New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 884388 link

Starred by 2 users

Issue metadata

Status: Assigned
Owner:
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 3
Type: Task



Sign in to add a comment

Placeholder for LayoutNG bisects

Project Member Reported by cbiesin...@chromium.org, Sep 14

Issue description

 ChromiumPerfFyi/linux-perf-fyi/blink_perf.layout_ng / floats_10_1000 / floats_10_1000.html

Test: ChromiumPerfFyi/linux-perf-fyi/blink_perf.layout_ng/floats_10_1000/floats_10_1000.html
Value: 372.98 (± 2.72525)
Point ID: 590597
Time added: 2018-09-12T06:16:47.000Z
Chromium Commit Position range: 590547 - 590597
Chromium Git Hash range: abf553f - 31d71fd
V8 Commit Position: 33b726d
WebRTC Git Hash range: 375d35e - bfd412e

^ Got faster here, then got slower here:

Test: ChromiumPerfFyi/linux-perf-fyi/blink_perf.layout_ng/floats_10_1000/floats_10_1000.html
Value: 423.88 (± 3.46006)
Point ID: 590950
Time added: 2018-09-13T09:45:49.000Z
Chromium Commit Position range: 590933 - 590950
V8 Commit Position: c95e4bd
WebRTC Git Hash: 1f87ec6
 
Components: Blink>Layout
Owner: cbiesin...@chromium.org
Labels: -Type-Bug Type-Task
Status: Available (was: Untriaged)
Project Member

Comment 6 by bugdroid1@chromium.org, Sep 18

The following revision refers to this bug:
  https://chromium.googlesource.com/catapult/+/460fd889fcdbf4ea879cfc9a6d7df169b9d5bf5d

commit 460fd889fcdbf4ea879cfc9a6d7df169b9d5bf5d
Author: Simon <simonhatch@chromium.org>
Date: Tue Sep 18 00:04:34 2018

Dashboard - Add new linux perf fyi name to Pinpoint approved list.

Bug: chromium:884388
Change-Id: I1deb7c3a3e40a54fb0285054441bd5363b41d36a
Reviewed-on: https://chromium-review.googlesource.com/1228637
Commit-Queue: Dave Tu <dtu@chromium.org>
Reviewed-by: Dave Tu <dtu@chromium.org>

[modify] https://crrev.com/460fd889fcdbf4ea879cfc9a6d7df169b9d5bf5d/dashboard/dashboard/static/pinpoint.html

Cc: simonhatch@chromium.org
@simon, should this be working now? I just got another error:
"            "message": "Builder linux_perf_bisect not found"
"
Project Member

Comment 9 by bugdroid1@chromium.org, Sep 18

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/8c622763720e5a3c01667cada476dca0191d31e9

commit 8c622763720e5a3c01667cada476dca0191d31e9
Author: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Date: Tue Sep 18 02:47:23 2018

Roll src/third_party/catapult 9f7d7304ef4c..460fd889fcdb (1 commits)

https://chromium.googlesource.com/catapult.git/+log/9f7d7304ef4c..460fd889fcdb


git log 9f7d7304ef4c..460fd889fcdb --date=short --no-merges --format='%ad %ae %s'
2018-09-18 simonhatch@chromium.org Dashboard - Add new linux perf fyi name to Pinpoint approved list.


Created with:
  gclient setdep -r src/third_party/catapult@460fd889fcdb

The AutoRoll server is located here: https://autoroll.skia.org/r/catapult-autoroll

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

If the roll is causing failures, please contact the current sheriff, who should
be CC'd on the roll, and stop the roller if necessary.

CQ_INCLUDE_TRYBOTS=luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel

BUG=chromium:884388
TBR=sullivan@chromium.org

Change-Id: Ia29a7ddba3fbaeba848978495ca1814e00f77b71
Reviewed-on: https://chromium-review.googlesource.com/1229507
Reviewed-by: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Commit-Queue: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#591923}
[modify] https://crrev.com/8c622763720e5a3c01667cada476dca0191d31e9/DEPS

This needed a redeploy of the dashboard, I deployed and restarted your bisect.
Thanks, Simon!
Cc: kaznacheev@google.com afdo-chr...@skia-buildbots.google.com.iam.gserviceaccount.com
Status: Assigned (was: Available)
📍 Found significant differences after each of 2 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/14e19350e40000

Roll AFDO from 71.0.3550.0_rc-r1 to 71.0.3551.0_rc-r1 by afdo-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/31022095acde4a72f6920d20f68394e89f404032
370.7 → 423.2 (+52.51)

Rescale Chrome app icons to comply with Material Design guidelines by kaznacheev@google.com
https://chromium.googlesource.com/chromium/src/+/6ae79b485b51f0459ae24ad157f6435f258082b0
423.2 → 424.7 (+1.518)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
Cc: -simonhatch@chromium.org -kaznacheev@google.com e...@chromium.org
The material design thing is likely irrelevant

The AFDO thing is interesting, if unfortunate :/

+eae, fyi
AFDO?
AFDO: Automatic FDO; see FDO & PGO.
FDO: Feedback-Directed Optimization; see AFDO & PGO.

