- 31 Aug, 2018 40 commits
-
-
Koji Ishii authored
Following bot results are included. 9294 9295 9302 9305 9308 9311 9313 9314 9315 9318 9328 9331 9336 9340 9343 9347 9348 9350 4 lines were removed and 3 lines were deflaked by consecutive results since 9313. TBR=eae@chromium.org, mstensho@chromium.org NOTRY=true Bug: 591099 Cq-Include-Trybots: luci.chromium.try:linux_layout_tests_layout_ng Change-Id: I962c5e9c148d84415cbca5013f8262eb696b31cd Reviewed-on: https://chromium-review.googlesource.com/1195255 Commit-Queue: Koji Ishii <kojii@chromium.org> Reviewed-by:
Koji Ishii <kojii@chromium.org> Cr-Commit-Position: refs/heads/master@{#587967}
-
Jan Wilken Dörrie authored
This reverts commit d576f34a. Reason for revert: Likely culprit of https://crbug.com/879491. Original change's description: > Make chrome.proxy.onProxyError extension API work under the Network Service. > > This adds an optional "ProxyErrorClient" client to the NetworkContext which > receives notifications of errors in the PAC script, as well as notifications > of failed URL loads that may have been proxy related. > > Bug: 876568,851609 > Cq-Include-Trybots: luci.chromium.try:linux_mojo > Change-Id: I645ef91875135fcc81a7bd113442a0b65c661e65 > Reviewed-on: https://chromium-review.googlesource.com/1192123 > Reviewed-by: Matt Menke <mmenke@chromium.org> > Reviewed-by: Will Harris <wfh@chromium.org> > Commit-Queue: Eric Roman <eroman@chromium.org> > Cr-Commit-Position: refs/heads/master@{#587812} TBR=eroman@chromium.org,mmenke@chromium.org,wfh@chromium.org Change-Id: I286c7f768b04b6e47f1836abc459a33c7b394498 No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: 876568, 851609 Cq-Include-Trybots: luci.chromium.try:linux_mojo Reviewed-on: https://chromium-review.googlesource.com/1199043Reviewed-by:
Jan Wilken Dörrie <jdoerrie@chromium.org> Commit-Queue: Jan Wilken Dörrie <jdoerrie@chromium.org> Cr-Commit-Position: refs/heads/master@{#587966}
-
Arthur Sonzogni authored
The current behavior of chrome is the following one: --- When a javascript navigation is triggered, if it happens: * before the load event, replace the current history entry. * after the load event, append a new history entry. --- The first case is sometimes called "javascript-redirect". The current set of tests uses two kind of tests. One with a timeout of 2 seconds. This one is expected to happens after the load event. Another one with a timeout of 0 seconds. We don't really know this one is expected to happens. This is flaky. It usually happens after the load event, but it is non deterministic. This CLs removes: * /http/tests/history/redirect-js-*-0-seconds}.html * /http/tests/history/redirect-js-*-2-seconds}.html And completes: * /http/tests/history/redirect-js-*-before-load}.html * /http/tests/history/redirect-js-*-after-load}.html It modifies the tests so that they really depend on the load event, not on timing, which is not reliable. Bug: 862591, 831155 Change-Id: I3d568019c8616e6b52b9db554ad95ed312bd82d9 Reviewed-on: https://chromium-review.googlesource.com/1185094 Commit-Queue: Arthur Sonzogni <arthursonzogni@chromium.org> Reviewed-by:
Nate Chapin <japhet@chromium.org> Cr-Commit-Position: refs/heads/master@{#587965}
-
Yoshifumi Inoue authored
This patch introduces |InclusiveAncestorsOf()| for |NGPaintFragment| tree to reduce common pattern for ease for traversing ancestors. Cq-Include-Trybots: luci.chromium.try:linux_layout_tests_layout_ng;luci.chromium.try:linux_layout_tests_slimming_paint_v2;master.tryserver.blink:linux_trusty_blink_rel Change-Id: I60d7f70f694c98094f290e759471e52f550c0710 Reviewed-on: https://chromium-review.googlesource.com/1196387Reviewed-by:
-
Ioana Pandele authored
Bug: 879486, 835234 Change-Id: Ife5ece610ec7f3173caf368681e3f3338c96354b Reviewed-on: https://chromium-review.googlesource.com/1195483 Commit-Queue: Ioana Pandele <ioanap@chromium.org> Reviewed-by:
Vasilii Sukhanov <vasilii@chromium.org> Reviewed-by:
Steven Holte <holte@chromium.org> Cr-Commit-Position: refs/heads/master@{#587963}
-
Eriko Kurimoto authored
This reverts commit 695d7853. This commit is just removing arrays already stored in EventType. Also, it is called only when chromevox developer option flag is enabled. This option is WIP and the flag is disabled in default. This is irrelevant to webview benchmark. TBR=dtseng@chromium.org,nednguyen@google.com,yawano@chromium.org,sarakato@chromium.org Bug: 878965 Change-Id: Id2ab67061bd39597a2ebac0fedb2d7f00e6494ee Reviewed-on: https://chromium-review.googlesource.com/1198649 Commit-Queue: Eriko Kurimoto <elkurin@google.com> Reviewed-by:
Yuki Awano <yawano@chromium.org> Cr-Commit-Position: refs/heads/master@{#587962}
-
Sergio Villar Senin authored
SimpleURLLoader does not return the response body by default for most of error situations. The failing code was unconditionally dereferencing the std::unique_ptr<std::string<>> holding the response body and so, causing a crash. Bots didn't complain when the CL landed because apparently log levels greater than 0 are filtered out for unit tests (the line crashing was a VLOG(1)). That's why an unit test is not provided. Bug: 876627 Change-Id: Ibbc87955343fbab069cbc90b926593ffe6568ecd Reviewed-on: https://chromium-review.googlesource.com/1196888Reviewed-by:
Matt Menke <mmenke@chromium.org> Reviewed-by:
Alexander Alekseev <alemate@chromium.org> Commit-Queue: Sergio Villar <svillar@igalia.com> Cr-Commit-Position: refs/heads/master@{#587961}
-
Tetsui Ohkubo authored
SetIsNested should only be called when MessageView is in MessageCenterView. However, there was minor exception that it's called on Aero enabled Windows. As a result, shadow will be removed on Aero enabled Windows (Win 7). The shadow was already removed on Win8 & 10, so it should be an acceptable minor change. TEST=manual BUG=none Change-Id: I1d252553cce7da5027ad0c8d5ce629351ab1ff2c Reviewed-on: https://chromium-review.googlesource.com/1196290 Commit-Queue: Tetsui Ohkubo <tetsui@chromium.org> Reviewed-by:
Evan Stade <estade@chromium.org> Reviewed-by:
Yoshiki Iguchi <yoshiki@chromium.org> Cr-Commit-Position: refs/heads/master@{#587960}
-
Jan Wilken Dörrie authored
This reverts commit 9e96a32b. Reason for revert: Likely culprit of https://crbug.com/879469. Original change's description: > [media] Maintain DecoderSelector state throughout decoder selection. > > This CL changes DecoderStream to maintain an instance of DecoderSelector > continually. This enabled DecoderSelector to maintain > DecryptingDemuxerStream state and the blacklist internally. > > An immediate consequence is that DecoderSelector can now always try the > full list of potential decoders each time that selection is triggered. > This fixes the GPU->GPU changeType() transition, and provides a > foundation to build GPU fallforward on top of. > > Another consequence is that fallback can occur more than once for a > stream. It is no longer possible to loop forever doing fallbacks, so this > is a robustness improvement in addition to being a requirement for > changeType(). > > Bug: 877673 > 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: I76d1bed1c914e1ec58a6653d7200fbb67f70971a > Reviewed-on: https://chromium-review.googlesource.com/1188978 > Commit-Queue: Dan Sanders <sandersd@chromium.org> > Reviewed-by: Dale Curtis <dalecurtis@chromium.org> > Cr-Commit-Position: refs/heads/master@{#587908} TBR=dalecurtis@chromium.org,xhwang@chromium.org,sandersd@chromium.org Change-Id: I6836b3e04b56591289fd973a5e89092b70e923fe No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: 877673 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 Reviewed-on: https://chromium-review.googlesource.com/1199082Reviewed-by:
-
Sergei Datsenko authored
