Jason Orendorff
0859395dae
Bug 1148652, part 1 - Move array-specific special cases to the top of NativeDefineProperty; update ArraySetLength to be able to cope with incomplete attrs. r=efaust.
2015-03-23 14:32:30 -05:00
Jason Orendorff
86ac038c99
Bug 1148568 - In JSON.parse with a reviver callback, ignore failure when defining properties. r=Waldo.
...
Switch to StandardDefineProperty while we're at it. I don't think the "CreateDataProperty" operation given in the spec is quite worth making into a function on its own, since the standard almost exclusively uses it in cases where it "can't fail", and (a) that is unpossible for us; (b) this is not one of those cases anyway.
2015-03-23 14:32:30 -05:00
Jason Orendorff
3ecf4c2cf0
Bug 1147660, part 5 - Split the part of NativeDefineProperty that updates the object into a separate function again. r=efaust.
2015-03-23 14:32:29 -05:00
Jason Orendorff
89f5cd6bcd
Bug 1147660, part 4 - Change NativeDefineProperty to use a PropertyDescriptor internally instead of a bunch of variables. This is a little ugly at first but it'll get better. r=efaust.
2015-03-23 14:32:29 -05:00
Jason Orendorff
19abf30ebb
Bug 1147660, part 3 - Rearrange NativeDefineProperty so that special cases are all dispensed with, and ES6 checks done, by the time we start thinking about how to update the object. r=efaust.
2015-03-23 14:32:29 -05:00
Jason Orendorff
67eb239ca7
Bug 1147660, part 2 - Merge DefinePropertyOrElement into NativeDefineProperty, making one long function we can refactor. r=efaust.
2015-03-23 14:32:28 -05:00
Jason Orendorff
6ad7566906
Bug 1147660, part 1 - Refactor NativeDefineProperty to put DefinePropertyOrElement's only call site right at the end. No change in behavior. r=efaust.
2015-03-23 14:32:28 -05:00
Jason Orendorff
15a7168cac
Bug 1142828 - Refactor to avoid having [[Set]]-specific weird cases in the [[DefineOwnProperty]] code. r=Waldo.
...
NativeSetExistingDataProperty is only moved in this changeset; its body is not modified.
2015-03-23 14:32:28 -05:00
Jason Orendorff
ee03b2d425
Bug 1138499, part 3 - Flip JS_CHECK_ACCESSOR_FLAGS from a blacklist to a whitelist. r=Waldo.
2015-03-23 16:44:10 -05:00
Jason Orendorff
54a7a21f2a
Bug 1138499, part 2 - Strengthen assertComplete() to require that both [[Get]] and [[Set]] be present on accessor properties. r=Waldo.
2015-03-23 14:32:27 -05:00
Jason Orendorff
fc99a92115
Bug 1138499, part 1 - Assert some basic rules on property descriptors on entry to DefineProperty and exit from GetOwnPropertyDescriptor. r=Waldo.
2015-03-23 14:32:27 -05:00
Jason Orendorff
149449abad
Bug 1138499, part 0 - Fix code spuriously using JSPROP_READONLY when defining an accessor property. r=Waldo.
2015-03-23 16:36:43 -05:00
Jason Orendorff
6ffcb02f8a
Bug 1145636 - Update jsapi-tests/README. (The functional change is where it talks about the Makefile; that was stale.) r=Waldo.
2014-10-31 16:58:18 -05:00
Wes Kocher
945253bc5d
Backed out changeset 429f4e9d44dc (bug 1149811) for test failures CLOSED TREE
2015-04-01 10:25:04 -07:00
Wes Kocher
c506200d80
Backed out 8 changesets (bug 679939) for test failures
...
Backed out changeset 4a1a45bc093a (bug 679939)
Backed out changeset 5336ae0fe919 (bug 679939)
Backed out changeset 4cbcb40e0ef4 (bug 679939)
Backed out changeset 3bc68cf4610b (bug 679939)
Backed out changeset ec2d863a3df9 (bug 679939)
Backed out changeset 4425a55312c6 (bug 679939)
Backed out changeset d3793e00b41f (bug 679939)
Backed out changeset 872edd1aedf7 (bug 679939)
2015-04-01 10:24:42 -07:00
Hannes Verschore
0e2adeb3f5
Bug 1138881 - IonMonkey: Allow fixing andor blocks which have MFilterTypeSet, r=bhackett
2015-03-25 16:15:28 +01:00
Hannes Verschore
bb370db44e
Bug 1138881 - IonMonkey: Improve types at AndOr, r=bhackett
2015-03-25 16:15:27 +01:00
Boris Zbarsky
21814fccdd
Bug 679939 part 8. Drop the now-unused compileAndGo from CompileOptions. r=luke
2015-04-01 12:05:29 -04:00
Boris Zbarsky
3773cc6f5c
Bug 679939 part 7. Drop the now-unused JSScript::compileAndGo. r=luke
2015-04-01 12:05:29 -04:00
Boris Zbarsky
edfec71b46
Bug 679939 part 6. Drop function-cloning uses of compileAndGo, since it no longer affects the bytecode. r=luke
2015-04-01 12:05:29 -04:00
Boris Zbarsky
44fe165422
