- 13 Apr, 2020 10 commits
-
-
Wanming Lin authored
This CL upstreams most device orientation tests to WPT. - Moves device orientation tests from web_tests/device_orientation/, web_tests/http/tests/security/powerfuFeatureRestrictions/, to web_tests/external/wpt/orientation-event with some necessary updates to adapt to WPT test framework. - Retains two page-visibility.html tests as which could not be automated in WPT at present and adds two similar manual tests there to keep test coverage sufficient. - Leaves existing tests as is, further improvement (e.g. remove duplicate tests, use EventWatcher, etc.) will be arranged in a follow-up. - Updates test paths in some profiles. Bug: 816462 Change-Id: I1b91168c5362fc6e059b2529dd4a1d8119f5dad6 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2135059 Commit-Queue: Wanming Lin <wanming.lin@intel.com> Reviewed-by:
Reilly Grant <reillyg@chromium.org> Cr-Commit-Position: refs/heads/master@{#758443}
-
Yuki Shiino authored
Bug: 839389 Change-Id: I802b01943d6c6a5a178c40f516ac367d43981c82 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2145424 Commit-Queue: Yuki Shiino <yukishiino@chromium.org> Reviewed-by:
Hitoshi Yoshida <peria@chromium.org> Cr-Commit-Position: refs/heads/master@{#758442}
-
Christian Biesinger authored
Change-Id: I5dbd39bf616354d715968b591912abbb427d5954 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2146214 Auto-Submit: Christian Biesinger <cbiesinger@chromium.org> Reviewed-by:
Gyuyoung Kim <gyuyoung@igalia.com> Commit-Queue: Christian Biesinger <cbiesinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#758441}
-
Igor Makarov authored
- added .style.yapf file - reformat *.py files Bug: 1051750 Change-Id: I97258536ef73d8b714cd1ccbba6ab2fc20b7d23b Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2132193Reviewed-by:
Kent Tamura <tkent@chromium.org> Commit-Queue: Kent Tamura <tkent@chromium.org> Cr-Commit-Position: refs/heads/master@{#758440}
-
Koji Ishii authored
This patch removes unused function |NGBoxFragmentPainter:: PaintLineBoxItem|. This became unused when r731835 crrev.com/c/1994556 switched line box painter to |PaintBlockFlowContents|. Bug: 982194 Change-Id: I51f9f1661f2462c6faba55e646c251e9992ecc78 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2143889 Auto-Submit: Koji Ishii <kojii@chromium.org> Commit-Queue: Kent Tamura <tkent@chromium.org> Reviewed-by:
Yoshifumi Inoue <yosin@chromium.org> Cr-Commit-Position: refs/heads/master@{#758439}
-
Yuki Shiino authored
Bug: 839389 Change-Id: I20ca170b3536309e887f97a50710e075cd1caaed Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2142669 Commit-Queue: Hitoshi Yoshida <peria@chromium.org> Reviewed-by:
-
Igor Makarov authored
- added .style.yapf file - reformat *.py files Bug: 1051750 Change-Id: I8288aa519ff5a4634728b16005d46377a7a2cdc5 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2141988Reviewed-by:
-
Timothy Gu authored
This reverts commit 103ac117. It appears that we were overly optimistic. There can be intermediate states where Element::GetIdAttribute() is called from Element::UpdateId(), which violates the DCHECK() put into place in the reverted commit. Original change's description: > Filter out empty ID in attribute changed callback > > Per spec [1], empty ID attributes must be treated identically as > nonexistent. This CL implements these steps faithfully, and results in > DocumentFragment::getElementById() always returning nullptr when the > empty string is given, thus fixing bug 1068580. > > Additionally, previously Document::getElementById() explicitly returned > nullptr when the empty string or the null string is given. The empty > string handling is no longer necessary after this CL, and thus removed. > (The null string handling is still necessary for internal C++ code that > relies on the short-circuiting behavior). > > [1]: https://dom.spec.whatwg.org/#concept-id > > Change-Id: Ia65024efef9e5ef54d9db233590a17b372282ed1 > Fixed: 1068580 > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2140054 > Reviewed-by: Kent Tamura <tkent@chromium.org> > Commit-Queue: Timothy Gu <timothygu@chromium.org> > Cr-Commit-Position: refs/heads/master@{#757310} Fixed: 1069728 Change-Id: I70b2fd8d14456ec0b063cb18e5982384bd1285bf Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2146924Reviewed-by:
-
Yuki Shiino authored
Bug: 839389 Change-Id: I95db56fa113354486742149024cd34173364eb46 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2142074Reviewed-by:
-
chrome-release-bot authored
