2c66b75bbb7fe94c60b4a9dbecb8e375facacfd4: Merge m-c to inbound, a=merge
Wes Kocher <wkocher@mozilla.com> - Wed, 08 Jun 2016 17:09:04 -0700 - rev 341241
Push 6389 by raliiev@mozilla.com at Mon, 19 Sep 2016 13:38:22 +0000
Merge m-c to inbound, a=merge
451483416eed6a8be15a6e2253321c9e99b45b04: Bug 1276688 part 1 - Handle entries arrays where we don't have a property value at the 0.0/1.0 offset; r=heycam
Brian Birtles <birtles@gmail.com> - Thu, 09 Jun 2016 08:12:52 +0900 - rev 341240
Push 6389 by raliiev@mozilla.com at Mon, 19 Sep 2016 13:38:22 +0000
Bug 1276688 part 1 - Handle entries arrays where we don't have a property value at the 0.0/1.0 offset; r=heycam MozReview-Commit-ID: GCCF4taOC6z
511a2389ca48a403386aeb8f0025f26bab1e882e: Bug 1276132 - Make more Widevine class constructors explicit to keep gcc happy. r=jwwang
Chris Pearce <cpearce@mozilla.com> - Tue, 31 May 2016 12:05:24 +1200 - rev 341239
Push 6389 by raliiev@mozilla.com at Mon, 19 Sep 2016 13:38:22 +0000
Bug 1276132 - Make more Widevine class constructors explicit to keep gcc happy. r=jwwang Fixes compile errors on Linux when Widevine is compiled. MozReview-Commit-ID: 19qQw02CqdQ
91b3cdd0640a29eefc8619bf06e431dafe22c6ed: Bug 1276132 - Don't show EME 'enable' UI for non-visible keysystems. r=gijs
Chris Pearce <cpearce@mozilla.com> - Tue, 31 May 2016 12:04:12 +1200 - rev 341238
Push 6389 by raliiev@mozilla.com at Mon, 19 Sep 2016 13:38:22 +0000
Bug 1276132 - Don't show EME 'enable' UI for non-visible keysystems. r=gijs This means we'll only show the EME UI for keysystems that are explicitly turned on in the build config, or those that are enabled after the build via prefs. MozReview-Commit-ID: 7VTp1BJSZFy
09b9972e36f9fb309a1f86918beeae26d1888aa6: Bug 1276132 - Remove use of #ifdef MOZ_{KEYSYSTEM}_EME in dom/media code. r=jwwang
Chris Pearce <cpearce@mozilla.com> - Tue, 31 May 2016 11:00:55 +1200 - rev 341237
Push 6389 by raliiev@mozilla.com at Mon, 19 Sep 2016 13:38:22 +0000
Bug 1276132 - Remove use of #ifdef MOZ_{KEYSYSTEM}_EME in dom/media code. r=jwwang Instead of controlling visibility of EME keysystems by build config, do it by preference. This means keysystems can be turned on easier. MozReview-Commit-ID: Ky1zrHPubOJ
69052d4e3dbb457219dad902a1ff9f1d98300c57: Bug 1276132 - Rename media.gmp-*.forcevisible to media.gmp-*.visible, and set it when keysystems are enabled. r=spohl
Chris Pearce <cpearce@mozilla.com> - Tue, 31 May 2016 10:59:44 +1200 - rev 341236
Push 6389 by raliiev@mozilla.com at Mon, 19 Sep 2016 13:38:22 +0000
Bug 1276132 - Rename media.gmp-*.forcevisible to media.gmp-*.visible, and set it when keysystems are enabled. r=spohl Repurpose the media.gmp-*.forcevisible pref to control whether the corresponding GMP is visible in the addons manager UI. The pref has to be true for the GMP to be usable. The pref is enabled and not hidden when the corresponding EME keysystem is enabled in the mozconfig. This means users can turn on EME without needing to recompile their build; they just need to create a hidden pref. This will be useful for CDM developers, and users on platforms where we've not enabled EME yet but users want to test it (Linux). We also need to change the GMPUtils.isPluginHidden() accessor so that plugins are considered hidden if the "visible" pref is false OR we're on an unsupported platform. This ensures that we must be on a supported OS and the visibility pref is true before GMPs appear in the addon list. A consequence of the isPluginHidden() change is that we also need to add a "force-supported" pref to override the checks that refuse to load the GMPs on various platform versions, so that the unit tests pass. MozReview-Commit-ID: h6CwLDkvFW
