Commit Graph

29050 Commits

Author SHA1 Message Date
Terrence Cole
77df39999e Bug 916993 - Handlify the public JSString APIs; r=bz,jonco
--HG--
extra : rebase_source : 35c5d7832d794ce37be622db3140309d2605c212
2013-09-16 18:33:40 -07:00
Jeff Walden
ddb31f55b2 Bug 912701 - Expand some Intl self-hosting assertion messages to aid in debugging-by-eye. r=trivial
--HG--
extra : rebase_source : 736ee90d527ea2b44cc3929e0ba7a40947842a0b
2013-09-16 14:26:56 -07:00
Jeff Walden
04d13243c9 Bug 916394 - Remove the |it| object from the shell: everything it tests is better-tested with a JSAPI test these days, and it's largely redundant. r=jorendorff
--HG--
extra : rebase_source : f97d63fdf89a0eacf0f2e091fc3f64aa590371eb
2013-09-12 15:45:43 -07:00
Bobby Holley
8f78b2faeb Bug 905926 - Remove workaround in xpcshell. r=billm 2013-09-17 09:46:33 -07:00
Bobby Holley
c8872347c9 Bug 905926 - Be more explicit about GCing twice in the nsXPConnect destructor. r=billm
In the current setup, we'll end up GCing once when destroying the SafeJSContext,
and then once again immediately after in the explicit GC we do before destroying
the XPCJSRuntime. It would be nice to avoid the first GC entirely, but we
currently need both to avoid leaking.

All in all, these patches cause us to GC three times during shutdown, rather
than twice as we did before, because the second GC was rolled together with
the runtime destruction GC when we destroyed the last JSContext. There are a
number of ways to eliminate these, at least in opt builds, but mccr8 thinks
it probably doesn't matter, since there shouldn't be much left in the heap after
the second GC.

We can probably get away with eliminating rambo GC entirely at some point. But
this might become irrelevant for the browser if we end up doing bug 662444.

It would also be interesting to see what, if anything, the rambo GC actually
collects. We might even be able to get away with asserting that all the zones
are gone and removing the GC entirely.

We also take the opportunity to kill mOwnSafeJSContext, which no longer holds
any meaning.
2013-09-17 09:46:32 -07:00
Bobby Holley
c0d1afa8de Bug 905926 - Stop using hasContexts(). r=billm,jonco
This patch is a joint effort between billm and myself. I'm marking myself as
the author because I worked on it more recently, which means that Bill should
probably review it again in addition to jonco, and self-review is frowned upon.

There were a few places where we relied on !rt->hasContexts() to mean
"this is the last GC". That was never really valid, and it especially isn't
valid with the previous patch. This patch replaces the check everywhere it's
used in the GC.
2013-09-17 09:46:32 -07:00
Bill McCloskey
a3fd52315b Bug 905926 - Move rambo GC to runtime destruction. r=jonco 2013-09-17 09:46:32 -07:00
Bobby Holley
42c719e380 Bug 905926 - Remove watchpoints on all compartments during shutdown, not just the cx compartment. r=billm 2013-09-17 09:46:32 -07:00
Bobby Holley
ac7034ec92 Bug 905926 - Invoke JS_DestroyRuntime before we totally tear down the XPCJSRuntime. r=billm 2013-09-17 09:46:31 -07:00
Bobby Holley
672d08bf20 Bug 905926 - Stop creating an ephemeral JSContext during XPConnect shutdown. r=billm
This interacts badly with our attempts to shut things down nicely, and happens
to be entirely unused. the primary impact of it ends up being that the cx
destruction triggers a GC, which we depend on. But we can do that manually.
2013-09-17 09:46:31 -07:00
Bobby Holley
d09c3fcc99 Bug 905926 - Explicitly track whether a context has been created. r=billm
The current mechanism is pretty sloppy, and falls over if the context count ever
drops to zero and then increases again. Let's do this right.
2013-09-17 09:46:31 -07:00
Bill McCloskey
c6adde8957 Bug 905926 - Fix small bugs. r=jonco
This patch fixes two small issues.