(per https://www.chromium.org/glossary)

I think it's the type of link-time optimization where it uses a profile to optimize the binary. But here it caused a fairly big regression for us :/
It likely doesn't run with LayoutNG so that isn't unexpected.
Cc: engedy@chromium.org
📍 Found significant differences after each of 2 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/1682ee97640000

Adjust transport list icon sizes and padding. by engedy@chromium.org
https://chromium.googlesource.com/chromium/src/+/6a6ca375d312e822a65ee997443ed7bc316cc310
418.9 → 422.1 (+3.2)

Roll AFDO from 71.0.3552.0_rc-r1 to 71.0.3552.4_rc-r1 by afdo-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/d10c090934c3426ce78d7bec283b252e1770c8ba
422.1 → 446.1 (+23.96)

Assigning to sheriff gbiv@chromium.org because "Roll AFDO from 71.0.3552.0_rc-r1 to 71.0.3552.4_rc-r1" is a roll.

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
Cc: palmer@chromium.org
📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/1401c88f640000

[PartitionAlloc] Add a flag to allow callers to zero-fill allocations. by palmer@chromium.org
https://chromium.googlesource.com/chromium/src/+/7bd9d693667d585c44d2d29666860c678c19bc83
1995 → 1906 (-89.4)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
Cc: -engedy@chromium.org -afdo-chr...@skia-buildbots.google.com.iam.gserviceaccount.com -palmer@chromium.org
(No need to pay attention to this bug, palmer/engedy)
Cc: ikilpatrick@chromium.org parastoog@google.com
📍 Found significant differences after each of 2 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/14a60db8e40000

[AF] New source of data validation in profiles. by parastoog@google.com
https://chromium.googlesource.com/chromium/src/+/dcd91e38a3a480bd9b19681af961a1fe5cd78967
3231 → 3258 (+27.39)

[LayoutNG] Fix %-resolution for ancestors of table cells. by ikilpatrick@chromium.org
https://chromium.googlesource.com/chromium/src/+/b0eec41c88383fa0254776d1cfba127b4a821a64
3258 → 424.7 (-2834)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
Cc: -parastoog@google.com
Ian -- fyi, that patch caused a massive regression for blink_perf.layout_ng/nested-percent-height-tables
Should I ignore this?
Parastoog -- yes, please ignore
😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/16c32f24e40000

Buildbucket says the build completed successfully, but Pinpoint can't find the isolate hash.
Cc: momon@google.com
📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/15e8aa47640000

Fix :scope selector matching in the case of DocumentFragment/ShadowRoot by momon@google.com
https://chromium.googlesource.com/chromium/src/+/2728052d428eb088be94db48fb475317ac011a47
421.1 → 374.8 (-46.28)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
Cc: -momon@google.com
Owner: ikilpatrick@chromium.org
Interesting, that selector matching patch seems to have made floats_10_1000 quite a bit faster!

I think the only thing to do with this bug at this point is to figure out why Ian's table patch had such negative impact.
Cc: dgro...@chromium.org
📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/1144b7a8e40000
Cc: afdo-chr...@skia-buildbots.google.com.iam.gserviceaccount.com
📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/12c4c584e40000

Roll AFDO from 71.0.3557.0_rc-r1 to 71.0.3558.0_rc-r1 by afdo-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/71634802f4f53b023cf6220a0ae89e79efed28fa
1889 → 1813 (-76.22)

Assigning to sheriff gbiv@chromium.org because "Roll AFDO from 71.0.3557.0_rc-r1 to 71.0.3558.0_rc-r1" is a roll.

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
Cc: cbiesin...@chromium.org rsesek@chromium.org
📍 Found significant differences after each of 2 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/13451697640000

mac: Delete the Cocoa download started animation. by rsesek@chromium.org
https://chromium.googlesource.com/chromium/src/+/c663200d7f97d8caf68e39c0e1ef4ff5ffa6e852
60.87 → 59.16 (-1.704)

[layoutng] Don't create a break token for every line when computing content size by cbiesinger@chromium.org
https://chromium.googlesource.com/chromium/src/+/d67d6aa0ea16efd0df6d1b40ed38348d4b916d86
59.54 → 57.49 (-2.051)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
Cc: -rsesek@chromium.org -afdo-chr...@skia-buildbots.google.com.iam.gserviceaccount.com
rsesek, feel free to ignore

Neither of those patches make sense. Maybe they just messed with PGO and increased the runtime that way
There's a thread I started yesterday on AFDO and its interaction with benchmarks like this.

http://g/chrome-speed/2lyfeApzVB8

and also https://crbug.com/889742 where gbiv is collecting problematic cases. Plesae chip in.
📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/1782dcf0e40000
Cc: yutak@chromium.org reillyg@google.com tsabolcec@google.com horo@chromium.org h...@chromium.org p...@webrtc.org jmad...@chromium.org
📍 Found significant differences after each of 7 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/117578af640000

Revert "[media] Fix video to canvas GPU-CPU race." by jmadill@chromium.org
https://chromium.googlesource.com/chromium/src/+/b39e1696cad4b7b661a91fe12e4fcc0aef45a9d3
1572 → 1534 (-37.66)

AEC3: Remove killswitches in AecState by peah@webrtc.org
https://webrtc.googlesource.com/src/+/f4801a1909dce2350a0941a43380880f9255e241
1529 → 1518 (-11.37)

Move most of compositor thread's initialization to Platform. by yutak@chromium.org
https://chromium.googlesource.com/chromium/src/+/88c843775504c8beb480aae0f62c02791ecbb108
1523 → 1538 (+15.21)

Reland "Send SXG accept header of navigation only to limited origins for Origin Trial." by horo@chromium.org
https://chromium.googlesource.com/chromium/src/+/6e5bd03fd2f66773d617fe5214fcf32f8a617df4
1533 → 1512 (-21.51)

