8346b9dfaf
Bug 928336. Make defining unforgeable properties on objects faster by just copying them from an unforgeable holder object. r=peterv
2015-03-03 21:01:58 -05:00
b6bc15079a
Bug 1134537 - Delete nsGnomeVFSService. r=nfroyd
2015-02-21 14:49:13 -07:00
4495d81efb
Bug 1131445 - variadic NS_NewRunnableMethodWithArgs with storage&passing argument type decorators. r=waldo
...
--HG--
extra : rebase_source : 3c2104c19bb37624ec32a1ca4c50235e7cefd953
2015-03-01 18:12:00 +01:00
237067b8b5
Bug 1138770 - Make memory reporters that have been stashed for testing eligible for unregistration. r=mccr8.
...
--HG--
extra : rebase_source : 06dcca55a7a47e02d9b43289472c76b810663727
2015-03-02 18:55:42 -08:00
7f88bc2704
Bug 1137464 - add missing include for nsICloneableInputStream.h to nsStreamUtils.cpp; r=bkelly
2015-02-26 16:35:09 -05:00
19fa938cf7
Bug 1050035 (part 1, attempt 2) - Lazily allocate PLDHashTable::mEntryStore. r=froydnj.
...
This makes zero-element hash tables, which are common, smaller, and also avoids
unnecessary malloc/free pairs.
I did some measurements during some basic browsing of a few sites. I found that
35% of all live tables were empty with a few tabs open. And cumulatively, for
the whole session, 45% of tables never had an element added to them.
--HG--
extra : rebase_source : 306bb50f250c09aa03a5e4822f41d6f605d76a1d
2015-02-01 14:56:33 -08:00
c611e2a6a8
Backed out 2 changesets (bug 1133939) because of bug 1136453
...
Backed out changeset 212080d51fb7 (bug 1133939)
Backed out changeset 27de4b553912 (bug 1133939)
2015-02-25 13:26:41 -05:00
e4d15a45bb
Bug 1135773 - Initialize QI return values to null in the cycle collector. r=smaug
2015-02-25 09:44:10 -08:00
8979effb6c
Merge m-c to inbound. a=merge
2015-02-23 12:33:38 -05:00
26cdf50eea
Update configs. IGNORE BROKEN CHANGESETS CLOSED TREE NO BUG a=release ba=release
2015-02-23 10:22:56 -05:00
285d84d141
Bug 1134030 - Add WindowsAddressSpaceReporter. code=njn,dmajor. r=dmajor,njn.
2015-02-20 13:13:16 +11:00
9f9db94841
Bug 1133939 P2 Add tests validating nsPipeOutputStream AsyncWait behavior. r=froydnj
2015-02-21 09:51:17 -05:00
d3eb781bf1
Bug 1133939 P1 Free buffer resources when an nsPipeInputStream is closed. r=froydnj
...
* * *
Bug 1133939 P1 interdiff 005 don't delete segments still being written
2015-02-21 09:51:17 -05:00
9e8395b4c9
Bug 742194 part 1. Add support for throwing uncatchable exceptions to Web IDL bindings. People keep asking for this. r=khuey
2015-02-20 23:58:36 -05:00
34d6e2d69c
Backed out changesets 4c2b179b71ae and c96050cdedd5 (bug 1133939) for asserts.
2015-02-20 20:00:19 -05:00
1eb9f13e73
Bug 1133939 P2 Add tests validating nsPipeOutputStream AsyncWait behavior. r=froydnj
2015-02-20 18:16:04 -05:00
78491bc003
Bug 1133939 P1 Free buffer resources when an nsPipeInputStream is closed. r=froydnj
2015-02-20 18:16:04 -05:00
795381b915
Bug 1135081 - Fix typo in nsTString::AssertValidDepedentString. r=froydnj
2015-02-20 08:02:00 -05:00
c90cdd2eef
Bug 1134537 - Only support GIO in nsLocalFileUnix. r=froydnj
2015-02-19 12:58:58 -07:00
8ad291fb4d
Bug 1130142, part 4 - Use reinterpret_cast for PR_SetThreadPrivate. r=froydnj
2015-02-20 09:38:47 -08:00
af29279350
Bug 1130142, part 3 - Use HashNumber method in more places in trace refcounting. r=froydnj
...