(1) It tightens an assertion in finalizeNow. Stuff finalized in finalizeNow
    should never be finalized off the main thread, so we shouldn't have to
    worry about BFS_JUST_FINISHED.

(2) If the only compartment that's left to GC is the atoms compartment, then
    we're not collecting it. I just moved the |any| check down below the
    code that decides whether to schedule the atoms compartment for GC.
2013-09-17 09:46:31 -07:00
Bobby Holley
20e00df790 Bug 916983 - Stop allowing writes for named access to cross-origin properties. r=bz 2013-09-17 09:46:30 -07:00
Bobby Holley
8ee4b2529c Bug 860494 - Tests. r=bz 2013-09-17 09:46:30 -07:00
Ehsan Akhgari
a1158bad50 Bug 917267 - Remove some unneeded code in xpconnect; r=bholley 2013-09-17 12:38:56 -04:00
Brian Hackett
74105e653d Bug 916914 - Remove ScriptAnalysis::ionInlineable(), r=jandem. 2013-09-17 08:39:01 -06:00
Ed Morley
c39b834022 Merge mozilla-central and inbound 2013-09-17 15:30:21 +01:00
Alex Keybl
63a42a7016 Merging in version bump NO BUG 2013-09-17 09:38:37 -04:00
Luke Wagner
95d9db5695 Bug 909709 - Reuse MOZ_LINKER IsSignalHandlingBroken to disable asm.js signal handlers (r=glandium)
--HG--
extra : rebase_source : 8f9daa6efb8ccd3344fa789a98a3fff93bbdd68b
2013-09-16 12:58:38 -05:00
Luke Wagner
b72efdcc4f Bug 909709 - Fix # indentation in AsmJSSignalHandlers.cpp (r=glandium)
--HG--
extra : rebase_source : b610d4904e7a052216189796ea2ced52097c6a3b
2013-09-16 12:58:37 -05:00
Tooru Fujisawa
3cd097471b Bug 762363 - ES6 spread-call syntax: f(...args). r=jorendorff. 2013-09-13 18:32:21 +09:00
Hannes Verschore
816764a48e Bug 916752 - IonMonkey: TypeBarrier will be typed when always bails, r=jandem 2013-09-17 15:52:28 +02:00
Dan Gohman
fd13c82c8b Bug 916167 - IonMonkey: Use movaps/movapd instead of movss/movsd for register-to-register moves to avoid partial register dependencies. Also, movaps is smaller :-). r=jandem 2013-09-16 07:53:28 -07:00
Jon Coppeard
5f0a69190d Bug 916795 - Fix error building browser with GGC enabled r=njn 2013-09-17 12:26:10 +01:00
Hannes Verschore
715a3845ac Bug 916846 - Baseline: Don't create dense getelem stub for negative numbers, r=jandem 2013-09-17 11:55:19 +02:00
Eddy Bruel
e3d5361231 Bug 763993 - Clean up source map API; r=jimb 2013-09-16 18:35:21 +02:00
Hannes Verschore
5c5501a9d8 Bug 911822 - TraceLogging: Update start, stop logging positions of the interpreter, r=till 2013-09-17 10:29:00 +02:00
Hannes Verschore
1aadcb4372 Bug 915244 - Tracelogging: Enable logging the gc background thread, r=till 2013-09-17 10:27:58 +02:00
Benjamin Bouvier
078dfffbea Bug 916662: Fix failed assertions and correctness errors on ARM; r=shu
--HG--
extra : rebase_source : 6c8aa8715282dc8773764faf59d75191451f88aa
2013-09-16 19:48:30 -07:00
Christian Holler
b89849ff48 Bug 877437 - Handle OOM when ion compiling. r=jdemooij
--HG--
extra : rebase_source : 48e8bb4c68f7d60614ec3ccf0b764bf08263944d
2013-09-17 09:21:28 +02:00