[Password Manager] Determine status of PasswordStore::Init by tsabolcec@google.com
https://chromium.googlesource.com/chromium/src/+/bde4f91e4491d129e2ceace8839be036d1b1f5c2
1515 → 1527 (+11.44)

Revert "Roll clang 342523:343189." by hans@chromium.org
https://chromium.googlesource.com/chromium/src/+/7dcaa8f8a20e48e280bd0a0a67f080dc481e3325
1516 → 1527 (+11.12)

Replace the extensions URLRequestContext with only a CookieStore by reillyg@google.com
https://chromium.googlesource.com/chromium/src/+/78e17e333d8971eac8b93e74c3390072ec593f8f
1527 → 1540 (+12.69)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
Cc: -reillyg@google.com -p...@webrtc.org -tsabolcec@google.com -h...@chromium.org -horo@chromium.org -yutak@chromium.org -jmad...@chromium.org
all of y'all who have just been cc'd, please ignore
📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/111d78af640000
Cc: braveyao@chromium.org chromium...@skia-public.iam.gserviceaccount.com tbansal@chromium.org
📍 Found significant differences after each of 3 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/127cb42ce40000

Skip cookie check for origins when sending client hints by tbansal@chromium.org
https://chromium.googlesource.com/chromium/src/+/a0c1fc317df31881810430d0acd5907b5bd0130e
floats_100_100_nested: 61.32 → 60.78 (-0.5368)

Roll src/third_party/skia 55aea84baa53..49fdd7ad1875 (6 commits) by chromium-autoroll@skia-public.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/4e3fda3318564ac6316ee6ffe0356acc943059ad
floats_100_100_nested: 60.95 → 61.1 (+0.1524)

[desktopCapture] stop capture when screen is locked. by braveyao@chromium.org
https://chromium.googlesource.com/chromium/src/+/ab81ed65a4d788b68cccfa6c25817864bf9c9a50
floats_100_100_nested: 60.18 → 60.68 (+0.4995)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
Cc: -chromium...@skia-public.iam.gserviceaccount.com -braveyao@chromium.org -tbansal@chromium.org
(please ignore)
😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/16d20eb0e40000

An error occured while connecting to the server: Connection closed unexpectedly by server at URL: https://chrome-isolated.appspot.com/_ah/api/isolateservice/v1/retrieve
Cc: donghee....@gmail.com dgozman@chromium.org
📍 Found significant differences after each of 2 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/1198e2c4e40000

[DevTools] Rework provisional load explicit calls to probes by dgozman@chromium.org
https://chromium.googlesource.com/chromium/src/+/15fd8ac6e4078448cd312e24750abef691b23ca7
floats_10_1000: 448.7 → 453.2 (+4.478)

css-scroll-snap: scroll-snap-align keywords should not be reversed by donghee.na92@gmail.com
https://chromium.googlesource.com/chromium/src/+/a31d6d6cf095f81e5a14c4845d2cddff92e539f9
floats_10_1000: 452 → 483.4 (+31.42)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
Cc: a...@chromium.org jsaul@google.com robertph...@google.com orinj@chromium.org verwa...@chromium.org rbpotter@chromium.org aleventhal@chromium.org mstarzinger@chromium.org chromium...@skia-public.iam.gserviceaccount.com cduvall@chromium.org dpranke@chromium.org chrishtr@chromium.org egdaniel@google.com w...@chromium.org csmartdalton@google.com jvanverth@google.com yunqingwang@google.com enne@chromium.org tommycli@chromium.org v8-ci-au...@chops-service-accounts.iam.gserviceaccount.com pbos@chromium.org pwnall@chromium.org wangxianzhu@chromium.org kjlubick@google.com mmenke@chromium.org angiebird@google.com
📍 Found significant differences after each of 29 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/16bf7118e40000

Omnibox: Query in Omnibox - Fix Cut/Copy of query to clipboard by tommycli@chromium.org
https://chromium.googlesource.com/chromium/src/+/ba2e073e2c7964bdeb7e71129dd1f985b572a977
floats_100_100_nested: 59.05 → 60.69 (+1.64)

Remove ability to sample textures in vertex or geometry shaders. by egdaniel@google.com
https://skia.googlesource.com/skia/+/d50d6579d1c00de65b947c0531fa04c043729e49
floats_100_100_nested: 60.39 → 60.95 (+0.5651)

Exercise newly exposed SkYUVIndex-based SkImage factories (take 2) by robertphillips@google.com
https://skia.googlesource.com/skia/+/51c89e41c05b7f28e448c7a36c994ef562313333
floats_100_100_nested: 60.73 → 61.31 (+0.5777)

Move dumpJSON behind the SK_ENABLE_DUMP_GPU flag by kjlubick@google.com
https://skia.googlesource.com/skia/+/f4def3488589108fca0fa94e605377f2bb59cf90
floats_100_100_nested: 60.78 → 61.7 (+0.9255)

ccpr: Delete the stashed atlas texture before allocating a new one by csmartdalton@google.com
https://skia.googlesource.com/skia/+/29b11d605f252d6e632302f172d9c8f4080435d9
floats_100_100_nested: 61.98 → 61.62 (-0.3649)

[omnibox] Connect first omnibox Pedal into autocomplete system by orinj@chromium.org
https://chromium.googlesource.com/chromium/src/+/76893bb36cffd276951b4209df61a5fe93b64e1d
floats_100_100_nested: 61.22 → 60.21 (-1.009)

