I couldn't find the toolchain guarantees doc the other day, so as a first step
link to it here. We should probably reorganize the docs a bit -- the perf
regression bit in updating_clang.md looks like it belongs in the sheriffing
doc, and the "which files belong in the main package" should arguably move to
the toolchain guarantees doc.
Bug: none
Change-Id: I6b9018748c47f5942a970e8bcc76bcf1107784e7
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2576583
Reviewed-by: Hans Wennborg <hans@chromium.org>
Commit-Queue: Nico Weber <thakis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#834199}
This patch implements chromium.cast.DataReset in CastRunner. Upon
request to reset persistent state, we:
- Block launching new components in the main context
- Move all data under /cache to a staging directory that is deleted
upon next startup of CastRunner
Bug: 1146474, 1146480
Test: run_cast_runner_integration_tests
Test: verified manually on device that a file in the cache directory of CastRunner is gone after DataReset and reboot
Change-Id: I713133540302a1c0535c778e5957e2c87e0f669b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2537840
Commit-Queue: Przemek Pietrzkiewicz <ppi@google.com>
Reviewed-by: Wez <wez@chromium.org>
Reviewed-by: Benjamin Lerman <qsr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#834193}
The std::map in StatsResponse::Report is only used to iterate though
the elements. Profiling on Mac shows that a replacing the std::map
with a std::vector shaves off about 40% CPU time spent in
StatsResponse::Report in some common video conferencing scenarios.
Bug: 1156090
Change-Id: I3070000bb534deb18fb0fd9b18b83a947fae64a9
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2576240
Reviewed-by: Henrik Boström <hbos@chromium.org>
Commit-Queue: Ali Tofigh <alito@chromium.org>
Cr-Commit-Position: refs/heads/master@{#834189}
GetDeviceAttributeUpdatePermission resolves the current user by OAuth
token and then checks if that user may change the device's attributes.
This is only sensible for manual enrollment where a user has actually
authenticated and we have an OAuth token for the user.
For enrollment modes without explicit user auth, such as
attestation-based enrollment or enrollment-token-based enrollment, the
enrolling user is not known so this check is not possible.
For now, assume that device attributes should not be changed in this,
as there is no way to find out if the enrolling user is permitted to
change them.
Bug: 942013
Change-Id: I5f7e08ca2af223d0dbcac3839c6eb2b2f45e62b2
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2566820
Commit-Queue: Pavol Marko <pmarko@chromium.org>
Reviewed-by: Denis Kuznetsov [CET] <antrim@chromium.org>
Cr-Commit-Position: refs/heads/master@{#834188}
The new ThumbStripMediator handles observing the regular/incognito
WebStateLists to add the ViewRevealingVerticalPanHandler as a
scrollViewProxy observer. This allows it to get scroll delegate
callbacks from the active webstate.
This CL also slightly fixes the BVC thumb strip hide/reveal behavior to
correctly change the scroll view's content insets as well as the
content offset.
Bug: 1094335
Change-Id: I006a3e72a06b1e327ce0984cf84bc69c2e95fe7a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2563556
Reviewed-by: Gauthier Ambard <gambard@chromium.org>
Reviewed-by: Chris Lu <thegreenfrog@chromium.org>
Commit-Queue: Robbie Gibson <rkgibson@google.com>
Cr-Commit-Position: refs/heads/master@{#834183}
Command:
ui/file_manager/base/tools/modules.sh [1] [2]
[1] build directory (e.g. out/Release)
[2] path of JS file to convert (relative to chromium/src)
What does the script do:
1. Update BUILD.gn to enable conversion of JS file.
2. Add exports.
3. Run closure compiler.
4. Parse closure compiler output to retrieve unknown types/variables/functions.
6. Parse these dependencies to find relevant exported types/variables/functions.
7. When found, update BUILD.gn and JS file.
Made sure that:
The script doesn't add any irrelevant import/export.
No duplicate lines are added.
Steps 3 to 7 are ran until the closure compiler output
'X error(s), Y warning(s), Z% typed' doesn't change. The closure compiler
is ran at least 3 times (except if no imports are found), sometimes more.
