Cykesiopka
83c04b6586
Bug 1085074 - Part 3 - Update inadequately sized Delegated Signer cert. r=briansmith
2014-12-07 20:42:00 +01:00
Cykesiopka
ee0a49c7ee
Bug 1085074 - Part 2 - Use explicit bit sizes for key size cert file names. r=briansmith
2014-12-07 20:41:00 +01:00
Cykesiopka
b42aa85de9
Bug 1085074 - Part 1 - Use adequate/OK and inadequate/notOK to refer to sizes for key size tests. r=briansmith
2014-12-07 20:23:00 +01:00
Douglas Crosher
2a5e793b4d
Bug 1080477 - Jit: recognize and optimize a singleton object typed array access. r=bhackett
2014-12-02 23:12:31 +11:00
Douglas Crosher
f15787c757
Bug 1080477 - Jit x86: optimize heap access with a masked pointer plus a constant, using range analysis. r=bhacket
2014-12-06 20:39:17 +11:00
Jan-Ivar Bruaroey
afad2eeeee
Bug 1053407
- return DOMError rather than string in peerConnection callbacks. r=bz, r=jesup
2014-11-26 15:26:37 -05:00
Christoph Kerschbaumer
71fcd888ff
Bug 1104623 - Inherit security flags if channel already has a loadinfo attached in NS_NewChannel (r=bz,sworkman)
2014-12-07 20:25:46 -08:00
Andrea Marchesini
174e98dcb2
Bug 1104623 - Inherit security flags if channel already has a loadinfo attached in NS_NewChannel -tests (r=ckerschb)
2014-12-07 20:24:18 -08:00
Jan de Mooij
32755a86bf
bug 1108411 - Fix IonBuilder::getPropTryArguments* to not abort the arguments analysis. r=bhackett
2014-12-07 20:52:52 -08:00
Jan de Mooij
ab5814dd5a
Bug 1108289 - Whitelist String object resolve hook for non-integers so that our ICs work. r=bhackett
2014-12-07 20:03:03 -08:00
Bobby Holley
aa27b00ccd
Backed out changeset c405b4271e64 (bug 1097823)
2014-12-07 17:09:12 -08:00
Bobby Holley
4867c99ce8
Backed out changeset 8dfbe042e946 (bug 1097823)
2014-12-07 17:09:11 -08:00
Bobby Holley
b55769853e
Backed out changeset b56e5e762329 (bug 1097823)
2014-12-07 17:09:09 -08:00
Bobby Holley
8327697375
Backed out changeset b76636e9f49a (bug 1097823)
2014-12-07 17:09:08 -08:00
Bobby Holley
60f876f46a
Backed out changeset ab6ddc8ac28a (bug 1097823)
2014-12-07 17:09:06 -08:00
Bobby Holley
5864562a3f
Backed out changeset 47c97b8b490c (bug 1097823)
2014-12-07 17:09:04 -08:00
Bobby Holley
d6b8543ed2
Backed out changeset 1ea671313ad3 (bug 1097823)
2014-12-07 17:09:03 -08:00
Bobby Holley
31f9bf9d0b
Backed out changeset 979bec024c07 (bug 1097823)
2014-12-07 17:08:19 -08:00
Bobby Holley
da23a71bbe
Bug 1097823 - Followup to remove the assertions cpearce asked for because they don't hold. rpending=cpearce
2014-12-07 15:41:46 -08:00
Chris Lord
cccccfadb3
Bug 1106132 - Signal more error types in browser element. r=fabrice
2014-12-05 11:06:25 +00:00
Bobby Holley
7b4b4838ca
Bug 1097823 - Followup to avoid null-derefing when promises have already been rejected during shutdown. rpending=cpearce
2014-12-07 15:08:47 -08:00
Bobby Holley
3871f53bf6
Bug 1097823 - Followup fix to ensure that the borrowed task queue doesn't get destroyed before subdecoders shut down. rpending=cpearce
2014-12-07 13:25:11 -08:00
Bobby Holley
b5f3305d3c
Bug 1097823 - Replace AudioDecodeRendezvous with promise-y goodness. r=cpearce,karlt
2014-12-07 11:21:35 -08:00
Bobby Holley
61900e1d2d
Bug 1097823 - Use MediaPromises for RequestAudioData and RequestVideoData. r=cpearce
2014-12-07 11:21:35 -08:00
Bobby Holley
ecd1b4cbb2
Bug 1097823 - Implement Media Promises. r=cpearce
2014-12-07 11:21:34 -08:00
Bobby Holley
6318733591
Bug 1097823 - Implement MediaTaskQueue::ForceDispatch. r=cpearce
...
