- 15 Aug, 2018 40 commits
-
-
John Budorick authored
Currently, luci-logdog-dev isn't letting anything write to project chromium. Bug: 868525 Change-Id: I970ccc0b6959f3d5830ca007df963a6edaf6693a Reviewed-on: https://chromium-review.googlesource.com/1176079Reviewed-by:
Nodir Turakulov <nodir@chromium.org> Commit-Queue: John Budorick <jbudorick@chromium.org> Cr-Commit-Position: refs/heads/master@{#583301}
-
Siddhartha authored
The test fails at shutdown because of contextual_suggestions.EnabledStateMonitor trying to access the ProfileSyncService at shutdown. So, restore the original service at tearDown. Change-Id: I02e7f14caf44ed9c0d4dc5a70d0f353a4428e34f Reviewed-on: https://chromium-review.googlesource.com/1170311 Commit-Queue: Siddhartha S <ssid@chromium.org> Reviewed-by:
Theresa <twellington@chromium.org> Cr-Commit-Position: refs/heads/master@{#583300}
-
Kristi Park authored
This reverts commit 31f1f7d3. Reason for revert: Bypasses the form's required fields when clicking "Done" Original change's description: > [NTP] Fix cancel button submitting when adding custom link > > "Cancel" was calling the form's submit event, which triggered the add > link flow. > > Bug: 874018 > Change-Id: Iad822d3884484f5dc0d9f3e6667c9d4c9e0a2728 > Reviewed-on: https://chromium-review.googlesource.com/1174908 > Reviewed-by: Fernando Serboncini <fserb@chromium.org> > Commit-Queue: Kristi Park <kristipark@chromium.org> > Cr-Commit-Position: refs/heads/master@{#583004} TBR=fserb@chromium.org,kristipark@chromium.org Change-Id: Ie605d0a5d5a5043fbb76d30eb98b7565818e41f7 No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: 874018 Reviewed-on: https://chromium-review.googlesource.com/1174775 Commit-Queue: Kristi Park <kristipark@chromium.org> Reviewed-by:
Kristi Park <kristipark@chromium.org> Cr-Commit-Position: refs/heads/master@{#583299}
-
skia-chromium-autoroll authored
https://skia.googlesource.com/skia.git/+log/137ca523a36e..c97a339cd601 git log 137ca523a36e..c97a339cd601 --date=short --no-merges --format='%ad %ae %s' 2018-08-15 ruiqimao@google.com added optimizations to speed up skinning Created with: gclient setdep -r src/third_party/skia@c97a339cd601 The AutoRoll server is located here: https://autoroll.skia.org Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+/master/autoroll/README.md If the roll is causing failures, please contact the current sheriff, who should be CC'd on the roll, and stop the roller if necessary. CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel;luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel TBR=kjlubick@chromium.org Change-Id: I53a79a12a6554d57de6157d9d305c3829305a31b Reviewed-on: https://chromium-review.googlesource.com/1175293Reviewed-by:
skia-chromium-autoroll <skia-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com> Commit-Queue: skia-chromium-autoroll <skia-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com> Cr-Commit-Position: refs/heads/master@{#583298}
-
Ella Ge authored
pointerevent fractional coordinates was shipped from M64, but offsetx/y were left behind. This CL makes them aligned with other coordinates, expose fractions for pointerevents and rounded version for mouseevents. Intent to Ship: https://groups.google.com/a/chromium.org/d/msg/blink-dev/tl2sOodhtCs/oe0ZMNXPCAAJ Bug: 867609 Change-Id: I31af4ce01fa820e3281932e0b0a7e8bd5a6c0042 Reviewed-on: https://chromium-review.googlesource.com/1151534 Commit-Queue: Ella Ge <eirage@chromium.org> Reviewed-by:
Dave Tapuska <dtapuska@chromium.org> Reviewed-by:
Navid Zolghadr <nzolghadr@chromium.org> Cr-Commit-Position: refs/heads/master@{#583297}
-
Scott Violet authored
Rather than Env::GetInstance(). This ensures the right instance is used. BUG=847992 TEST=covered by tests Change-Id: I9b75458c1e4d21402ef2e58833c57ff9a289b7a0 Reviewed-on: https://chromium-review.googlesource.com/1176062Reviewed-by:
Michael Wasserman <msw@chromium.org> Commit-Queue: Scott Violet <sky@chromium.org> Cr-Commit-Position: refs/heads/master@{#583296}
-
John Abd-El-Malek authored
Bug: 859594 Cq-Include-Trybots: luci.chromium.try:linux_mojo Change-Id: I5026f51b6bc005215827de6d8d4a15b870b42cf4 Reviewed-on: https://chromium-review.googlesource.com/1175981Reviewed-by:
Matt Menke <mmenke@chromium.org> Commit-Queue: John Abd-El-Malek <jam@chromium.org> Cr-Commit-Position: refs/heads/master@{#583295}
-
Max Moroz authored
This instrumentation flag is recommended by libFuzzer authors, as it may enable various different instrumentation options which are actual and recommended for use. The flag is implemented in clang, so we won't have to update individual instrumentation flags anymore. See crbug.com/764514 and https://github.com/google/oss-fuzz/issues/832 AFL still supports only trace-pc-guard instrumentation, so we have to use separate compilation flags for AFL and libFuzzer, the same way as we do in OSS-Fuzz. Previous reverts: 1) https://chromium-review.googlesource.com/c/chromium/src/+/852826 The issue with sancov report generation is not resolved yet. We can either land this and break coverage on CF for some time, until I resolve crbug.com/818467 OR we can wait until crbug.com/818467 gets resolved first. I'd prefer landing this now and fixing coverage on CF after that. 2) https://chromium-review.googlesource.com/c/chromium/src/+/1087431 The issue with linker trying to link in symbols which were not used and thus were not properly instrumented has been reported to the sanitizers project and fixed on LLVM side: https://github.com/google/sanitizers/issues/971. 3) https://chromium-review.googlesource.com/c/chromium/src/+/1110348 Dead code stripping led to libFuzzer complaining about a mismatch of the size of coverage PC tables: https://crbug.com/856239. Fixed upstream: https://reviews.llvm.org/rL336941. Bug: 764514 Change-Id: Ic19a6c27ee37143f3ca126e3bc57c9182e962750 Reviewed-on: https://chromium-review.googlesource.com/1175851Reviewed-by:
Abhishek Arya <inferno@chromium.org> Reviewed-by:
Jonathan Metzman <metzman@chromium.org> Commit-Queue: Max Moroz <mmoroz@chromium.org> Cr-Commit-Position: refs/heads/master@{#583294}
-
Xida Chen authored
TBR=wjmaclean@chromium.org NOTRY=true Bug: 874500 Change-Id: I16f3b600c597b3243c5ad5723b7f845cc0e76e61 Reviewed-on: https://chromium-review.googlesource.com/1176080Reviewed-by:
Xida Chen <xidachen@chromium.org> Commit-Queue: Xida Chen <xidachen@chromium.org> Cr-Commit-Position: refs/heads/master@{#583293}
-
Annie Sullivan authored
These stories depend on tabs and/or omnibox. The fact that they don't run on WebView is intentional, not a bug. So remove them from expectations and clarify that they shouldn't run on WebView in the code. Bug: 865464,865466,865471,867568 Change-Id: I9bbd6a4a87aaa423bdf8d106d3de2d64cf6c8d52 Reviewed-on: https://chromium-review.googlesource.com/1174700 Commit-Queue: Annie Sullivan <sullivan@chromium.org> Commit-Queue: Juan Antonio Navarro Pérez <perezju@chromium.org> Reviewed-by:
Juan Antonio Navarro Pérez <perezju@chromium.org> Reviewed-by:
Ned Nguyen <nednguyen@google.com> Cr-Commit-Position: refs/heads/master@{#583292}
-
Patrick Noland authored