Omnibox: Query in Omnibox - Fix CurrentTextIsURL and match generation by tommycli@chromium.org
https://chromium.googlesource.com/chromium/src/+/555e0ca1e1eff5f08f4380325f8bc84e6b805423
floats_100_100_nested: 60.34 → 60.66 (+0.3237)

Fix DataPipe(capacity) constructor not to double-allocate. by wez@chromium.org
https://chromium.googlesource.com/chromium/src/+/7bec6951df50f0828bc7b154376e056603b10ae9
floats_100_100_nested: 59.53 → 60.94 (+1.406)

Cleanup: Pass scoped_refptr by value in //storage. by pwnall@chromium.org
https://chromium.googlesource.com/chromium/src/+/6ea3da840db396048d6f491a93345c3c181ee56b
floats_100_100_nested: 62.14 → 60.82 (-1.326)

Quota: Implement TODO in quota_client.h. by pwnall@chromium.org
https://chromium.googlesource.com/chromium/src/+/bc7fb3ba96f7f86637d02139b6bd0c856ce7aa52
floats_100_100_nested: 60.9 → 61.56 (+0.6634)

Roll src/third_party/chromite 39a66e9fc8df..8fd9d735fdea (1 commits) by chromium-autoroll@skia-public.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/a3b871e005a2abc753519358874e663546aa316a
floats_100_100_nested: 62.48 → 60.57 (-1.908)

Revert "Remove perf_expectations" by dpranke@chromium.org
https://chromium.googlesource.com/chromium/src/+/a230a9900389fef6bc948cf69b1c8d37836f3359
floats_100_100_nested: 60.57 → 61.38 (+0.8047)

Fix nullptr crash in ExtensionUninstallDialogView by pbos@chromium.org
https://chromium.googlesource.com/chromium/src/+/06c6145d4613eedf2f6ca20708352910f41526ad
floats_100_100_nested: 61.38 → 60.05 (-1.329)

[wasm] Initial implementation of catch-all decoding. by mstarzinger@chromium.org
https://chromium.googlesource.com/v8/v8/+/8cc0761841695ec74d240609a3d22f7074c1bd30
floats_100_100_nested: 60.19 → 60.95 (+0.7548)

[parser] Fix single-expression arrow function scoping by verwaest@chromium.org
https://chromium.googlesource.com/v8/v8/+/af34c6c236748b97cc949f04f5323c22e6b0d5c1
floats_100_100_nested: 60.95 → 61.17 (+0.2215)

Update V8 to version 7.1.288. by v8-ci-autoroll-builder@chops-service-accounts.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/2b5354f1f94ce7e4e88f43ed92ae1a3c02f2668a
floats_100_100_nested: 61.03 → 59.25 (-1.784)

Revert "Fix bugs and improve perf in Touch Bar text suggestions." by avi@chromium.org
https://chromium.googlesource.com/chromium/src/+/6068cf6b0b7348f09085a3256588f831653b4a52
floats_100_100_nested: 59.86 → 60.26 (+0.4024)

Fix a WebRtcLogUploader threading bug. by mmenke@chromium.org
https://chromium.googlesource.com/chromium/src/+/fbb7d2e309369c8539c63851da1e49d3e13a2018
floats_100_100_nested: 61.01 → 61.61 (+0.5954)

[Autofill] Action post-review comments from 1255723 by jsaul@google.com
https://chromium.googlesource.com/chromium/src/+/4f4542eae187ee47e39291b9c5f1ac835eb32e6d
floats_100_100_nested: 61.67 → 59.97 (-1.696)

Fix DataReductionProxyBrowsertests on official builds by cduvall@chromium.org
https://chromium.googlesource.com/chromium/src/+/d76b93b3c4c2c106b197d6bc53c092e41085aba8
floats_100_100_nested: 59.97 → 61.2 (+1.235)

[LayoutNG] Separate code for computing visual and layout overflows. by chrishtr@chromium.org
https://chromium.googlesource.com/chromium/src/+/2231aea5dcda383dd2c924e7750a02a3f88c783c
floats_100_100_nested: 60.66 → 62.71 (+2.052)

Table-related properties such as cell coordinates on layout tables by aleventhal@chromium.org
https://chromium.googlesource.com/chromium/src/+/170d265340ac18ec07903ba1d02dcb09ac656288
floats_100_100_nested: 62.23 → 61.96 (-0.2732)

[css-flex] Refactor FlexLayoutAlgorithm creation by cbiesinger@chromium.org
https://chromium.googlesource.com/chromium/src/+/d05aeb611fba445e954232c2485bdf65e4e7410a
floats_100_100_nested: 61.62 → 61.76 (+0.1316)

[PE] Check and fix under-invalidation of GeometryMapper cache by wangxianzhu@chromium.org
https://chromium.googlesource.com/chromium/src/+/9cdbb8acce4b90a37cf9f022bda2be2e6d9215a5
floats_100_100_nested: 61.76 → 61.27 (-0.4892)

Clean up chrome://devices by rbpotter@chromium.org
https://chromium.googlesource.com/chromium/src/+/5aba2f8667a9a70f5f1b0eee6ce8514b3f234a3b
floats_100_100_nested: 62.06 → 61.61 (-0.4468)

Changes to facilitate accurate sub-pel motion search by yunqingwang@google.com
https://chromium.googlesource.com/webm/libvpx/+/c5586bfa849475825e248588da2adafeb6be1cb8
floats_100_100_nested: 61.76 → 62.11 (+0.352)