This is necessary to have strong guarantees that promises will be resolved.
While we're flushing the task queue, normal dispatch starts to fail,
meaning that we can't dispatch promise resolution. We have 3 options to handle
this:
(A) Never respond to the promise.
(B) Invoke the Resolve/Reject callback synchronously if dispatch fails.
(C) Prevent dispatch from failing.
(C) seems like the option least likely to violate invariants if we can get away
with it. Promise resolution is unlikely to be a heavyweight task in the way that
a decode task might be, so this should hopefully be ok.
Note that this still doesn't help for bonafide task queue shutdown. It's up to
consumers to tear down their MediaPromiseHolders before the task queues are shut
down.
2014-12-07 11:21:34 -08:00
Bobby Holley
14e9457ae3
Bug 1097823 - Move NotDecodedReason from RequestSampleCallback to MediaDecoderReader. r=cpearce
2014-12-07 11:21:34 -08:00
Sotaro Ikeda
ab264f3007
Bug 1100110 - Fix ScaleMode handling of ImageLayer and HwcComposer2D r=mattwoodrow
2014-12-06 17:12:50 -08:00
Tom Schuster
87d731b076
Bug 1108261 - Use new to create cross compartment typed array. no-r,testonly
2014-12-06 12:18:12 -08:00
Bobby Holley
455f15535d
Bug 1103324 - Instantiate an AutoNoJSAPI when dispatching IPC messages. r=billm
2014-12-06 09:51:14 -08:00
Jan Varga
7a79af88ee
Bug 771288 - Multiprocess FileHandle support (FileHandle on PBackground); Part 1 - Remove unused stuff; r=bent
2014-12-06 07:31:53 -08:00
Jonathan Watt
da852fb016
Bug 1083079 - Prevent the size of TiledDrawTarget from being zero due to its tiles having negative offsets. r=mattwoodrow
2014-12-05 03:00:24 -08:00
Ryan VanderMeulen
df6811b023
Backed out changeset af9f873b8bc6 (bug 1107416) for non-unified bustage.
2014-12-06 02:06:56 -05:00
Dan Glastonbury
dd19bb1f83
Bug 1097416 - [WebGL2] Enable WebGL 2 if all required features are available. r=jgilbert
...
--HG--
extra : rebase_source : 696e99670b8184593ea8ca9aa4becb1c4d86cad6
extra : source : e2ff0e2bcf21067b6a121e50e4f290227aedc6c6
2014-10-27 09:36:26 +10:00
Chris Peterson
28b15aa1fc
Bug 1107416 - Add static_assert reminder to increment XDR_BYTECODE_VERSION_SUBTRAHEND when extending js.msg. r=shu
2014-12-05 00:19:40 -08:00
Steve Fink
204cae5a93
Bug 1107936 - Do not define os for fuzzing-safe shell
...
--HG--
extra : rebase_source : 8ec7747fee2f1117f649edd4d0470ff63cb6e102
2014-12-05 18:01:22 -08:00
Brian Hackett
61ff37cf13
Bug 1107496 - Add PlainObject subclass, r=luke.
