Saurabh Anand
f902b20205
Bug 795780 - Remove unnecessary rooting in proxy methods. r=nnethercote.
...
--HG--
extra : rebase_source : ff593ecc2082dea276f7fe09734394dce595a20c
2012-10-14 17:46:28 +05:30
Siddharth Agarwal
377ab9eb07
Back out bug 585011 (making cl.py a pymake native command) because of bug 794490.
2012-10-14 18:54:18 -07:00
Ehsan Akhgari
6b1b2c0d54
Bug 801344 - Make sure that JSRuntime::initSelfHosting never returns an uninitialized value; r=till
2012-10-14 11:12:03 -04:00
Peter Van der Beken
12f507617a
Fix for bug 799465 (Add complete support for non-nsISupports objects in new DOM bindings) - fix CC traversal and wrapper preservation. r=bz.
...
--HG--
rename : dom/tests/mochitest/chrome/file_bug800817.xul => dom/tests/mochitest/chrome/file_bug799299.xul
rename : dom/tests/mochitest/chrome/test_bug800817.xul => dom/tests/mochitest/chrome/test_bug799299.xul
extra : rebase_source : bed68ffa15fe5933994c33fbd94ad2b597b5b1b9
2012-09-26 20:12:15 +02:00
Aryeh Gregor
45b3e80460
Bug 799917 - Make document.characterSet always lowercase; r=bz,f=hsivonen,smontagu
2012-10-14 10:13:19 +02:00
Aryeh Gregor
2f24871362
Bug 799905 - .URL and .compatMode should be defined on Document, not HTMLDocument; r=bz
2012-10-11 14:48:31 +02:00
Jim Blandy
0a2a9c19f9
Bug 743311: Implement Debugger.prototype.findAllGlobals. r=jorendorff
2012-10-13 16:04:41 -07:00
Jim Blandy
2935cf9083
Bug 743311: Implement Debugger.prototype.onNewGlobalObject. r=jorendorff
2012-10-13 16:04:41 -07:00
Jim Blandy
db40a2832d
Bug 743311: Add option to shell 'evaluate' to catch termination, for tests. r=jorendorff
2012-10-13 16:04:41 -07:00
Jim Blandy
ffc6457899
Bug 796073: Identify debuggees only by Debugger.Object instances and CCWs that can be unwrapped securely. r=luke
...
This patch affects behavior that a lot of debugger tests rely on; I've
tried to update the tests without losing coverage of behavior that is still
supported:
- A prior patch in this series removes uses of addDebuggee to generate
Debugger.Object instances referring to random objects, using
makeDebuggeeValue instead.
- The test debug/Debugger-debuggees-07.js is deleted, because it's testing
for the very behavior we're removing. Other tests are trimmed to remove
usage that is no longer supported.
- The new test debug/Debugger-debuggees-17.js checks that we reject objects
that don't designate debuggees. The existing test
Debugger-debuggees-06.js checks that non-object values are properly
rejected.
- The new test debug/Debugger-debuggees-18.js checks that globals are
correctly identified regardless of how we designate them.
2012-10-13 16:04:41 -07:00
Jim Blandy
b755c3b751
Bug 796073: Use Debugger.Object.prototype.makeDebuggeeValue where appropriate, not Debugger.prototype.addDebuggee. r=luke
...
Many tests use Debugger.prototype.addDebuggee to produce Debugger.Object
instances referring to random objects. Now we have
Debugger.Object.prototype.makeDebuggeeValue to do that, which is more
careful about wrappers. This patch changes all the existing tests that
abuse addDebuggee in this way (well, they didn't have any choice when they
were written) to use makeDebuggeeValue instead, in preparation for
tightening up addDebuggee to only accept global objects.
2012-10-13 16:04:41 -07:00
Jim Blandy
9e7714919a
Bug 795119: Complain coherently when evalInGlobal{,withBindings} is passed a CCW to a global, not a global. r=luke
2012-10-13 16:04:16 -07:00
Jim Blandy
789197fda9
Bug 799272: Implement Debugger.Object.prototype.unwrap. r=luke
2012-10-13 16:04:16 -07:00
Jim Blandy
520b89592a
Bug 799272: Implement Debugger.Object.prototype.global. r=luke
2012-10-13 16:04:16 -07:00
Jacek Szpot
0ebb25a85a
Bug 800106: replace NS_ALWAYS_INLINE with MOZ_ALWAYS_INLINE; r=ehsan
2012-10-13 17:52:10 +02:00
Jan Beich
a9328aa582
Bug 801295 - mincore() call has incorrect type on tier3 (non-Linux). r=jlebar
...
