c0c14c758588be6ca6b58492d54736df95f38506: Bug 1384688 - Add a pausedRanges field to the profile JSON of each process which lists time ranges during which the profiler was paused or collecting a profile. r?njn draft
Markus Stange <mstange@themasta.com> - Mon, 31 Jul 2017 14:44:35 -0400 - rev 618570
Push 71382 by bmo:mstange@themasta.com at Mon, 31 Jul 2017 18:46:43 +0000
Bug 1384688 - Add a pausedRanges field to the profile JSON of each process which lists time ranges during which the profiler was paused or collecting a profile. r?njn MozReview-Commit-ID: 6nzt7uHTLVm
7458f2c4b12a0848462a17d750fbc6d932313917: Bug 1384688 - Pass aIsShuttingDown == true when obtaining shutdown profiles. r?njn draft
Markus Stange <mstange@themasta.com> - Fri, 28 Jul 2017 16:16:54 -0400 - rev 618569
Push 71382 by bmo:mstange@themasta.com at Mon, 31 Jul 2017 18:46:43 +0000
Bug 1384688 - Pass aIsShuttingDown == true when obtaining shutdown profiles. r?njn MozReview-Commit-ID: Aw79gYQgkqy
983d412b7d802ad58fb47cee3e4dcc0f1aecfadb: Bug 1384688 - Add an aIsShuttingDown argument to profiler_get_profile() and to profiler_stream_json_for_this_process(). r?njn draft
Markus Stange <mstange@themasta.com> - Mon, 31 Jul 2017 14:23:13 -0400 - rev 618568
Push 71382 by bmo:mstange@themasta.com at Mon, 31 Jul 2017 18:46:43 +0000
Bug 1384688 - Add an aIsShuttingDown argument to profiler_get_profile() and to profiler_stream_json_for_this_process(). r?njn When set to true, the resulting profile will have a non-null meta.shutdownTime field which is set to current time. Non-shutdown profiles also get that field, but it's null for them. MozReview-Commit-ID: 1vpmhBR8rC6
ab29eb3d0d9dc83daa5843e4cef66fe58d9d2267: Bug 1105109 - ScrollInputMethod telemetry for APZ autoscrolling. r=kats draft
Botond Ballo <botond@mozilla.com> - Fri, 28 Jul 2017 16:32:42 -0400 - rev 618567
Push 71381 by bballo@mozilla.com at Mon, 31 Jul 2017 18:42:38 +0000
Bug 1105109 - ScrollInputMethod telemetry for APZ autoscrolling. r=kats MozReview-Commit-ID: 5ICwYnZTcm9
a85b20059ae904cc18241e128974bbe8ee060468: Bug 1105109 - Put APZ autoscrolling behind a pref. r=kats draft
Botond Ballo <botond@mozilla.com> - Fri, 28 Jul 2017 19:01:52 -0400 - rev 618566
Push 71381 by bballo@mozilla.com at Mon, 31 Jul 2017 18:42:38 +0000
Bug 1105109 - Put APZ autoscrolling behind a pref. r=kats For now, the pref is off by default. MozReview-Commit-ID: BtzmFGIAGey
0d96639d873d12e63d981b3bc50b08360fb10d59: Bug 1105109 - Notify content when APZ is handling an autoscroll. r=kats draft
Botond Ballo <botond@mozilla.com> - Wed, 26 Jul 2017 19:32:57 -0400 - rev 618565
Push 71381 by bballo@mozilla.com at Mon, 31 Jul 2017 18:42:38 +0000
Bug 1105109 - Notify content when APZ is handling an autoscroll. r=kats MozReview-Commit-ID: BeuZt30fMpn
c5957ad858a31fc923af8ae18c7f1dea0864e46d: Bug 1355455 - Replace shared icons with Photon versions. r=dao draft
Nihanth Subramanya <nhnt11@gmail.com> - Mon, 31 Jul 2017 01:36:43 +0530 - rev 618564
Push 71380 by nhnt11@gmail.com at Mon, 31 Jul 2017 18:39:21 +0000
Bug 1355455 - Replace shared icons with Photon versions. r=dao MozReview-Commit-ID: EpXQOAFjsds
5f6041cf802f0d6320453d51d6d134a387010e3e: Bug 1385407 - Add an extra selector for the urlbar-icon within the star-button-box now that the urlbar-icon for the bookmark star isn't a direct child of urlbar-icons. r?gijs draft
Jared Wein <jwein@mozilla.com> - Mon, 31 Jul 2017 11:39:05 -0400 - rev 618563
Push 71379 by bmo:jaws@mozilla.com at Mon, 31 Jul 2017 18:34:03 +0000