# This is an automated release commit. # Do not revert without consulting chrome-pmo@google.com. NOAUTOREVERT=true TBR=lakpamarthy@chromium.org Change-Id: Ia981b9a58c02eff74658022b67e9113ee04fa9ca Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2146925Reviewed-by:
Chrome Release Bot (LUCI) <chrome-official-brancher@chops-service-accounts.iam.gserviceaccount.com> Cr-Commit-Position: refs/heads/master@{#758434}
-
- 12 Apr, 2020 18 commits
-
-
Anthony Fandrianto authored
Currently, a CIPD staging directory is populated by copying specified `sources` into it, and producing a .yaml manifest which reflects the copied files. This change refactors the template to invoke a pytho script to do the above instead of in GN, using the `yaml` library to write the file data in the proper format. Also make the license filepath explicitly passed through sources. Renames "archive" -> "package" and "manifest" -> "package definition" to be consistent with CIPD semantics (crbug.com/1042819#c8). Finally, have the script emit a depfile for all files tracked by the .yaml manifest. This refactor is desirable, as we want to create CIPD packages with more complicated rules than simply "copy and include these files" (see crrev.com/c/1925446, which extends the functionality of the template to allow for the inclusion of directories). Wriiting yamls with a library reduces the error-proneness of trying to recreate yaml's format in pure GN. Also, emitting a depfile ensure that we track and rebuild if the CIPD package is modified. Currently the potential danger of this change is limited in scope to fuchsia, as they are the only users of the template. I've checked locally that the produced CIPD packages for fuchsia are identical prior to this change. Bug: fuchsia:41443 Change-Id: If1b27d6a8fbbffdf767aaa4230bf7b527a553170 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2135186 Commit-Queue: Anthony Fandrianto <atyfto@google.com> Reviewed-by:
Bruce Dawson <brucedawson@chromium.org> Reviewed-by:
Kevin Marshall <kmarshall@chromium.org> Cr-Commit-Position: refs/heads/master@{#758433}
-
chromium-internal-autoroll authored
https://chrome-internal.googlesource.com/chrome/src-internal.git/+log/16297237f8cc..fd5072e45780 Created with: gclient setdep -r src-internal@fd5072e45780 If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://skia-autoroll.corp.goog/r/src-internal-chromium-autoroll Please CC nhiroki@google.com,reillyg@google.com on the revert to ensure that a human is aware of the problem. To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+/master/autoroll/README.md Cq-Include-Trybots: luci.chrome.try:linux-chromeos-chrome Bug: None Tbr: nhiroki@google.com,reillyg@google.com Change-Id: I2e96594a981f8509b1b853c3a450f98c78ae7269 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2146846Reviewed-by:
chromium-internal-autoroll <chromium-internal-autoroll@skia-corp.google.com.iam.gserviceaccount.com> Commit-Queue: chromium-internal-autoroll <chromium-internal-autoroll@skia-corp.google.com.iam.gserviceaccount.com> Cr-Commit-Position: refs/heads/master@{#758432}
-
chromium-autoroll authored
If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://autoroll.skia.org/r/fuchsia-sdk-chromium-autoroll Please CC cr-fuchsia+bot@chromium.org on the revert to ensure that a human is aware of the problem. To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+/master/autoroll/README.md Cq-Include-Trybots: luci.chromium.try:fuchsia-arm64-cast;luci.chromium.try:fuchsia-x64-cast Tbr: cr-fuchsia+bot@chromium.org Change-Id: Ia016e89d022b541d676003bf1ea16865e37cc6f3 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2146844Reviewed-by:
chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com> Commit-Queue: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com> Cr-Commit-Position: refs/heads/master@{#758431}
-
Bo Liu authored
It is unsafe and generally not possible to delete the profile data of a profile that has already been "loaded". The only safe implementation is to mark the profile data as deleted, wait until the next time browser process starts, ensure the profile data is not loaded, and find a time to delete the data. This CL implements this for weblayer. * Deletion marker are empty files in <data>/profiles_to_delete matching the directory names of profiles. If <data>/profiles_to_delete/foo exists, then <data>/profiles/foo is marked for deletion. * If client create with a name (say "foo") matching a deleted profile, will try to linearly check for directory with a suffix (eg "foo.1", "foo.2", etc) until a directory that is not marked for deletion is found. "." is not part of a valid profile