Fix bug in prepare_nb_full_mvs by angiebird@google.com
https://chromium.googlesource.com/webm/libvpx/+/e49ef1476db5083e002863c7c2b8b5188c298d48
floats_100_100_nested: 62.11 → 60.85 (-1.259)

Revert "Use transfer buffer for small transfer cache entries" by enne@chromium.org
https://chromium.googlesource.com/chromium/src/+/6bbcf5cf66c5e9540c93857d2fac6a357b273884
floats_100_100_nested: 60.65 → 61.29 (+0.6365)

More SkImage_GpuYUV updates by jvanverth@google.com
https://skia.googlesource.com/skia/+/8bbce0e29e335a16e7ee80830b4647a7096dc6c3
floats_100_100_nested: 61.26 → 62.24 (+0.9797)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
Cc: -mmenke@chromium.org
My cl was a 1-line crash fix.  If the line were ever hit, we'd unconditionally crash.  Since we didn't go from crashing to not crasahing, my CL is irrelevant here.
Cc: -cduvall@chromium.org
My change is test only, so also irrelevant.
Cc: erikc...@chromium.org fmalita@chromium.org herb@google.com
📍 Found significant differences after each of 12 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/16f7f8dae40000

Omnibox: Query in Omnibox - Fix Cut/Copy of query to clipboard by tommycli@chromium.org
https://chromium.googlesource.com/chromium/src/+/ba2e073e2c7964bdeb7e71129dd1f985b572a977
floats_20_100: 188.6 → 190.3 (+1.757)

[skottie] Optional path control points by fmalita@chromium.org
https://skia.googlesource.com/skia/+/2c06e146978959573a77aa51be5577d811bbf63b
floats_20_100: 190.3 → 188.9 (-1.421)

Remove ability to sample textures in vertex or geometry shaders. by egdaniel@google.com
https://skia.googlesource.com/skia/+/d50d6579d1c00de65b947c0531fa04c043729e49
floats_20_100: 188.7 → 190.5 (+1.793)

Move dumpJSON behind the SK_ENABLE_DUMP_GPU flag by kjlubick@google.com
https://skia.googlesource.com/skia/+/f4def3488589108fca0fa94e605377f2bb59cf90
floats_20_100: 189.9 → 191.7 (+1.853)

Remove old code by moving to new API in bitmap text by herb@google.com
https://skia.googlesource.com/skia/+/f3a3864290ddfef846b7b032be5845110ae8466c
floats_20_100: 191.9 → 189.8 (-2.1)

Revert "Remove ability to sample textures in vertex or geometry shaders." by egdaniel@google.com
https://skia.googlesource.com/skia/+/6e2625d9ae89f6709a606bcf1a15b35741393e02
floats_20_100: 189.6 → 190.8 (+1.211)

Roll src/third_party/skia ba651682ae85..8ad991da4c04 (28 commits) by chromium-autoroll@skia-public.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/fb1e210df14d369b4be7a856c100d31b75c4e275
floats_20_100: 189.8 → 189.1 (-0.7464)

[omnibox] Connect first omnibox Pedal into autocomplete system by orinj@chromium.org
https://chromium.googlesource.com/chromium/src/+/76893bb36cffd276951b4209df61a5fe93b64e1d
floats_20_100: 189.4 → 190.3 (+0.8623)

Fix DataPipe(capacity) constructor not to double-allocate. by wez@chromium.org
https://chromium.googlesource.com/chromium/src/+/7bec6951df50f0828bc7b154376e056603b10ae9
floats_20_100: 189.6 → 191.2 (+1.617)

Roll src/third_party/chromite 39a66e9fc8df..8fd9d735fdea (1 commits) by chromium-autoroll@skia-public.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/a3b871e005a2abc753519358874e663546aa316a
floats_20_100: 191.3 → 190 (-1.235)

[LayoutNG] Separate code for computing visual and layout overflows. by chrishtr@chromium.org
https://chromium.googlesource.com/chromium/src/+/2231aea5dcda383dd2c924e7750a02a3f88c783c
floats_20_100: 190 → 196.6 (+6.528)

Add documentation for parsing test results. by erikchen@chromium.org
https://chromium.googlesource.com/chromium/src/+/b83e010af90a3be542cbae52a4f0aa8d10b3ec37
floats_20_100: 196.8 → 195.8 (-1.006)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
Cc: -mstarzinger@chromium.org
My CL is on an experimental feature not enabled in production yet.
Cc: -verwa...@chromium.org
Cc: -pbos@chromium.org
Cc: -tommycli@chromium.org -a...@chromium.org -aleventhal@chromium.org -angiebird@google.com -chromium...@skia-public.iam.gserviceaccount.com -orinj@chromium.org -wangxianzhu@chromium.org -robertph...@google.com -w...@chromium.org -fmalita@chromium.org -egdaniel@google.com -enne@chromium.org -dpranke@chromium.org -donghee....@gmail.com -dgozman@chromium.org -v8-ci-au...@chops-service-accounts.iam.gserviceaccount.com -erikc...@chromium.org -kjlubick@google.com -rbpotter@chromium.org -pwnall@chromium.org -yunqingwang@google.com -csmartdalton@google.com -jvanverth@google.com -herb@google.com -chrishtr@chromium.org -jsaul@google.com
Cc: sunxd@chromium.org chromium...@skia-public.iam.gserviceaccount.com altimin@chromium.org
📍 Found significant differences after each of 3 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/1495bfade40000

[PaintTouchActionRects] Record overflow paint rect as hit test rect by sunxd@chromium.org
https://chromium.googlesource.com/chromium/src/+/4294591ceb97fab1465d9490484041a708a9ba38
floats_100_100_nested: 57.11 → 56.69 (-0.4131)

