Closed Bug 607202 Opened 14 years ago Closed 11 years ago

IonMonkey: slower than v8 on PeaceKeeper-derived testcase from bug 606650

Categories

(Core :: JavaScript Engine, defect)

Product:
Core
Component:
JavaScript Engine
Platform:
x86
macOS
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

()

Status:
RESOLVED WORKSFORME

People

(Reporter: bzbarsky, Unassigned)

References

(
URL
)

Details

(Keywords: perf, Whiteboard: [jsperf])

Direct urls: https://bugzilla.mozilla.org/attachment.cgi?id=485477 https://bugzilla.mozilla.org/attachment.cgi?id=485478 JM is 5-10x slower than TM here, and 2-3x slower than V8.
Whiteboard: [jsperf]
Keywords: perf
Firefox 21 1125ms 639ms Chrome 24 621ms 484ms
The second of these testcases is fine, now: we're on par with current Chrome Canary and faster than Safari. On the first one, however, we're still ~2x as slow as Chrome: Nightly 2013-07-18: 576ms 269ms Chrome Canary (30.0.1569.0): 288ms 271ms Safari 6.0.5: 911ms 296ms
URL: https://bugzilla.mozilla.org/attachme...
Summary: Figure out why JM is slower than TM and competition on the two testcases attached in bug 606650 → IonMonkey: slower than v8 on PeaceKeeper-derived testcase from bug 606650
Nightly 26.0a1 (2013-09-15) Time: 158ms Time: 208ms Chrome 29 Time: 652ms Time: 497ms This testcases are basically the same as those Jan de Mooij fixed from https://bugzilla.mozilla.org/show_bug.cgi?id=606648? Anyway, this bug is super fixed.
Nice! Thanks for testing.
Status: NEW → RESOLVED
Closed: 11 years ago
Depends on: 914132
Resolution: --- → WORKSFORME
You need to log in before you can comment on or make changes to this bug.