name, so there is no risk of collision. * Now DestroyAndDeleteDataFromDisk will mark the profile for deletion, and only remove data with BrowsingDataRemover. It will not attempt to delete the profile directory. * Nuke and Mark are run in the same SequencedTaskRunner to ensure Nuke is called first in a process before any mark. This ensures Nuke will never remove a newly marked Profile during the current process's lifetime. * Nuke will delete the data and cache directories first, and only remove the marker if both succeeds. This ensures there is no race in reusing a profile path. Add weblayer_unittests and add some unit tests for profile_disk_operations. Bug: 1065585 Change-Id: I4b936c9a1f31da56612095bb1088b93232c5751f Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2140101 Commit-Queue: Bo <boliu@chromium.org> Reviewed-by:
Scott Violet <sky@chromium.org> Cr-Commit-Position: refs/heads/master@{#758430}
-
CQ_INCLUDE_TRYBOTS=luci.chrome.try:chromeos-betty-pi-arc-chrome CQ_INCLUDE_TRYBOTS=luci.chrome.try:chromeos-eve-compile-chrome CQ_INCLUDE_TRYBOTS=luci.chrome.try:chromeos-kevin-compile-chrome BUG=762641 TBR=chrome-os-gardeners@google.com Change-Id: I5715a8a0e6ee4fc53ea793f7eef832deab377e71 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2146780Reviewed-by:
ChromeOS bot <3su6n15k.default@developer.gserviceaccount.com> Reviewed-by:
Achuith Bhandarkar <achuith@chromium.org> Commit-Queue: ChromeOS bot <3su6n15k.default@developer.gserviceaccount.com> Commit-Queue: Achuith Bhandarkar <achuith@chromium.org> Cr-Commit-Position: refs/heads/master@{#758429}
-
Rainhard Findling authored
Bug: 1015841 Change-Id: I2554be24bccc2c7fe2dfa76ecc380e0b98a7d32c Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2135757Reviewed-by:
Esmael Elmoslimany <aee@chromium.org> Commit-Queue: Rainhard Findling <rainhard@chromium.org> Cr-Commit-Position: refs/heads/master@{#758428}
-
Guido Urdaneta authored
Bug: 1065838 Change-Id: Id87979155a33bb8bc6daf18555fc4135ab21f3d9 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2144276Reviewed-by:
Marina Ciocea <marinaciocea@chromium.org> Commit-Queue: Guido Urdaneta <guidou@chromium.org> Cr-Commit-Position: refs/heads/master@{#758427}
-
chromium-autoroll authored
https://skia.googlesource.com/skia.git/+log/639083a98906..6dfc55454671 git log 639083a98906..6dfc55454671 --date=short --first-parent --format='%ad %ae %s' 2020-04-12 skia-recreate-skps@skia-swarming-bots.iam.gserviceaccount.com Update SKP version Created with: gclient setdep -r src/third_party/skia@6dfc55454671 If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://autoroll.skia.org/r/skia-autoroll Please CC kjlubick@google.com on the revert to ensure that a human is aware of the problem. To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+/master/autoroll/README.md Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux-blink-rel;luci.chromium.try:linux-chromeos-compile-dbg;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel Bug: None Tbr: kjlubick@google.com Change-Id: I100cc5c489812eb5e5605588090b9b4dc22c5b79 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2146839Reviewed-by:
-
chromium-autoroll authored
https://skia.googlesource.com/skia.git/+log/459ecee2cbdc..639083a98906 git log 459ecee2cbdc..639083a98906 --date=short --first-parent --format='%ad %ae %s' 2020-04-12 skia-recreate-skps@skia-swarming-bots.iam.gserviceaccount.com Update Go Deps Created with: gclient setdep -r src/third_party/skia@639083a98906 If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://autoroll.skia.org/r/skia-autoroll Please CC kjlubick@google.com on the revert to ensure that a human is aware of the problem. To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+/master/autoroll/README.md Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux-blink-rel;luci.chromium.try:linux-chromeos-compile-dbg;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel Bug: None Tbr: kjlubick@google.com Change-Id: I0e606ff64dc0076f4d111a5ffe1b8f5826e52e41 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2146521Reviewed-by:
-
chromium-autoroll authored
https://chromium.googlesource.com/devtools/devtools-frontend.git/+log/30c590849d00..3f138576af56 git log 30c590849d00..3f138576af56 --date=short --first-parent --format='%ad %ae %s' 2020-04-12 devtools-ci-autoroll-builder@chops-service-accounts.iam.gserviceaccount.com Update DevTools Chromium DEPS. Created with: gclient setdep -r src/third_party/devtools-frontend/src@3f138576af56 If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://autoroll.skia.org/r/devtools-frontend-chromium Please CC devtools-waterfall-sheriff-onduty@grotations.appspotmail.com on the revert to ensure that a human is aware of the problem. To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+/master/autoroll/README.md Bug: None Tbr: devtools-waterfall-sheriff-onduty@grotations.appspotmail.com Change-Id: I85f35ca5b07bdf91fef842e16acee941a736b2dc Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2146519Reviewed-by:
-
chromium-autoroll authored
If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://autoroll.skia.org/r/fuchsia-sdk-chromium-autoroll Please CC cr-fuchsia+bot@chromium.org on the revert to ensure that a human is aware of the problem. To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+/master/autoroll/README.md Cq-Include-Trybots: luci.chromium.try:fuchsia-arm64-cast;luci.chromium.try:fuchsia-x64-cast Tbr: cr-fuchsia+bot@chromium.org Change-Id: Icf4c30a505f4fa79d8b766acb47ecd226cdf9328 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2146520Reviewed-by:
-
chromium-autoroll authored
This CL may cause a small binary size increase, roughly proportional to how long it's been since our last AFDO profile roll. For larger increases (around or exceeding 100KB), please file a bug against gbiv@chromium.org. Additional context: https://crbug.com/805539 Please note that, despite rolling to chrome/android, this profile is used for both Linux and Android. If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://autoroll.skia.org/r/afdo-chromium-autoroll Please CC gbiv@chromium.org on the revert to ensure that a human is aware of the problem. To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+/master/autoroll/README.md Tbr: gbiv@chromium.org Change-Id: I9c397002854c1ced07ae994031d1dd8fa0225687 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2146518Reviewed-by:
-
chrome-release-bot authored
# This is an automated release commit. # Do not revert without consulting chrome-pmo@google.com. NOAUTOREVERT=true TBR=benmason@chromium.org Change-Id: Ibf4c1b01da77ee16dbe8deb4dfb51eed980a8cdb Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2144444Reviewed-by:
-
Chromium WPT Sync authored
Using wpt-import in Chromium 6e02a18d. With Chromium commits locally applied on WPT: d0ad7cf8 "UpgradeInsecureRequest: Add WPTs for static-import in dedicated/shared workers" c53a40f4 "Update size.attributes.parse.trailingjunk" 86131ac4 "Add a test for ReplaceTrack that verifies video track content." 380347e7 "[css-masking] Migrate clip-path-columns-svg-clippath-usou.html to WPT" Note to sheriffs: This CL imports external tests and adds expectations for those tests; if this CL is large and causes a few new failures, please fix the failures by adding new lines to TestExpectations rather than reverting. See: https://chromium.googlesource.com/chromium/src/+/master/docs/testing/web_platform_tests.md NOAUTOREVERT=true TBR=robertma No-Export: true Change-Id: Ic30608748251aeff165a89057b3c9ac5f4903ea5 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2144705Reviewed-by:
WPT Autoroller <wpt-autoroller@chops-service-accounts.iam.gserviceaccount.com> Commit-Queue: WPT Autoroller <wpt-autoroller@chops-service-accounts.iam.gserviceaccount.com> Cr-Commit-Position: refs/heads/master@{#758420}
-
Xiaoqian Dai authored
Bug: 1066467 Change-Id: I23114012b994684004a5553fea74f31795030bd9 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2144841Reviewed-by:
Sammie Quon <sammiequon@chromium.org> Commit-Queue: Xiaoqian Dai <xdai@chromium.org> Cr-Commit-Position: refs/heads/master@{#758419}
-
Henrique Ferreiro authored
Migrate this test out of third_party/blink/web_tests/css3/masking/ and into the WPT-specific directory, adding links to the relevant specs and a test assertion describing its purpose. Bug: 1063749 Change-Id: I207d3241ef7b13e13a3b25e5582109840bf217b3 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2143819 Commit-Queue: Henrique Ferreiro <hferreiro@igalia.com> Reviewed-by:
Stephen McGruer <smcgruer@chromium.org> Reviewed-by:
Philip Rogers <pdr@chromium.org> Cr-Commit-Position: refs/heads/master@{#758418}
-
Ahmed Mehfooz authored
Bug: 1035647 Change-Id: Ifeaadb55c5865a041b2461db17baf7397340c673 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2142577 Commit-Queue: Ahmed Mehfooz <amehfooz@chromium.org> Reviewed-by:
Steven Holte <holte@chromium.org> Reviewed-by:
Tim Song <tengs@chromium.org> Cr-Commit-Position: refs/heads/master@{#758417}
-
Sorin Jianu authored
