159957bbef8f0c8c4adcc461d3fd25a3dbf19b74: Bug 1466606 - Calculate the current GetSample index when needed. r=bryce, a=RyanVM
Jean-Yves Avenard <jyavenard@mozilla.com> - Wed, 06 Jun 2018 15:14:24 +0200 - rev 808925
Push 113536 by nthomas@mozilla.com at Thu, 21 Jun 2018 04:40:01 +0000
Bug 1466606 - Calculate the current GetSample index when needed. r=bryce, a=RyanVM The assumption that when calling GetNextRandomAccessPoint the next GetSample index would always be known was incorrect. It assumed that the call to GetNextRandomAccessPoint would always be preceded by a call to GetSample. This is not always how the MediaSourceDemuxer called it. MozReview-Commit-ID: H1MyPpDEytk
8a24c581a1820b0619f34a5f2f6fd240d83cb3f5: Bug 1401883 - Don't hold unnecessary references to the Windows app shell. r=jimm, a=RyanVM
Ting-Yu Chou <janus926@gmail.com> - Wed, 06 Jun 2018 11:05:18 -0400 - rev 808924
Push 113536 by nthomas@mozilla.com at Thu, 21 Jun 2018 04:40:01 +0000
Bug 1401883 - Don't hold unnecessary references to the Windows app shell. r=jimm, a=RyanVM When runnables are posted to the main thread's event loop, the event loop notifies any thread observers that this has been done. The app shell registers itself as just such a runnable, and posts messages to the native event loop that processing native events needs to be done. On Windows, this posting takes an extra reference to the app shell, to keep it alive until the message is processed by the native event loop, since app shell code needs to be invoked during that processing. The processing releases this extra reference, so everything stays balanced. Except that it's possible for messages to be posted to the native event loop, and then browser shutdown happens. Those messages are not processed and the associated references taken are not released. This imbalance means that in debug builds, we appear to be leaking the app shell, and that leaking results in intermittent oranges. This intermittent orange has manifested itself in a variety of ways over the years, depending on how big the app shell itself was (since that changes the number of bytes leaked) and how many leak-checked things the app shell was holding on to. This bug is merely the latest manifestation; the last serious work on analyzing the phenomenon and fixing it was done in bug 1220517. The solution proposed in that bug was that we simply stop the extra reference counting; when the app shell is destroyed normally, we shouldn't be processing the native event loop any more anyway. So even if the native event loop is holding (freed) pointers to the app shell, we'd never execute the callback and perform a use-after-free. Reading through the code suggests that this *ought* to be the case, but the potential for shooting ourselves in the foot seems awfully high. In any event, this is not a problem unique to Windows; we have seen this same sort of thing happen on OS X. See nsAppShell::ProcessGeckoEvents in widget/cocoa/nsAppShell.mm. Here we propose a slightly different solution: we keep track of the number of native event callbacks we have scheduled, incrementing when we schedule, and decrementing when we actually run one. When the app shell is destroyed, we simply set the number of outstanding events to zero, and we prohibit the callback from accessing the app shell if there are no outstanding events. This solution is analogous to dropping the extra reference counting, but avoids potential badness if we do wind up processing native events after the app shell is destroyed.
05f93f4378c67c9a3e608edf11c6ef6dde18e091: Bug 1462979 - Broker HttpEndRequestA for plugin process. r=jimm, a=RyanVM
David Parks <dparks@mozilla.com> - Tue, 29 May 2018 12:27:23 -0700 - rev 808923
Push 113536 by nthomas@mozilla.com at Thu, 21 Jun 2018 04:40:01 +0000
Bug 1462979 - Broker HttpEndRequestA for plugin process. r=jimm, a=RyanVM Some file uploads fail (terminate early) without this function. Its omission from the original system was unintentional.
f5f970316cb959932af5ef2c7e99b0da98ead7b3: Bug 1462979 - Fix printf formatting in FunctionHook logging. r=jimm, a=RyanVM
David Parks <dparks@mozilla.com> - Tue, 29 May 2018 10:49:24 -0700 - rev 808922
Push 113536 by nthomas@mozilla.com at Thu, 21 Jun 2018 04:40:01 +0000
Bug 1462979 - Fix printf formatting in FunctionHook logging. r=jimm, a=RyanVM Use %ls instead of %S, which is now checked by an assertion.
