All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=703774

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgrOTlqAoM


Bot(s) for this bug's original alert(s):

android-nexus5

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8984485213283017472

=== Auto-CCing suspected CL author zmo@chromium.org ===

Hi zmo@chromium.org, the bisect results pointed to your CL, please take a look at the
results.


=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : zmo
  Commit : 9b5537f92d5910a563811ce14208d7cb8b45b377
  Date   : Mon Mar 20 23:01:56 2017
  Subject: Expose GL_VENDOR/GL_RENDERER strings to internal clients of GPU command buffer.

Bisect Details
  Configuration: android_nexus5_perf_bisect
  Benchmark    : smoothness.gpu_rasterization.tough_filters_cases
  Metric       : frame_times/http___web.archive.org_web_20150502135732_http___ie.microsoft.com_testdrive_Performance_Pirates_Default.html
  Change       : 8.48% | 40.4461386579 -> 43.8741455086

Revision             Result                   N
chromium@458188      40.4461 +- 0.45366       6      good
chromium@458219      40.6049 +- 0.495397      6      good
chromium@458220      43.8002 +- 0.434328      6      bad       <--
chromium@458221      43.8722 +- 0.287438      6      bad
chromium@458223      43.9151 +- 0.662745      6      bad
chromium@458227      43.8717 +- 0.608588      6      bad
chromium@458235      43.9402 +- 0.170957      6      bad
chromium@458250      43.8741 +- 0.544586      6      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http...web.archive.org.web.20150502135732.http...ie.microsoft.com.testdrive.Performance.Pirates.Default.html smoothness.gpu_rasterization.tough_filters_cases

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8984485213283017472

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6177132989906944


| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Speed>Bisection.  Thank you!

Bian, can you take a look at this? Assuming skia going down a different path than before, which you said it shouldn't?

Yep, I will investigate.

I'm having trouble running the perf test. I run

tools/perf/run_benchmark -v --browser=android-chromium --pageset-repeat=1 --also-run-disabled-tests --story-filter=http...web.archive.org.web.20150502135732.http...ie.microsoft.com.testdrive.Performance.Pirates.Default.html smoothness.gpu_rasterization.tough_filters_cases


The web.archive.org page never finishes loading on the device. The tail of the output looks like:


INFO:devil.utils.cmd_helper:[host]> /usr/local/google/home/bsalomon/android-sdk-linux/platform-tools/adb -s 094cda84 shell '( dumpsys SurfaceFlinger --latency SurfaceView );echo %$?'
WARNING:root:Exception was raised in the with statement block, the end of interaction record is not marked.
INFO:root:*************** BROWSER STANDARD OUTPUT ***************
INFO:root:Cannot get standard output on Android
INFO:root:*********** END OF BROWSER STANDARD OUTPUT ************
INFO:root:********************* BROWSER LOG *********************
INFO:root:No log file
INFO:root:***************** END OF BROWSER LOG ******************
INFO:devil.utils.cmd_helper:[host]> /usr/local/google/home/bsalomon/android-sdk-linux/platform-tools/adb -s 094cda84 shell '( /system/bin/screencap -p /data/local/tmp/temp_file-eec5d74ca14ce.png );echo %$?'
INFO:devil.utils.cmd_helper:[host]> /usr/local/google/home/bsalomon/android-sdk-linux/platform-tools/adb -s 094cda84 shell '( dumpsys SurfaceFlinger --latency SurfaceView );echo %$?'
INFO:devil.utils.cmd_helper:[host]> /usr/local/google/home/bsalomon/android-sdk-linux/platform-tools/adb -s 094cda84 pull /data/local/tmp/temp_file-eec5d74ca14ce.png /tmp/tmpu8KF9U.png
INFO:devil.utils.cmd_helper:[host]> /usr/local/google/home/bsalomon/android-sdk-linux/platform-tools/adb -s 094cda84 shell 'rm -f /data/local/tmp/temp_file-eec5d74ca14ce.png'
Traceback (most recent call last):
  File "/usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/telemetry/telemetry/internal/story_runner.py", line 88, in _RunStoryAndProcessErrorIfNeeded
    state.RunStory(results)
  File "/usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/common/py_trace_event/py_trace_event/trace_event_impl/decorators.py", line 52, in traced_function
    return func(*args, **kwargs)
  File "/usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/telemetry/telemetry/page/shared_page_state.py", line 299, in RunStory
    self._current_page.Run(self)
  File "/usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/telemetry/telemetry/page/__init__.py", line 112, in Run
    self.RunPageInteractions(action_runner)
  File "/usr/local/google/home/bsalomon/src/clank/src/tools/perf/page_sets/tough_filters_cases.py", line 27, in RunPageInteractions
    'document.getElementById("benchmarkButtonText").click()')
  File "/usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/common/py_trace_event/py_trace_event/trace_event_impl/decorators.py", line 75, in traced_function
    return func(*args, **kwargs)
  File "/usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/telemetry/telemetry/internal/actions/action_runner.py", line 236, in EvaluateJavaScript
    return self._tab.EvaluateJavaScript(*args, **kwargs)
  File "/usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/common/py_trace_event/py_trace_event/trace_event_impl/decorators.py", line 75, in traced_function
    return func(*args, **kwargs)
  File "/usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/telemetry/telemetry/internal/browser/web_contents.py", line 167, in EvaluateJavaScript
    return self._inspector_backend.EvaluateJavaScript(*args, **kwargs)
  File "/usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/common/py_trace_event/py_trace_event/trace_event_impl/decorators.py", line 75, in traced_function
    return func(*args, **kwargs)
  File "/usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/telemetry/telemetry/internal/backends/chrome_inspector/inspector_backend.py", line 37, in inner
    return func(inspector_backend, *args, **kwargs)
  File "/usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/telemetry/telemetry/internal/backends/chrome_inspector/inspector_backend.py", line 244, in EvaluateJavaScript
    return self._runtime.Evaluate(expression, context_id, timeout)
  File "/usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/telemetry/telemetry/internal/backends/chrome_inspector/inspector_runtime.py", line 54, in Evaluate
    description=details.get('exception', {}).get('description'))
EvaluateException: UncaughtTypeError:
TypeError: Cannot read property 'click' of null
    at <anonymous>:1:47