Bug 679939 part 5. Stop using the compileAndGo script flag in the bytecode emitter. r=luke
2015-04-01 12:05:29 -04:00
Boris Zbarsky
0a2cfae2a5
Bug 679939 part 4. Set the isRunOnce compile flag as needed. r=luke
2015-04-01 12:05:28 -04:00
Boris Zbarsky
cc23d228f1
Bug 679939 part 3. Add a CompileOptions flag for indicating that the script should be compiled runOnce. r=luke
2015-04-01 12:05:28 -04:00
Boris Zbarsky
6404539703
Bug 679939 part 2. Disallow execution of global/eval scripts that are flagged runOnce and have already run. r=luke
2015-04-01 12:05:28 -04:00
Boris Zbarsky
4a066c2428
Bug 679939 part 1. Extend the hasRunOnce/treatAsRunOnce setup to global and eval scripts. r=luke
2015-04-01 12:05:28 -04:00
Boris Zbarsky
3ec3e636ca
Bug 1149811. When XDR-decoding a non-lazy function that can be lazified, we need to set up a backpointer from its LazyScript to its JSScript. r=luke
2015-04-01 12:05:28 -04:00
Jan de Mooij
0b62bfe44f
Bug 1146836 part 2 - Use Vectors instead of malloc in emitSwitch and fix an old bug. r=luke
2015-04-01 16:20:29 +02:00
Jan de Mooij
9e23e7dbd5
Bug 1146836 part 1 - Cleanup BytecodeEmitter::emitSwitch. r=luke
2015-04-01 16:20:26 +02:00
David Rajchenbach-Teller
9c731361b4
Bug 674779 - Per-component CPU monitoring, low-level. r=blassey, r=jandem
2015-03-31 12:40:52 +02:00
Tooru Fujisawa
4dd9b63efd
Bug 1054755 - Part 3: Use IsRegExp in String.prototype.{contains,startsWith,endsWith}. r=till
2015-04-01 18:34:03 +09:00
Tooru Fujisawa
0c3893a2a1
Bug 1054755 - Part 2: Implement IsRegExp. r=till
2015-04-01 18:34:03 +09:00
Tooru Fujisawa
2c6deea24c
Bug 1054755 - Part 1: Add Symbol.match. r=till
2015-04-01 18:34:03 +09:00
Tooru Fujisawa
a1ce97e60c
Bug 1054755 - Part 0: Rename existing IsRegExp to IsRegExpObject. r=till
2015-04-01 18:34:03 +09:00
Tooru Fujisawa
8179cb9219
Bug 1005110 - Warn about unreachable expression after semicolon-less return. r=efaust
2015-04-01 18:34:03 +09:00
Tooru Fujisawa
0ec843f005
Bug 1148899 - Rename isExprClosure/setIsExprClosure/EXPR_CLOSURE to isExprBody/setIsExprBody/EXPR_BODY. r=shu
2015-04-01 18:34:02 +09:00
Tooru Fujisawa
df49e8f6e7
Bug 1149015 - Part 3: Remove some use of expression closure from jit-test r=luke
2015-04-01 18:34:02 +09:00
Tooru Fujisawa
163479c9aa
Bug 1149015 - Part 2: Remove some use of expression closure from jstests js1_8_5/extensions. r=luke
2015-04-01 18:34:02 +09:00
Tooru Fujisawa
7bd7692f49
Bug 1149015 - Part 1: Remove some use of expression closure from jstests ecma_7/. r=sfink
2015-04-01 18:34:02 +09:00
Jon Coppeard
7c7275d219
Bug 1149135 - Don't create HashMapEntry on stack when adding to a HashMap r=luke
2015-04-01 10:11:06 +01:00
Jon Coppeard
9e18ab0f78
Bug 1146696 - Don't assume there are no arenas available after last ditch GC. r=terrence
2015-03-30 11:03:35 +01:00
Jon Coppeard
67ba4cd61f
Bug 1145997 - Suppress GC in debugger GC callback r=sfink
2015-04-01 10:09:56 +01:00
Chris Peterson
51b6b7c866
Bug 1125698 - Add ExpandErrorArguments() varargs helper function and remove uninitialized dummy
va_list variable. r=Waldo
2015-01-25 18:38:28 -08:00
Mike Hommey
ccd2a9b975
Bug 1134920 - Use moz_xmalloc/moz_xrealloc/free instead of nsMemory::Alloc/Realloc/Free. r=nfroyd
2015-04-01 13:51:45 +09:00
L. David Baron
900854f647
Backed out changeset e942133b1f9b (bug 1149769) for jittest-2, jsreftest, and spidermonkey test failures, on a CLOSED TREE.
2015-03-31 19:37:54 -07:00
Eric Faust
ef2ada318b
Bug 1149769 - Split reflect-parse.js reftest into its own subdirectory. (r=Waldo)
2015-03-31 18:29:10 -07:00
Tom Schuster
b6e24288b0
Bug 1147005 - Change JSAddPropertyOp signature. r=jorendorff,peterv
2015-03-28 14:47:02 +01:00
Sean Stangl
2468d8703c
Bug 1148375 - Ignore unhandled Elements. r=jandem
2015-03-30 13:34:19 -07:00
Sean Stangl
9bdb826506
Bug 1146619 - Rename "Prolog" to "Prologue" everywhere. r=efaust
2015-03-23 14:13:26 -07:00
Boris Zbarsky
530c951ee0
Bug 1149280 part 1. Make nullprincipal creation faster. r=smaug
2015-03-31 13:11:00 -04:00
Boris Zbarsky
6eea18cc58
Bug 1148973. When skipping shape guards in Ion common getter/setter code because the object has a non-configurable property, first verify that its current shape matches the shape we're using to compile our code. r=jandem
2015-03-30 23:44:01 -04:00