86a5a856fa1755aabc4db9c7c02f1d7587894015: Bug 1463329 - Iterate over a copy of VRManagerChild::mDisplays to guard against modifications to mDisplays during the loop. r=mccr8, a=RyanVM
Kearwood Gilbert <kgilbert@mozilla.com> - Wed, 30 May 2018 14:07:13 -0700 - rev 808921
Push 113536 by nthomas@mozilla.com at Thu, 21 Jun 2018 04:40:01 +0000
Bug 1463329 - Iterate over a copy of VRManagerChild::mDisplays to guard against modifications to mDisplays during the loop. r=mccr8, a=RyanVM
5d9a445906a939dc8e3839e00d4a5d1d7933af58: Bug 1461619 - Don't update atoms marking bitmaps in parallel due to potential races. r=sfink, a=RyanVM
Jon Coppeard <jcoppeard@mozilla.com> - Mon, 04 Jun 2018 11:18:27 +0100 - rev 808920
Push 113536 by nthomas@mozilla.com at Thu, 21 Jun 2018 04:40:01 +0000
Bug 1461619 - Don't update atoms marking bitmaps in parallel due to potential races. r=sfink, a=RyanVM
1f8fb784f0d7200144fc9ade7f06d41ed818262a: No bug - Tagging 5b5e620e33ee59c1820d1bef687f015d63c2185c with FIREFOX_61_0b13_BUILD1, FIREFOX_61_0b13_RELEASE a=release CLOSED TREE
"Mozilla Releng Treescript <release+treescript@mozilla.org>" - Tue, 12 Jun 2018 18:36:48 +0000 - rev 808919
Push 113536 by nthomas@mozilla.com at Thu, 21 Jun 2018 04:40:01 +0000
No bug - Tagging 5b5e620e33ee59c1820d1bef687f015d63c2185c with FIREFOX_61_0b13_BUILD1, FIREFOX_61_0b13_RELEASE a=release CLOSED TREE
f87145007afc1b796ecda2edf8c92083d04348d3: No bug - Tagging 5b5e620e33ee59c1820d1bef687f015d63c2185c with FENNEC_61_0b13_BUILD1, FENNEC_61_0b13_RELEASE a=release CLOSED TREE
"Mozilla Releng Treescript <release+treescript@mozilla.org>" - Tue, 12 Jun 2018 18:33:45 +0000 - rev 808918
Push 113536 by nthomas@mozilla.com at Thu, 21 Jun 2018 04:40:01 +0000
No bug - Tagging 5b5e620e33ee59c1820d1bef687f015d63c2185c with FENNEC_61_0b13_BUILD1, FENNEC_61_0b13_RELEASE a=release CLOSED TREE
d036894cc7cdb19c5033944624f166326f23241c: Automatic version bump CLOSED TREE NO BUG a=release
"Mozilla Releng Treescript <release+treescript@mozilla.org>" - Tue, 12 Jun 2018 18:30:40 +0000 - rev 808917
Push 113536 by nthomas@mozilla.com at Thu, 21 Jun 2018 04:40:01 +0000
Automatic version bump CLOSED TREE NO BUG a=release
e5dd3357937b185b1ac960ba987d446bf11f8686: No bug - Tagging 5b5e620e33ee59c1820d1bef687f015d63c2185c with DEVEDITION_61_0b13_BUILD1, DEVEDITION_61_0b13_RELEASE a=release CLOSED TREE
"Mozilla Releng Treescript <release+treescript@mozilla.org>" - Tue, 12 Jun 2018 18:30:34 +0000 - rev 808916
Push 113536 by nthomas@mozilla.com at Thu, 21 Jun 2018 04:40:01 +0000
No bug - Tagging 5b5e620e33ee59c1820d1bef687f015d63c2185c with DEVEDITION_61_0b13_BUILD1, DEVEDITION_61_0b13_RELEASE a=release CLOSED TREE
d0d7e628c4530cf2aeb7b26ccb4ce07594109eae: no bug - Bumping Firefox l10n changesets DONTBUILD r=release a=l10n-bump
L10n Bumper Bot <release+l10nbumper@mozilla.com> - Tue, 12 Jun 2018 06:22:11 -0700 - rev 808915
Push 113536 by nthomas@mozilla.com at Thu, 21 Jun 2018 04:40:01 +0000
no bug - Bumping Firefox l10n changesets DONTBUILD r=release a=l10n-bump be -> 2a2bbc5b6564 bg -> f946829d6a95 br -> c6ec6dc3c9d8 cak -> 2e587e9198fd cs -> 0329aa9865d4 cy -> 9e3bec44527f da -> 3c2beb3bab29 de -> 06a86c49aad1 dsb -> 20d4d1538346 el -> 853394eb26c1 es-AR -> 87554d50cd14 es-CL -> ffb0e185e65e es-ES -> c36b07eab954 es-MX -> 8ab45f33d132 fa -> 810d2845425b fr -> b8983c32d952 fy-NL -> 72f035f4b2bf
1e708201f647e3115edf76b6992ff02daad02f01: no bug - Bumping Fennec l10n changesets DONTBUILD r=release a=l10n-bump
L10n Bumper Bot <release+l10nbumper@mozilla.com> - Tue, 12 Jun 2018 06:22:05 -0700 - rev 808914