2e602689376d22362b4bfc556af2c3b6c39dce5f: Bug 1270357 Implement runtime.connectNative() r=kmag
Andrew Swan <aswan@mozilla.com> - Wed, 08 Jun 2016 13:51:01 -0700 - rev 341235
Push 6389 by raliiev@mozilla.com at Mon, 19 Sep 2016 13:38:22 +0000
Bug 1270357 Implement runtime.connectNative() r=kmag MozReview-Commit-ID: Fo4BxEo3xus
e1979d928c7c4eb08b6a5303b5533ce5a3318ee1: Bug 1270357 Add a test hook to Subprocess.jsm r=kmag
Andrew Swan <aswan@mozilla.com> - Tue, 31 May 2016 16:15:53 -0700 - rev 341234
Push 6389 by raliiev@mozilla.com at Mon, 19 Sep 2016 13:38:22 +0000
Bug 1270357 Add a test hook to Subprocess.jsm r=kmag MozReview-Commit-ID: Or2EOAL1eC
e0f9b54febc83597415fe8d056ef7b4e8939f03d: Bug 1270357 Add isDeeply to mochitest eslintrc r=kmag
Andrew Swan <aswan@mozilla.com> - Tue, 31 May 2016 16:16:37 -0700 - rev 341233
Push 6389 by raliiev@mozilla.com at Mon, 19 Sep 2016 13:38:22 +0000
Bug 1270357 Add isDeeply to mochitest eslintrc r=kmag MozReview-Commit-ID: 6TsspFtPiuZ
fb084575df12b88ac718f023454f3f5d500ab388: Bug 1271574 - Purposefully leak the XUL_APP_FILE string passed to putenv. r=bsmedberg
Mike Hommey <mh+mozilla@glandium.org> - Thu, 02 Jun 2016 08:44:16 +0900 - rev 341232
Push 6389 by raliiev@mozilla.com at Mon, 19 Sep 2016 13:38:22 +0000
Bug 1271574 - Purposefully leak the XUL_APP_FILE string passed to putenv. r=bsmedberg Before bug 552864, the string was created with PR_smprintf, and PR_SetEnv'ed (which, under the hood, just calls putenv). PR_smprintf was allocating the string on the heap. Now, it's allocated on the stack, and still putenv'ed. putenv kind of takes ownership of the strings it's being passed, so stack allocated strings are dangerous to use. It looks like we've been fairly lucky that it worked, presumably because compilers would keep the stack frame with the variable, but that's not guaranteed to happen, and in some case, doesn't. So we strdup the string and purposefully leak it instead, which matches what happened before bug 552864, and is the only "sane" way to use putenv.
302da31353a0a8ca6d65e01e869d98a589d34b62: Backed out changeset e8b366f388f6 (bug 1278985) for build bustage on a CLOSED TREE
Sebastian Hengst <archaeopteryx@coole-files.de> - Thu, 09 Jun 2016 00:24:15 +0200 - rev 341231
Push 6389 by raliiev@mozilla.com at Mon, 19 Sep 2016 13:38:22 +0000
Backed out changeset e8b366f388f6 (bug 1278985) for build bustage on a CLOSED TREE
2dbc26e0764692cd9d6e9d6c01ad21063de00dbe: Bug 1277298 - Store the 'scrollgrab' property in a NODE_FLAG_BIT rather than a new field of nsGenericHTMLElement. r=smaug
Botond Ballo <botond@mozilla.com> - Mon, 06 Jun 2016 17:45:46 -0400 - rev 341230
Push 6389 by raliiev@mozilla.com at Mon, 19 Sep 2016 13:38:22 +0000
Bug 1277298 - Store the 'scrollgrab' property in a NODE_FLAG_BIT rather than a new field of nsGenericHTMLElement. r=smaug MozReview-Commit-ID: 7DV33kG2KNt
df61facc5659fca43f03e367cd899382f5f9cfce: Bug 1279006 - Clean up webaudioeditor node destruction tests, and fix another race condition. (r=bgrins)