This CL removes the hardcoded list of files in the updater's setup. At runtime, the setup copies all files unpacked by the metainstaller, or uses a build-generated dependency file to determine what files it needs to install, if `updater --install` is run from the `out` directory of the build. The CL introduces a new command line argument `--install-from-out-dir`, which is used in addition to other command line arguments of `--install`. This CL fixes the issue where --install could not find the `uninstall.cmd` file when run from the `out` dir. Bug:1069264,1054685 Change-Id: I03382b6d60fc71352a650e90b9abad6a85f085e9 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2146082Reviewed-by:
Joshua Pawlicki <waffles@chromium.org> Commit-Queue: Sorin Jianu <sorin@chromium.org> Cr-Commit-Position: refs/heads/master@{#758416}
-
- 11 Apr, 2020 12 commits
-
-
chromium-internal-autoroll authored
https://chrome-internal.googlesource.com/chrome/src-internal.git/+log/607540bcad8c..16297237f8cc Created with: gclient setdep -r src-internal@16297237f8cc If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://skia-autoroll.corp.goog/r/src-internal-chromium-autoroll Please CC nhiroki@google.com,reillyg@google.com on the revert to ensure that a human is aware of the problem. To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+/master/autoroll/README.md Cq-Include-Trybots: luci.chrome.try:linux-chromeos-chrome Bug: None Tbr: nhiroki@google.com,reillyg@google.com Change-Id: I2e7fa54ff227d5104892511d915ee86ad48490c4 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2146514Reviewed-by:
-
David Bokan authored
This reverts commit 77eb87e6. Reason for revert: Relanding now that we have data from UMA Original change's description: > Revert "Overlay scrollbars work correctly without overflow" > > This reverts commit 2c6eb18b. > > Reason for revert: Reverting temporarily to confirm effect on metrics > > Original change's description: > > Overlay scrollbars work correctly without overflow > > > > Overlay scrollbars - the type that fade in/out, not overflow: overlay - > > should only ever be visible when the content is scrolling. Therefore, if > > the content cannot scroll, we should never see an overlay scrollbar. > > This is different from classic (i.e. non-overlay) scrollbars. An > > overflow:scroll scroller requires showing a (disabled) scrollbar even > > when there is no overflow. Today, when an overlay scrollbar is created > > we check after-the-fact in UpdateAfterLayout whether there is overflow > > and, if there isn't, we remove overlay scrollbars. This is how the above > > mentioned behavior for |overflow:scroll| overlays is implemented. > > > > However, this doesn't account for different kinds of updates. > > UpdateAfterStyleChange must also recompute scrollbar states and didn't > > have this no-overflow removal logic. If content caused a style update > > that didn't require a layout, an overlay scrollbar would be attached in > > this case. In the attached bug, the scrollbar is visible because the > > fade out (on the main thread) is controlled by a timer that causes > > scrollbars to be disabled (which causes them not to paint). In addition > > to incorrectly creating the scrollbar, UpdateAfterStyleChange does not > > compute the scrollbar enabled state so it defaults to true. Without > > being able to scroll, the fade out timer is never started. > > > > This CL fixes the issue by moving the overflow requirement for overlay > > scrollbars into ComputeScrollbarExistence so that we account for it no > > matter the point at which we recompute scrollbar existence. We also > > refactor ComputeScrollbarExistence to make clear the distinction between > > overflow dependent and independent computations. > > > > A considered alternative was to always create the scrollbar but ensure > > its enabled state is correctly set. This would have unified the > > lifetimes of overlay and non-overlay scrollbars. However, it requires we > > prevent creating composited layers for scrollbars that are overlay but > > have no overflow since the compositor fades all of a scroller's > > scrollbars together so such scrollbars would become visible if the other > > axis is scrollable. We couldn't use the enabled state for this because > > it is flipped every time the scrollbars fade in/out so that would cause > > layers to constanty be created/destroyed so we'd have to introduce more > > complexity to compositing decisions. > > > > This change caused three tests to break (note: unit tests generally use > > mock overlay scrollbars): > > > > RepaintScrollableAreaLayersInMainThreadScrolling: > > > > A style change like this does not require invalidating paint on > > overlay