Push 113536 by nthomas@mozilla.com at Thu, 21 Jun 2018 04:40:01 +0000
no bug - Bumping Fennec l10n changesets DONTBUILD r=release a=l10n-bump be -> 2a2bbc5b6564 bg -> f946829d6a95 bn-BD -> 4cbd6a35df3e br -> c6ec6dc3c9d8 cak -> 2e587e9198fd da -> 3c2beb3bab29 de -> 06a86c49aad1 dsb -> 20d4d1538346 el -> 853394eb26c1 es-AR -> d3474d08c608 es-CL -> ffb0e185e65e es-MX -> 8ab45f33d132 fa -> 810d2845425b fy-NL -> 72f035f4b2bf
57c8a2867732fa7281ae624eb5b49638f48be54d: No bug - Tagging 873262bdac1b9da93249e77d53aefdf2f57d84de with FENNEC_61_0b12_BUILD1, FENNEC_61_0b12_RELEASE a=release CLOSED TREE
"Mozilla Releng Treescript <release+treescript@mozilla.org>" - Mon, 11 Jun 2018 16:03:14 +0000 - rev 808913
Push 113536 by nthomas@mozilla.com at Thu, 21 Jun 2018 04:40:01 +0000
No bug - Tagging 873262bdac1b9da93249e77d53aefdf2f57d84de with FENNEC_61_0b12_BUILD1, FENNEC_61_0b12_RELEASE a=release CLOSED TREE
5b5e620e33ee59c1820d1bef687f015d63c2185c: Bug 1453093 - Move accessible actor instantiation and panel tab highlighting into tool startup component. r=ochameau, a=RyanVM DEVEDITION_61_0b13_BUILD1 DEVEDITION_61_0b13_RELEASE FENNEC_61_0b13_BUILD1 FENNEC_61_0b13_RELEASE FIREFOX_61_0b13_BUILD1 FIREFOX_61_0b13_RELEASE
Yura Zenevich <yura.zenevich@gmail.com> - Mon, 11 Jun 2018 09:44:18 -0400 - rev 808912
Push 113536 by nthomas@mozilla.com at Thu, 21 Jun 2018 04:40:01 +0000
Bug 1453093 - Move accessible actor instantiation and panel tab highlighting into tool startup component. r=ochameau, a=RyanVM MozReview-Commit-ID: F02RgSyupUQ
44883c246d62061847e07dcba4fd88f55eb45362: Bug 1467852 Align LoadInfo::mServiceWorkerTaintingSynthesized handling with other service worker fields. r=valentin a=abillings
Ben Kelly <ben@wanderview.com> - Sun, 10 Jun 2018 18:44:53 -0700 - rev 808911
Push 113536 by nthomas@mozilla.com at Thu, 21 Jun 2018 04:40:01 +0000
Bug 1467852 Align LoadInfo::mServiceWorkerTaintingSynthesized handling with other service worker fields. r=valentin a=abillings
13af68f9e6d0b394945dfe041b4ef5108abfddb9: Bug 1465694 - Add tab event telemetry for Savant Shield study; r=dao, a=RyanVM
Bianca Danforth <bdanforth@mozilla.com> - Wed, 06 Jun 2018 03:44:48 -0700 - rev 808910
Push 113536 by nthomas@mozilla.com at Thu, 21 Jun 2018 04:40:01 +0000
Bug 1465694 - Add tab event telemetry for Savant Shield study; r=dao, a=RyanVM These probes will register and record (for the duration of the study only): * When a tab opens. * When a tab closes. * When a tab is selected. MozReview-Commit-ID: BvknEH0ofS0
63275d610ea3f9c82e2d88b923443beaa0fbc311: Bug 1465685 - Add login_form probe for Savant Shield study; r=MattN, a=RyanVM
Bianca Danforth <bdanforth@mozilla.com> - Thu, 07 Jun 2018 03:24:09 -0700 - rev 808909
Push 113536 by nthomas@mozilla.com at Thu, 21 Jun 2018 04:40:01 +0000
Bug 1465685 - Add login_form probe for Savant Shield study; r=MattN, a=RyanVM This probe will register and record (for the duration of the study only): * When a login form is loaded * When a login form is submitted (excluding the case from unresolved bug 1287202) The login_form probe has an 'extra' field called 'flow_id'. This value associates actions that occur in the same tab. It should be noted that for several reasons, we should expect a higher than 1:1 ratio between 'load' and 'submit' events: * Some sites, like Google and Amazon, have a two-step login process, and each step fires its own 'load' event. Only the second step fires a 'submit' event. * Some sites, like Facebook and Twitter, fire multiple 'load' events on the same page. * A common pattern for unsuccessful logins is for the site to redirect to a page with a login form. This would be a 'load' --> 'submit' --> 'load' series. * Unlike 'load', the 'submit' event fires only when the Password Manager is enabled and the user is in a non-private window. 'Load' events will have a 'canRecordSubmit' key in the 'extra' field which will be true if a submit event for that form can be recorded. MozReview-Commit-ID: LOMDSN6tgRV