Reland "[base] Use TaskRunner instead of MessageLoop in FileDescriptorWatcher" by altimin@chromium.org
https://chromium.googlesource.com/chromium/src/+/6247f6a91ade61d198c618966cdb2669beab981b
floats_100_100_nested: 55.48 → 55.1 (-0.3743)

Roll src/third_party/skia f10ea2108809..ff78aaa431ed (1 commits) by chromium-autoroll@skia-public.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/a183cbf3ccf5b9f59595ebd3e698b24d26b390c1
floats_100_100_nested: 55.1 → 56 (+0.8974)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
Cc: -chromium...@skia-public.iam.gserviceaccount.com -sunxd@chromium.org -altimin@chromium.org
(please ignore)
Cc: andruud@chromium.org lizeb@chromium.org mastiz@chromium.org davidben@chromium.org clemensh@chromium.org sa...@chromium.org chromium...@skia-public.iam.gserviceaccount.com jhawkins@chromium.org reed@google.com qnnguyen@chromium.org hongjunchoi@chromium.org caseq@chromium.org djacobo@chromium.org jongdeok...@navercorp.com v8-ci-au...@chops-service-accounts.iam.gserviceaccount.com weidongg@chromium.org pbos@chromium.org martiniss@chromium.org peter.wm...@gmail.com collinbaker@chromium.org
📍 Found significant differences after each of 24 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/11e5c523e40000

[layoutng] Give the baseline request vector an inline capacity by cbiesinger@chromium.org
https://chromium.googlesource.com/chromium/src/+/8f0dba841629a98154dbc3186a062f692bf69f00
flexbox-with-stretch-layout: 18.28 → 17.26 (-1.026)

Create TouchSelectionMenuViews to separate Menu by djacobo@chromium.org
https://chromium.googlesource.com/chromium/src/+/9a0f2af84c082c3deccd2355e6c82bffda0ec6e6
flexbox-with-stretch-layout: 17.15 → 17.67 (+0.5192)

Fix arrow button issue in tablet mode by weidongg@chromium.org
https://chromium.googlesource.com/chromium/src/+/edccc9b7128aee9bb555c1c03985ed4be6bb6c06
flexbox-with-stretch-layout: 17.7 → 17.11 (-0.5926)

Extract instrumentation methods from RenderFrameDevToolsAgentHost into devtools_instrumentation by caseq@chromium.org
https://chromium.googlesource.com/chromium/src/+/f2d4ff7a9f2bc8e67334a95ea7b169a8158d5ac7
flexbox-with-stretch-layout: 17.12 → 17.65 (+0.5281)

add helpers to SkFont (and SkTypeface) by reed@google.com
https://skia.googlesource.com/skia/+/7d7c68339d158203d2eb791954c7f6a4de75652e
flexbox-with-stretch-layout: 17.64 → 17.83 (+0.1866)

Roll src/third_party/skia 68825776f4b4..c09dc8ac723f (12 commits) by chromium-autoroll@skia-public.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/833fa18f15c01a76c42dda0f4a3ddb8926b71716
flexbox-with-stretch-layout: 17.79 → 17.29 (-0.495)

Move focus to web contents when DownloadShelfView is closed by collinbaker@chromium.org
https://chromium.googlesource.com/chromium/src/+/99bd43cf24bd1c38c6d12b180696f8d4c1c094d6
flexbox-with-stretch-layout: 17.29 → 17.87 (+0.5745)

Wire DevTools instrumentation signals for browser agent through devtools_instrumentation by caseq@chromium.org
https://chromium.googlesource.com/chromium/src/+/f02829c1bd86c78390b7ab04a75f4adff7e60d56
flexbox-with-stretch-layout: 17.86 → 17.35 (-0.5133)

Smart Lock: Remove SetTrialRun method. by jhawkins@chromium.org
https://chromium.googlesource.com/chromium/src/+/0c491d560e79043e204bdf81126fa4fd54f211f6
flexbox-with-stretch-layout: 17.35 → 17.66 (+0.3124)

cros: Attempt externalBinaryUnlock on lid-open while locked. by qnnguyen@chromium.org
https://chromium.googlesource.com/chromium/src/+/da08b742817387a26a035a7fd493f36a4740cb87
flexbox-with-stretch-layout: 17.63 → 17.52 (-0.1103)

Remove win10_chromium_x64_rel_ng from the CQ by martiniss@chromium.org
https://chromium.googlesource.com/chromium/src/+/60b53b5d929f5c3d5059aee2d800b7a7177596aa
flexbox-with-stretch-layout: 17.52 → 17.69 (+0.1704)

[iwyu] Split off allocation-builder-inl.h by clemensh@chromium.org
https://chromium.googlesource.com/v8/v8/+/9299702afc1a5ad3dbd70af5844f6a755ade4ce3
flexbox-with-stretch-layout: 17.54 → 17.35 (-0.1913)

[wasm] Turn field into local variable by clemensh@chromium.org
https://chromium.googlesource.com/v8/v8/+/d99a1b950ec952c328d9f27e2f0ff824021fb866
flexbox-with-stretch-layout: 17.42 → 17.75 (+0.3296)

[builtins] Cap and grow Array.p.join's internal buffer. by peter.wm.wong@gmail.com
https://chromium.googlesource.com/v8/v8/+/2f382887d48777a737696242186bc15025db9d81
flexbox-with-stretch-layout: 17.76 → 17.28 (-0.481)