Bug 1385407 - Add an extra selector for the urlbar-icon within the star-button-box now that the urlbar-icon for the bookmark star isn't a direct child of urlbar-icons. r?gijs This change in markup was necessary to implement bug 1352063. I have audited all other CSS to make sure other selectors looking for this icon still apply. This was the only one found that needed to be updated. The change to browser_startup_images.js is not actually a new image that is loaded at startup, but a revert of 767224f031ac (bug 1352063) which removed that section (the test failing due to that section no longer applying should have called out what was necessary to fix this bug). MozReview-Commit-ID: 6O1iUUjJ0dh
2945e67416de4dfc66db6d8329555017d4835559: Bug 1373739 - Make macOS compatible with the headless WPT changes draft
Michael Smith <michael@spinda.net> - Fri, 28 Jul 2017 23:20:21 -0700 - rev 618562
Push 71378 by bmo:lists@spinda.net at Mon, 31 Jul 2017 18:33:52 +0000
Bug 1373739 - Make macOS compatible with the headless WPT changes MozReview-Commit-ID: BeDPpG6G6Iy
97084f51284d53ad8ff85a99943cc20bf7475ad0: Bug 1373739 - Set up automated test running for WPT in headless mode. r?jgraham draft
Michael Smith <michael@spinda.net> - Wed, 21 Jun 2017 15:14:05 -0700 - rev 618561
Push 71378 by bmo:lists@spinda.net at Mon, 31 Jul 2017 18:33:52 +0000
Bug 1373739 - Set up automated test running for WPT in headless mode. r?jgraham Currently these are enabled for Linux and Windows, using the corresponding WPT symbols with an "H" suffix to match the Mn->MnH convention followed by the Marionette tests. For Linux, the tests are set to run with a simulated 1600x1200 screen resolution, which matches the Docker configuration. The default 1366x768 for headless mode results in some unexpected failures due to windows not being able to reach the full sizes the web platform tests expect. For Windows, the simulated screen resolution is set to 1024x768, which is what I've determined to be the resolution that Windows VMs run under within our testing infrastructure. Setting a different screen resolution causes some web platform tests to unexpectedly pass (in particular, several in open-features-non-integer-height.html). A "headless" variable is added to the environment that the WPT metadata (.ini) files can access, indicating whether or not the test is running in headless mode. This is used to mark the reftest bdi-neutral-wrapped.html as an expected pass on Linux under headless mode: it normally fails (spuriously) due to differences in scrollbar sizing, but scrollbars aren't rendered in headless mode. MozReview-Commit-ID: 3usazOoxx3q
b7e40b5a98be92a12a225afe624b16b702b07d94: Bug 1373739 - Headless: use native look-and-feel on Win, hardcoded on Linux. r?bdahl draft
Michael Smith <michael@spinda.net> - Wed, 05 Jul 2017 19:41:21 -0700 - rev 618560
Push 71378 by bmo:lists@spinda.net at Mon, 31 Jul 2017 18:33:52 +0000
Bug 1373739 - Headless: use native look-and-feel on Win, hardcoded on Linux. r?bdahl MozReview-Commit-ID: 3CB1n7ccy86
589b1cafc69b75deac53a790878dfda0d72d3a83: Bug 1373739 - Hook HeadlessSound and HeadlessScreenHelper into Windows widgets. r?bdahl draft
Michael Smith <michael@spinda.net> - Tue, 11 Jul 2017 14:01:00 -0700 - rev 618559
Push 71378 by bmo:lists@spinda.net at Mon, 31 Jul 2017 18:33:52 +0000
Bug 1373739 - Hook HeadlessSound and HeadlessScreenHelper into Windows widgets. r?bdahl MozReview-Commit-ID: 1KOlPyLmJaY
766dc502d40d64ac41bcc4b3b3c207ee8838e7ee: Bug 1373739 - Make headless compositing Windows-compatible, in addition to Linux. r?dvander draft