--HG--
extra : rebase_source : 3201ac0b491a8cb40f916276e5b34a14d3122dc7
2012-10-13 09:34:00 -04:00
Till Schneidereit
70ab7b00c6
Bug 784400 - Move all intrinsics initialization from the globalObject->intrinsicsHolder object to the runtime's selfHostingGlobal. r=luke
2012-10-11 14:43:21 +02:00
Till Schneidereit
8cf23858fa
Bug 784400 - Enable cloning of object literals within functions. r=luke
2012-10-11 11:53:02 +02:00
Till Schneidereit
9c3f9146e0
Bug 784400 - Give self-hosted code access to standard builtins. r=luke
2012-08-28 04:58:00 +02:00
Till Schneidereit
8af01ce667
Bug 784620 - Support reading self-hosted JS from a file set via environment variable. r=terrence
2012-10-11 11:19:43 +02:00
Benjamin Peterson
e4dfaf6857
Bug 779118 - Try to be more robust about waiting for compression to finish. r=luke
2012-10-13 09:24:13 -04:00
Benjamin Peterson
85e07532ff
Bug 791465 - Fix processDirectives' ASI emulation. code=André Bargull r=benjamin
2012-10-13 09:23:53 -04:00
Benjamin Peterson
3d2f961aae
No bug - Readd executable bit. rs=terrence
...
--HG--
extra : rebase_source : c3a77fb2fd8d6b1bae65efdd2bbc481f139b91e9
2012-10-12 14:44:20 -04:00
Andrew Quartey
464807b57d
Bug 792581 - part 16: Replace LL_L2UI macro with uint32_t cast. r=ehsan
2012-10-12 13:29:11 -04:00
Justin Lebar
ddc9c65e5f
Bug 798510 - Part 2: Fix bug in JSRuntime::sizeOfExplicitNonHeap() where, if execAlloc_ is null, we don't measure the stack space. rs=njn
2012-10-12 10:26:06 -04:00
Justin Lebar
9ed3356569
Bug 798510 - Part 1: Use mincore to correctly calculate js::StackSpace::sizeOfCommitted() on *nix. r=luke,njn
...
Previously, we assumed all memory was committed, thus vastly overstating the memory used by our stack.
2012-10-12 10:26:06 -04:00
Ed Morley
8fc88262ba
Backout 3044539fec87 (bug 799465),2dbcd6d16b43 (bug 798264), f7019f73a5f0 (bug 711628), a484a3a904da, 7154061ddc00 & 7e7fc42021c1 (bug 799465) for burning
2012-10-12 14:45:38 +01:00
Peter Van der Beken
52e6c5623d
Fix for bug 799465 (Add complete support for non-nsISupports objects in new DOM bindings) - fix CC traversal and wrapper preservation. r=bz.
...
--HG--
extra : rebase_source : 42b0e6db4b418901f253d66e7323ba5a7cd83eea
2012-09-26 20:12:15 +02:00
Nicholas Cameron
404ba9b65d
Bug 734668; reviewer's changes. r=Ms2ger,roc
...
--HG--
rename : content/canvas/src/nsCanvasRenderingContext2DAzure.cpp => content/canvas/src/CanvasRenderingContext2D.cpp
extra : rebase_source : 827a45dd8cc2bdc77e850e8b1d1e23e9f2f72f7d
2012-10-12 22:42:53 +13:00
Nicholas Cameron
80df51e492
Bug 734668; remove canvas quickstubs. r=Ms2ger
...
--HG--
extra : rebase_source : ceefaf29933c58a0605d60a599fddf31b0c1b229
2012-10-12 22:42:53 +13:00
Nicholas Nethercote
0a7dd0f914
Bug 800481 - Remove unnecessary second arg from HashTable::hash{1,2}. r=luke.
