/06/02 at 21:06:58, nednguyen wrote: > lgtm with nits > > https://codereview.chromium.org/1104053006/diff/260001/tools/telemetry/telemetry/web_perf/metrics/blob_timeline.py > File ...

5 years, 6 months ago (2015-06-02 22:12:52 UTC) #45

dmurph

Actually, it would be worse than that. Every interaction in a benchmark would have two ...

5 years, 6 months ago (2015-06-02 22:18:24 UTC) #46

dmurph

Based on outside conversation, added None state and then filter it out in our benchmark. ...

5 years, 6 months ago (2015-06-02 22:55:45 UTC) #47

nednguyen

On 2015/06/02 22:55:45, dmurph wrote: > Based on outside conversation, added None state and then ...

5 years, 6 months ago (2015-06-02 22:59:44 UTC) #48

dmurph

The patchset sent to the CQ was uploaded after l-g-t-m from eakuefner@chromium.org, kinuko@chromium.org Link to ...

5 years, 6 months ago (2015-06-02 23:15:34 UTC) #50