scrollbars' compositing layers since the compositor can deal > > with resizes on its own. However, this was causing an invalidation > > because of the bug fixed here: the style change adds a horizontal > > scrollbar and the layout removes it. Because it isn't composited, it > > causes the whole PaintLayer to be invalidated[1]. This causes the > > scrollbar layer to also need a repaint[2]. Since we no longer cause > > this creation and destruction of the horizontal scrollbar, we don't > > get this invalidation. > > > > This test now resizes the overflow so that the vertical scrollbar > > needs to be disabled which does cause an invalidation. > > > > [1] https://cs.chromium.org/chromium/src/third_party/blink/renderer/core/paint/paint_layer_scrollable_area.cc?l=2845&rcl=33a9a3956034a765124f034a8f7d86f19c9c17e6 > > [2] https://cs.chromium.org/chromium/src/third_party/blink/renderer/core/paint/compositing/composited_layer_mapping.cc?l=2099&rcl=01e24ce806680bf99abf117e77d30d9566c72121 > > > > TraverseNonCompositingDescendantsInPaintOrder: > > > > The source of the failure here was that the |overflow:scroll| div no > > longer has overlay scrollbars after its style is updated (but before > > it does layout). When it performs PaintLayer::UpdateSelfPaintingLayer > > it will no longer be self painting[3]. Since it no longer has a self > > painting layer it will be added to the parent's visual overflow. This > > larger rect means that during overlap testing, the scroller's sibling > > (#stacked-child-of-composited-non-stacking-context) and its parent > > will overlap and so the sibling will require compositing. Finally, > > because it is now composited, it is now a paint invalidation container > > and so it won't be invalidated from the container element[5]. > > > > The fix here is to simply add overflow to the scrolling div so that it > > continues to cause a self painting layer in [3]. > > > > [3] https://cs.chromium.org/chromium/src/third_party/blink/renderer/core/paint/paint_layer.cc?l=3054&rcl=fa18fef1b31a878fed89e2a543c3be0af2f57623 > > [4] https://cs.chromium.org/chromium/src/third_party/blink/renderer/core/layout/layout_box.cc?l=5780&rcl=0650ec2415a0f2583db48d11c6d8b6a31ff448c4 > > [5] https://cs.chromium.org/chromium/src/third_party/blink/renderer/core/paint/object_paint_invalidator.cc?l=70&rcl=d9d7df43f1718c900afc470371b382c8dca085b6 > > > > mock-scrollbars.html: > > > > This was failing only on the legacy LayoutNG disabled bot so I didn't > > dig too deeply into it. The failure here was that the inner box would > > layout as if the parent had non-overlay scrollbars. I believe this is > > because the mock scrollbars were being enabled at load, after layout > > was already completed. I think this used to work because the scrollbar > > was being added and then removed which would cause a layout > > invalidation which no longer happens, thus we need to ensure mock > > scrollbars are enabled before we produce the first layout. > > > > Bug: 1039501 > > Change-Id: Ic68522a4fdba8ffaa2c5ef508736c9a94eac7682 > > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2119933 > > Commit-Queue: David Bokan <bokan@chromium.org> > > Reviewed-by: Xianzhu Wang <wangxianzhu@chromium.org> > > Cr-Commit-Position: refs/heads/master@{#754808} > > TBR=wangxianzhu@chromium.org,bokan@chromium.org > > # Not skipping CQ checks because original CL landed > 1 day ago. > > Bug: 1039501 > Change-Id: Ifa1851da28d080686d7444db56ea660638f469d6 > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2140310 > Reviewed-by: David Bokan <bokan@chromium.org> > Reviewed-by: Stephen Chenney <schenney@chromium.org> > Commit-Queue: David Bokan <bokan@chromium.org> > Cr-Commit-Position: refs/heads/master@{#757192} TBR=wangxianzhu@chromium.org,bokan@chromium.org,schenney@chromium.org # Not skipping CQ checks because original CL landed > 1 day ago. Bug: 1039501,1068351,1069574 Change-Id: I5c272b3a2f32bd2fab09bfaa8b0895ffc3f4261c Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2146451Reviewed-by:
David Bokan <bokan@chromium.org> Commit-Queue: David Bokan <bokan@chromium.org> Cr-Commit-Position: refs/heads/master@{#758414}
-
David Bokan authored
