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

Issue 732494 link

Starred by 1 user

Issue metadata

Status: Archived
Owner:
Last visit > 30 days ago
Closed: Oct 16
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 3
Type: Bug



Sign in to add a comment

Speedometer1.0 benchmark should be using the copy of Speedometer1.0 in third_party/webkit/PerformanceTest

Project Member Reported by nedngu...@google.com, Jun 12 2017

Issue description

third_party/webkit/PerformanceTest includes both Speedometer1.0 & Speedometer2.0 benchmarks. We want to switch speedometer benchmark to use the copy of Speedometer1.0 in third_party/webkit/PerformanceTest instead of WPR to ensure consistency.
 
Owner: ashleymarie@chromium.org
Status: Assigned (was: Untriaged)
Project Member

Comment 2 by sheriffbot@chromium.org, Jul 12 2017

Labels: Hotlist-Google
Since I'm making a change to the story here anyways, I'm going to rename the story from http://browserbench.org/Speedometer/ to Speedometer to be consistent with Speedometer2
This will involve a naming migration from */*/speedometer/*/http___browserbench.org_Speedometer_ to */*/speedometer/*/Speedometer once the change is in and collecting new metrics
Cc: u...@chromium.org
Project Member

Comment 5 by bugdroid1@chromium.org, Aug 16 2017

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/6437bba4ee92ecb4e526c3882eae6338d9d89ed6

commit 6437bba4ee92ecb4e526c3882eae6338d9d89ed6
Author: Ashley Enstad <ashleymarie@chromium.org>
Date: Wed Aug 16 16:32:45 2017

Using speedometer in third_party/webkit/PerformanceTest

This change switches speedometer benchmark to use the copy of
Speedometer1.0 in third_party/webkit/PerformanceTest instead of WPR

BUG= chromium:732494 

Change-Id: Idb299ae5f23178d4c1fa8a4569d40d7d88632e04
Reviewed-on: https://chromium-review.googlesource.com/616802
Commit-Queue: Ashley Enstad <ashleymarie@chromium.org>
Reviewed-by: Ned Nguyen <nednguyen@google.com>
Cr-Commit-Position: refs/heads/master@{#494816}
[modify] https://crrev.com/6437bba4ee92ecb4e526c3882eae6338d9d89ed6/tools/perf/benchmarks/speedometer.py
[delete] https://crrev.com/9143943d0b9f60fafe156b76bab7d3ea44f31722/tools/perf/page_sets/data/speedometer.json
[delete] https://crrev.com/9143943d0b9f60fafe156b76bab7d3ea44f31722/tools/perf/page_sets/data/speedometer_000.wprgo.sha1

Status: Fixed (was: Assigned)
Project Member

Comment 7 by bugdroid1@chromium.org, Sep 5 2017

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/667e31b9c1e0ecc0ef85aa6c4fe6cce89ff56d42

commit 667e31b9c1e0ecc0ef85aa6c4fe6cce89ff56d42
Author: Ned Nguyen <nednguyen@google.com>
Date: Tue Sep 05 19:54:18 2017

Revert "Using speedometer in third_party/webkit/PerformanceTest"

This reverts commit 6437bba4ee92ecb4e526c3882eae6338d9d89ed6.

Reason for revert: suspect this is actually running speedometer2 benchmark.

BUG:756990

Original change's description:
> Using speedometer in third_party/webkit/PerformanceTest
> 
> This change switches speedometer benchmark to use the copy of
> Speedometer1.0 in third_party/webkit/PerformanceTest instead of WPR
> 
> BUG= chromium:732494 
> 
> Change-Id: Idb299ae5f23178d4c1fa8a4569d40d7d88632e04
> Reviewed-on: https://chromium-review.googlesource.com/616802
> Commit-Queue: Ashley Enstad <ashleymarie@chromium.org>
> Reviewed-by: Ned Nguyen <nednguyen@google.com>
> Cr-Commit-Position: refs/heads/master@{#494816}

TBR=nednguyen@google.com,ashleymarie@chromium.org

# Not skipping CQ checks because original CL landed > 1 day ago.

Bug:  chromium:732494 
Change-Id: I7a991f6d99f857b15f94466d2655280902b346b2
Reviewed-on: https://chromium-review.googlesource.com/650547
Reviewed-by: Ned Nguyen <nednguyen@google.com>
Commit-Queue: Ned Nguyen <nednguyen@google.com>
Cr-Commit-Position: refs/heads/master@{#499718}
[modify] https://crrev.com/667e31b9c1e0ecc0ef85aa6c4fe6cce89ff56d42/tools/perf/benchmarks/speedometer.py
[add] https://crrev.com/667e31b9c1e0ecc0ef85aa6c4fe6cce89ff56d42/tools/perf/page_sets/data/speedometer.json
[add] https://crrev.com/667e31b9c1e0ecc0ef85aa6c4fe6cce89ff56d42/tools/perf/page_sets/data/speedometer_000.wprgo.sha1

Cc: bccheng@chromium.org llozano@chromium.org
Status: Unconfirmed (was: Fixed)
https://chromeperf.appspot.com/report?sid=e1cc606a7f00451a54cb68b3bd60b94956f5142c4d0797ff410e0993a01d1302

This was causing the score regression in R62, and R62 is still running Speedometer2. Should we merge the revert to R62 too?
Cc: hablich@chromium.org
I did a visual inspection and it turns out that both the fast and slow versions are running speedometer 1.0. However, the slower version runs 450 tests, while the old and faster version runs only 210 tests, hence the almost doubled total runtime, which is used as the key metrics for the benchmark.

Adding Michael as we happened to be discussing in a separate email thread about whether we should report "Runs/Minute" as the final result like what the UI does.
Project Member

Comment 10 by sheriffbot@chromium.org, Oct 16

Status: Archived (was: Unconfirmed)
Issue has not been modified or commented on in the last 365 days, please re-open or file a new bug if this is still an issue.

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

Sign in to add a comment