INFO:root:Chrome version: 3047
INFO:devil.utils.cmd_helper:[host]> /usr/local/google/home/bsalomon/android-sdk-linux/platform-tools/adb -s 094cda84 shell '( rm -f /data/local/chrome-trace-config.json );echo %$?'
INFO:devil.utils.cmd_helper:[host]> /usr/local/google/home/bsalomon/android-sdk-linux/platform-tools/adb -s 094cda84 shell '( dumpsys SurfaceFlinger --latency SurfaceView );echo %$?'
INFO:devil.utils.cmd_helper:[host]> /usr/local/google/home/bsalomon/android-sdk-linux/platform-tools/adb -s 094cda84 shell '( dumpsys SurfaceFlinger --latency SurfaceView );echo %$?'
INFO:devil.utils.cmd_helper:[host]> /usr/local/google/home/bsalomon/android-sdk-linux/platform-tools/adb -s 094cda84 shell '( ps | grep -F surfaceflinger; echo "PIPESTATUS: ${PIPESTATUS[@]}" );echo %$?'
INFO:devil.utils.cmd_helper:[host]> /usr/local/google/home/bsalomon/android-sdk-linux/platform-tools/adb -s 094cda84 shell '( log -p i -t Telemetry '"'"'END http://web.archive.org/web/20150502135732/http://ie.microsoft.com/testdrive/Performance/Pirates/Default.html'"'"' );echo %$?'
INFO:devil.utils.cmd_helper:[host]> /usr/local/google/home/bsalomon/android-sdk-linux/platform-tools/adb -s 094cda84 shell '( ps | grep -F org.chromium.chrome; echo "PIPESTATUS: ${PIPESTATUS[@]}" );echo %$?'
INFO:devil.utils.cmd_helper:[host]> /usr/local/google/home/bsalomon/android-sdk-linux/platform-tools/adb -s 094cda84 shell '( ls /root );echo %$?'
INFO:devil.utils.cmd_helper:[host]> /usr/local/google/home/bsalomon/android-sdk-linux/platform-tools/adb -s 094cda84 shell '( cd /sys/devices/system/cpu && for CPU in cpu0 cpu1 cpu2 cpu3; do test -e "${CPU}/cpufreq/scaling_available_governors" && grep -q interactive ${CPU}/cpufreq/scaling_available_governors && test -e "$CPU/cpufreq/scaling_governor" && echo interactive > "$CPU/cpufreq/scaling_governor"; echo -n "%~%$?%~%"; done );echo %$?'
INFO:devil.android.perf.perf_control:Successfully set cpufreq/scaling_governor to 'interactive' on: cpu0 cpu1
INFO:devil.utils.cmd_helper:[host]> /usr/local/google/home/bsalomon/android-sdk-linux/platform-tools/adb -s 094cda84 shell '( start mpdecision );echo %$?'
INFO:devil.utils.cmd_helper:[host]> /usr/local/google/home/bsalomon/android-sdk-linux/platform-tools/adb -s 094cda84 shell '( p=org.chromium.chrome;if [[ "$(ps)" = *$p* ]]; then am force-stop $p; fi );echo %$?'
INFO:devil.utils.cmd_helper:[host]> /usr/local/google/home/bsalomon/android-sdk-linux/platform-tools/adb -s 094cda84 forward --remove tcp:43368
INFO:devil.utils.cmd_helper:[host]> /usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/devil/bin/deps/linux2/x86_64/forwarder_host --adb=/usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/devil/bin/deps/linux2/x86_64/bin/adb --serial-id=094cda84 --unmap 35841
ERROR:devil.android.forwarder:`/usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/devil/bin/deps/linux2/x86_64/forwarder_host --adb=/usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/devil/bin/deps/linux2/x86_64/bin/adb --serial-id=094cda84 --unmap 35841` exited with 1:
[0322/115842:ERROR:host_forwarder_main.cc(477)] ERROR: could not unmap port.
ERROR: Existing controllers:


[  FAILED  ] http://web.archive.org/web/20150502135732/http://ie.microsoft.com/testdrive/Performance/Pirates/Default.html (45405 ms)
[  PASSED  ] 0 tests.
[  FAILED  ] 1 test, listed below:
[  FAILED  ]  http://web.archive.org/web/20150502135732/http://ie.microsoft.com/testdrive/Performance/Pirates/Default.html

1 FAILED TEST

WARNING:root:Failed to extract previous results from HTML output
View result at file:///usr/local/google/home/bsalomon/src/clank/src/tools/perf/results.html
INFO:devil.utils.cmd_helper:[host]> /usr/local/google/home/bsalomon/android-sdk-linux/platform-tools/adb -s 094cda84 shell '( dumpsys battery );echo %$?'
INFO:root:Charging status checked at exit.

Ken, do you have any tips for how to run the perf tests? I've encountered this same issue with run_benchmark when working on other perf regressions over the past N > 6 months and have never figured it out. In the past I've always found some other way to reproduce the problem.

In this case I've tried just running the Pirate Marrrk benchmark from archive.org manually but I'm not seeing a consistent difference in the benchmark score. This probably just means the benchmark is finishing in about the same time but with fewer frames rendered with the change.

I had been using a N7v2 thinking it would be close enough. I got my hands on a N5 and now on the device it never attempts to load the page. It stays on about:blank. The N7 I was using was running Android 5.1.1. The N5 is running 6.0.1.

Sorry, I'm not an expert on the run_benchmark harness. (To be honest, the last time I investigated a perf regression I also wasn't able to reproduce the problem locally.)

Annie or Ned: can you help Brian figure out why this benchmark doesn't run correctly locally? Thanks.

Also CC'ing nduca for help.

Can you try to follow go/telemetry-device-setup to see if that fix the issues you are having?

I followed the instructions, including the provisioning script, but I get the same behavior. The site never fully loads when run run_benchmark. I attached a log of run_benchmark.

Deleted: log.txt 49.7 KB

log.txt 49.7 KB View Download

Are you running this on Mac or Linux workstation as the host? Note that we only support Linux as the host.

Hmhh, actually from the log, it seems like the step of installing the fake certificate was failing:

Traceback (most recent call last):
  File "/usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/telemetry/telemetry/internal/platform/network_controller_backend.py", line 141, in _InstallTestCa
    self._platform_backend.InstallTestCa(self._wpr_ca_cert_path)
  File "/usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/telemetry/telemetry/internal/platform/android_platform_backend.py", line 558, in InstallTestCa
    self._device_cert_util.install_cert(overwrite_cert=True)
  File "/usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/telemetry/third_party/web-page-replay/adb_install_cert.py", line 191, in install_cert
    if self._is_cert_installed():
  File "/usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/telemetry/third_party/web-page-replay/adb_install_cert.py", line 151, in _is_cert_installed
    return (self._adb_su_shell('ls', self.android_cacerts_path).strip() ==
  File "/usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/telemetry/third_party/web-page-replay/adb_install_cert.py", line 98, in _adb_su_shell
    build_version_sdk = int(self._get_property('ro.build.version.sdk'))
ValueError: invalid literal for int() with base 10: "adb server version (32) doesn't match this client (36); killing...\n* daemon started successfully *\n22"

In this case, because certificate installation failed, Telemetry has to launch the browser with "--ignore-certificate-error".

@bsalomon: when you run the test locally, did you see the butter bar that warn about certificate error? If not, then this is probably another case of Telemetry failed to pass flags to Chrome Android.