2012-10-11 16:36:29 -07:00
Ehsan Akhgari
1b602e3962
Backout bug 798510 because of xpcshell orange
2012-10-11 20:17:33 -04:00
Trevor Saunders
3d3592cd63
bug 798595 - remove useless prmem.h includes r=ehsan
2012-10-04 02:14:06 -04:00
Justin Lebar
2b6b69da90
Bug 798510 - Part 2: Fix bug in JSRuntime::sizeOfExplicitNonHeap() where, if execAlloc_ is null, we don't measure the stack space. rs=njn
2012-10-11 18:55:53 -04:00
Justin Lebar
d90a8e1023
Bug 798510 - Part 1: Use mincore to correctly calculate js::StackSpace::sizeOfCommitted() on *nix. r=luke,njn
...
Previously, we assumed all memory was committed, thus vastly overstating the memory used by our stack.
2012-10-11 18:55:53 -04:00
David Anderson
3266eed6fc
Don't explicitly unbox before GetPropertyCache (bug 799907, r=djvj).
2012-10-11 13:47:50 -07:00
Bobby Holley
c30bf840f9
Bug 799348 - When creating a new content window, create the surrounding chrome docshell as system. r=bz
2012-10-11 19:44:20 +02:00
Nicolas B. Pierron
be98caba91
Bug 797977 - Rename StackIter::fp() to StackIter::interpFrame(). r=luke
2012-10-10 20:50:07 -07:00
Sean Stangl
5304d41f92
Bug 798946 - Confer with TI about the return type of Math.random(). r=dvander
2012-10-10 17:07:53 -07:00
Rafael Ávila de Espíndola
6bf8982169
Bug 799807 - update reference to llvm bug. r=dvander.
2012-10-10 10:31:02 -04:00
Bobby Holley
0403fed8cc
Bug 797204 - Remove API. r=mrbkap
...
\o/
2012-10-10 11:01:26 +02:00
Bobby Holley
220736fd0b
Bug 797204 - Consolidate XBL filename hack logic into nsContentUtils. r=mrbkap
...
These two functions do the exact same thing, as far as I can tell.
2012-10-10 11:01:26 +02:00
Bobby Holley
055fcd4448
Bug 789224 - Remove chrome check in xpc_EvalInSandbox. r=mrbkap
...
This function is also called to evaluate javascript:// URIs, at which point the chrome caller assumption isn't actually valid. Let's just do the modern thing here, and fix up the caller that breaks.
2012-10-10 11:01:26 +02:00
foudfou
3a23c1b328
Bug 793433 - Fix crash when nsAppShellService::RegisterTopLevelWindow called with invalid arguments. r=bholley
...
--HG--
extra : rebase_source : 2f3be0e474c727ea5a58177809bf3af1f34477ec
2012-10-09 11:56:58 +02:00
Marty Rosenberg
0755fd0e7b
Change the method we use to update ranges on phis so narrowing works properly (bug 799282, r=jandem)
2012-10-10 01:25:03 -04:00
Marty Rosenberg
7240ad13d1
The fix bitand to not compute bogus ranges (bug 799793, r=dvander)
2012-10-10 01:25:03 -04:00
Nicholas Nethercote
a67ecb43f6
Bug 793967 - Refactor Proxy::getElementIfPresent(). r=ejpbruel.
...
--HG--
extra : rebase_source : 5832487556791f6fc36b840d6306875c2c3e386a
2012-10-07 22:39:54 -07:00
Terrence Cole
acbbc1562e
Bug 799777 - Unravel the threaded interpreter; r=luke
...
With the JITs, this is no longer buying us performance proportional to its
complexity.
2012-10-09 17:26:13 -07:00
Terrence Cole
e7ed15573e
Bug 798624 - Specialize low-level character access to JSStableString; r=luke, rs=Waldo
...
Implements JSStableString::chars() and pushes StableCharPtr into the interface
at several key choke-points. This will solidly enforce the requirement that we
have uninlined jschar*s in places that we may GC with a jschar* on the stack.
--HG--
extra : rebase_source : 122785b474af371ed22d43a6bfcb600c05440405
2012-10-08 15:04:36 -07:00
Ehsan Akhgari
7d0ac4e6bc
Bug 799432 - Backout changeset 30ab4e1d4874 and changeset dbcc29e9fcf6 from bug 777292 since they're no longer necessary with bug 795238
2012-10-09 16:01:24 -04:00