http://tinderbox.mozilla.org/showlog.cgi?log=Mozilla-Inbound/1310505722.1310508257.28018.gz Rev3 WINNT 6.1 mozilla-inbound debug test mochitests-3/5 on 2011/07/12 14:22:02 s: talos-r3-w7-046 { TEST-UNEXPECTED-FAIL | automationutils.processLeakLog() | leaked 23384 bytes during test execution TEST-UNEXPECTED-FAIL | automationutils.processLeakLog() | leaked 13 instances of AsyncStatement with size 48 bytes each (624 bytes total) TEST-UNEXPECTED-FAIL | automationutils.processLeakLog() | leaked 1 instance of AsyncStatementParams with size 20 bytes TEST-UNEXPECTED-FAIL | automationutils.processLeakLog() | leaked 1 instance of BackstagePass with size 24 bytes TEST-UNEXPECTED-FAIL | automationutils.processLeakLog() | leaked 1 instance of CalculateFrecencyFunction with size 12 bytes TEST-UNEXPECTED-FAIL | automationutils.processLeakLog() | leaked 1 instance of Connection with size 112 bytes } "Analyze the leak" link says "No DOMWINDOWs leaked!"

I think this is a duplicate of bug 593711.

I'm not so sure -- in that bug, a MatchAutoCompleteFunction is leaked, whereas here a CalculateFrecencyFunction is leaked. Is that unimportant?

(previous comment applies to the leak in bug 593711 comment 0. I see that the log most recently posted there had a CalculateFrecencyFunction leaked, so maybe it's broadened since it was filed.)

All of the more recent builds starred as bug 593711 leak a CalculateFrecencyFunction. It looks like the exact contents of that leak have changed since bug 593711 comment 0.