This should not change any behavior.
2015-02-20 09:38:47 -08:00
138983aa4c
Bug 1130142, part 2 - Don't explicitly coerce pointer to bool in nsTraceRefCnt. r=froydnj
2015-02-20 09:38:47 -08:00
d0ba9119f7
Bug 1130142, part 1 - Don't truncate pointers to 32-bit values in refcount logging. r=froydnj
2015-02-20 09:38:46 -08:00
a423210355
Bug 1132802 - Add more assertions to pldhash.cpp. r=froydnj.
...
The use of |new| in PL_NewDHashTable() is necessary to avoid the new
assertions in Init() from failing.
--HG--
extra : rebase_source : 63cf962ce146142b72ffa0d6fcd3d8af1ec88bca
2015-02-11 20:24:33 -08:00
6bf8ed2521
Bug 1127201 (attempt 2, part 3) - Remove the NS_ABORT_IF_FALSE macro. r=Waldo.
...
--HG--
extra : rebase_source : b0ff8671df31c6da4a2063ede611b3c554540950
2015-02-04 20:07:09 -08:00
99d941db3c
bug 1133075 - add more MOZ_OVERRIDE r=froydnj, waldo, jrmuizel
2015-02-18 17:03:30 -05:00
3bd556b4ad
Bug 467729 - Add a PackageKit XPCOM API. r=karlt
2015-02-16 23:59:00 -05:00
3aff974cb2
Bug 1117242 followup. Rev the nsIStackFrame IID so things get rebuilt correctly. r=bustage
2015-02-13 20:24:20 -05:00
ead84e4acf
Bug 1117242 - SavedFrame objects should do principal checks for every accessor; r=jimb,jandem,bz
2015-02-13 09:21:50 -08:00
e6f06e5cfc
Bug 1127151. Part 4: Update TimeStamp::FromSystemTime with windows testing. r=kats
2015-02-13 17:17:28 +08:00
716e7de1ef
Bug 1130434- xpcom: add support for mips n64 abi. r=froydnj
2015-02-06 08:28:00 -05:00
9b07a3a0af
Bug 1131436 - WindowsDllInterceptor is always failure with --disable-optimize and --enable-debug. r=dmajor
2015-02-12 15:07:57 +09:00
8fc460a98a
Bug 1131199, part 3 - Update comments relating to PLDHashTableOp::initEntry. r=mccr8.
...
DONTBUILD because this is a comment-only change.
--HG--
extra : rebase_source : 51501a06196814dbdabbd8b170ec41ccbf6ed2d0
2015-02-11 14:24:47 -08:00
fa624bca8a
Bug 1131901 (part 2) - Make CCGraphBuilder::AddNode() infallible. r=mccr8.
...
--HG--
extra : rebase_source : cfbef5e7c783c1abb1e9bb69dcc5ef14d3c96369
2015-02-02 23:58:42 -08:00
3a7b0a9f57
Bug 1131901 (part 1) - 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 : 526d96ab65e4d7d71197b90d086d19fbdd79b7b5
2015-02-02 14:48:58 -08:00
5c46ccc050
Bug 1100398 P5 Provide NS_CloneInputStream() factory method in nsStreamUtils.h. r=froydnj
2015-02-10 23:55:43 -05:00
39a8291465
Bug 1100398 P4 Make nsPipeInputStream cloneable. r=froydnj
2015-02-10 23:55:43 -05:00
a3fc23c873
Bug 1100398 P3 Make nsStorageStream's input streams cloneable. r=froydnj
2015-02-10 23:55:43 -05:00
a7631559d4
Bug 1100398 P2 Make nsStringInputStream cloneable. r=froydnj
2015-02-10 23:55:43 -05:00
3e70bce6b1
Bug 1100398 P1 Add the nsICloneableInputStream interface. r=froydnj
2015-02-10 23:55:43 -05:00
b4a93b9b3d
Bug 1100398 P0 Break out helper routines to support gtests for more stream types. r=froydnj
2015-02-10 23:55:43 -05:00
87e4b4c13f
Bug 1036515 - narrow the scope of unlocking mMonitor in nsTimerImpl::PostTimerEvents; r=bsmedberg
2015-02-06 16:19:36 -05:00
31ba9aaed9
Bug 1131199, part 2 - Make PLDHashtInitEntry infallible. r=froydnj
...