Change-Id: I20cd1fe7eeb2eac30d7913599a8dc1285ae9023e ******************************Resource Sizes Diff****************************** Monochrome.apk_Dex +1,728 entries fields +2,366 entries methods +581 entries types +1,193 entries strings Monochrome.apk_Breakdown (+169,703 bytes) +3,241 bytes Zip Overhead +1,510 bytes Non-compiled Android resources size +12,528 bytes Native code size +129 bytes Package metadata size +148,695 bytes Java code size +3,600 bytes Compiled Android resources size Monochrome.apk_Specifics +523,907 bytes normalized apk size +3 zip entries file count +12,528 bytes main lib size +354,204 bytes main dex size Change-Id: I20cd1fe7eeb2eac30d7913599a8dc1285ae9023e Reviewed-on: https://chromium-review.googlesource.com/1175218Reviewed-by:Filip Gorski <fgorski@chromium.org> Commit-Queue: Patrick Noland <pnoland@chromium.org> Cr-Commit-Position: refs/heads/master@{#583291}
-
Xiaocheng Hu authored
The test currently asserts that, in list-based hit test results, floats precede other descendants of the block container. However, according to spec: - Floats are painted earlier than other descendants [1] - Elements should be in reverse paint order in hit test result [2] Hence, this patch changes the test to assert that non-float descendants come first in hit test result. It also marks it failure in legacy and pass in LayoutNG. [1] https://www.w3.org/TR/CSS22/zindex.html#painting-order [2] https://drafts.csswg.org/cssom-view/#dom-document-elementsfrompoint Cq-Include-Trybots: luci.chromium.try:linux_layout_tests_layout_ng Change-Id: I8778d68b4d47f616a653e51f745c10c6b96cfe72 Reviewed-on: https://chromium-review.googlesource.com/1173398Reviewed-by:
Koji Ishii <kojii@chromium.org> Reviewed-by:
Emil A Eklund <eae@chromium.org> Commit-Queue: Emil A Eklund <eae@chromium.org> Cr-Commit-Position: refs/heads/master@{#583290}
-
Xida Chen authored
Flakiness dashboard shows that it has been flaky for a while TBR=wjmaclean@chromium.org NOTRY=true Bug: 874497 Change-Id: I017096a3a33ba69cab80cf9f78caa59bd4672319 Reviewed-on: https://chromium-review.googlesource.com/1175939Reviewed-by:
-
Dominik Röttsches authored
Preparation for hooking up Blink on Android to the font unique name matching service in RenderProcessHost. Bug: 870223 Change-Id: I89bd326e78dbd1bdf989265ca6f8820f8d242621 Reviewed-on: https://chromium-review.googlesource.com/1160648Reviewed-by:
Avi Drissman <avi@chromium.org> Reviewed-by:
Philip Jägenstedt <foolip@chromium.org> Commit-Queue: Dominik Röttsches <drott@chromium.org> Cr-Commit-Position: refs/heads/master@{#583288}
-
Scott Violet authored
MASH_DEPRECATED is never set anymore. BUG=640773 TEST=none Change-Id: Iab913023f56e7f832cb987b10b5725d55ccc1ead Reviewed-on: https://chromium-review.googlesource.com/1176072Reviewed-by:
Xiyuan Xia <xiyuan@chromium.org> Commit-Queue: Scott Violet <sky@chromium.org> Cr-Commit-Position: refs/heads/master@{#583287}
-
Scott Violet authored
This fails in single-process mash mode because it uses EventMonitorAura, which uses Env::GetInstance(). BUG=874481 TEST=test only change Change-Id: I1f71b68db1a1b52c8e1c2ecc6c87828eea131700 Reviewed-on: https://chromium-review.googlesource.com/1176067 Commit-Queue: Scott Violet <sky@chromium.org> Reviewed-by:
-
Dominic Mazzoni authored
Bug: 852735 Change-Id: I56752deea548f5aa01f9ae3d71f3551841737809 Reviewed-on: https://chromium-review.googlesource.com/1175489Reviewed-by:
Alice Boxhall <aboxhall@chromium.org> Commit-Queue: Dominic Mazzoni <dmazzoni@chromium.org> Cr-Commit-Position: refs/heads/master@{#583285}
-
Xida Chen authored
TBR=wjmaclean@chromium.org NOTRY=true Bug: 874491 Change-Id: Idda7c1bbd55468c33dc6bf2aaafceeb16eb699c5 Reviewed-on: https://chromium-review.googlesource.com/1176075Reviewed-by:
-
Quan Nguyen authored
Change-Id: I24e7febdedac576c7a8e77c5412b657383e065bd Reviewed-on: https://chromium-review.googlesource.com/1176064Reviewed-by:
Jacob Dufault <jdufault@chromium.org> Commit-Queue: Quan Nguyen <qnnguyen@chromium.org> Cr-Commit-Position: refs/heads/master@{#583283}
-
Weidong Guo authored