Michael Smith <michael@spinda.net> - Thu, 06 Jul 2017 17:45:34 -0700 - rev 618558
Push 71378 by bmo:lists@spinda.net at Mon, 31 Jul 2017 18:33:52 +0000
Bug 1373739 - Make headless compositing Windows-compatible, in addition to Linux. r?dvander To make the HeadlessCompositorWidget work under Windows as well as Linux, I had to change the way that I hooked it into the existing CompositorWidget system. Under GTK, the CompositorWidgetInitData and CompositorWidgetDelegate types provided the information needed by the headless compositor widget already (the widget client size). On Windows, however, the definitions of these types differ, and the client size information is simply retrieved from the platform APIs when needed. After this patch, CompositorWidgetDelegate is renamed to PlatformCompositorWidgetDelegate, and a new base class called CompositorWidgetDelegate is added with "AsPlatformSpecificDelegate()" and "AsHeadlessCompositorWidget()" methods. In non-headless mode, widgets use AsPlatformSpecificDelegate() to access the Windows- and GTK-specific delegate APIs. In headless mode, AsHeadlessCompositorWidget() is used to access the singular CompositorWidget implementation for headless. Meanwhile, the CompositorWidgetInitData IPDL type is made into a union which always contains a headless-specific HeadlessCompositorWidgetInitData struct and under GTK and Windows also contains an {X11,Win}CompositorWidgetInitData struct. This also includes a small patch to ensure that the GPU process and hardware-accelerated compositing are always disabled under headless mode. These features weren't activated by default in the Linux environments I tested in, but did end up activating (and then promptly crashing Firefox) when I tested on Windows. MozReview-Commit-ID: CocPoHBDV7H
59ea624757e29f9afe25d06f74b27982958f283b: Bug 1373739 - Make headless widgets hidden by default, matching other platforms. r?bdahl draft
Michael Smith <michael@spinda.net> - Tue, 11 Jul 2017 13:57:25 -0700 - rev 618557
Push 71378 by bmo:lists@spinda.net at Mon, 31 Jul 2017 18:33:52 +0000
Bug 1373739 - Make headless widgets hidden by default, matching other platforms. r?bdahl It seems that my reworking of how the z-index events are sent in the patch pertaining to active window tracking made it so that we can have headless widgets default to being invisible again, which matches the behavior on all other platforms. This then allows us to avoid applying size mode change side effects while a window/widget is invisible, matching the behavior on Windows and allowing the relevant web platform tests to pass despite the low screen resolution with which we run Windows VMs in testing (eg. open-features-non-integer-height.html). MozReview-Commit-ID: 3xnaEn7RImP
429c505d782d2279c10a5eac7f3aca68e05e65f1: Bug 1373739 - Enable Marionette "screenshot" tests in headless. r?bdahl draft
Michael Smith <michael@spinda.net> - Thu, 06 Jul 2017 13:01:17 -0700 - rev 618556
Push 71378 by bmo:lists@spinda.net at Mon, 31 Jul 2017 18:33:52 +0000
Bug 1373739 - Enable Marionette "screenshot" tests in headless. r?bdahl These pass now (on my machine) after filling in look & feel values for headless mode. MozReview-Commit-ID: DpT74JbcMoF
e9617df237d28b7b6331ab7a323d61a17585f70c: Bug 1373739 - Use ClientLayerManager in headless mode. r?jrmuizel draft
Michael Smith <michael@spinda.net> - Wed, 21 Jun 2017 08:50:02 -0700 - rev 618555
Push 71378 by bmo:lists@spinda.net at Mon, 31 Jul 2017 18:33:52 +0000