Bug: 1133186
Change-Id: I1528c8187bc1476d6add1bcb14a0d4fd1b64a4fe
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2550328
Commit-Queue: Jeremie Boulic <jboulic@chromium.org>
Reviewed-by: Luciano Pacheco <lucmult@chromium.org>
Cr-Commit-Position: refs/heads/master@{#834181}
Adjust the domain placeholder names, descriptions and screenshots in
IDS_SETTINGS_UPDATE_REQUIRED_EOL_BANNER_ONE_WEEK and
IDS_SETTINGS_UPDATE_REQUIRED_EOL_BANNER_DAYS to prepare them for
FlexOrgs and make it easier for the translators to understand the usage
of the texts.
Additionally replace calls to
BrowserPolicyConnectorChromeOS::GetEnterpriseDisplayDomain with calls to
BrowserPolicyConnectorChromeOS::GetEnterpriseDomainManager.
Bug: b:172683940, b:172683948
Test: Manual
Change-Id: I1102446a9d065638a8dd008a751340f6abb6060c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2571878
Reviewed-by: Brian Malcolm <bmalcolm@chromium.org>
Reviewed-by: James Cook <jamescook@chromium.org>
Commit-Queue: Jana Grill <janagrill@chromium.org>
Cr-Commit-Position: refs/heads/master@{#834177}
ToolbarViewWithExtensionsToolbarMenuTest did not correctly disable
Ephemeral Guest profiles when they were enabled by default.
This issue is fixed.
Bug: 1125474
Change-Id: I4f277c3477d76cf3fee51bcd80c283b2929e2ea5
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2573104
Reviewed-by: Bret Sepulveda <bsep@chromium.org>
Commit-Queue: Ramin Halavati <rhalavati@chromium.org>
Cr-Commit-Position: refs/heads/master@{#834173}
De/RecommitSystemPages() have similar properties under POSIX and
Fuchsia, since both use over-commit, but the API implementations differ
in their optimizations. Update the Fuchsia implementations with the same
optimizations as provided for POSIX.
Bug: 766882
Change-Id: I34b98db27ed85543e53b3231ec118d0a2409b454
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2558122
Commit-Queue: Bartek Nowierski <bartekn@chromium.org>
Reviewed-by: Chris Palmer <palmer@chromium.org>
Reviewed-by: Wez <wez@chromium.org>
Auto-Submit: Bartek Nowierski <bartekn@chromium.org>
Cr-Commit-Position: refs/heads/master@{#834172}
OOF descendants of a fieldset should not be contained by the fieldset
because they should be scrolled together with the fieldset contents.
This CL makes an anonymous fieldset content box an OOF container by
adding |filter: brightness(100%)| or |position: relative| if the
fieldset can contain OOF objects.
We need to add a special handling for OOF descendants of a LEGEND to
FindAncestorByPredicate() because such OOF descendants are contained
by fieldsets in NG fragment trees. Without this change, we had some
CHECK failures.
Bug: 1153042
Change-Id: I993922cfd373a9e0558bc50edb009680f0f1244a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2569140
Reviewed-by: Morten Stenshorne <mstensho@chromium.org>
Commit-Queue: Kent Tamura <tkent@chromium.org>
Cr-Commit-Position: refs/heads/master@{#834169}
No change to logic. This inlines BuildInfo.isAtLeast{Q,R} and
BuildInfo.targetsAtLeast{Q,R} (and their C++ equivalents) in
/base/android/java/src/org/chromium/base/process_launcher.
The methods themselves will be removed in a follow-up CL once all call
sites have been removed.
This CL was uploaded by git cl split.
R=rmcilroy@chromium.org
Bug: 1106920
Test: autoninja -C out/Default monochrome_public_apk
Change-Id: I74be1508173db531826fb0b8cfdc953642740989
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2575373
Auto-Submit: Nate Fischer <ntfschr@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Commit-Queue: Ross McIlroy <rmcilroy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#834167}
This change replaces network::IsUrlPotentiallyTrustworthy with
network::IsOriginPotentiallyTrustworthy in the Touch To Fill controller.
