All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=894728

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=7159d905cc0bf555774b0128a39ce625b9e7794a9f933fa1330e9c88be9f3fd4


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

Android Nexus5 Perf

speedometer - Benchmark documentation link:
  None

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/17d4246ce40000

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/1358395ae40000

Issue 894729 has been merged into this issue.

 Issue 894731  has been merged into this issue.

Started this pinpoint job for a test which looks less noisy than some others: https://pinpoint-dot-chromeperf.appspot.com/job/1358395ae40000 

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/1358395ae40000

Roll AFDO from 71.0.3576.0_rc-r1 to 71.0.3577.0_rc-r1 by chromium-autoroll@skia-public.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/6fb1a0860e442c7f9707c2003d219cb03b3dff81
jQuery-TodoMVC: 5756 → 6454 (+698.3)

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

Benchmark documentation link:
  None

Bisect found this commit:

Roll AFDO from 71.0.3576.0_rc-r1 to 71.0.3577.0_rc-r1

This CL may cause a small binary size increase, roughly proportional
to how long it's been since our last AFDO profile roll. For larger
increases (around or exceeding 100KB), please file a bug against
gbiv@chromium.org. Additional context:  https://crbug.com/805539 

Please note that, despite rolling to chrome/android, this profile is
used for both Linux and Android.

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


TBR=gbiv@chromium.org

I think that's the largest regression I've seen from an AFDO profile roll (for speedometer, at least)

I'll check on Monday to see if a later roll fixes this (we do them daily). If r598701 ended up in the branch, I'll push to revert it there, since I've heard 0 complaints about the roll before this one.

Thanks

Sounds like the branch was made at r599034, so this is our 'final' profile for 71. That's exactly what I didn't want. Revert coming soon :)

Hi,

I'd like to merge this revert into M71 to fix this 9% speedometer regression: https://chromium-review.googlesource.com/c/chromium/src/+/1279231

We had the previous profile in for ~a day, and I've received no complaints about it, so this should be quite safe from both a functionality and performance standpoint.

Also, a "you didn't do anything obviously wrong," on my CL would be appreciated. This is my first merge request into a release branch of Chromium, and I've tripped over Gerrit's UI a few times in the past. :)

Thank you!

Your change meets the bar and is auto-approved for M71. Please go ahead and merge the CL to branch 3578 manually. Please contact milestone owner if you have questions.
Owners: benmason@(Android), kariahda@(iOS), kbleicher@(ChromeOS), govind@(Desktop)

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

Poked a milestone owner for the informal LGTM I requested.

Looks like the charts have recovered with the AFDO roll in r599330. Will close this out once the merge is done.

📍 Found significant differences after each of 2 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/17d4246ce40000

Roll AFDO from 71.0.3576.0_rc-r1 to 71.0.3577.0_rc-r1 by chromium-autoroll@skia-public.iam.gserviceaccount.com
https://chromium.googlesource.com/chromium/src/+/6fb1a0860e442c7f9707c2003d219cb03b3dff81
jQuery-TodoMVC: 2.382e+04 → 2.654e+04 (+2715)

Replace FTN::IsDescendantOf by RFH::IsDescendantOf by arthursonzogni@chromium.org
https://chromium.googlesource.com/chromium/src/+/969a794e014a8851806cc6792d7b453a070c1de7
jQuery-TodoMVC: 2.654e+04 → 2.635e+04 (-186.3)

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

Benchmark documentation link:
  None

In my CL, I just moved a trivial function from A to B. So I don't think we got a performance improvement because of this.
The pinpoint graph doesn't look special in my CL position.

Removing myself.

Graphs are (still) recovered and revert is submitted to M71.

Thanks again!

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/c96c54b23eb0f17e9e5c3fb6e0a84591de656170

commit c96c54b23eb0f17e9e5c3fb6e0a84591de656170
Author: George Burgess <gbiv@chromium.org>
Date: Tue Oct 16 05:42:47 2018

Revert "Roll AFDO from 71.0.3576.0_rc-r1 to 71.0.3577.0_rc-r1"

This reverts commit 6fb1a0860e442c7f9707c2003d219cb03b3dff81.