Eric Faust <efaustbmo@gmail.com> - Wed, 08 Jun 2016 14:54:53 -0700 - rev 341229
Push 6389 by raliiev@mozilla.com at Mon, 19 Sep 2016 13:38:22 +0000
Bug 1279006 - Clean up webaudioeditor node destruction tests, and fix another race condition. (r=bgrins)
30174708069d124a762b8fbd617f4fc0dc17a5b8: Bug 1278530 Sanity check ids derived from signatures r=kmag
Andrew Swan <aswan@mozilla.com> - Wed, 08 Jun 2016 14:38:35 -0700 - rev 341228
Push 6389 by raliiev@mozilla.com at Mon, 19 Sep 2016 13:38:22 +0000
Bug 1278530 Sanity check ids derived from signatures r=kmag MozReview-Commit-ID: Afb7fSwlvVe
cfc53646cd69d2355d046bde9930eb1ae97a71cf: Bug 1278322 - Remove the gfxContext::Mask overload that takes a gfxASurface. r=eflores
Jonathan Watt <jwatt@jwatt.org> - Thu, 02 Jun 2016 11:13:30 +0100 - rev 341227
Push 6389 by raliiev@mozilla.com at Mon, 19 Sep 2016 13:38:22 +0000
Bug 1278322 - Remove the gfxContext::Mask overload that takes a gfxASurface. r=eflores
1bfdc24b25401e27abd6ae071be69d0ffcb95f2f: Bug 1278989 - Make parent outer window available on TabParent. r=mrbkap
Mike Conley <mconley@mozilla.com> - Wed, 27 Apr 2016 15:40:36 -0400 - rev 341226
Push 6389 by raliiev@mozilla.com at Mon, 19 Sep 2016 13:38:22 +0000
Bug 1278989 - Make parent outer window available on TabParent. r=mrbkap MozReview-Commit-ID: 9nS1zQkGmu3
e8b366f388f6e41f9f4663ab89eab383eea8609e: Bug 1278985 - Remove unused URL arg from CreateChromeWindow2. r=mrbkap
Mike Conley <mconley@mozilla.com> - Wed, 27 Apr 2016 16:19:25 -0400 - rev 341225
Push 6389 by raliiev@mozilla.com at Mon, 19 Sep 2016 13:38:22 +0000
Bug 1278985 - Remove unused URL arg from CreateChromeWindow2. r=mrbkap This isn't used at all internally, and was probably used by people embedding Gecko back when Gecko was more embeddable. For now, it's just noise - I figure we can remove it. MozReview-Commit-ID: I1H2WZeoHTJ
b598ce1977bc54968e485184a16d410b231f2843: Bug 1277695 Add sanity checks for addons with no id r=kmag
Andrew Swan <aswan@mozilla.com> - Tue, 07 Jun 2016 16:23:17 -0700 - rev 341224
Push 6389 by raliiev@mozilla.com at Mon, 19 Sep 2016 13:38:22 +0000
Bug 1277695 Add sanity checks for addons with no id r=kmag The patch for bug 1277965 (enable ADDON_SIGNING by default) fixes the actual problem here but this patch adds a series of sanity checks in the add-on manager and the test case, so any similar problem in the future should be easier to troubleshoot. MozReview-Commit-ID: JGQk9yCVFkQ
8b7926661e4a89b25210539d3f4485e7fad75f9a: Bug 1274777 - Check for oveflow. r=peterv
Eric Rahm <erahm@mozilla.com> - Wed, 08 Jun 2016 13:54:05 -0700 - rev 341223
Push 6389 by raliiev@mozilla.com at Mon, 19 Sep 2016 13:38:22 +0000
Bug 1274777 - Check for oveflow. r=peterv
b3930a21b6edbf8e3138ca739f1b7ead13ecf703: Bug 1197954 - Implement Linux hardware vsync using GLX_SGI_video_sync. r=mchang,lsalzman
Andrew Comminos <andrew@comminos.com> - Mon, 09 May 2016 11:35:58 -0400 - rev 341222
Push 6389 by raliiev@mozilla.com at Mon, 19 Sep 2016 13:38:22 +0000
Bug 1197954 - Implement Linux hardware vsync using GLX_SGI_video_sync. r=mchang,lsalzman MozReview-Commit-ID: DPeqxPl0NSh
(0) -300000 -100000 -30000 -10000 -3000 -1000 -300 -100 -50 -20 +20 +50 +100 +300 +1000 +3000 +10000 +30000 +100000 tip