970001 - Valgrind detects leak - 1 bytes are definitely lost (direct) involving sourcePolicy

Status: RESOLVED FIXED

(Core :: JavaScript Engine, defect)

Description

[Gary Kwong [:gkw] [:nth10sd] (NOT official MoCo now)]

Attachment: stack

evaluate("", ({
    sourcePolicy: ''
}))

shows "1 bytes in 1 blocks are definitely lost" on m-c changeset cafe909f7e07 with --ion-parallel-compile=off.

valgrind --vex-iropt-register-updates=allregs-at-mem-access --leak-check=full --smc-check=all-non-file ./js-opt-64-dm-vg-ts-er-linux-cafe909f7e07 --ion-parallel-compile=off testcase.js

Not sure who should get involved here, setting needinfo from jorendorff as a start.

My configure flags are:

AR=ar sh ./configure --enable-optimize=-O1 --disable-debug --enable-profiling --enable-gczeal --enable-debug-symbols --enable-methodjit --enable-type-inference --disable-tests --enable-more-deterministic --enable-exact-rooting --enable-valgrind --with-ccache --enable-threadsafe <other NSPR options>

Comment 1

[Jason Orendorff [:jorendorff]]

Yeah, this is trivial, patch coming.

Comment 2

[Nicholas Nethercote [inactive]]

I don't think I've ever seen a 1 byte leak before. LOL.

Comment 3

[Jason Orendorff [:jorendorff]]

Attachment: bug-970001-leak-v1.patch

I found a few other leaky-in-error-cases uses of JS_Encode*, too, and switched them to JSAutoByteString.

Comment 4

[Till Schneidereit [:till]]

Comment on attachment 8374124 [details] [diff] [review]
bug-970001-leak-v1.patch

Review of attachment 8374124 [details] [diff] [review]:
-----------------------------------------------------------------

nice

Comment 5

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/6deadaf3af44

Comment 6

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/mozilla-central/rev/6deadaf3af44