ext4 encryption on ChromeOS doesn't allow access to inode attributes for files in encrypted dir, so we need to indicate files that are ok to remove on low disk space with extended attributes too. BUG=chromium:879003 Change-Id: Ibf917ac255c93ea4e971b2a622b51e452257a4c7 Reviewed-on: https://chromium-review.googlesource.com/1196288Reviewed-by:
Stuart Langley <slangley@chromium.org> Reviewed-by:
Sam McNally <sammc@chromium.org> Commit-Queue: Sergei Datsenko <dats@chromium.org> Cr-Commit-Position: refs/heads/master@{#587958}
-
src-internal-chromium-autoroll authored
https://chrome-internal.googlesource.com/chrome/src-internal.git/+log/049f97004fb6..4458ecb45717 Created with: gclient setdep -r src-internal@4458ecb45717 The AutoRoll server is located here: https://autoroll-internal.skia.org/r/src-internal-chromium-autoroll 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=mmoss@chromium.org Change-Id: I9d73ca67ce74dd2589031d190940f47e8be96b1d Reviewed-on: https://chromium-review.googlesource.com/1198390Reviewed-by:
src-internal-chromium-autoroll <src-internal-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com> Commit-Queue: src-internal-chromium-autoroll <src-internal-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com> Cr-Commit-Position: refs/heads/master@{#587957}
-
Sam McNally authored
Bug: 879425 Change-Id: Ibcb7b71ca7424cd085c26fde7693b48b7473ba74 Reviewed-on: https://chromium-review.googlesource.com/1198703 Commit-Queue: Sam McNally <sammc@chromium.org> Reviewed-by:
Joel Hockey <joelhockey@chromium.org> Cr-Commit-Position: refs/heads/master@{#587956}
-
Yutaka Hirano authored
- Replace |url.ProtocolIsData() || origin->CanRequest(url)| with |origin->CanReadContent(url)|. - Remove cross-origin settings for data URLs in workers as data URLs are now considered as same-origin. This is a follow up for [1]. 1: https://chromium-review.googlesource.com/c/chromium/src/+/1160126 Bug: 870173 Change-Id: Ie95e82c57882e4ce7d52e8ab0607c81e4b18c171 Reviewed-on: https://chromium-review.googlesource.com/1177472 Commit-Queue: Yutaka Hirano <yhirano@chromium.org> Reviewed-by:
Hiroshige Hayashizaki <hiroshige@chromium.org> Cr-Commit-Position: refs/heads/master@{#587955}
-
Chromite Chromium Autoroll authored
https://chromium.googlesource.com/chromiumos/chromite.git/+log/108fea28a159..1940d7df3f6d git log 108fea28a159..1940d7df3f6d --date=short --no-merges --format='%ad %ae %s' 2018-08-31 athilenius@google.com Changed oauth2 refresh logs wording Created with: gclient setdep -r src/third_party/chromite@1940d7df3f6d The AutoRoll server is located here: https://autoroll.skia.org/r/chromite-chromium-autoroll 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=chrome-os-gardeners@chromium.org Change-Id: I955a3e687d08241d060b75b0a444f6dce9fd0145 Reviewed-on: https://chromium-review.googlesource.com/1198404Reviewed-by:
Chromite Chromium Autoroll <chromite-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com> Commit-Queue: Chromite Chromium Autoroll <chromite-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com> Cr-Commit-Position: refs/heads/master@{#587954}
-
Mike West authored
Several cookie tests are failing under content shell because it defaults to blocking third-party cookies. We should invert that default in order to correctly run web platform tests. Bug: 843945 Change-Id: Ifead36d9b890ea603ea14538caff56ad579cb83e Reviewed-on: https://chromium-review.googlesource.com/1196547Reviewed-by:
Friedrich Horschig [CEST] <fhorschig@chromium.org> Commit-Queue: Mike West <mkwst@chromium.org> Cr-Commit-Position: refs/heads/master@{#587953}
-
Eyor Alemayehu authored
Currently, we send an empty AssistantBundle if we don't have the assistant extra and the assistant tree. Made a change to prevent that from happening. Bug: b:113528361 Change-Id: I37d8cef85cd0631e61a756ed4e6683d274b0a877 Reviewed-on: https://chromium-review.googlesource.com/1195925 Commit-Queue: Eyor Alemayehu <eyor@google.com> Reviewed-by:
Xiaohui Chen <xiaohuic@chromium.org> Cr-Commit-Position: refs/heads/master@{#587952}
-
Hajime Hoshi authored
RenderThreadImpl::IdleHandler called only ReleaseFreeMemory on background tabs, which releases memory that is already free. We think this just affects the memory usage figures at the task manager. Futhermore, ReleaseFreeMemory is also called via OnPurgeMemory, so this call seems unnecessarily redundant. This CL removes RenderThreadImpl::IdleHandler as we expect this doesn't affect the real world usage. In case perf bots complain, we'll revert this fix. Bug: 878700 Change-Id: Ic7513837709598ea169ebf2a85454ba211678434 Reviewed-on: https://chromium-review.googlesource.com/1196284Reviewed-by:
Takashi Sakamoto <tasak@google.com> Reviewed-by:
Kentaro Hara <haraken@chromium.org> Reviewed-by:
Alexander Timin <altimin@chromium.org> Reviewed-by:
Devlin <rdevlin.cronin@chromium.org> Reviewed-by:
Kinuko Yasuda <kinuko@chromium.org> Commit-Queue: Hajime Hoshi <hajimehoshi@chromium.org> Cr-Commit-Position: refs/heads/master@{#587951}
-
Yutaka Hirano authored
The fetch spec says data URLs should be treated as same-origin, but currently Chrome treats it as cross-origin, and attaches "access-control-allow-origin: *" to every response to a request to data URLs. Let's align with the spec. See [1] for details. There are some modules using FetchRequestMode::kSameOrigin for which it's not obvious whether handling data URLs as same-origin is correct. 1. SVG - a. ExternalSVGResource::Load - b. SVGUseElement::UpdateTargetReference 2. XSLT - a. XSLStyleSheet::LoadChildSheet - b. DocLoaderFunc 2a is not affected by this change because of LibXSLT implementation. For 2b I wrote a test. The test result is aligned with Firefox. 1: https://docs.google.com/document/d/1_BD15unoPJVwKyf5yOUDu5kie492TTaBxzhJ58j1rD4/edit Bug: 870173 Cq-Include-Trybots: luci.chromium.try:linux_layout_tests_slimming_paint_v2;luci.chromium.try:linux_mojo;master.tryserver.blink:linux_trusty_blink_rel Change-Id: I666c9cb46c83e0e6b290e03b5b999db501c20541 Reviewed-on: https://chromium-review.googlesource.com/1160126 Commit-Queue: Yutaka Hirano <yhirano@chromium.org> Reviewed-by:
Takashi Toyoshima <toyoshim@chromium.org> Reviewed-by:
Kouhei Ueno <kouhei@chromium.org> Reviewed-by:
-
Sergei Datsenko authored
BUG=chromium:879051 Change-Id: I3e9000b6bdf7e48683010ba5e039bd6b79219910 Reviewed-on: https://chromium-review.googlesource.com/1196285Reviewed-by:
Luciano Pacheco <lucmult@chromium.org> Commit-Queue: Luciano Pacheco <lucmult@chromium.org> Cr-Commit-Position: refs/heads/master@{#587949}
-
calamity authored
This reverts commit 7f98222a. Reason for revert: Suspected of causing BookmarksTestCase/testCachePositionIsRecreated to fail. https://ci.chromium.org/p/chromium/builders/luci.chromium.ci/ios-uirefresh-simulator/2187 Original change's description: > [ios] Changes Bookmark cache to use Table row instead of offset > > Bookmarks caches the tableView scroll offset, so whenever > bookmarks is closed and re-opened, we maintain the scroll offset > and the user is in the same position as before dismissal. > > As part of supporting dynamic type on Bookmarks this approach is > no longer valid since: > -The row height might change due to the size of the font. > -The row height is not calculated unless the cell is dequeued, > this means that when the content offset is set to the cached value > (when re-oppening Bookmarks) the position offset is correct, but > the content being displayed are not. This happens because the cells > that weren't displayed when setting the offset manually used the > estimatedRowHeight instead of their real height. This messes up the > cache value. > > In order to support dynamic height/type this CL changes the cache > so it stores the tableView row instead of the tableView scroll > offset. It also updates the pref