Benjamin Bouvier
26252ed8ba
Bug 1113338: Generalize AsmJS{Load,Store}Heap to handle partial loads; r=luke
...
--HG--
extra : rebase_source : 14f1292bcaef32933e4a13136165973657cabeec
2015-01-28 13:30:32 +01:00
Benjamin Bouvier
de9ea0eeeb
Bug 1113338: Rename viewType into accessType everywhere; r=luke
...
--HG--
extra : rebase_source : 90af0e88aea7d95493ec1a00571f91bc37ffa9af
2015-01-26 19:57:24 +01:00
Benjamin Bouvier
f6b43dea9e
Bug 1113338: Throw on out-of-bounds SIMD heap accesses in asm.js; r=luke
...
--HG--
extra : rebase_source : a645d89059764f5caa2fed6a156afa7e31b5589c
2015-01-28 09:31:07 +01:00
Benjamin Bouvier
ca2f09c9a8
Bug 1113338: Move asm.js SIMD.load/store tests into their own file; r=luke
...
--HG--
extra : rebase_source : e80632b2412514c4c5d0bc1d7365f782602db6a7
2015-01-26 19:12:24 +01:00
Nicolas B. Pierron
120a757f5f
Bug 1112163 - Make Ion frame size a multiple of the JitStackAlignment. r=bbouvier
2015-02-05 21:41:50 +01:00
Nicolas B. Pierron
164a9c3547
Bug 1126375 - Fix assertJitStackInvariant to account for details of x86 rectifier frames. r=bbouvier
2015-02-05 20:12:19 +01:00
Bobby Holley
2bbc2b39d3
Bug 1126911 - Tests. r=gabor
2015-02-05 11:07:40 -08:00
Bobby Holley
e8247b28ad
Bug 1126911 - Special-case all chrome objects in wrapper selection. r=gabor
2015-02-05 11:07:40 -08:00
Bobby Holley
f10144e7c3
Bug 1126911 - Fix rest of test suite. r=gabor
2015-02-05 11:07:40 -08:00
Victor Carlquist
f845777b8a
Bug 1127929 - IonMonkey: Inline SIMD.int32x4.and calls. r=nbp
2015-02-05 15:13:14 +01:00
Benjamin Bouvier
5e878ba66f
Bug 1129416: Make FuncConvertBits and SIMD ctors GC sane in the interpreter; r=jonco
...
--HG--
extra : rebase_source : ec70d33848e00824152de4d45c0703f237571153
2015-02-04 19:28:50 +01:00
Douglas Crosher
5f61fd7dea
Bug 1129348 - JIT: handle immediate addresses in the disassembler. r=sunfish
2015-02-04 20:45:53 +11:00
Jan de Mooij
1e8c1994e1
Bug 1125784 - Remove JS_ConvertArguments and JS_ConvertValue. r=jorendorff
2015-02-05 12:01:11 +01:00
Jan de Mooij
25559b44d6
Bug 1128535 followup - Fix Dromaeo DOM Query regression. r=efaust
2015-02-05 12:01:07 +01:00
Jon Coppeard
9339aabbc1
Bug 1129424 - Remove obsolete check that shape's parent hasn't been reallocated r=terrence
2015-02-05 10:45:03 +00:00
Jon Coppeard
c8418b9e2e
Bug 1129233 - Take care to only use AutoDisable guards in an appropriate scope r=terrence
2015-02-05 10:44:32 +00:00
Shu-yu Guo
8854feaf73
Bug 1128061 - Check unhandlable OOM when finishing off-thread Ion compiles. (r=terrence)
2015-02-04 16:07:04 -08:00
Nicholas Nethercote
b5913e0b3d
Bug 1050035 (part 4) - Make PL_DHashTableAdd() infallible by default, and add a fallible alternative. r=froydnj.
...
I kept all the existing PL_DHashTableAdd() calls fallible, in order to be
conservative, except for the ones in nsAtomTable.cpp which already were
followed immediately by an abort on failure.
--HG--
extra : rebase_source : eeba14d732077ef2e412f4caca852de6b6b85f55
2015-02-02 14:48:58 -08:00
Nicholas Nethercote
f325953043
Bug 1050035 (part 3) - Remove PL_NewDHashTable() and PL_DHashTableDestroy(). r=froydnj.
...
Because they are now just equivalent to |new PLDHashTable()| +
PL_DHashTableInit() and PL_DHashTableFinish(t) + |delete t|, respectively.
They're only used in a handful of places and obscure things more than they
clarify -- I only recently worked out exactly how they different from Init()
and Finish().
--HG--
extra : rebase_source : c958491447523becff3e01de45a5d2d227d1ecd3
2015-02-01 20:36:52 -08:00
Nicholas Nethercote
9a36fdbde4
Bug 1050035 (part 2) - Remove the fallible version of PL_DHashTableInit(). r=froydnj,mrbkap.
...
Because it's no longer needed now that entry storage isn't allocated there.
(The other possible causes of failures are much less interesting and simply
crashing is a reasonable thing to do for them.)
This also makes PL_DNewHashTable() infallible.