Also, drop the unused table argument.
2015-02-11 09:46:40 -08:00
63e3218e4c
Back out changesets 2fcef6b54be7, 2be07829fefc, 66dfe37b8532, df3fcd2be8fd, 0a436bce77a6 (bug 1050035) for causing intermittent crashes and assertion failures.
...
--HG--
extra : rebase_source : eb30be83c3143c6c203585a80a18f180025efaba
2015-02-10 14:39:49 -08:00
fa0e01bfea
Bug 1129877 - Add a templatized move constructor for nsRefPtr. v1 r=froydnj
2015-02-10 13:45:02 -08:00
5a1bc25cf4
Bug 680547 bustage fix: Make #pragmas be #ifndef __clang__. No review.
2015-02-10 16:24:54 +11:00
e5f1b47ba8
Bug 680547 - Compile Linux 64-bit NS_InvokeByIndex with -mno-avx to allow compiling with -march=native on new hardware, or similar -march flags. r=froydnj
...
As explained in bug 1111355, having avx enabled appears to change the
alignment behavior of alloca (apparently adding an extra 16 bytes) of
padding/alignment (and using 32-byte alignment instead of 16-byte). The
suggestion of using __bultin_alloca_with_align in bug 1111355 didn't fix
the problem, so this seems to be the best available workaround, given
that this code, which should perhaps better be written in assembly, is
written in C++.
Interestingly, this is NOT fixed by #pragma GCC target ("arch=x86-64").
(I determined the (undocumented) name for the default -march value on
x86_64 from the gcc source code (gcc/config/i386/i386.c, function
ix86_option_override_internal, code that sets opts->x_ix86_arch_string .)
I confirmed that this sets the same macros based on the empty diff
between the output of 'gcc -E -dM -x c++ /dev/null' and 'gcc -E -dM -x
c++ -march=x86-64 /dev/null', which was not an empty diff for other
-march values (e.g., k8).)
I confirmed that the push_options and pop_options actually work by
putting the push/pop pair around a different (earlier) function, and
testing that this did not fix the bug (with the pop_options before
NS_InvokeByIndex).
See the gcc documentation at:
https://gcc.gnu.org/onlinedocs/gcc-4.9.0/gcc/Function-Specific-Option-Pragmas.html
https://gcc.gnu.org/onlinedocs/gcc-4.9.0/gcc/Function-Attributes.html
https://gcc.gnu.org/onlinedocs/gcc-4.9.0/gcc/i386-and-x86-64-Options.html
--HG--
extra : transplant_source : %DA%7CJ4H%DE%80%15%84%0D%116%85Q%9A%F9%2C%D1v%16
2015-02-10 16:04:46 +11:00
242708cf72
Bug 1127201 (attempt 2, part 1) - Replace most NS_ABORT_IF_FALSE calls with MOZ_ASSERT. r=Waldo.
...
--HG--
extra : rebase_source : 488e401ff87e31a2074c4108c4df0572d9536667
2015-02-09 14:34:50 -08:00
d3826daa16
Back out Bug 1127201 (part 2) for various problems.
2015-02-06 15:04:32 -08:00
c11d7db0a4
Back out Bug 1127201 (part 3) for various problems.
2015-02-06 15:04:00 -08:00
Boris Zbarsky
Alex Henrie
Gerald Squelart
Nicholas Nethercote
Nathan Froyd
Ehsan Akhgari
Andrew McCreight
Ryan VanderMeulen
ffxbld
Ben Kelly
Tom Tromey
Trevor Saunders
Frédéric Wang
Nick Fitzgerald
Mason Chang
James Cowgill
Makoto Kato
Bobby Holley
L. David Baron
Andrew McCreight