I have been using a Goobuntu workstation. I did not see a butter bar about a certificate error.

I filed  issue 704782  about commandline flag failure.

Actually, the problem seems to be the forwarder using a different version of adb than the rest of Telemetry. Most adb commands appear to be using:

  /usr/local/google/home/bsalomon/android-sdk-linux/platform-tools/adb

But the line when setting the forwarder shows:

INFO:devil.android.forwarder:Forwarding using commands: [[u'--adb=/usr/local/google/home/bsalomon/src/clank/src/third_party/catapult/devil/bin/deps/linux2/x86_64/bin/adb', '--serial-id=018e0bb120c8b857', '--map', '0', '54044']]

+jbudorick

bsalomon: do you have an ANDROID_SDK_ROOT environment variable set?

devil jumps through a few hoops to find the adb path to use for AdbWrapper: https://codesearch.chromium.org/chromium/src/third_party/catapult/devil/devil/android/sdk/adb_wrapper.py?rcl=0d592b9399db95b8e8b935d45ad238a14aa4a12d&l=58

It looks for:
 - a configured adb that exists locally
 - platform-tools/adb within a configured android_sdk that exists locally
 - a configured adb that it has to download

In your case, it appears to be using the second. Unfortunately, the forwarder *just* looks for a configured adb (https://codesearch.chromium.org/chromium/src/third_party/catapult/devil/devil/android/forwarder.py?rcl=0d592b9399db95b8e8b935d45ad238a14aa4a12d&l=133) rather than deferring to AdbWrapper.GetAdbPath. That's a bug; I'll send out a CL in a bit.

I do. I have ANDROID_SDK_ROOT=/usr/local/google/home/bsalomon/android-sdk-linux and the adb there is on my path:

bsalomon:~/src/clank/src ▶ which adb
/usr/local/google/home/bsalomon/android-sdk-linux/platform-tools/adb

yep, as expected. https://codereview.chromium.org/2775793002/ should make the forwarder use the same version of adb as everything else.

I remove the environment var and took adb off my path and now the benchmark runs correctly. Thanks for your help with this! And I'm glad to hear that it'll be fixed.

not sure where bugdroid is, but that fix rolled into chromium in https://chromium.googlesource.com/chromium/src/+/7522fb00d86cf9436d2af703fe17a0c1471b706a

The following revision refers to this bug:
  https://skia.googlesource.com/skia/+/1b52df33927c00d9c5aee63f7239dd77bce15e03

commit 1b52df33927c00d9c5aee63f7239dd77bce15e03
Author: Brian Salomon <bsalomon@google.com>
Date: Sat Mar 25 01:12:53 2017

Remove draw-as-clear workaround for Adreno 3xx in Chrome.

This caused a performance regression after Skia started seeing the real GL strings.

BUG= chromium:703774 

Change-Id: Ib0878062670b5ea51a005718a64a58d89fbaa42c
Reviewed-on: https://skia-review.googlesource.com/10165
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>

[modify] https://crrev.com/1b52df33927c00d9c5aee63f7239dd77bce15e03/src/gpu/gl/GrGLCaps.cpp

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/9f26840d0d0cb457306ee4e589011bb8ad97436e

commit 9f26840d0d0cb457306ee4e589011bb8ad97436e
Author: skia-deps-roller@chromium.org <skia-deps-roller@chromium.org>
Date: Sat Mar 25 09:41:12 2017

Roll src/third_party/skia/ dd1b4e94c..1b52df339 (1 commit)

https://skia.googlesource.com/skia.git/+log/dd1b4e94c429..1b52df33927c

$ git log dd1b4e94c..1b52df339 --date=short --no-merges --format='%ad %ae %s'
2017-03-24 bsalomon Remove draw-as-clear workaround for Adreno 3xx in Chrome.

Created with:
  roll-dep src/third_party/skia
BUG= 703774 


Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md

If the roll is causing failures, see:
http://www.chromium.org/developers/tree-sheriffs/sheriff-details-chromium#TOC-Failures-due-to-DEPS-rolls


CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel
TBR=fmalita@chromium.org

Change-Id: Id21c276163440a86050730e413391d6d81f8180c
Reviewed-on: https://chromium-review.googlesource.com/459090
Reviewed-by: Skia Deps Roller <skia-deps-roller@chromium.org>
Commit-Queue: Skia Deps Roller <skia-deps-roller@chromium.org>
Cr-Commit-Position: refs/heads/master@{#459648}
[modify] https://crrev.com/9f26840d0d0cb457306ee4e589011bb8ad97436e/DEPS