For security reasons, text fragments must only be activated when navigated with a user gesture. However, browser initiated navigations (e.g. user typing in the omnibox, bookmarks) don't have the user gesture bit set despite being initiated by the user (see discussion in https://crrev.com/c/2132673 for details). Because of this limitation, text fragment code explicitly checked if the navigation was browser initiated, assuming that such navigations are always user activated. However, history navigations are a special case. They're intentionally considered to be browser initiated, even if they originate from renderer script (e.g. `history.back()`). This meant that our check above would allow script to use the history API to activate a text fragment without a user gesture. This CL explicitly forbids activating a text fragment if the navigation is of history type. This is a trivial change (in terms of UX) because a history navigation will restore the scroll position to where the user left off so the text fragment scroll is already clobbered. This change prevents a transient scroll that will be undone. Note: we had an explicit test for this case that failed to catch the failure. The reason was that the test was checking that the fragment wasn't activated by checking that the scroll offset after a navigation is 0. However, the text fragment's scroll would be clobbered (assuming by history scroll restoration) so this check would erroneously pass. We fix it in this CL by using a scroll listener so that we can tell a scroll occurred even if it is later restored. Bug: 1042986 Change-Id: Ia0ad9a8adcda2250603e6a7dd2b386193be2a6e6 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2135407Reviewed-by:
Charlie Reis <creis@chromium.org> Reviewed-by:
Daniel Cheng <dcheng@chromium.org> Reviewed-by:
Nick Burris <nburris@chromium.org> Commit-Queue: David Bokan <bokan@chromium.org> Cr-Commit-Position: refs/heads/master@{#758413}
-
David Bokan authored
https://crrev.com/48614c8 added a DCHECK in Page so that we ensure all code can access a ScrollbarTheme. This is because unit tests on Android run without a platform theme engine. If a MockScrollbarTheme isn't installed, any execution that requests a scrollbar theme will crash. A scrollbar theme is essential to style and layout so this should always be available. However, this caused tests to break, revealing that some tests are running without a mock theme. It's unclear why those tests are flakily passing some runs and why the CL cleared the trybots in the first place; the issue is 100% reproducible locally. This CL fixes all blink unit tests to run with a mock theme. We default a DummyPageHolder to use a MockOverlayScrollbar theme and manually install it on other tests that don't use this class. A few tests create multiple DummyPageHolders, these required a small fix to ensure the old one is destructed before the new one is created so that we maintain proper nesting of ScopedMockOverlayScrollbars. One test in ImageDocumentTest required updated expectations as the test now runs with overlay scrollbars, changing the required expected offset. Bug: 1068595 Change-Id: I034c768e4b353dd0082ec5d568a0de073a28413b Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2141213 Commit-Queue: David Bokan <bokan@chromium.org> Reviewed-by:
Xianzhu Wang <wangxianzhu@chromium.org> Cr-Commit-Position: refs/heads/master@{#758412}
-
Jeroen Dhollander authored
This reverts commit e3f29527. Reason for revert: Rolldep of internal changes failed Original change's description: > Uprev LibAssistant to 1.46s > > The current LibAssistant branch (1.42c) is no longer maintained, > and also does not contain the latest features. > > * This will only be landed after M83 branches > * This depends on the internal changes to land first > > Note: > Bug: b/152777569 > Change-Id: Ic34b6ca3ba1d793c719ebb58ab9766e943fc5ec6 > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2134617 > Commit-Queue: Jeroen Dhollander <jeroendh@google.com> > Reviewed-by: Xiaohui Chen <xiaohuic@chromium.org> > Reviewed-by: Tao Wu <wutao@chromium.org> > Cr-Commit-Position: refs/heads/master@{#758408} TBR=xiaohuic@chromium.org,wutao@chromium.org,jeroendh@google.com Change-Id: I022550bcb93f4201a63aa2b28f64f6ccb32e5005 No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: b/152777569 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2146450Reviewed-by:
Jeroen Dhollander <jeroendh@google.com> Commit-Queue: Jeroen Dhollander <jeroendh@google.com> Cr-Commit-Position: refs/heads/master@{#758411}
-
Sadrul Habib Chowdhury authored
It is possible for a draw to succeed, but a frame to still not be submitted. This can happen when nothing blocks the draw, but still no frame was submitted because there was no damage. In such cases, mark the frame as deliberately dropped, rather than a frame that was aborted while waiting on the main-thread. BUG=790761 Change-Id: I410151858f7f2dd48b066cfc0ec05c29aac2e6d1 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2144422 Commit-Queue: Sadrul Chowdhury <sadrul@chromium.org> Reviewed-by:
Sunny Sachanandani <sunnyps@chromium.org> Cr-Commit-Position: refs/heads/master@{#758410}
-
chromium-internal-autoroll authored
https://chrome-internal.googlesource.com/chrome/src-internal.git/+log/f63c870256d3..607540bcad8c Created with: gclient setdep -r src-internal@607540bcad8c If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://skia-autoroll.corp.goog/r/src-internal-chromium-autoroll Please CC nhiroki@google.com,reillyg@google.com on the revert to ensure that a human is aware of the problem. To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+/master/autoroll/README.md Cq-Include-Trybots: luci.chrome.try:linux-chromeos-chrome Bug: None Tbr: nhiroki@google.com,reillyg@google.com Change-Id: I97eb54295428b91c0e1e02c9a5892ad525e8e22a Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2146271Reviewed-by:
-
Jeroen Dhollander authored
The current LibAssistant branch (1.42c) is no longer maintained, and also does not contain the latest features. * This will only be landed after M83 branches * This depends on the internal changes to land first Note: Bug: b/152777569 Change-Id: Ic34b6ca3ba1d793c719ebb58ab9766e943fc5ec6 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2134617 Commit-Queue: Jeroen Dhollander <jeroendh@google.com> Reviewed-by:
Xiaohui Chen <xiaohuic@chromium.org> Reviewed-by:
Tao Wu <wutao@chromium.org> Cr-Commit-Position: refs/heads/master@{#758408}
-
chromium-autoroll authored
https://swiftshader.googlesource.com/SwiftShader.git/+log/d37179337d95..9d2fd9c161ab git log d37179337d95..9d2fd9c161ab --date=short --first-parent --format='%ad %ae %s' 2020-04-11 bclayton@google.com LLVMReactor: Replace deprecated CreateCall overload Created with: gclient setdep -r src/third_party/swiftshader@9d2fd9c161ab If this roll has caused a breakage, revert this CL and stop the roller using the controls here: https://autoroll.skia.org/r/swiftshader-chromium-autoroll Please CC swiftshader-team+autoroll@google.com on the revert to ensure that a human is aware of the problem. To report a problem with the AutoRoller itself, please file a bug: https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+/master/autoroll/README.md Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_chromium_msan_rel_ng;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:linux-swangle-try-x64;luci.chromium.try:win-swangle-try-x86 Bug: None Tbr: swiftshader-team+autoroll@google.com Change-Id: I6f3b782dfb9c11022741566183e13160807c98da Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2146269Reviewed-by:
-
Philip Rogers authored
CSS Filter should create a containing block for fixed and absolute position except for the root[1]. Will-change: filter should do the same[2]. [1] https://drafts.fxtf.org/filter-effects/#FilterProperty [2] https://drafts.csswg.org/css-will-change/#will-change Bug: 960953 Change-Id: I9de08e812ffd818bd907f078cbe31e918c19333c Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2136667Reviewed-by:
-
Peng Huang authored
The crash is because the vulkan device lost. Fix the problem by always calling context_state_->MakeCurrent() in SkiaOutputSurfaceImplOnGpu::MakeCurrent(), it will check vulkan device lost by testing gr_context_->abandoned(), and if the device is lost, we will skip the draw to avoid crash. Bug: 1068161 Change-Id: I128df218958c0fb0d8f48e8533678f90ceab270b Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2146235 Commit-Queue: Vasiliy Telezhnikov <vasilyt@chromium.org> Auto-Submit: Peng Huang <penghuang@chromium.org> Reviewed-by:
Vasiliy Telezhnikov <vasilyt@chromium.org> Cr-Commit-Position: refs/heads/master@{#758405}
-
Rouslan Solomakhin authored
Before this patch, an unowned function pointer would be invoked asynchronously with a reference to the possibly freed reference to the browser context, which could cause use after free in certain circumstances. This patch makes the browser context own the callback and binds the function with a weak pointer, so freeing the browser context invalidates the weak pointer, which cancels the callback execution. After this patch, freeing the browser context aborts the asynchronous callback that dereferences the browser context, so the use after free is prevented. Bug: 1065298 Change-Id: Id6de3099a55c4505e94a8a6d21fb25d6d2b34c6c Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2144311Reviewed-by:
Danyao Wang <danyao@chromium.org> Commit-Queue: Rouslan Solomakhin <rouslan@chromium.org> Cr-Commit-Position: refs/heads/master@{#758404}
-
