Bug 132508

Summary:

REGRESSION(r167889): ~1.5% Regression in DYEBench

Product:

WebKit

Reporter:

Ryosuke Niwa <rniwa>

Component:

JavaScriptCore

Assignee:

Nobody <webkit-unassigned>

Status:

NEW ---

Severity:

Normal

CC:

barraclough, ggaren, mhahnenberg

Priority:

Keywords:

Regression

Version:

528+ (Nightly build)

Hardware:

Unspecified

OS:

Unspecified

Attachments:

Description	Flags
A/B test result	none
More testing	none

Description Ryosuke Niwa 2014-05-02 18:15:26 PDT

https://perf.webkit.org/#mode=charts&chartList=%5B%5B%22mac-mountainlion%22%2C%22DoYouEvenBench%2FFull%3ATime%3ATotal%22%5D%2C%5B%22mac-mavericks%22%2C%22DoYouEvenBench%2FFull%3ATime%3ATotal%22%5D%5D&zoom=%5B1398632363755.7725%2C1398728150893.5771%5D

Comment 1 Ryosuke Niwa 2014-05-02 18:15:57 PDT

Created attachment 230727 [details]
A/B test result

Comment 2 Ryosuke Niwa 2014-05-02 18:18:43 PDT

Looks like EmberJS-TodoMVC/CompletingAllItems regressed the most.

Comment 3 Ryosuke Niwa 2014-05-02 20:07:18 PDT

Created attachment 230733 [details]
More testing

s_maxForgivenDeletes=1 is ~1% regression but s_maxForgivenDeletes=3 is still 1.5-2% regression :(

So we might need to do something a little more clever than just tweaking the numbers here.