[Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.9.2a1pre) Gecko/20081209 Minefield/3.2a1pre] (home, optim default) (W2Ksp4) (http://hg.mozilla.org/mozilla-central/rev/85507cfcdda8) No leak. [Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.9.2a1pre) Gecko/20081210 Minefield/3.2a1pre] (home, optim default) (W2Ksp4) (http://hg.mozilla.org/mozilla-central/rev/29f1fa84f8fd) [Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.9.2a1pre) Gecko/20081211 Minefield/3.2a1pre] (home, optim default) (W2Ksp4) (http://hg.mozilla.org/mozilla-central/rev/0d2bceefc012) Leaks. This timeframe includes bug 467971 checkin...

[Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.9.2a1pre) Gecko/20081211 Minefield/3.2a1pre] (home, optim default) (W2Ksp4) (http://hg.mozilla.org/mozilla-central/rev/0d2bceefc012)

this leak is funny, if i comment out the full test we still leak!

Bug 452899 could be related

'wanted1.9.1=?' now that bug 467971 got landed on the 1.9.1 branch :-/

the leak is actually due to 2 causes: 1. wrong test finish (this is understandable) 2. params binding. This is not understandable, if i bind using BindInt64Parameter we don't leak, if i bind using params["param"] or params.param then we leak. Shawn any idea about that? Hwv i think that if that strange behaviour needs to be investigated it needs probably a new storage bug, this should be fine to take as it is.

i'm removing wanted1.9.1 flags requests because this will make 1.9.1 as soon as baked on trunk, so there's no need to nag drivers

Comment on attachment 353019 [details] [diff] [review] patch r=sdwilsh Please file a bug in storage with steps to reproduce. The current tests in storage that test this do not leak, so something weird is going on.

sorry i pasted wrong clipboard :\ http://hg.mozilla.org/mozilla-central/rev/6021935375b1

(In reply to comment #7) > Please file a bug in storage with steps to reproduce. The current tests in > storage that test this do not leak, so something weird is going on. i'm trying to build one, if successfull i'll file that bug.

filed bug 469972 on the strange leak behaviour

[Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.9.2a1pre) Gecko/20081219 Minefield/3.2a1pre] (home, optim default) (W2Ksp4) (http://hg.mozilla.org/mozilla-central/rev/9410ad10c6f7) V.Fixed

[Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.9.2a1pre) Gecko/20081221 SeaMonkey/2.0a3pre] (home, optim default) (W2Ksp4) (http://hg.mozilla.org/mozilla-central/rev/b839ff0630c6 +http://hg.mozilla.org/comm-central/rev/2a4c4c1f0feb + bug 469606 patch) Confirming Fixed with SeaMonkey too.

Was bug 473845 causing this leak ? Should we back the workaround part out ?

the code is correct so i think there's no need to backout it from 2 branches, leaks fix can be verified without the need to touch this. Plus this fixes another leak due to not calling finish_test();

Bug 473845 was likely a major cause of this.