There was an unrelated performance improvement and then after the Skia roll there was a additional improvement commensurate with the original regression.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/b9ee54948a13bebf39f76a3cac615aa986beb3c4

commit b9ee54948a13bebf39f76a3cac615aa986beb3c4
Author: skia-deps-roller@chromium.org <skia-deps-roller@chromium.org>
Date: Tue Mar 28 01:54:32 2017

Roll src/third_party/skia/ 61e990297..2084ffa9a (43 commits)

https://skia.googlesource.com/skia.git/+log/61e990297a5c..2084ffa9a3fa

$ git log 61e990297..2084ffa9a --date=short --no-merges --format='%ad %ae %s'
2017-03-27 bsalomon Revert "Fix SkImage_Gpu::getROPixels"
2017-03-23 benjaminwagner Omit ImageStorageLoad on IntelBayTrail Ubuntu
2017-03-27 egdaniel Minor fix to VkWrapTest
2017-03-27 mtklein ok: add basic unit test support
2017-03-27 csmartdalton Add a simple tool for isolating a path from an SKP
2017-03-27 mtklein quiet some annoying logging in unit tests
2017-03-27 jvanverth Add height function to shadow tessellators.
2017-03-27 herb Change SkMemory to the more accurately named SkMalloc.
2017-03-27 bsalomon Split GrPipelineInput into separate color and coverage types, the latter of which is just an enum.
2017-03-27 mtklein ok: alias 565 -> sw:ct=565
2017-03-27 robertphillips Remove GrFragmentProcessor-derived class' GrTexture-based ctors
2017-03-27 kjlubick Update ATT Galaxy S7 to Android N
2017-03-27 robertphillips Add new proxy-based DetermineDomainMode w/ test
2017-03-26 mtklein ok: handle concurrent crashes better
2017-03-27 bsalomon Fully initialize array of previous edges in GrGLConvexPolyEffect to please MSAN
2017-03-24 iannucci Manually roll recipes (trivial)
2017-03-24 iannucci Manually roll recipes
2017-03-24 iannucci Manually roll recipes (trivial)
2017-03-26 mtklein focus hidden-visibility flags
2017-03-24 iannucci Manually roll recipes
2017-03-24 iannucci Manually roll recipes (trivial)
2017-03-26 update-skps Update SKP version
2017-03-25 mtklein ok: refactor Src/Dst interactions
2017-03-25 mtklein ok: basic Vias
2017-03-25 mtklein Spin off non-core parts of ok into their own files.
2017-03-25 reed Revert "Revert "simplify api to bitmapcache""
2017-03-24 bsalomon Remove draw-as-clear workaround for Adreno 3xx in Chrome.
2017-03-24 fmalita Revert "simplify api to bitmapcache"
2017-03-24 mtklein ok: introduce the idea of Dsts
2017-03-24 reed simplify api to bitmapcache
2017-03-24 benjaminwagner Omit tests causing IntelIris540 Vulkan to crash/fail.
2017-03-24 jvanverth Try a different calculation for computing convexity.
2017-03-24 brianosman Revert "Remove texture sampling from GrConfigConversionEffect"
2017-03-24 reed quick-fix to surface dox
2017-03-24 brianosman Fix SkImage_Gpu::getROPixels
2017-03-24 msarett Delete SkColorSpace::ColorSpaceFlags
2017-03-24 bsalomon Add function to perform color fragment processor elimination while performing FP analysis.
2017-03-24 egdaniel Update Linux Intel Release driver with patch to fix Vulkan
2017-03-24 bsalomon Run angle in 8 sample msaa on desktop
2017-03-24 jvanverth Disable convex check for now.
2017-03-24 bsalomon Revert "Revert "Create new inset algorithm for spot shadows""
2017-03-24 bsalomon Revert "Create new inset algorithm for spot shadows"
2017-03-23 mtklein ok: fix an assert from unbalanced save/restores

Created with:
  roll-dep src/third_party/skia
BUG= 703774 


Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md

If the roll is causing failures, see:
http://www.chromium.org/developers/tree-sheriffs/sheriff-details-chromium#TOC-Failures-due-to-DEPS-rolls


CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel
TBR=stephana@chromium.org

Change-Id: If7df898d4d3cb03d6d4511276a395d0e59677881
Reviewed-on: https://chromium-review.googlesource.com/461227
Reviewed-by: Skia Deps Roller <skia-deps-roller@chromium.org>
Commit-Queue: Skia Deps Roller <skia-deps-roller@chromium.org>
Cr-Commit-Position: refs/heads/master@{#459979}
[modify] https://crrev.com/b9ee54948a13bebf39f76a3cac615aa986beb3c4/DEPS