Changes: In homecher, folder or search result page is closed when an app is launched or a search result is opened. Bug: 853846 Change-Id: I92edf732b01bb9c5b0f97ff2206bd09b2b6130ca Reviewed-on: https://chromium-review.googlesource.com/1175417 Commit-Queue: Weidong Guo <weidongg@chromium.org> Reviewed-by:
-
depot-tools-chromium-autoroll authored
https://chromium.googlesource.com/chromium/tools/depot_tools.git/+log/52c721180793..af03ae194f78 git log 52c721180793..af03ae194f78 --date=short --no-merges --format='%ad %ae %s' 2018-08-15 recipe-roller@chromium.org Roll recipe dependencies (trivial). Created with: gclient setdep -r src/third_party/depot_tools@af03ae194f78 The AutoRoll server is located here: https://depot-tools-chromium-roll.skia.org Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+/master/autoroll/README.md If the roll is causing failures, please contact the current sheriff, who should be CC'd on the roll, and stop the roller if necessary. TBR=agable@chromium.org Change-Id: I567c98c6cbdae16c6b753b9dcff1cd958d3a7d60 Reviewed-on: https://chromium-review.googlesource.com/1174805Reviewed-by:
depot-tools-chromium-autoroll <depot-tools-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com> Commit-Queue: depot-tools-chromium-autoroll <depot-tools-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com> Cr-Commit-Position: refs/heads/master@{#583281}
-
Robert Ogden authored
This CL just adds the strings and drawables for the Previews badge. The actual triggering and interaction with the native code will come in another CL. Bug: 871839 Change-Id: Iad6d5543492bf78e9b8f54e525d979b66217eefb Reviewed-on: https://chromium-review.googlesource.com/1173266Reviewed-by:
Ted Choc <tedchoc@chromium.org> Reviewed-by:
Ryan Sturm <ryansturm@chromium.org> Commit-Queue: Robert Ogden <robertogden@chromium.org> Cr-Commit-Position: refs/heads/master@{#583280}
-
bsheedy authored
Adds render tests for JavaScript confirm and prompt dialogs in the VR browser. Additionally, cleans the test file up a bit and adds the golden image directory as a data dependency so that the tests will actually fail on a mismatch instead of constantly producing new golden images. Bug: 810847 Change-Id: I6e4ea25cf6fb84f73ed3a7d90fb02eb5adb9e783 Reviewed-on: https://chromium-review.googlesource.com/1174990Reviewed-by:
Amirhossein Simjour <asimjour@chromium.org> Commit-Queue: Brian Sheedy <bsheedy@chromium.org> Cr-Commit-Position: refs/heads/master@{#583279}
-
Scott Violet authored
NativeWidgetPrivate defined three related functions for show: . Show() . ShowMaximizedWithBounds() . ShowWindowWithState() The actual implementation of these generally funnelled into a single function. To make matters equally interesting DesktopWindowTreeHost had these variants: 1. WindowTreeHost::Show(): this comes by virture of DesktopWindowTreeHost having (or more commonly subclassing) a WindowTreeHost. 2. ShowWindowWithState() 3. ShowMaximizedWithBounds() As mentioned, 1 comes from WindowTreeHost. WindowTreeHost::Show() invokes the pure virtual method WindowTreeHost::ShowImpl(). The implementation of ShowImpl() is typically taken to mean show with a state of ui::SHOW_STATE_NORMAL. Because ShowImpl() is taken to mean SHOW_STATE_NORMAL, the implementations of DesktopWindowTreeHost::ShowWindowWithState()/ShowMaximizedWithBounds() can *not* call WindowTreeHost::Show() (otherwise we would end up in a loop of Show() -> ShowImpl()-> ShowWindowWithState() -> Show()...). This leads to the following pattern: 1. DesktopWindowTreeHost implementations of Show* do not call WindowTreeHost, but do the work in WindowTreeHost::Show(). 