http://tinderbox.mozilla.org/showlog.cgi?log=Firefox/1261438070.1261443241.5034.gz Linux mozilla-central opt test everythingelse on 2009/12/21 15:27:50 s: moz2-linux-slave25 REFTEST TEST-UNEXPECTED-FAIL | file:///builds/slave/mozilla-central-linux-opt-unittest-everythingelse/build/jsreftest/tests/jsreftest.html?test=js1_5/extensions/regress-371636.js | Numeric sort performance Expected value '(tint/tstr < 3)=true', Actual value '(tint/tstr < 3)=false' item 1 Pretty sure that's the first time, since it's easy enough to file that I wouldn't expect that I'd have skipped over it.
I think we should give up on all jsreftests that attempt to compute or compare orders of complexity. Nearly all of the intermittent jsreftests have been ones attempting to make some sort of perf evaluation. We're not going to eliminate enough machine noise for them to be completely reliable, and they're a constant nuisance to everyone -- if only to track down that an "orange" is "ignorable" -- in the meantime.
Status: NEW → RESOLVED
Last Resolved: 9 years ago
Resolution: --- → DUPLICATE
Duplicate of bug: 528284
No longer blocks: 438871
You need to log in before you can comment on or make changes to this bug.