The small difference between those two APIs is irrelevant for Touch To
Fill, and the latter matches the semantics of
TouchToFillView::IsOriginSecure better.
Bug: 1012717, 1153336
Change-Id: I06553ac01aeae540426e6df30631e03db0ccfe60
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2574743
Commit-Queue: Jan Wilken Dörrie <jdoerrie@chromium.org>
Commit-Queue: Friedrich [CET] <fhorschig@chromium.org>
Auto-Submit: Jan Wilken Dörrie <jdoerrie@chromium.org>
Reviewed-by: Friedrich [CET] <fhorschig@chromium.org>
Cr-Commit-Position: refs/heads/master@{#834163}
No change to logic. This inlines BuildInfo.isAtLeast{Q,R} and
BuildInfo.targetsAtLeast{Q,R} (and their C++ equivalents) in
/components/browser_ui/photo_picker/android.
The methods themselves will be removed in a follow-up CL once all call
sites have been removed.
This CL was uploaded by git cl split.
R=finnur@chromium.org
Bug: 1106920
Test: autoninja -C out/Default monochrome_public_apk
Change-Id: Ib3278a63d0a27967cb873aa265b504bc5dd7107d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2575382
Auto-Submit: Nate Fischer <ntfschr@chromium.org>
Reviewed-by: Finnur Thorarinsson <finnur@chromium.org>
Commit-Queue: Finnur Thorarinsson <finnur@chromium.org>
Cr-Commit-Position: refs/heads/master@{#834162}
The new page set needs extra Chrome flags; pipe those through the
superclass constructors.
The sample that's the source for the new page set has multiple JS files;
change update_webrtc_cases to concatenate all JS files to create the
target JS file. Also, fix the URL for the video file (reusing the video
file for trivial_full_screen).
`./tools/perf/run_benchmark run webrtc --browser=canary` passes locally
on Mac and runs the new stories. `./tools/perf/run_benchmark run webrtc
--browser=system` also passes, correctly skipping the stories since
they use features that are new in M89.
Change-Id: I732b3c967469c7b83b940faf2f9486d909430ee3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2566935
Commit-Queue: Ben Wagner aka dogben <benjaminwagner@google.com>
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
Reviewed-by: John Chen <johnchen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#834161}
This CL makes layout of disclosure-* equivalent to the default layout
of ::-webkit-details-marker.
* DisclosureSymbolSize()
Make the symbol size same as 0.66em.
* ListMarker::WidthOfSymbol()
1px gaps are not necessary.
* ListMarker::InlineMarginsForInside()
No margin for inline-start.
0.4em margin for inline-end.
* ListMarker::RelativeSymbolMarkerRect()
No margin for inline-start.
Align the marker bottom and the baseline
* ListMarkerPainter::PaintSymbol() should take a non-pixel-snapped
rectangle. It matches to the current behavior of LayoutDetailsMarker
and DetailsMarkerPainter.
This CL will reduce the number of rebaseline files significantly when
SummaryListItem flag is enabled.
Bug: 590014
Change-Id: I600ab534a05c08bca482a4316270b9e9b9a3c651
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2577077
Commit-Queue: Kent Tamura <tkent@chromium.org>
Commit-Queue: Koji Ishii <kojii@chromium.org>
Auto-Submit: Kent Tamura <tkent@chromium.org>
Reviewed-by: Koji Ishii <kojii@chromium.org>
Cr-Commit-Position: refs/heads/master@{#834154}
This CL converts uses of base::Callback to their proper {Once|Repeating}
equivalents in the chrome/browser/sync_file_system.
Specifically, LocalFileSyncService, LocalFileSyncContext and
RootDeleteHelper are updated.
Bug: 1152272
Change-Id: I7955714c5b000d77dab4a21e463161f1038f0e44
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2575047
Auto-Submit: Thomas Guilbert <tguilbert@chromium.org>
Commit-Queue: Kinuko Yasuda <kinuko@chromium.org>
Reviewed-by: Kinuko Yasuda <kinuko@chromium.org>
Cr-Commit-Position: refs/heads/master@{#834153}
