All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=729760

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgxqGc6AoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgxvSz5ggM


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

android-nexus6

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8977590803099886448

=== BISECT JOB RESULTS ===
Bisect failed for unknown reasons

Please contact the team (see below) and report the error.


Bisect Details
  Configuration: android_nexus6_perf_bisect
  Benchmark    : blink_perf.layout
  Metric       : fixed-grid-lots-of-stretched-data/fixed-grid-lots-of-stretched-data


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 blink_perf.layout

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8977590803099886448

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5880507436367872


| 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!

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8977586154814253920

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8977585667542236640

=== BISECT JOB RESULTS ===
Bisect failed for unknown reasons

Please contact the team (see below) and report the error.


Bisect Details
  Configuration: android_nexus6_perf_bisect
  Benchmark    : blink_perf.layout
  Metric       : fixed-grid-lots-of-stretched-data/fixed-grid-lots-of-stretched-data


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 blink_perf.layout

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8977585667542236640

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5880507436367872


| 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!

=== BISECT JOB RESULTS ===
Bisect was unable to run to completion

Error: INFRA_FAILURE

The bisect was able to narrow the range, you can try running with:
  good_revision: 958dbb23d9fb6ef967f76d9cda0153a6cb05f8cd
  bad_revision : d560ed865c053d87594aca8a72637593117d5d1b

If failures persist contact the team (see below) and report the error.


Bisect Details
  Configuration: android_nexus6_perf_bisect
  Benchmark    : blink_perf.layout
  Metric       : fixed-grid-lots-of-stretched-data/fixed-grid-lots-of-stretched-data

Revision             Result                  N
chromium@475340      192.469 +- 3.5172       6      good
chromium@475371      191.386 +- 3.70863      6      good
chromium@475386      192.191 +- 3.67647      6      good
chromium@475394      186.731 +- 6.60673      6      good
chromium@475401      166.491 +- 2.6029       6      bad
chromium@475462      168.535 +- 4.78267      6      bad
chromium@475584      165.558 +- 4.05435      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 blink_perf.layout

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8977586154814253920

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=4749331430637568


| 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!

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8977529051202602912

=== BISECT JOB RESULTS ===
Bisect was unable to run to completion

Error: INFRA_FAILURE

The bisect was able to narrow the range, you can try running with:
  good_revision: 958dbb23d9fb6ef967f76d9cda0153a6cb05f8cd
  bad_revision : d560ed865c053d87594aca8a72637593117d5d1b

If failures persist contact the team (see below) and report the error.


Bisect Details
  Configuration: android_nexus6_perf_bisect
  Benchmark    : blink_perf.layout
  Metric       : fixed-grid-lots-of-stretched-data/fixed-grid-lots-of-stretched-data

Revision             Result                  N
chromium@475340      192.469 +- 3.5172       6      good
chromium@475371      191.386 +- 3.70863      6      good
chromium@475386      192.191 +- 3.67647      6      good
chromium@475394      186.731 +- 6.60673      6      good
chromium@475401      166.491 +- 2.6029       6      bad
chromium@475462      168.535 +- 4.78267      6      bad
chromium@475584      165.558 +- 4.05435      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 blink_perf.layout

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8977529051202602912

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=4749331430637568


| 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!

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8971003983102629408

=== Auto-CCing suspected CL author jfernandez@igalia.com ===

Hi jfernandez@igalia.com, 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 : jfernandez
  Commit : 5389373c6dec30d783eb46b4c8190720f411a8a7
  Date   : Tue May 30 00:56:52 2017
  Subject: [css-align] Don't resolve 'auto' values for computed style.

Bisect Details
  Configuration: win_x64_perf_bisect
  Benchmark    : blink_perf.layout
  Metric       : fixed-grid-lots-of-data/fixed-grid-lots-of-data
  Change       : 11.63% | 1122.02480912 -> 991.522438608

Revision             Result                  N
chromium@475368      1122.02 +- 11.2738      6      good
chromium@475385      1121.77 +- 11.7547      6      good
chromium@475393      1101.61 +- 10.2124      6      good
chromium@475397      1125.98 +- 5.52094      6      good
chromium@475399      1115.92 +- 6.77463      6      good
chromium@475400      1030.21 +- 6.79047      6      bad       <--
chromium@475401      991.522 +- 18.1016      6      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests blink_perf.layout

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8971003983102629408


For feedback, file a bug with component Speed>Bisection

I'm working on this now.

I have a lot of doubts about this issue. I sent a perf try job in https://crrev.com/c/621886 which should neutralize the changes made in the culprit found by the bisect.

The suspicious change indeed add some overhead to the layout logic by moving the auto value resolution to the layout phase, instead of doing it at style resolution. However, the speculative fix changes the affected performance tests so that they have non-auto Self Alignment values. Hence, we should have recovered the performance level before the regression. 

However, the perf try job show almost no improvement. The perf graphs also show that the mentioned tests have recovered the values achieved before the regression:

https://chromeperf.appspot.com/report?sid=aa70da91458061cc4f4f5f0f5c494231160add664bf35b2a976cff6e5c73f02c&start_rev=472695&end_rev=485158

I have sent a speculative fix to the perf try jobs, but for some reason I can't get the results:

https://crrev.com/c/649147

I've just realized that this is probably duplicated from  issue #729236 

I confirmed locally that reverting the change in r478722 which fixed  issue #729236  causes a regression around 7% - 11% . Hence, I think we can close this bug now.