2014-12-05 18:52:28 -07:00
Nick Fitzgerald
5bfb7008b9
Bug 1103817 - Handle cross compartment references to JSScripts in JS_TraceIncomingCCWs. r=terrence
2014-12-05 17:11:30 -08:00
Ryan VanderMeulen
ba53ff1161
Backed out 5 changesets (bug 1098126
) for Windows web-platform-test failures.
...
Backed out changeset 7cf7a6ea28c2 (bug 1098126
)
Backed out changeset 21aab02fb41c (bug 1098126
)
Backed out changeset 55dba2167b36 (bug 1098126
)
Backed out changeset 73c2707c4f60 (bug 1098126
)
Backed out changeset 4d6a52c5fb75 (bug 1098126
)
2014-12-05 20:09:54 -05:00
Ryan VanderMeulen
0d016bc066
Merge m-c to inbound. a=merge
2014-12-05 19:16:21 -05:00
Ryan VanderMeulen
5e0f31c0c5
Merge fx-team to m-c. a=merge
2014-12-05 19:10:22 -05:00
Ryan VanderMeulen
5f0aa6245d
Merge inbound to m-c. a=merge
2014-12-05 19:02:53 -05:00
Hannes Verschore
b6b0fc882e
Bug 1105684: Add testcase, r=bbouvier
2014-12-05 16:01:56 -08:00
Bas Schouten
3eaeba266b
Bug 1102499
: Use WARP by default when D3D11 is available. r=BenWa
...
This will cause us to use WARP for composition on systems with Direct3D 11 when hardware acceleration is blacklisted or we're unable to find a powerful enough graphics card. It also allows us to force WARP on, in that situation WARP will also be used for Direct2D content rendering if Direct2D is enabled. Essentially this will ensure all systems running Windows 7 or higher always use D3D11 as their composition backend.
2014-12-06 00:02:19 +00:00
Bas Schouten
76753982e0
Bug 1107299: Only block shared surface usage when that is broken. r=jrmuizel
...
Currently when the usage of surface sharing is broken on certain drivers, be block D3D11 entirely and fall back to Basic composition. There's no real benefit in this case from disabling D3D11 entirely, and we might as well just use it with software composition like we do when we don't have Direct2D, and in addition to that disable texture sharing for webGL, falling back to slow WebGL which again is no worse than the situation when using Basic or D3D9 composition.
2014-12-06 00:02:18 +00:00
Bas Schouten
184e84297b
Bug 1107297: Only recomposite the damaged rect with D3D11. r=jrmuizel
...
This patch makes us behave similarly to when we're using the BasicCompositor, essentially we will clip all drawing to the final display to the area that was labeled as invalid. When DXGI 1.2 is available we will then also report the damaged area to the Present call so that can then be used to minimize the amount of bits that actually need to be blitted to the screen. Since we're no longer recompositing the whole screen this means we should also only clear the damaged area of the window.
2014-12-06 00:02:18 +00:00
L. David Baron
fb22e3fc16
Bug 1089417 patch 8 - Only drop MediumFeaturesChanged on the floor if we've never computed style before, rather than never computed style using this rule processor. r=heycam
...
This depends on patches 4 and 7.
2014-12-05 15:50:01 -08:00
L. David Baron
b3e14ccac7
Bug 1089417 patch 7 - Save the previous media query cache key on the rule processor. r=heycam
...
This depends on patches 5 and 6, and is needed for patch 8.
2014-12-05 15:50:00 -08:00
L. David Baron
da55abe7bc
Bug 1089417 patch 6 - Pass the previous CSS rule processor to the constructor of the new one (when we replace one with another). r=heycam
...
This is needed for patch 7.
2014-12-05 15:50:00 -08:00
L. David Baron
fccbbb258f
Bug 1089417 patch 5 - Add method to save the current media query result cache key from a rule processor. r=heycam
...
This is needed for patch 7.
2014-12-05 15:50:00 -08:00