values so they store an int instead > of a double. > > It also deletes BookmarkHomeSharedState cellHeightPt and updates > the estimatedRowHeight value to 65. > > This CL doesn't introduce dynamic height yet, that will be done in > a follow up CL. > > Bug: 869671 > Cq-Include-Trybots: luci.chromium.try:ios-simulator-cronet;luci.chromium.try:ios-simulator-full-configs > Change-Id: I6cd5f8baa8b385d3045ab5e9d0ff54a9d8202b63 > Reviewed-on: https://chromium-review.googlesource.com/1157684 > Commit-Queue: Sergio Collazos <sczs@chromium.org> > Reviewed-by: Kurt Horimoto <kkhorimoto@chromium.org> > Reviewed-by: Mark Cogan <marq@chromium.org> > Cr-Commit-Position: refs/heads/master@{#587821} TBR=marq@chromium.org,kkhorimoto@chromium.org,sczs@chromium.org Change-Id: I14635c3f977857714a76feefae6535d8df0cf35b No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: 869671 Cq-Include-Trybots: luci.chromium.try:ios-simulator-cronet;luci.chromium.try:ios-simulator-full-configs Reviewed-on: https://chromium-review.googlesource.com/1198647Reviewed-by:
calamity <calamity@chromium.org> Commit-Queue: calamity <calamity@chromium.org> Cr-Commit-Position: refs/heads/master@{#587948}
-
Hiroki Nakagawa authored
This is a follow-up CL for https://chromium-review.googlesource.com/c/chromium/src/+/1194807 For a while, we need to keep 3 shared worker script loading paths: classic, S13nServiceWorker, and NetworkService. To clarify which function arguments are used for which loading path, we should give more specific names to them. Bug: 715632 Change-Id: I5c075cd72f89d7ca761861db02be0b9146cc215e TBR: kinuko@chromium.org Reviewed-on: https://chromium-review.googlesource.com/1198663 Commit-Queue: Hiroki Nakagawa <nhiroki@chromium.org> Reviewed-by:
Matt Falkenhagen <falken@chromium.org> Cr-Commit-Position: refs/heads/master@{#587947}
-
Yoichi Osato authored
For LegacyLayout, we have to do character-range-based partial invalidation for selections because a LayoutText is too big to be fully invalidates for selection. In the other hand, we're invalidating each NGPaintfragment. So no further grain invalidation, which is based on character range that we're doing for legacy, is not needed. Bug: 708452 Cq-Include-Trybots: luci.chromium.try:linux_layout_tests_layout_ng Change-Id: Id011ced00ee2de43efdc9bf35f7ad922b0cf1078 Reviewed-on: https://chromium-review.googlesource.com/1198744Reviewed-by:
Xianzhu Wang <wangxianzhu@chromium.org> Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org> Cr-Commit-Position: refs/heads/master@{#587946}
-
Stuart Langley authored
Update DriveNotificationManager so that it can register for, and process, XMPP push notifications for team drives. We can register for changes in COSMO by prefixing "TD:" to the team drive ID when we register for the invalidation. To support this change we: - Add OnTeamDrivesUpdated to FileSystemObserver to flow changes in the team drives that the user can access to observers. - Allow DriveNotificationManager to keep track of the users team drive ID's and register for notifications from them. - Change DriveNotificationManager to generate a different event if the update is triggered by the polling timer or a push notification. - Change CheckForUpdates to accept a set of id's which will be the team drive id's that were recieved from the invalidation service. - Provide a glue class in DriveIntegrationService to pass team drive updates to DriveNotificationManager. - Addes file system test cases. Follow ups: - Fix DriveNotificationManager so we can write unit tests against it. Write the tests for registering for and processing invalidations. - Look at batching in DriveNotifcationManager, as invalidation service can send many rapid invalidations for the same object. - Fix up the polling timer logic in DriveNotificationManager, and adjust the timeouts so we do not spam the drive backends with needless update checks if the user has many team drives. Bug: 837456, 783184, 867166 Change-Id: I3fd80d41902bdd7f6c8f090346acbbf64f9dafdb Reviewed-on: https://chromium-review.googlesource.com/1195256Reviewed-by:
-
Tarun Bansal authored
Add more UMA to track the performance and correctness of the resource loading hints experiment. Specifically, this CL records UMA on how many page hints are received via component updater, and how frequently were the resource hints available at the time of page commit. Bug: 878129 Change-Id: I8bc3dd1e6d7c80e4cefc34ebb6119665d5fbee5e Reviewed-on: https://chromium-review.googlesource.com/1197108 Commit-Queue: Tarun Bansal <tbansal@chromium.org> Reviewed-by:
Ryan Sturm <ryansturm@chromium.org> Cr-Commit-Position: refs/heads/master@{#587944}
-
Peter Kotwicz authored
As part of fixing BUG 817263, I am going to add more methods to WebApkUtils. This CL splits out the host browser selection logic into its own class. Bug=817263 Change-Id: I60a2246959ecb008d69937fbe62b37b2769b99ce Reviewed-on: https://chromium-review.googlesource.com/1184980 Commit-Queue: Peter Kotwicz <pkotwicz@chromium.org> Reviewed-by:
Xi Han <hanxi@chromium.org> Cr-Commit-Position: refs/heads/master@{#587943}
-
Wenzhao Zang authored
The space info button in the Files app redirects to the “Storage management” section in Settings, which has already been hidden in Demo Mode. So it makes sense to hide this button as well. In Demo Mode (a public session), the cryptohome runs in RAM. As a result, the Files app only considers available RAM when showing free disk space. Therefore, customers in stores might think the device only has 4 GB of disk space. We should hide the disk space stat until there's a better solution, e.g. show an explanatory message ("disk space is limited in Demo Mode" or something). Bug: 868747 Change-Id: I8b138bbf0b864acaf38a51659cccf7ad3d198c16 Reviewed-on: https://chromium-review.googlesource.com/1198882Reviewed-by:Naoki Fukino <fukino@chromium.org> Commit-Queue: Wenzhao (Colin) Zang <wzang@chromium.org> Cr-Commit-Position: refs/heads/master@{#587942}
-
Michael Giuffrida authored
Remove --enable-demo-mode flag and always enable Demo Mode when ARC++ is supported. The --enable-offline-demo-mode flag will be removed separately for ease of reverting in case that feature ends up being delayed. Bug: 812037 Change-Id: Ie3ac7a98af21d3fab3e3c7f2aaff0c71a2bcaf08 Reviewed-on: https://chromium-review.googlesource.com/1195116 Commit-Queue: Michael Giuffrida <michaelpg@chromium.org> Commit-Queue: Alexander Alekseev <alemate@chromium.org> Reviewed-by:
Aga Wronska <agawronska@chromium.org> Cr-Commit-Position: refs/heads/master@{#587941}
-
jonross authored
LayerWithRealCompositorTest.ReportMetrics was racy due to the fact that LayerAnimationElement and SchedulerStateMachine's timing and task queues cannot be mocked out. This left the test vulnerable to differences in timing on bots. The test had a certain expectation of the order of events, which is not actually a guarantee. An animation's frame could be generated before it entered its wait loop. Similarly a animation could be delayed in executing for so long that not enough frames were generated to actually create a report. This updates the test to mock out time, so that it can reliably control the order of events. The test has also been moved to LayerAnimatorTest.ReportMetrics. LayerWithRealCompositorTest utilizes a TestCompositorHost in order to control the Compositor throughout the tests. This was done by having CompositorTestSuite always create a ScopedTaskEnvironment. Unfortunately this causes a thread race with base::TimeTicks::Now, as this thread starts enqueuing work before an individual test can mock out time. This change also updates CompositorTestSuite to not create the ScopedTaskEnvironment. Instead each test suite which desires to mock out the time of task runners, must explicitly create one. Thus allowing tests which need not a full Compositor thread, can instead be able to mock time separately. TBR=sadrul@chromium.org TEST=LayerWithRealCompositorTest.ReportMetrics LayerAnimatorTest.ReportMetrics Bug: 709080 Change-Id: Ia12cdcdf2b6c67c8abc21399eae66cdb1d0d1eb1 Reviewed-on: https://chromium-review.googlesource.com/1157234 Commit-Queue: Jonathan Ross <jonross@chromium.org> Reviewed-by:
Robert Flack <flackr@chromium.org> Cr-Commit-Position: refs/heads/master@{#587940}
-
Lei Zhang authored
Change-Id: Ib785f3bab568f15a4d970e185cd6c6eedd6e806d Reviewed-on: https://chromium-review.googlesource.com/1196236 Commit-Queue: Lei Zhang <thestig@chromium.org> Reviewed-by:
Dirk Pranke <dpranke@chromium.org> Cr-Commit-Position: refs/heads/master@{#587939}
-
Charlie Harrison authored
Bug: None Change-Id: I41c66bdab7c27bc3b1df30576b8116143383eb14 Reviewed-on: https://chromium-review.googlesource.com/1195832Reviewed-by:
Robert Kaplow (slow) <rkaplow@chromium.org> Commit-Queue: Charlie Harrison <csharrison@chromium.org> Cr-Commit-Position: refs/heads/master@{#587938}
-
Yutaka Hirano authored
CORSURLLoaderFactory rejects a request whose credentials mode and load flags contradict each other. The load flags rely on blink::ResourceRequest::allow_stored_credentials_, so let's configure the member when out-of-blink CORS is enabled. Now we have response tainting, so use it to configure load_flags. This changes the existing behavior (even when out-of-blink CORS is disabled), which leads to test changes. See https://github.com/whatwg/fetch/issues/169 for the corresponding spec change. Bug: 870173 Cq-Include-Trybots: luci.chromium.try:linux_mojo Change-Id: I8163ce45780b0b6bf9e8ee61e8bee0c433ea0967 Reviewed-on: https://chromium-review.googlesource.com/1174080 Commit-Queue: Yutaka Hirano <yhirano@chromium.org> Reviewed-by:
-
Dongheun Kang authored
Bug: n/a Change-Id: Iffd7f0bece86b9438b285da3a53ceea00104b77d Reviewed-on: https://chromium-review.googlesource.com/1195383 Commit-Queue: Stefan Zager <szager@chromium.org> Reviewed-by:
Stefan Zager <szager@chromium.org> Reviewed-by:
David Bokan <bokan@chromium.org> Cr-Commit-Position: refs/heads/master@{#587936}
-
Yutaka Hirano authored
blink::Resource has a field cors_status_ to store the result of CORS check. This CL replaces the concept with "response type", a concept defined in the fetch spec as https://fetch.spec.whatwg.org/#concept-response-type. Actually, that field has existed for long for service worker, but this CL starts using it for network fetch. Using response type enables us to have the consistent logic among several implementations, namely, CORSURLLoader, ResourceLoader and ThreadableLoader. https://html.spec.whatwg.org/#cors-same-origin and https://html.spec.whatwg.org/#cors-cross-origin are now implemented so various resource clients should use them. Side effects: - blink::ResourceLoader now dispatches a ResourceError with a CORS error for CORS violations instead of an abort error. - Now CORS erros from CORSURLLoader are printed on the devtools console. Bug: 736308, 878339 Cq-Include-Trybots: luci.chromium.try:linux_mojo Change-Id: I8777904654c0478d092ad90a2a919d5eb5dcc2bd Reviewed-on: https://chromium-review.googlesource.com/1154246 Commit-Queue: Yutaka Hirano <yhirano@chromium.org> Reviewed-by:
Fredrik Hubinette <hubbe@chromium.org> Reviewed-by:
-
Wenzhao Zang authored
Bug: 879310 Change-Id: I65ac6b822e1fadd9440189a3e11c6d4fa118ec8f Reviewed-on: https://chromium-review.googlesource.com/1198762 Commit-Queue: Alexander Alekseev <alemate@chromium.org> Reviewed-by:
-
Timothy Gu authored