--HG--
extra : rebase_source : 848cc9bbdfe434525857183b8370d309f3acbf49
2015-02-01 20:19:08 -08:00
Tom Schuster
324fce84c8
Bug 1127494 - Remove proto parameter from JS_DefineObject. r=Waldo,bz
2015-02-04 22:50:17 +01:00
Shu-yu Guo
49af4f5b4c
Bug 1127156 - Attach optimization info to frames in profiler. (r=BenWa)
2015-02-04 13:41:04 -08:00
Shu-yu Guo
e0aee2d0b6
Bug 1127156 - Rework optimization tracking JSAPI to be more usable from the profiler. (r=djvj)
2015-02-04 13:40:02 -08:00
Shu-yu Guo
72874b667c
Bug 1128653 - Make the native code bounds of optimization regions closed. (r=djvj)
2015-02-04 13:40:01 -08:00
Till Schneidereit
7a4a2758f6
Bug 1127169 - Use self-hosting internal List type for TypedObject functions' internal bookkeeping. r=nmatsakis
...
--HG--
extra : rebase_source : de08f4c63f55b91e35066a94ce4f05b9f3028a3b
2015-01-29 16:40:49 +01:00
Jon Coppeard
9ffcc5d9e5
Backed out 2 changesets (bug 1122640) for jsapi test failures on a CLOSED TREE
...
Backed out changeset 52a98fafd551 (bug 1122640)
Backed out changeset 8c11a58bf243 (bug 1122640)
2015-02-04 18:30:18 +00:00
Jon Coppeard
b25dbc2107
Bug 1122640 - Fix inbound bustage due to missing 'explicit' on single arg constructor r=me on a CLOSED TREE
2015-02-04 16:50:57 +00:00
Jon Coppeard
e859cbf0d5
Bug 1122640 - Free nursery huge slots off main thread r=terrence
2015-02-04 16:12:06 +00:00
Tooru Fujisawa
103cb3319e
Bug 1108930 - Part 7: Warn when Map/Set/WeakMap are called without new. r=evilpie
2015-02-05 00:59:16 +09:00
Tooru Fujisawa
80ed789950
Bug 1108930 - Part 6: Call WeakMap with new in xpconnect. r=mrbkap
2015-02-05 00:59:15 +09:00
Tooru Fujisawa
47879ec69e
Bug 1108930 - Part 5: Call Map and Set with new in devtools. r=fitzgen
2015-02-05 00:59:15 +09:00
Tooru Fujisawa
28581b0f50
Bug 1108930 - Part 3: Call WeakMap with new. r=evilpie
2015-02-05 00:59:14 +09:00
Tooru Fujisawa
07f776a866
Bug 1108930 - Part 2: Call Set with new. r=evilpie
2015-02-05 00:59:14 +09:00
Tooru Fujisawa
396ef79ba6
Bug 1108930 - Part 1: Call Map with new. r=evilpie
2015-02-05 00:59:13 +09:00
Hannes Verschore
df64911442
Bug 1129377 - IonMonkey: Don't specialize Float32 with Double during inlined calls, r=bbouvier
2015-02-04 13:51:35 +01:00
Tom Schuster
df9f317f40
Bug 1129153 - Change [[Delete]] to follow ES6. r=Waldo
2015-02-04 13:23:06 +01:00
Lars T Hansen
1c6671b9f4
Bug 1129351 - critical section around MIPS simulator variables. r=jandem
2015-02-04 12:39:00 +01:00
Carsten "Tomcat" Book
ad4fcd1ca7
Backed out changeset 00b251cb7bb1 (bug 1128653) for XPCshell Linux Debug Test Failures
2015-02-04 12:35:54 +01:00
Carsten "Tomcat" Book
05d151a695
Backed out changeset cde091bad9e8 (bug 1127156)
2015-02-04 12:35:20 +01:00
Carsten "Tomcat" Book
b301968cb8
Backed out changeset fbf1b0195d80 (bug 1127156)
2015-02-04 12:35:15 +01:00
Benjamin Bouvier
c21d1dfd53
Bug 1124485: Clampped float32-to-int32 conversions should bailout on ARM; r=mjrosenb
...
--HG--
extra : rebase_source : 67ae33deac75722e8d62caf81c2e6cbe69b83232
2015-02-04 11:55:53 +01:00
Jan de Mooij
cba2aabb0a
Bug 1128528 - Don't unnecessarily require flat strings, to avoid wasting a ton of memory in pathological cases. r=luke
2015-02-04 11:16:02 +01:00
Lars T Hansen
25760866f7
Bug 1128452 - reintroduce the simulator lock on ARM. r=jandem
2015-02-04 10:34:36 +01:00
Lars T Hansen
c744efc8b6
Bug 1128449 - MIPS simulator: split the constructor to allow OOM signaling. r=jandem
2015-02-04 10:34:34 +01:00
Benjamin Bouvier
ab8d6cff23
Bug 1123404: Disable temporarily bitselect-select.js SIMD test; r=Waldo
...
--HG--
extra : amend_source : e3f8a281821532ce1b4e50ce784a887ee4d264a0
2015-02-03 22:20:50 +01:00
Lars T Hansen
645953ed9b
Bug 1074237 - evolve the JS engine futex API. r=luke
2015-02-04 09:47:31 +01:00
Ms2ger
731bff666c
Bug 1125901 - Part d: Remove unused nsIXPConnect.holdObject; r=bholley
...
It is now unused, so can be removed.
2015-02-04 09:41:41 +01:00
Tooru Fujisawa
cc87fe9c9c
Bug 1129285 - Fix findReferences-01.js to follow bug 1125930. r=jandem
2015-02-04 17:11:46 +09:00
Shu-yu Guo
302d60a584
Bug 1127156 - Attach optimization info to frames in profiler. (r=BenWa)
2015-02-03 23:51:36 -08:00
Shu-yu Guo
08256840f7
Bug 1127156 - Rework optimization tracking JSAPI to be more usable from the profiler. (r=djvj)
2015-02-03 23:51:36 -08:00