Update V8 to version 7.2.179. by v8-ci-autoroll-builder@chops-service-accounts.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/9394cf24c478f68eb848061e856dbff3b9daa988
flexbox-with-stretch-layout: 17.28 → 17.15 (-0.1316)

Clean up AnimatedAppMenuIcon experiment by pbos@chromium.org
https://chromium.googlesource.com/chromium/src/+/a2ea8886cb4baf0b900279e4b22c9225fc0ce038
flexbox-with-stretch-layout: 17.16 → 17.7 (+0.5349)

Classify match contents considering words in input text. by jongdeok.kim@navercorp.com
https://chromium.googlesource.com/chromium/src/+/1825303c8b22c57096121c5a030e584f13687595
flexbox-with-stretch-layout: 17.55 → 17.25 (-0.2991)

Fix Sync.ModelTypeEntityChange2 undercounting initial updates by mastiz@chromium.org
https://chromium.googlesource.com/chromium/src/+/d5398c5cd3256cb51ea09936d897aa882d349ed5
flexbox-with-stretch-layout: 17.27 → 17.82 (+0.5465)

Rewrite NameToPseudoCompare to avoid dummy_key. by davidben@chromium.org
https://chromium.googlesource.com/chromium/src/+/1bf36a0584562dc9662804a91d03105de1b6977a
flexbox-with-stretch-layout: 17.84 → 17.69 (-0.145)

Roll AFDO from 72.0.3595.0_rc-r1 to 72.0.3596.0_rc-r1 by chromium-autoroll@skia-public.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/cd069ae57be6a2f0d420777a3d91f39f9d6ed7a2
flexbox-with-stretch-layout: 17.67 → 17.09 (-0.5799)

Fix an off-by-one in Drive invalidations. by sammc@chromium.org
https://chromium.googlesource.com/chromium/src/+/9438e4bb60af2ba72ab8ef50f77d4610da2ad848
flexbox-with-stretch-layout: 17.07 → 17.69 (+0.6158)

<b>[css-properties-values-api] Compute <time> values to seconds.</b> by andruud@chromium.org
https://chromium.googlesource.com/chromium/src/+/c2d00575676e5532dbc4841916ce513caa630e0d
flexbox-with-stretch-layout: 17.69 → 17.43 (-0.2557)

Remove first_list_item_view_ in HoverListView by hongjunchoi@chromium.org
https://chromium.googlesource.com/chromium/src/+/0ac254b33b26cbc3164873a9dc160e127f49c476
flexbox-with-stretch-layout: 17.36 → 17.56 (+0.2026)

blink/bindings: Compress ParkableStrings in the background. by lizeb@chromium.org
https://chromium.googlesource.com/chromium/src/+/370d3c5b036bdc5e9cf1513a1316e6b128228fa6
flexbox-with-stretch-layout: 17.61 → 17.01 (-0.6034)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/144ff87ee40000

entity is too big
Cc: -mastiz@chromium.org -davidben@chromium.org -pbos@chromium.org -chromium...@skia-public.iam.gserviceaccount.com -peter.wm...@gmail.com -ikilpatrick@chromium.org -weidongg@chromium.org -andruud@chromium.org -cbiesin...@chromium.org -caseq@chromium.org -collinbaker@chromium.org -jhawkins@chromium.org -martiniss@chromium.org -v8-ci-au...@chops-service-accounts.iam.gserviceaccount.com -jongdeok...@navercorp.com -hongjunchoi@chromium.org -e...@chromium.org -qnnguyen@chromium.org -dgro...@chromium.org -clemensh@chromium.org -reed@google.com -sa...@chromium.org -lizeb@chromium.org -djacobo@chromium.org
Owner: cbiesin...@chromium.org
(please ignore)
So for my reference:
- Per the last comment, my baseline patch made flexbox-with-stretch-layout slower (??) but Koji's NGBaseline rewrite fixed that
- Multicol tests got slower a couple of times due to AFDO (603366, 603020)
Cc: vasi...@chromium.org jdoerrie@chromium.org r...@igalia.com futhark@chromium.org tommycli@chromium.org hta@chromium.org m...@chromium.org ma...@igalia.com dgozman@chromium.org
📍 Found significant differences after each of 9 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/110d94a1e40000

[css-grid] Fix percentages in relative offsets for grid items by rego@igalia.com
https://chromium.googlesource.com/chromium/src/+/6298f794c7a71b2bd3f3efe137add7ceac5ae38d
fixed-grid-lots-of-stretched-data: 1455 → 1371 (-83.63)

Make :host account for its own pseudo-class specificity. by futhark@chromium.org
https://chromium.googlesource.com/chromium/src/+/674ef6ac2499fb51d82e752ccf723fd76c505caf
fixed-grid-lots-of-stretched-data: 1373 → 1389 (+16.27)

Deprecate "Smart Lock" in the native UI on desktop. by vasilii@chromium.org
https://chromium.googlesource.com/chromium/src/+/9a98a17c068a33fac936c1eecde53fd82ba2781f
fixed-grid-lots-of-stretched-data: 1385 → 1375 (-10.58)

Legacy audio diverting without UMAs. by miu@chromium.org
https://chromium.googlesource.com/chromium/src/+/8cb1798d0f73fee9161fb9e26db3ab000788ca19
fixed-grid-lots-of-stretched-data: 1377 → 1370 (-6.437)