2. DesktopWindowTreeHost::ShowImpl() calls Show with a particular state. In an attempt to simplify things I've introduced a single function on NativeWidgetPrivate and DesktopWindowTreeHost. There is still the quirk that the code deals with WindowTreeHost::Show() being called outside of through Widget. This is unfortunate, but I kept it. I also made each DesktopWindowTreeHost implementation responsible for showing the appropriate windows. This means we don't end up with the double show that was causing problems. I also added some minor debugging code to ash. My knee jerk reaction is this pain/complexity is because WindowTreeHost has a public Show() function. If WindowTreeHost did not have a Show() function this code would be slightly less subtle. Rethinking WindowTreeHost::Show() is for another day. BUG=873701 TEST=covered by tests Change-Id: Id3525e7bc61d741cd7239fc466ca0ac16a19eab6 Reviewed-on: https://chromium-review.googlesource.com/1174910 Commit-Queue: Scott Violet <sky@chromium.org> Reviewed-by:
-
Fredrik Söderquist authored
Set 'mix-blend-mode' to 'normal' on the stroked rects that make up the "edge cover". Hopefully this avoids aliasing artifacts on the edge cover itself. In blending-002.svg, localize the use of (non-auto) isolation to the content of the test. BUG=866928 Change-Id: I269e01a4d814f7fd8e202ffdf71c1a1d5dc48347 Reviewed-on: https://chromium-review.googlesource.com/1175825Reviewed-by:
Stephen Chenney <schenney@chromium.org> Commit-Queue: Fredrik Söderquist <fs@opera.com> Cr-Commit-Position: refs/heads/master@{#583277}
-
Joe Downing authored
This change prevents a hang when connecting to a Windows machine which has curtain mode and 3d display mode enabled. The root cause of the hang is that creating a D3D11 device takes several seconds so function calls which used to take a few milliseconds now take 5+ seconds. I did not see a mechanism to control this display mode programatically so I am adding code which can detect the condition and the logic to prevent D3D usage when 3D mode is enabled. Of note is the change in BasicDesktopEnvironment where I requery the D3D apis. DesktopEnvironmentOptions are queried in Session 0 and then passed on to the Desktop process. There is a non-obvious problem with this as many D3D methods cannot be queried or will not return reliable results. This results in a base set of options being set in Session 0 (which includes user / experiment overrides) and then passed into a session where D3D will return new values. In this CL, I requery the D3D APIs and update the capturer options if they should be disabled. BUG=836007 Change-Id: Ib7a6e61fa7987bd093e93a6100b97cc49e3aff05 Reviewed-on: https://chromium-review.googlesource.com/1164403 Commit-Queue: Joe Downing <joedow@chromium.org> Reviewed-by:
Jamie Walch <jamiewalch@chromium.org> Cr-Commit-Position: refs/heads/master@{#583276}
-
Tibor Goldschwendt authored
Bug: 873826 Change-Id: I9381fb88efd83933e30d5c14050f50500f241469 Reviewed-on: https://chromium-review.googlesource.com/1175084Reviewed-by:
Michael Thiessen <mthiesse@chromium.org> Reviewed-by:
-
Xida Chen authored
See failure here: https://ci.chromium.org/buildbot/chromium.webkit/WebKit%20Mac10.13%20%28retina%29/1465 TBR=wjmaclean@chromium.org NOTRY=true Bug: 869726 Change-Id: Id5e2c77e6c6eb001dbd2c8fcd1b9d5e3df02fb66 Reviewed-on: https://chromium-review.googlesource.com/1176061Reviewed-by:
-
Peter E Conn authored
Bug: 873178 Change-Id: I700d6d710474b356e32f58b435d78e7e24cfb0ff Reviewed-on: https://chromium-review.googlesource.com/1174537Reviewed-by:
Paul Jensen <pauljensen@chromium.org> Reviewed-by:
Bernhard Bauer <bauerb@chromium.org> Commit-Queue: Peter Conn <peconn@chromium.org> Cr-Commit-Position: refs/heads/master@{#583273}
-
Amirhossein Simjour authored
API is upgraded. Optional method calls are updated to check for nullptr first. Change-Id: Ibfe5bfc6b8e1f34e138ec3b9efcab981911d8491 Reviewed-on: https://chromium-review.googlesource.com/1175098Reviewed-by:
Aldo Culquicondor <acondor@chromium.org> Commit-Queue: Amirhossein Simjour <asimjour@chromium.org> Cr-Commit-Position: refs/heads/master@{#583272}
-
Xida Chen authored
See failure here: https://ci.chromium.org/buildbot/chromium.webkit/WebKit%20Mac10.13%20%28retina%29/1461 TBR=wjmaclean@chromium.org NOTRY=true Bug: 869726 Change-Id: Ia74cfb2b6dfcbf5b7ca2b331701d6e2453d9090d Reviewed-on: https://chromium-review.googlesource.com/1175849 Commit-Queue: Xida Chen <xidachen@chromium.org> Reviewed-by:
-
sawtelle authored
Currently, assigning custom controls to a video through a call to setPictureInPictureCustomControls() passes in the data but does not show the buttons on the PiP window. This CL makes it so that up to 2 buttons are added to the PiP window on the left and the right of the play/pause button when this call is made. Note: the desired icon that is passed in is not currently being assigned to these new buttons, and they are currently using the play and pause icons as defaults, for left and right respectively. This will be updated to work properly in a future patch - this is just to get the basic functionality set up. Bug: 864271 Change-Id: I42d9cf47dcd22dd481675b1990e1bb9ae2265055 Reviewed-on: https://chromium-review.googlesource.com/1139110 Commit-Queue: Sadrul Chowdhury <sadrul@chromium.org> Reviewed-by:
apacible <apacible@chromium.org> Reviewed-by:
Mounir Lamouri <mlamouri@chromium.org> Cr-Commit-Position: refs/heads/master@{#583270}
-
Mike Reed authored
Refactoring to support https://skia-review.googlesource.com/c/skia/+/147113 Change-Id: Ic4d430ed29b4edab3febab4a4fe811881b5617e1 Reviewed-on: https://chromium-review.googlesource.com/1175438Reviewed-by:
Evan Stade <estade@chromium.org> Commit-Queue: Mike Reed <reed@chromium.org> Cr-Commit-Position: refs/heads/master@{#583269}
-
skia-chromium-autoroll authored
https://skia.googlesource.com/skia.git/+log/49bb9c17d713..137ca523a36e git log 49bb9c17d713..137ca523a36e --date=short --no-merges --format='%ad %ae %s' 2018-08-15 robertphillips@google.com Remove prePathMatrix from SkDevice::drawPath 2018-08-15 csmartdalton@google.com Consolidate the scissor enums into one "GrScissorTest" enum Created with: gclient setdep -r src/third_party/skia@137ca523a36e The AutoRoll server is located here: https://autoroll.skia.org Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+/master/autoroll/README.md If the roll is causing failures, please contact the current sheriff, who should be CC'd on the roll, and stop the roller if necessary. CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel;luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel TBR=kjlubick@chromium.org Change-Id: I818d430b5622960332e554fd4155b3e3cdb9bc3a Reviewed-on: https://chromium-review.googlesource.com/1174803Reviewed-by:
-
Alexander Timin authored
In addition to existing limit of 200 messages in a single task also limit it to 50 milliseconds to avoid starvation. This logic will be removed when message-per-task mojo dispatching will be implemented. This patch also modifies maps_pixel_test to use requestIdleCallback for detecting the end of the page load instead of three requestAnimationFrames. BUG=867133,872239 R=mek@chromium.org,kbr@chromium.org Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel Change-Id: Ib2de605bc6d069c23c3a991edcc0dccc995a4632 Reviewed-on: https://chromium-review.googlesource.com/1167053 Commit-Queue: Alexander Timin <altimin@chromium.org> Reviewed-by:
Kenneth Russell <kbr@chromium.org> Reviewed-by:
Marijn Kruisselbrink <mek@chromium.org> Cr-Commit-Position: refs/heads/master@{#583267}
-
Elad Alon authored
The documentation of UploadList's ctor mentions a parameter which the ctor no longer appears to take. Bug: None Change-Id: I8656b0c76d684ee87c1baa26cd79104c7b50b358 Reviewed-on: https://chromium-review.googlesource.com/1174382Reviewed-by:
Scott Violet <sky@chromium.org> Commit-Queue: Elad Alon <eladalon@chromium.org> Cr-Commit-Position: refs/heads/master@{#583266}
-
Maksim Sisov authored
This patch makes it possible to run Wayland backend with a separate gpu process. That is, WaylandConnection is not available on the GPU process side and so buffers cannot be created that way. To overcome this, a Wayland dmabuf based approach is used: - On the gpu process side, gl::Surfaceless is used, which is implemented by GbmSurfaceless. It differs from the one provided by the Ozone/Gbm backend (the difference is that there are no frame buffers, because DRM render node is used instead.) Basically, on the GPU side, a gbm_bo buffer is created and its file descriptor is shared with the WaylandConnection, which then imports a wl_buffer and attaches it to a requested surface. TL:DR; - There are two communicating parties - WaylandConnection, which establishes a connection with Wayland, and a WaylandConnectionProxy, which serves the GPU process with a forwarding mojo connection to the WaylandConnection. In single process mode, WaylandConnectionProxy just forwards calls to WaylandConnection, and GLSurfaceEGL is used. In a multi process mode, the l::Surfaceless and dmabuf based approach is used instead. - The communication is established in the following way: when WaylandConnectionConnector receives a OnGpuServiceLaunched, it binds WaylandConnection and WaylandConnectionProxy interfaces, and calls to WaylandConnectionProxy to set a mojo ptr to the WaylandConnection. The WaylandConnectionProxy receives the pointer, stores it and also stores a task runner to make sure mojo is called on the right sequence. How buffers are created: - Basically, the GPU process doesn't know about Wayland. What it does is that it creates a GbmBuffer utilizing a DRM render node (it can't use a drm card device, because authentication requires master access), takes a gbm_bo file descriptor and shares it via a mojo call with the WaylandConnection, which then uses the zwp_linux_dmabuf interface to import a wl_buffer, which can be attached to a needed WaylandWindow and commit-ed to show contents. That is, the gbm framework is used to share buffers between the browser and the GPU process, where the browser process can read from them and the GPU process is able to read and write using gl commands. How buffers are committed: - GbmSurfaceless uses a normal procedure and submits frames on SwapBuffersAsync call: a frame is created and scheduled as an overlay plane, once scheduling is done, a widget is taken from that pending plane and WaylandConnection::ScheduleBufferSwap is called. In order to specify, which wl_buffer must be used, gbm bo handle is used as a buffer id. Once the WaylandConnection receives a buffer swap call, it finds a buffer by a buffer id and attaches that buffer to a WaylandWindow, which is also found by a widget passed to it. PresentationCallback is not used and will be enabled in the future. The missing feature is the software rasterization. That is, the gbm buffers cannot be mmap'ed using unix mmap method. Thus, it's impossible to get a mapped memory range and let the cpu write to it. I think this problem can be resolved in the downstream projects in different ways but these solutions are not generic enough to be upstreamed. Another missing feature is presentation feedback. This will be implemented in follow up CLs. COMPILE= compiles only with third_party minigbm now. For example, add use_intel_minigbm=true if intel gpu is ozone_platform_wayland=true. Using system libgbm will be available soon. TEST=run and compile ozone/wayland. Bug: 820047 Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel Change-Id: Id351edd9883a765d3e549543a50175299108c6ab Reviewed-on: https://chromium-review.googlesource.com/1057468Reviewed-by:
Antoine Labour <piman@chromium.org> Reviewed-by:
Ilya Sherman <isherman@chromium.org> Reviewed-by:
Tom Sepez <tsepez@chromium.org> Reviewed-by:
Robert Kroeger <rjkroege@chromium.org> Commit-Queue: Maksim Sisov <msisov@igalia.com> Cr-Commit-Position: refs/heads/master@{#583265}
-
Max Moroz authored
This reverts commit b0490220. Reason for revert: Breaks fuzzing builds: https://ci.chromium.org/p/chromium/builders/luci.chromium.ci/Libfuzzer%20Upload%20Linux%20UBSan/20928 Original change's description: > Update V8 to version 7.0.237. > > Summary of changes available at: > https://chromium.googlesource.com/v8/v8/+log/091d6f21..e756beb7 > > Please follow these instructions for assigning/CC'ing issues: > https://github.com/v8/v8/wiki/Triaging%20issues > > Please close rolling in case of a roll revert: > https://v8-roll.appspot.com/ > This only works with a Google account. > > CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel;luci.chromium.try:android_optional_gpu_tests_rel > > TBR=hablich@chromium.org,v8-waterfall-sheriff@grotations.appspotmail.com > > Change-Id: I4d77df256a6f6a4a905a2b93e11a4c946a21ce30 > Reviewed-on: https://chromium-review.googlesource.com/1175700 > Reviewed-by: V8 Autoroller <v8-ci-autoroll-builder@chops-service-accounts.iam.gserviceaccount.com> > Commit-Queue: V8 Autoroller <v8-ci-autoroll-builder@chops-service-accounts.iam.gserviceaccount.com> > Cr-Commit-Position: refs/heads/master@{#583252} TBR=hablich@chromium.org,v8-ci-autoroll-builder@chops-service-accounts.iam.gserviceaccount.com,v8-waterfall-sheriff@grotations.appspotmail.com Change-Id: Ica5decafaab4827b642edd106a99165a2ddbcb83 No-Presubmit: true No-Tree-Checks: true No-Try: true Cq-Include-Trybots: master.tryserver.blink:linux_trusty_blink_rel;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel;luci.chromium.try:android_optional_gpu_tests_rel Reviewed-on: https://chromium-review.googlesource.com/1174774Reviewed-by:
Max Moroz <mmoroz@chromium.org> Commit-Queue: Max Moroz <mmoroz@chromium.org> Cr-Commit-Position: refs/heads/master@{#583264}
-
Wan-Teh Chang authored
This does not change the current working directory of the script, and allows the subsequent 'git cl format' command to run in the correct directory. (The 'git cl format' command was running in the "${SRC}" directory incorrectly.) If 'git cl format' fails, point out what command failed in the error message. Allow the error output of 'git cl format' to be emitted. R=johannkoenig@google.com,jzern@google.com Change-Id: I18b7d8777a8ebb2cf09c32a35e7a5957fd1fa32a Reviewed-on: https://chromium-review.googlesource.com/1169701 Commit-Queue: Wan-Teh Chang <wtc@google.com> Reviewed-by:James Zern <jzern@google.com> Reviewed-by:
Johann Koenig <johannkoenig@google.com> Cr-Commit-Position: refs/heads/master@{#583263}
-
Scott Violet authored
The OverviewButtonTrayTests need to set the keyboard devices *after* AshTestBase::SetUp() is called. This is is important in single-process mash as Ash is the one creating the Env that the keyboard devices us. In non-single-process-mash AshTestSuite is creating the Env these tests use. The case ShellTest2.DontCrashWhenWindowDeleted is exercising is different enough for single-process mash that I created a separate test. Adding NOPRESUBMIT=true as this bot passed the trybots, and we're ok with this test including Env::GetInstance(). NOPRESUBMIT=true BUG=847992 TEST=test only changes TBR=jamescook@chromium.org Change-Id: I46fbfa9c78e8b28ac459b9863e34f3761a077c9a Reviewed-on: https://chromium-review.googlesource.com/1175848 Commit-Queue: Scott Violet <sky@chromium.org> Reviewed-by:
-