FrameLoader::StopAllLoaders() has this explicit note: Warning: stopAllLoaders can and will detach the LocalFrame out from under you. All callers need to either protect the LocalFrame or guarantee they won't in any way access the LocalFrame after stopAllLoaders returns. Check frame_'s existence after the call to prevent a NULL dereference. Bug: 879366 Change-Id: I1e537374f59fbad7b069f9de63cfa3b6b2b2b00c Reviewed-on: https://chromium-review.googlesource.com/1198022Reviewed-by:Kent Tamura <tkent@chromium.org> Reviewed-by:
Hayato Ito <hayato@chromium.org> Commit-Queue: Timothy Gu <timothygu@chromium.org> Cr-Commit-Position: refs/heads/master@{#587933}
-
pdfium-chromium-autoroll authored
https://pdfium.googlesource.com/pdfium.git/+log/d4e7ca034cbb..b2d25960b756 git log d4e7ca034cbb..b2d25960b756 --date=short --no-merges --format='%ad %ae %s' 2018-08-31 thestig@chromium.org Stop checking out GYP. Created with: gclient setdep -r src/third_party/pdfium@b2d25960b756 The AutoRoll server is located here: https://autoroll.skia.org/r/pdfium-autoroll 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=dsinclair@chromium.org Change-Id: I406bc25fb39940707da7f450ccac926b64613413 Reviewed-on: https://chromium-review.googlesource.com/1198386Reviewed-by:
pdfium-chromium-autoroll <pdfium-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com> Commit-Queue: pdfium-chromium-autoroll <pdfium-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com> Cr-Commit-Position: refs/heads/master@{#587932}
-
Sujie Zhu authored
Add billing customer number in the migration request. TBR=mathp@chromium.org Bug: 852904 Change-Id: Ia474eda34bc62982e724318f4048ecee6571cc14 Reviewed-on: https://chromium-review.googlesource.com/1198763Reviewed-by:
Jared Saul <jsaul@google.com> Commit-Queue: Sujie Zhu <sujiezhu@google.com> Cr-Commit-Position: refs/heads/master@{#587931}
-
Miguel Casas authored
ToT looks at RuntimeEnabledFeatures::Canvas2dImageChromiumEnabled() (enabled by default for Mac, disabled by default elsewhere) to use ImageCHROMIUM for Canvas 2D contexts or not. This is fine, but in the LowLatency it doesn't make sense not use that features, we always want to use ImageCHROMIUM (if we |want_acceleration| at all). Hence this CL: - ignores the RuntimeEnabledFeature if LowLatency (so all platforms will use ImageCHROMIUM for LowLatency -- if |want_acceleration|). - removes the associated code in runtime_features.cc and about_flags.cc, essentially undoing my previous crrev.com/c/1056168. (So that Mac can still control the use of ImageCHROMIUM for non-low-latency cases). Bonus cleanup: simplify |enable_web_gl_image_chromium| initialization. Bug: 878405 Change-Id: I91c5422fad82b1146af02c193546f924bbe5448b Reviewed-on: https://chromium-review.googlesource.com/1194473 Commit-Queue: Miguel Casas <mcasas@chromium.org> Reviewed-by:
Avi Drissman <avi@chromium.org> Reviewed-by:
Fernando Serboncini <fserb@chromium.org> Reviewed-by:
ccameron <ccameron@chromium.org> Cr-Commit-Position: refs/heads/master@{#587930}
-
Alex Deymo authored
The compression density of small thumbnails and large "photos" is expected to be different. This patch splits the recently added metric reporting the compressed density for JPEG images measured as "bits per pixel" into three groups according to the smallest dimension (width or height): * 100 to 399 pixels, * 400 to 999 pixels, and * 1000 or more pixels Test: Updated unittests Bug: None Cq-Include-Trybots: luci.chromium.try:linux_layout_tests_slimming_paint_v2;master.tryserver.blink:linux_trusty_blink_rel Change-Id: I1c3ccd6a404896e02c9d6a6cd978af8f213ec921 Reviewed-on: https://chromium-review.googlesource.com/1186886Reviewed-by:
Philip Rogers <pdr@chromium.org> Reviewed-by:
-
chrome-release-bot authored
# This is an automated release commit. # Do not revert without consulting chrome-pmo@google.com. NOAUTOREVERT=true TBR=mmoss@chromium.org Change-Id: I1a44dfd5f1535fa36388030b6291aac4f87667a0 Reviewed-on: https://chromium-review.googlesource.com/1198311Reviewed-by:
chrome-release-bot@chromium.org <chrome-release-bot@chromium.org> Cr-Commit-Position: refs/heads/master@{#587928}
-
