Closed Bug 491602 Opened 15 years ago Closed 13 years ago

Investigate JSGC performance on real workload

Categories

(Core :: JavaScript Engine, defect)

Product:
Core
Component:
JavaScript Engine
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

()

Status:
RESOLVED FIXED

People

(Reporter: bruce, Assigned: bruce)

Details

(Keywords: perf) 

User-Agent:       Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.0.10) Gecko/2009042315 Firefox/3.0.10
Build Identifier: 

sayrer suggests using gmail. Look at things such as the frequency of GCs, time spent in each phase, distribution of object sizes and fragmentation.

Reproducible: Always
Assignee: general → anygregor
This is supposed to be one of the first things my intern will work on. Bruce, could you post a couple of suggestions what exactly to measure? I already have a list of things for core JS I would like to know (mostly regarding values and types). We should add some GC-related metrics too.
(also, if you get around to this before Gregor arrives here next month, feel free to steal the bug :) )
Assignee: anygregor → bruce
Back to bruce. He will do the GC work first, and then gregor will extend that with GC-unrelated stats gathering.
Flags: wanted1.9.2?
Keywords: perf
Status: UNCONFIRMED → RESOLVED
Closed: 13 years ago
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.