Bug 1373739 - Use ClientLayerManager in headless mode. r?jrmuizel This fixes a series of issues uncovered by the Web Platform Tests. The most immediately noticeable symptom was that, during shutdown, PuppetWidgets on the client side attempted to delete shadow layers that were never created on the compositor side. The usage of BasicLayerManager in all other widgets due to headless mode meant that the PLayerTransactionParent was never initialized with a layer manager, and thus discarded all transaction messages it received but not delete layer messages. The effects of only using BasicLayerManager in headless mode also showed up in the web platform reftests, which ended up as blank white boxes in e10s mode as the compositor thread never received paint instructions. Switching over to using ClientLayerManager in headless mode causes these paint instructions to be relayed. In order to make ClientLayerManager work under headless mode, it was necessary to implement a HeadlessCompositorWidget and hook that up to the CompositorWidget creation function in widget/gtk. A follow-up patch will be necessary to hook up the same for the other supported widget platforms. MozReview-Commit-ID: 8vB3lrxP7iX
f16e32e298696e4a4ee04e5c7cc555ca15fc3e8f: Bug 1373739 - Re-enable chrome-switching tests on headless mode. r?bdahl draft
Michael Smith <michael@spinda.net> - Tue, 27 Jun 2017 16:42:41 -0700 - rev 618554
Push 71378 by bmo:lists@spinda.net at Mon, 31 Jul 2017 18:33:52 +0000
Bug 1373739 - Re-enable chrome-switching tests on headless mode. r?bdahl MozReview-Commit-ID: He19Xcq88vo
d3b6b2c41098a257e91d0752afb072f579e87584: Bug 1373739 - Simulate window activation events in headless mode. r?jrmuizel draft
Michael Smith <michael@spinda.net> - Wed, 21 Jun 2017 08:51:39 -0700 - rev 618553
Push 71378 by bmo:lists@spinda.net at Mon, 31 Jul 2017 18:33:52 +0000
Bug 1373739 - Simulate window activation events in headless mode. r?jrmuizel This is necessary for focus events to propagate correctly. Otherwise, /2dcontext/drawing-paths-to-the-canvas/drawFocusIfNeeded_004.html in the Web Platform Tests fails, for example. MozReview-Commit-ID: 69GItIaQWfr
3a12a8935934f4093ab273e8eb20ddf5b2c467a7: Bug 1373739 - Apply thumb styles directly to their orientation variants. r?mossop draft
Michael Smith <michael@spinda.net> - Thu, 22 Jun 2017 17:45:50 -0700 - rev 618552
Push 71378 by bmo:lists@spinda.net at Mon, 31 Jul 2017 18:33:52 +0000
Bug 1373739 - Apply thumb styles directly to their orientation variants. r?mossop Previously the min-height style from vertical thumbs was allowed to cascade down to horizontal thumbs. This caused broken scrollbar sizing behavior when running without a native (platform/widget toolkit) theme: horizontal scrollbars would have a non-zero height and width while vertical scrollbars would have a zero width and non-zero height. This broke one of the web platform tests in headless mode (/intersection-observer/root-margin.html). MozReview-Commit-ID: 4Y3jz77Z0UA
322c7fd08429ca92f49acaf0337a2d28bee710a0: Bug 1373739 - Constrain widget size to screen size in headless mode. r?jrmuizel draft
Michael Smith <michael@spinda.net> - Thu, 22 Jun 2017 15:31:32 -0700 - rev 618551
Push 71378 by bmo:lists@spinda.net at Mon, 31 Jul 2017 18:33:52 +0000
Bug 1373739 - Constrain widget size to screen size in headless mode. r?jrmuizel Every other widget backend ensures widget size is bounded to screen size, but this check was missing from headless. MozReview-Commit-ID: 6bKIAkdRxoO
(0) -300000 -100000 -30000 -10000 -3000 -1000 -300 -100 -50 -20 +20 +50 +100 +300 +1000 +3000 +10000 +30000 +100000 tip