41f2ad70c0eaa3abc0d8935fe1ec19a7a5e967a5: Bug 1465685 - Add Password Manager probes for Savant Shield study; r=MattN, a=RyanVM
Bianca Danforth <bdanforth@mozilla.com> - Thu, 07 Jun 2018 03:23:47 -0700 - rev 808908
Push 113536 by nthomas@mozilla.com at Thu, 21 Jun 2018 04:40:01 +0000
Bug 1465685 - Add Password Manager probes for Savant Shield study; r=MattN, a=RyanVM These probes will register and record (for the duration of the study only): * When the user is prompted by the Password Manager * When the user saves their login information through the Password Manager prompt * When the user updates their login information through the Password Manager prompt * When the user uses login information stored by the Password Manager Both the 'pwmgr' and 'pwmgr_use' probe have an 'extra' field called 'flow_id'. This is a tab identifier. For a given session, flow_id remains constant, even if the tab is moved to a different index within the same window. Tabs at the same index in different windows will have different flow_ids. MozReview-Commit-ID: CoBNl6lUQmH
ffc838d8250706fc915db3806a3581ca04f15018: Bug 1465697 - Add menu open probes for Savant Shield study; r=jaws, a=RyanVM
Bianca Danforth <bdanforth@mozilla.com> - Wed, 06 Jun 2018 06:26:38 -0700 - rev 808907
Push 113536 by nthomas@mozilla.com at Thu, 21 Jun 2018 04:40:01 +0000
Bug 1465697 - Add menu open probes for Savant Shield study; r=jaws, a=RyanVM These probes will register and record (for the duration of the study only): * When the library menu opens. * When the hamburger menu opens. * When the dotdotdot menu opens. These will be captured in existing and new windows. Note: The library panel is dynamically added and removed from the document (see PanelUI.js for where the panel is created and removed), so a listener can't be added to it in advance. However, the library menu "ViewShowing" event bubbles up to the navBar in its default location. A separate listener is needed if it is moved to the overflow panel via Hamburger > Customize. MozReview-Commit-ID: EBBBgXAQnxE
16a92f7f6fbaa86797d5698dd6e2be110c3b2c86: Bug 1465704 - Add follow_urlbar_link event telemetry for Savant Shield study; r=adw, a=RyanVM
Bianca Danforth <bdanforth@mozilla.com> - Wed, 06 Jun 2018 03:51:01 -0700 - rev 808906
Push 113536 by nthomas@mozilla.com at Thu, 21 Jun 2018 04:40:01 +0000
Bug 1465704 - Add follow_urlbar_link event telemetry for Savant Shield study; r=adw, a=RyanVM These probes will register and record (for the duration of the study only): * When the user clicks on a history link displayed by the urlbar * When the user clicks on a bookmark link displayed by the urlbar It should be noted that the same site can be multiple result types at the same time, so there will be times when a bookmark and/or history selection is captured as a result type other than bookmark or history (e.g. 'switchtab', 'autofill' or 'visiturl'). MozReview-Commit-ID: FEKVoaTIkiH
(0) -300000 -100000 -30000 -10000 -3000 -1000 -300 -100 -50 -20 +20 +50 +100 +300 +1000 +3000 +10000 tip