Reason for revert:  crbug.com/894728  . The plan is to not revert this
on master, since we'll probably roll past the regression. As it
stands, this is the 'final' profile for m71, though, and a 9%
speedometer regression isn't acceptable when yesterday's profile
appears just fine. :)

Original change's description:
> Roll AFDO from 71.0.3576.0_rc-r1 to 71.0.3577.0_rc-r1
>
> This CL may cause a small binary size increase, roughly proportional
> to how long it's been since our last AFDO profile roll. For larger
> increases (around or exceeding 100KB), please file a bug against
> gbiv@chromium.org. Additional context:  https://crbug.com/805539 
>
> Please note that, despite rolling to chrome/android, this profile is
> used for both Linux and Android.
>
> The AutoRoll server is located here: https://autoroll.skia.org/r/afdo-chromium-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.
>
>
> TBR=gbiv@chromium.org
>
> Change-Id: Ic2facf850ac74d8087cc599ce0a9a7351b330ce3
> Reviewed-on: https://chromium-review.googlesource.com/c/1275868
> 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@{#598701}

Bug:  894728 
Change-Id: I03686edca0cda64e161a32ea5a21ba84ea993eaa
Reviewed-on: https://chromium-review.googlesource.com/c/1279231
Reviewed-by: George Burgess <gbiv@chromium.org>
Cr-Commit-Position: refs/branch-heads/3578@{#34}
Cr-Branched-From: 4226ddf99103e493d7afb23a4c7902ee496108b6-refs/heads/master@{#599034}
[modify] https://crrev.com/c96c54b23eb0f17e9e5c3fb6e0a84591de656170/chrome/android/profiles/newest.txt

The following revision refers to this bug: 
https://chromium.googlesource.com/chromium/src.git/+/c96c54b23eb0f17e9e5c3fb6e0a84591de656170

Commit: c96c54b23eb0f17e9e5c3fb6e0a84591de656170
Author: gbiv@chromium.org
Commiter: gbiv@chromium.org
Date: 2018-10-16 05:42:47 +0000 UTC

Revert "Roll AFDO from 71.0.3576.0_rc-r1 to 71.0.3577.0_rc-r1"

This reverts commit 6fb1a0860e442c7f9707c2003d219cb03b3dff81.

Reason for revert:  crbug.com/894728  . The plan is to not revert this
on master, since we'll probably roll past the regression. As it
stands, this is the 'final' profile for m71, though, and a 9%
speedometer regression isn't acceptable when yesterday's profile
appears just fine. :)

Original change's description:
> Roll AFDO from 71.0.3576.0_rc-r1 to 71.0.3577.0_rc-r1
>
> This CL may cause a small binary size increase, roughly proportional
> to how long it's been since our last AFDO profile roll. For larger
> increases (around or exceeding 100KB), please file a bug against
> gbiv@chromium.org. Additional context:  https://crbug.com/805539 
>
> Please note that, despite rolling to chrome/android, this profile is
> used for both Linux and Android.
>
> The AutoRoll server is located here: https://autoroll.skia.org/r/afdo-chromium-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.
>
>
> TBR=gbiv@chromium.org
>
> Change-Id: Ic2facf850ac74d8087cc599ce0a9a7351b330ce3
> Reviewed-on: https://chromium-review.googlesource.com/c/1275868
> 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@{#598701}

Bug:  894728 
Change-Id: I03686edca0cda64e161a32ea5a21ba84ea993eaa
Reviewed-on: https://chromium-review.googlesource.com/c/1279231
Reviewed-by: George Burgess <gbiv@chromium.org>
Cr-Commit-Position: refs/branch-heads/3578@{#34}
Cr-Branched-From: 4226ddf99103e493d7afb23a4c7902ee496108b6-refs/heads/master@{#599034}