Disable layout test RTCDTMFSender-ontonechange (flaky) by hta@chromium.org
https://chromium.googlesource.com/chromium/src/+/c62ce43b9bbb46a2c3b5f272ebaf83fd452c225d
fixed-grid-lots-of-stretched-data: 1372 → 1386 (+14.46)

Remove StatsCollectionObserver by dgozman@chromium.org
https://chromium.googlesource.com/chromium/src/+/3e37dd0eadbde2851762d83544723eca9db95b03
fixed-grid-lots-of-stretched-data: 1386 → 1381 (-4.643)

[Password Manager] Get All Logins With Affiliation Info by jdoerrie@chromium.org
https://chromium.googlesource.com/chromium/src/+/9e057b75d8e0ca01ecbeead1a4553815527aedc7
fixed-grid-lots-of-stretched-data: 1382 → 1389 (+7.685)

Omnibox: Query in Omnibox / Elisions - Reveal full URL for double Ctrl+L by tommycli@chromium.org
https://chromium.googlesource.com/chromium/src/+/b9d74169d4da6045e7f4d1db87ae2b6d991f57f6
fixed-grid-lots-of-stretched-data: 1388 → 1373 (-14.45)

Migrate AdvancedProtectionStatusManager to the IdentityManager by mario@igalia.com
https://chromium.googlesource.com/chromium/src/+/81eb18725b56153da09f926a3f50ea73aae05d0a
fixed-grid-lots-of-stretched-data: 1372 → 1379 (+6.442)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
Cc: chromium...@skia-public.iam.gserviceaccount.com
📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/12b1a2d5e40000

Roll AFDO from 72.0.3591.0_rc-r1 to 72.0.3592.0_rc-r1 by chromium-autoroll@skia-public.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/acfd23af50e37e3d9087f6a66f8031b6c9261297
multicol_lots-of-text-autofill: 87.31 → 82.02 (-5.291)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/14ed28b9e40000

Roll AFDO from 72.0.3593.0_rc-r1 to 72.0.3594.0_rc-r1 by chromium-autoroll@skia-public.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/e0532d4ed62dfeab4628b44bce4ccc0ae1c3aa72
multicol_lots-of-text-autofill: 81.69 → 76.77 (-4.927)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
Cc: -tommycli@chromium.org -ma...@igalia.com -chromium...@skia-public.iam.gserviceaccount.com -m...@chromium.org -r...@igalia.com -hta@chromium.org -futhark@chromium.org -vasi...@chromium.org -jdoerrie@chromium.org -dgozman@chromium.org
(please ignore)
Cc: sorin@chromium.org gyuyoung...@lge.com
📍 Found significant differences after each of 2 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/16db9eade40000

<b>Fix UpdateClient XML serialization when <disabled> reasons are present.</b> by sorin@chromium.org
https://chromium.googlesource.com/chromium/src/+/05d610bfffc6a715c4d4aece4461032af065402d
attach-inlines-2: 8457 → 8432 (-25.45)

AppCache: Add command-line flags for default quota and disk size. by gyuyoung.kim@lge.com
https://chromium.googlesource.com/chromium/src/+/6efb4dae8ab5d42f5ffcd9b9185c3b7af72925e5
attach-inlines-2: 8437 → 8403 (-33.33)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
Cc: -sorin@chromium.org -gyuyoung...@lge.com
Please ignore.
Cc: reed@google.com
📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/15c8228be40000

start to use SkFont api (paint's equivalents are deprecated) by reed@google.com
https://chromium.googlesource.com/chromium/src/+/d90028160469664fccdec91f15b64202b91a2600
timeToFirstContentfulPaint:layout: 4277 → 4817 (+540.8)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/15a2a0f7e40000
Cc: szager@chromium.org
📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/1132a4f4140000

Disable ignore-pending-stylesheets mechanism by szager@chromium.org
https://chromium.googlesource.com/chromium/src/+/4f35b999c1e4159ded31f7c16b71ca3e0e7ddfee
Shapes_MultipleShapes: 12.2 → 28.42 (+16.22)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
Cc: -szager@chromium.org
#95 is known at issue 909964 according to Stefan, thanks!
Project Member

Comment 97 by 42576172...@developer.gserviceaccount.com, Yesterday (25 hours ago)

Project Member

Comment 98 by 42576172...@developer.gserviceaccount.com, Today (3 hours ago)

Cc: psla@webrtc.org jam@chromium.org chromium...@skia-public.iam.gserviceaccount.com maxlg@chromium.org
📍 Found significant differences after each of 4 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/16a0e585540000

Roll src/third_party/skia 19ec80fb37ce..a2187bf7629a (5 commits) by chromium-autoroll@skia-public.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/656390b64b64b04300b246ae4dada1d839684114
appending-text: 811.1 → 814.5 (+3.397)

Instantiate Domain Reliability code in NetworkContext so it works with and without the network service. by jam@chromium.org
https://chromium.googlesource.com/chromium/src/+/aec73c860b15a0873431e32913e7e536f929da68
appending-text: 815.4 → 803.6 (-11.82)

Fix data race in channel_send.cc by psla@webrtc.org
https://webrtc.googlesource.com/src/+/1eebec980898ad142d3b2e5d172c27aaf365aaa4
appending-text: 804 → 810.1 (+6.105)

[FCP++] Enable FCP++ metrics via finch experiment by maxlg@chromium.org
https://chromium.googlesource.com/chromium/src/+/4e243fc4a328acee8354c73e7c9df2facac9f3f7
appending-text: 811 → 921.3 (+110.4)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions

Sign in to add a comment