Jon Coppeard
02dde35b94 Bug 900756 - Ionmonkey (ARM): add float32 support r=mjrosenb 2013-09-15 00:48:10 +10:00
Nicholas Nethercote
65daa17f13 Bug 916531 - Backed out changeset 47e05e8df03b (bug 915482, part 4) for causing hangs on Windows. r=me.
--HG--
extra : rebase_source : 1c3d96a670c031137d984584449704f32f728c3f
2013-09-15 16:20:17 -07:00
David Zbarsky
ab12956b20 Bug 915813: Remove nsIDOMUserProximityEvent r=smaug 2013-09-14 14:59:51 -04:00
David Zbarsky
6a388bb846 Bug 915813: Convert DeviceLightEvent to webidl-only r=smaug 2013-09-14 14:59:50 -04:00
Makoto Kato
15bacec282 Bug 915973 - Need -FS option for parallel build on VS2013. r=ted 2013-09-14 19:17:27 +09:00
Shu-yu Guo
744281e9cf Bug 915495 - Support float32 optimization in typed array SetElem ICs. (r=bbouvier) 2013-09-14 01:45:42 -07:00
Wes Kocher
efff09d7ac Backed out changeset 422937706171 (bug 915495) 2013-09-13 23:10:43 -07:00
Shu-yu Guo
e470ed545c Bug 915495 - Support float32 optimization in typed array SetElem ICs. (r=bbouvier) 2013-09-13 22:45:42 -07:00
Jeff Walden
0c1c7845fd Bug 915805 - Refactor TryConvertFreeName to be more readable, and to not decide whether it can convert in one single huge |if|. r=jorendorff 2013-09-12 15:18:14 -07:00
Jan de Mooij
9ec09b844b Bug 916039 - Don't wrap |this| in strict mode eval. r=bhackett
--HG--
extra : rebase_source : 678d783f4cfe32a9b00db1a0bbb34940938bae0b
2013-09-13 17:51:50 +02:00
Dan Gohman
5b3bff6c46 Bug 915852 - IonMonkey: Add function positions to IONFLAGS=codegen output. r=jandem 2013-09-13 08:52:06 -07:00
Jason Orendorff
e2f528ffc2 Bug 913885 - Remove non-deterministic (and obsolete) warning when calling eval with multiple arguments, to reduce false alarms from differential testing. r=Waldo. 2013-09-13 09:52:53 -05:00
Jason Orendorff
38e866f22a Bug 892671 - Specify the tolerance in each new ES6 Math test. r=jandem. 2013-09-13 09:52:11 -05:00
Jason Orendorff
9323342ae1 Bug 897403 - "Assertion failure: !((attrs ^ shape->attrs) & 0x40) || !(attrs & 0x40)" with bound function proxy. r=Waldo. 2013-09-13 09:52:05 -05:00
Douglas Crosher
486538c27a Bug 880204 - Asm.js: support constant global variables and make their loads hoistable. r=luke 2013-09-13 17:35:38 +10:00
Hannes Verschore
14c5ab041d Bug 915171 - IonMonkey: Remove bogus assertion, r=jandem 2013-09-13 13:55:00 +02:00
Nathan Froyd
367732eac4 Bug 915848 - make check_spidermonkey_style.py work correctly with git; r=njn 2013-09-12 16:12:21 -04:00
Jan de Mooij
67b63c129d Bug 913749 - Mark fallible unbox instructions as guards. r=h4writer
--HG--
extra : rebase_source : f064535983782dcd52791f977f16a21d83603381
2013-09-13 10:09:11 +02:00
Nicholas Nethercote
5271b0e451 Bug 915482 (part 4) - Minimize vm/Shape-inl.h includes. r=terrence.
--HG--
extra : rebase_source : 973eb6668f38be4261ab3eef0b3645473bd3ebc2
2013-09-12 17:44:47 -07:00
Ryan VanderMeulen
183be64998 Merge m-c to inbound. 2013-09-12 22:30:22 -04:00